Multiple enzyme activities of flavivirus proteins.

Dengue viruses (DENV) have 5'-capped RNA genomes of (+) polarity and encode a single polyprotein precursor that is processed into mature viral proteins. NS2B, NS3 and NS5 proteins catalyse/activate enzyme activities that are required for key processes in the virus life cycle. The heterodimeric NS2B/NS3 is a serine protease required for processing. Using a high-throughput protease assay, we screened a small molecule chemical library and identified -200 compounds having > or = 50% inhibition. Moreover, NS3 exhibits RNA-stimulated NTPase, RNA helicase and the 5'-RNA triphosphatase activities. The NTPase and the 5'-RTPase activities of NS3 are stimulated by interaction with NS5. Moreover, the conserved, positively charged motif in DENV-2 NS3, 184RKRK, is required for RNA binding and modulates the RNA-dependent enzyme activities of NS3. To study viral replication, a variety of methods are used such as the in vitro RNA-dependent RNA polymerase assays that utilize lysates from DENV-2-infected mosquito or mammalian cells or the purified NS5 along with exogenous short subgenomic viral RNAs or the replicative intracellular membrane-bound viral RNAs as templates. In addition, a cell-based DENV-2 replicon RNA encoding a luciferase reporter is also used to examine the role of cis-acting elements within the 3' UTR and the RKRK motif in viral replication.

In vitro RNA synthesis from exogenous dengue viral RNA templates requires long range interactions between 5'- and 3'-terminal regions that influence RNA structure.

Viral replicases of many positive-strand RNA viruses are membrane-bound complexes of cellular and viral proteins that include viral RNA-dependent RNA polymerase (RdRP). The in vitro RdRP assay system that utilizes cytoplasmic extracts from dengue viral-infected cells and exogenous RNA templates was developed to understand the mechanism of viral replication in vivo. Our results indicated that in vitro RNA synthesis at the 3'-untranslated region (UTR) required the presence of the 5'-terminal region (TR) and the two cyclization (CYC) motifs suggesting a functional interaction between the TRs. In this study, using a psoralen-UV cross-linking method and an in vitro RdRP assay, we analyzed structural determinants for physical and functional interactions. Exogenous RNA templates that were used in the assays contained deletion mutations in the 5'-TR and substitution mutations in the 3'-stem-loop structure including those that would disrupt the predicted pseudoknot structure. Our results indicate that there is physical interaction between the 5'-TR and 3'-UTR that requires only the CYC motifs. RNA synthesis at the 3'-UTR, however, requires long range interactions involving the 5'-UTR, CYC motifs, and the 3'-stem-loop region that includes the tertiary pseudoknot structure.

Construction of a full length infectious clone for dengue-1 virus Western Pacific,74 strain.

The flavivirus dengue 1 Western Pacific,74 (DEN1 WP) virus has a positive-stranded RNA genome of 10,735 nucleotides. DEN1 WP genomic RNA was amplified into three overlapping fragments by RT-PCR. These fragments were assembled into a full-length cDNA clone in the yeast-E. coli shuttle vector pRS424, using homologous recombination in yeast. RNA produced by in vitro transcription of this clone was infectious upon electroporation into LLCMK2 cells, as shown by cytopathic effects and detection of viral antigens by indirect immunofluorescence, and by propagation of the virus released into the culture media. Biological properties of the transcript-derived virus, such as the pattern of dengue-specific protein synthesis and growth rate in LLCMK2 or C6/36 cells, resembled those of the parent DEN1 WP virus.

Identification of specific nucleotide sequences within the conserved 3'-SL in the dengue type 2 virus genome required for replication.

The flavivirus genome is a positive-stranded approximately 11-kb RNA including 5' and 3' noncoding regions (NCR) of approximately 100 and 400 to 600 nucleotides (nt), respectively. The 3' NCR contains adjacent, thermodynamically stable, conserved short and long stem-and-loop structures (the 3'-SL), formed by the 3'-terminal approximately 100 nt. The nucleotide sequences within the 3'-SL are not well conserved among species. We examined the requirement for the 3'-SL in the context of dengue virus type 2 (DEN2) replication by mutagenesis of an infectious cDNA copy of a DEN2 genome. Genomic full-length RNA was transcribed in vitro and used to transfect monkey kidney cells. A substitution mutation, in which the 3'-terminal 93 nt constituting the wild-type (wt) DEN2 3'-SL sequence were replaced by the 96-nt sequence of the West Nile virus (WN) 3'-SL, was sublethal for virus replication. An analysis of the growth phenotypes of additional mutant viruses derived from RNAs containing DEN2-WN chimeric 3'-SL structures suggested that the wt DEN2 nucleotide sequence forming the bottom half of the long stem and loop in the 3'-SL was required for viability. One 7-bp substitution mutation in this domain resulted in a mutant virus that grew well in monkey kidney cells but was severely restricted in cultured mosquito cells. In contrast, transpositions of and/or substitutions in the wt DEN2 nucleotide sequence in the top half of the long stem and in the short stem and loop were relatively well tolerated, provided the stem-loop secondary structure was conserved.

Mutagenesis of the NS3 protease of dengue virus type 2.

The flavivirus protease is composed of two viral proteins, NS2B and NS3. The amino-terminal portion of NS3 contains sequence and structural motifs characteristic of bacterial and cellular trypsin-like proteases. We have undertaken a mutational analysis of the region of NS3 which contains the catalytic serine, five putative substrate binding residues, and several residues that are highly conserved among flavivirus proteases and among all serine proteases. In all, 46 single-amino-acid substitutions were created in a cloned NS2B-NS3 cDNA fragment of dengue virus type 2, and the effect of each mutation on the extent of self-cleavage of the NS2B-NS3 precursor at the NS2B-NS3 junction was assayed in vivo. Twelve mutations almost completely or completely inhibited protease activity, 9 significantly reduced it, 14 decreased cleavage, and 11 yielded wild-type levels of activity. Substitution of alanine at ultraconserved residues abolished NS3 protease activity. Cleavage was also inhibited by substituting some residues that are conserved among flavivirus NS3 proteins. Two (Y150 and G153) of the five putative substrate binding residues could not be replaced by alanine, and only Y150 and N152 could be replaced by a conservative change. The two other putative substrate binding residues, D129 and F130, were more freely substitutable. By analogy with the trypsin model, it was proposed that D129 is located at the bottom of the substrate binding pocket so as to directly interact with the basic amino acid at the substrate cleavage site. Interestingly, we found that significant cleavage activity was displayed by mutants in which D129 was replaced by E, S, or A and that low but detectable protease activity was exhibited by mutants in which D129 was replaced by K, R, or L. Contrary to the proposed model, these results indicate that D129 is not a major determinant of substrate binding and that its interaction with the substrate, if it occurs at all, is not essential. This mutagenesis study provided us with an array of mutations that alter the cleavage efficiency of the dengue virus protease. Mutations that decrease protease activity without abolishing it are candidates for introduction into the dengue virus infectious full-length cDNA clone with the aim of creating potentially attenuated virus stocks.

Infectious RNA transcripts from full-length dengue virus type 2 cDNA clones made in yeast.

The dengue virus type 2 genomic RNA was amplified by reverse transcription-PCR and cloned as four cDNA fragments. We could not assemble these four fragments into full-length cDNA in Escherichia coli. The full-length dengue virus cDNA was constructed by homologous recombination in yeast, either as part of a yeast artificial chromosome or in a yeast-E. coli shuttle vector. Full-length cDNA clones were propagated once in E. coli to prepare useful quantities of DNA. In vitro transcription of these clones produced full-length RNA transcripts. Introduction of these transcripts into LLC-MK2 cells produced typical dengue infection, as judged by cytopathic effects and indirect immunofluorescence. Infectivity was sensitive to RNase digestion and was dependent on the presence of cap analog in the transcription reaction mixture. Virus in the medium was passaged on C6-36 cells to produce stocks, and these stocks had titers and plaque morphologies similar to those of the parental dengue virus type 2. Intracellular dengue virus RNA from cells infected with transcript-derived virus contained an introduced BstEII site, proving that infectivity was derived from RNA transcripts and not from contamination with parental dengue virus. Transcript-derived virus was comparable to dengue virus type 2 for growth and protein expression in tissue culture cells. Sequence analysis of the dengue virus cDNA in one full-length clone revealed only one unexpected silent mutation. By using yeast technology, it will be easy to introduce specific mutations into the dengue virus cDNA, allowing analysis of the virus phenotype in cells transfected with mutant transcripts.

A conserved internal hydrophobic domain mediates the stable membrane integration of the dengue virus capsid protein.

The mature flavivirus capsid protein (virion C) is commonly thought to be free in the cytoplasm of infected cells and to form a nucleocapsid-like complex with genomic RNA in mature virus particles. There is little sequence conservation among flavivirus virion C proteins, but they are similar in size (e.g., 99 amino acids [aa] for the dengue-4 [DEN4] C) and in bearing a net positive charge. In addition, we noted that C contained a conserved internal hydrophobic segment (spanning aa 45-65 in the DEN4 C). Results of in vivo expression and in vitro translation of wt and mutant forms of the DEN4 virion C demonstrated that the conserved internal hydrophobic segment in the DEN C functioned as a membrane anchor domain. Signal peptide function of this segment was also suggested by its requirement for the entry of C into membranes. Virion C was integrated in membranes in a "hairpin" conformation; positively charged segments amino- and carboxy-terminal to the hydrophobic signal-anchor segment were accessible to protease digestion in the "cytoplasm." The net positive charge in the amino-terminal extramembraneous portion of C (aa 1-44) was one determinant of the hairpin membrane orientation; a conserved positively charged residue within the hydrophobic segment (Arg-54 in the DEN4 C) was not.

Evidence that flavivirus NS1-NS2A cleavage is mediated by a membrane-bound host protease in the endoplasmic reticulum.

Previous deletion mutagenesis studies have shown that the flavivirus NS1-NS2A clevage requires the eight C-terminal residues of NS1, constituting the cleavage recognition sequence, and sequences in NS2A far downstream of the cleavage site. We now demonstrate that replacement of all of NS1 upstream of the cleavage recognition sequence with prM sequences still allows cleavage in vivo. Thus, other than the eight C-terminal residues, NS1 is dispensable for NS1-NS2A cleavage. However, deletion of the N-terminal signal sequence abrogated cleavage, suggesting that entry into the exocytic pathway is required. Cleavage in vivo was not blocked by brefeldin A, and cleavage could occur in vitro in the presence of dog pancreas microsomes, indicating that NS1-NS2A cleavage occurs in the endoplasmic reticulum. Four in-frame deletions in NS2A were cleavage defective in vitro, as were two mutants in which NS4A-NS4B sequences were substituted for NS2A, suggesting that most of NS2A is required. A series of substitution mutants were constructed in which all Asp, Cys, Glu, His, and Ser residues in NS2A were collectively replaced; all standard proteases require at least one of these residues in their active sites. No single mutant was cleavage defective, suggesting that NS2A is not a protease. Fractionation of the microsomes indicated that the lumenal contents were not required for NS1-NS2A cleavage. It seems most likely that NS1-NS2A cleavage is effected by a host membrane-bound endoplasmic reticulum-resident protease, quite possibly signalase, and that NS2A is required to present the cleavage recognition sequence in the correct conformation to the host enzyme for cleavage.

Processing of Japanese encephalitis virus non-structural proteins: NS2B-NS3 complex and heterologous proteases.

Processing of Japanese encephalitis (JE) virus non-structural (NS) proteins expressed by recombinant vaccinia viruses was analysed to characterize the responsible viral protease. Analysis of the processing of polyprotein NS2A-2B-3' containing the N-terminal 322 amino acids of NS3 revealed products consistent with cleavages at the predicted intergenic junctions as well as at one or possibly two sites within NS2A. Cleavage at the alternate site(s) containing the cleavage sequence motif within NS2A could possibly explain the production of the NS1' protein in JE virus-infected cells. Polyprotein NS2A-d2B-3' containing a large deletion within NS2B was cleavage-defective, despite the presence of the proposed NS3 protease domain. Cleavage of NS2A-d2B-3' was restored if NS2B or NS2A-2B was supplied in trans, providing evidence that NS2B is strictly required for NS3 proteolytic activity. NS2B- or NS3-specific sera raised against the bacterial TrpE fusion protein co-precipitated NS2B and NS3 or NS3' from the lysate of JE virus or recombinant virus-infected cells. Thus both protease components are associated as a complex, presumably representing the active JE virus protease. JE virus and the analogous dengue 4 (DEN-4) protease components were employed to examine the activity of heterologous proteases. The defective cleavage of JE virus NS2A-d2B-3' was complemented by heterologous DEN-4 NS2B, whereas the defective cleavage of DEN-4 NS2A-d2B-3' was not corrected by heterologous JE virus NS2B. This suggests that the heterologous JE virus NS2B-DEN-4 NS3 protease is not active, despite the considerable sequence conservation of NS2B and NS3 between the two viruses. The cleavage activity was restored by replacement of the C-terminal 80 amino acids of JE virus NS2B with the corresponding DEN-4 sequence, consistent with the notion that the C-terminal region contains amino acid residues for interaction with DEN-4 NS3.

Processing of flavivirus structural glycoproteins: stable membrane insertion of premembrane requires the envelope signal peptide.

The flavivirus structural proteins capsid (C), premembrane (prM), and envelope (E) are cleaved in that order from the N-terminus of the polyprotein by the ER intralumenal enzyme signal peptidase. The prM-E and E-NS1 junctions contain hydrophobic domains with both transmembrane and signal function. These domains reside at the C-termini of prM and E, respectively, after cleavage. We studied the functions of the 37-amino-acid C-terminus of the dengue virus type 4 (DEN4) prM (amino acids 243-279 of the DEN4 polyprotein) in the processing of prM and E. Hydrophobicity in this domain is interrupted by a conserved Arg residue (Arg-264) within a short amphipathic segment. Hydrophobic amino acids upstream from Arg-264 (aa 243-263) were presumed to constitute the membrane anchor for prM (the "tm" segment). Previous results had suggested that sequences downstream from Arg-264 (aa 265-279) constitute the E signal peptide. RNA transcripts prepared from wild-type (wt) and deletion-mutant DEN4 cDNAs encoding the prM signal peptide, prM, E, and the N-terminus of the nonstructural glycoprotein, NS1, were translated in rabbit reticulocyte lysate in the presence of microsomes. Processing of wt prM and E in vitro appeared to mimic processing occurring during flavivirus infection. Analysis of mutants confirmed the localization of the E signal peptide within residues 265 to 279. However, deletions within either the E signal peptide or the tm segment resulted in a defect in both membrane insertion of prM and cleavage of the prM-E junction. Membrane anchoring of prM appeared to be a two-step process requiring function of both the tm segment and the E signal peptide, and fully efficient prM-E cleavage was also dependent upon the integrity of both hydrophobic domains. We propose a model for the processing of the flavivirus structural glycoproteins based on these results.

Processing of dengue type 4 and other flavivirus nonstructural proteins.

Dengue type 4 (DEN4) and other flaviviruses employ host and viral proteases for polyprotein processing. Most proteolytic cleavages in the DEN4 nonstructural protein (NS) region are mediated by the viral NS2B-NS3 protease. The N-terminal third of NS3, containing sequences homologous to serine protease active sites, is the protease domain. To determine required sequences in NS2B, deletions were introduced into DEN4 NS2B-30% NS3 cDNA and the expressed polyproteins assayed for self-cleavage. A 40 amino acid segment within NS2B was essential. Sequence analysis of NS2B predicts that this segment constitutes a hydrophilic domain surrounded by hydrophobic regions. Hydrophobicity profiles of other flavivirus NS2Bs show similar patterns. Cleavage of DEN4 NS1-NS2A requires an octapeptide sequence at the NS1 C terminus and downstream NS2A. Comparison of the analogous octapeptide sequences among flaviviruses indicates a consensus cleavage sequence of (P8)/Met/Leu-Val-Xaa-Ser-Xaa-Val-Ala(P1), where Xaa are non-conserved amino acids. The effects on cleavage of amino acid substitutions in this consensus sequence were analyzed. Most substitutions of the conserved residues interfered with cleavage, whereas substitutions of non-conserved residues had little or no effect. These findings indicate that the responsible enzyme recognizes well-defined sequences at the cleavage site.

Deletion analysis of dengue virus type 4 nonstructural protein NS2B: identification of a domain required for NS2B-NS3 protease activity.

Most proteolytic cleavages in the nonstructural protein (NS) region of the flavivirus polyprotein are effected by a virus-encoded protease composed of two viral proteins, NS2B and NS3. The N-terminal 180-amino-acid-region of NS3 includes sequences with homology to the active sites of serine proteases, and there is evidence that this portion of NS3 can mediate proteolytic cleavages. In contrast, nothing is known about required sequences in NS2B. We constructed a series of deletion mutations in the NS2B portion of plasmid pTM/NS2B-30% NS3, which expresses dengue virus type 4 (DEN4) cDNA encoding NS2B and the N-terminal 184 residues of NS3 from the T7 RNA polymerase promoter. Mutant or wild-type plasmids were transfected into cells that had been infected with a recombinant vaccinia virus expressing T7 RNA polymerase, and the protease activities of the expressed polyproteins were assayed by examining the extent of self-cleavage at the NS2B-NS3 junction. The results identify a 40-amino-acid segment of NS2B (DEN4 amino acids 1396 to 1435) essential for protease activity. A hydrophobicity profile of DEN4 NS2B predicts this segment constitutes a hydrophilic domain surrounded by hydrophobic regions. Hydrophobicity profiles of the NS2B proteins of other flaviviruses show similar patterns. Amino acid sequence alignment of this domain of DEN4 NS2B with comparable regions of other proteins of flaviviruses indicates significant sequence conservation, especially at the N-terminal end. These observations suggest that the central hydrophilic domain of NS2B of these other flaviviruses will also prove to be essential for protease activity.

Dengue virus protein recognition by virus-specific murine CD8+ cytotoxic T lymphocytes.

The identification of the protein targets for dengue virus-specific T lymphocytes may be useful for planning the development of subunit vaccines against dengue. We studied the recognition by murine dengue virus-specific major histocompatibility complex class I-restricted, CD8+ cytotoxic T lymphocytes (CTL) of dengue virus proteins using recombinant vaccinia viruses containing segments of the dengue virus genome. CTL from H-2k mice recognized a single serotype-cross-reactive epitope on the nonstructural (NS) protein NS3. CTL from H-2b mice recognized a serotype-cross-reactive epitope that was localized to NS4a or NS4b. CTL from H-2d mice recognized at least three epitopes: a serotype-specific epitope on one of the structural proteins, a serotype-cross-reactive epitope on NS3, and a serotype-cross-reactive epitope on NS1 or NS2a. Our findings demonstrate the limited recognition of dengue virus proteins by CTL from three inbred mouse strains and the predominance of CTL epitopes on dengue virus nonstructural proteins, particularly NS3. Since human dengue virus-specific CTL show similar patterns of recognition, these findings suggest that nonstructural proteins should be considered in designing vaccines against dengue.

Mutational analysis of the octapeptide sequence motif at the NS1-NS2A cleavage junction of dengue type 4 virus.

We have previously shown that proper processing of dengue type 4 virus NS1 from the NS1-NS2A region of the viral polyprotein requires a hydrophobic N-terminal signal and the downstream NS2A. Results from deletion analysis indicate that a minimum length of eight amino acids at the C terminus of NS1 is required for cleavage at the NS1-NS2A junction. Comparison of this eight-amino-acid sequence with the corresponding sequences of other flaviviruses suggests a consensus cleavage sequence of Met/Leu-Val-Xaa-Ser-Xaa-Val-Xaa-Ala. Site-directed mutagenesis was performed to construct mutants of NS1-NS2A that contained a single amino acid substitution at different positions of the consensus cleavage sequence or at the immediate downstream position. Three to eight different substitutions were made at each position. A total of 50 NS1-NS2A mutants were analyzed for their cleavage efficiency relative to that of the wild-type dengue type 4 virus sequence. As predicted, nearly all substitutions at positions P1, P3, P5, P7, and P8, occupied by conserved amino acids, yielded low levels of cleavage, with the exception that Pro or Ala substituting for Ser (P5) was tolerated. Substitutions of an amino acid at the remaining positions occupied by nonconserved amino acids generally yielded high levels of cleavage. However, some substitutions at nonconserved positions were not tolerated. For example, substitution of Gly or Glu for Gln (P4) and substitution of Val or Glu for Lys (P6) each yielded a low level of cleavage. Overall, these data support the proposed cleavage sequence motif deduced by comparison of sequences among the flaviviruses. This study also showed that in addition to the eight-amino-acid sequence, the amino acid immediately following the NS1-NS2A cleavage site plays a role in cleavage.

Cleavage of the dengue virus polyprotein at the NS3/NS4A and NS4B/NS5 junctions is mediated by viral protease NS2B-NS3, whereas NS4A/NS4B may be processed by a cellular protease.

The cleavage mechanism utilized for processing of the NS3-NS4A-NS4B-NS5 domain of the dengue virus polyprotein was studied by using the vaccinia virus expression system. Recombinant vaccinia viruses vNS2B-NS3-NS4A-NS4B-NS5, vNS3-NS4A-NS4B-NS5, vNS4A-NS4B-NS5, and vNS4B-NS5 were constructed. These recombinants were used to infect cells, and the labeled lysates were analyzed by immunoprecipitation. Recombinant vNS2B-NS3-NS4A-NS4B-NS5 expressed the authentic NS3 and NS5 proteins, but the other recombinants produced uncleaved polyproteins. These findings indicate that NS2B is required for processing of the downstream nonstructural proteins, including the NS3/NS4A and NS4B/NS5 junctions, both of which contain a dibasic amino acid sequence preceding the cleavage site. The flavivirus NS4A/NS4B cleavage site follows a long hydrophobic sequence. The polyprotein NS4A-NS4B-NS5 was cleaved at the NS4A/NS4B junction in the absence of other dengue virus functions. One interpretation for this finding is that NS4A/NS4B cleavage is mediated by a host protease, presumably a signal peptidase. Although vNS3-NS4A-NS4B-NS5 expressed only the polyprotein, earlier results demonstrated that cleavage at the NS4A/NS4B junction occurred when an analogous recombinant, vNS3-NS4A-84%NS4B, was expressed. Thus, it appears that uncleaved NS3 plus NS5 inhibit NS4A/NS4B cleavage presumably because the putative signal sequence is not accessible for recognition by the responsible protease. Finally, recombinants that expressed an uncleaved NS4B-NS5 polyprotein, such as vNS4A-NS4B-NS5 or vNS4B-NS5, produced NS5 when complemented with vNS2B-30%NS3 or with vNS2B plus v30%NS3. These results indicate that cleavage at the NS4B/NS5 junction can be mediated by NS2B and NS3 in trans.

Both nonstructural proteins NS2B and NS3 are required for the proteolytic processing of dengue virus nonstructural proteins.

The cleavages at the junctions of the flavivirus nonstructural (NS) proteins NS2A/NS2B, NS2B/NS3, NS3/NS4A, and NS4B/NS5 share an amino acid sequence motif and are presumably catalyzed by a virus-encoded protease. We constructed recombinant vaccinia viruses expressing various portions of the NS region of the dengue virus type 4 polyprotein. By analyzing immune precipitates of 35S-labeled lysates of recombinant virus-infected cells, we could monitor the NS2A/NS2B, NS2B/NS3, and NS3/NS4A cleavages. A polyprotein composed of NS2A, NS2B, and the N-terminal 184 amino acids of NS3 was cleaved at the NS2A/NS2B and NS2B/NS3 junctions, whereas a similar polyprotein containing only the first 77 amino acids of NS3 was not cleaved. This finding is consistent with the proposal that the N-terminal 180 amino acids of NS3 constitute a protease domain. Polyproteins containing NS2A and NS3 with large in-frame deletions of NS2B were not cleaved at the NS2A/NS2B or NS2B/NS3 junctions. Coinfection with a recombinant expressing NS2B complemented these NS2B deletions for NS2B/NS3 cleavage and probably also for NS2A/NS2B cleavage. Thus, NS2B is also required for the NS2A/NS2B and NS2B/NS3 cleavages and can act in trans. Other experiments showed that NS2B was needed, apparently in cis, for NS3/NS4A cleavage and for a series of internal cleavages in NS3. Indirect evidence that NS3 can also act in trans was obtained. Models are discussed for a two-component protease activity requiring both NS2B and NS3.

Immunization of mice with recombinant vaccinia virus expressing authentic dengue virus nonstructural protein NS1 protects against lethal dengue virus encephalitis.

The protective immunity conferred by a set of recombinant vaccinia viruses containing the entire coding sequence of dengue virus type 4 nonstructural glycoprotein NS1 plus various flanking sequences was evaluated by using a mouse encephalitis model. Mice immunized with recombinant vNS1-NS2a, which expresses authentic NS1, were solidly protected against intracerebral dengue virus challenge. However, mice immunized with recombinants vNS1-15%NS2a and vRSVG/NS1-15%NS2a, which express aberrant forms of NS1, were only partially protected (63 to 67% survival rate). Serologic analysis showed that mice immunized with vNS1-NS2a developed high titers of antibodies to NS1 as measured by radioimmunoprecipitation, enzyme-linked immunosorbent assay, and complement-mediated cytolytic assays. In addition, a pool of sera from these animals was protective in a passive transfer experiment. Lower titers of NS1-specific antibodies were detected in sera of animals immunized with vNS1-15%NS2a or vRSVG/NS1-15%NS2a by all three assays. These data support the view that protection against dengue virus infection in mice may be mediated at least in part by NS1-specific antibodies through a mechanism of complement-mediated lysis of infected cells. Additionally, immunization with two recombinant viruses expressing authentic NS1 of dengue virus type 2 conferred partial protection (30-50%) against dengue virus type 2 challenge.

Dengue virus-specific cross-reactive CD8+ human cytotoxic T lymphocytes.

Stimulation with live dengue virus of peripheral blood mononuclear cells from a dengue virus type 4-immune donor generated virus-specific, serotype-cross-reactive, CD8+, class I-restricted cytotoxic T lymphocytes (CTL) capable of lysing dengue virus-infected cells and cells pulsed with dengue virus antigens of all four serotypes. These CTL lysed autologous fibroblasts infected with vaccinia virus-dengue virus recombinant viruses containing the E gene or several nonstructural dengue virus type 4 genes. These results demonstrate that both dengue virus structural and nonstructural proteins are targets for the cytotoxic T-cell-mediated immune response to dengue virus and suggest that serotype-cross-reactive CD8+ CTL may be important mediators of viral clearance and of virus-induced immunopathology during secondary dengue virus infections.

Proper processing of dengue virus nonstructural glycoprotein NS1 requires the N-terminal hydrophobic signal sequence and the downstream nonstructural protein NS2a.

Expression of dengue virus gene products involves specific proteolytic cleavages of a precursor polyprotein. To study the flanking sequences required for expression of the dengue virus nonstructural glycoprotein NS1, we constructed a series of recombinant vaccinia viruses that contain the coding sequence for NS1 in combination with various lengths of upstream and downstream sequences. The NS1 products expressed by these viruses in infected CV-1 cells were immune precipitated and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The data show that the 24-residue hydrophobic sequence preceding NS1 was necessary and sufficient for the production of glycosylated NS1 and that this sequence was cleaved from NS1 in the absence of most dengue virus proteins. This finding is consistent with previous proposals that this hydrophobic sequence serves as an N-terminal signal sequence that is cleaved by signal peptidase. The cleavage between the C terminus of NS1 and the downstream protein NS2a occurred when the complete NS2a was present. Recombinant viruses containing NS1 plus 15 or 49% of NS2a produced proteins larger than authentic NS1, indicating that the cleavage between NS1 and NS2a had not occurred. Failure of cleavage was not corrected by coinfection with a recombinant virus capable of cleavage. These results suggest that NS2a may be a cis-acting protease that cleaves itself from NS1, or NS2a may provide sequences for recognition by a specific cellular protease that cleaves at the NS1-NS2a junction.

Characterization of adenovirus particles made by deletion mutants lacking the fiber gene.

H2dl802, H2dl807, and H5dl1021 are defective deletion mutants of human adenovirus which do not make the capsid protein fiber yet which can make substantial amounts of virus particles. Virions made by the mutants contain very little fiber (which comes from helper virus contaminants in the deletion virus stocks): less than 6% as much as that contained by wild-type virions. This demonstrates that fiber is not an essential structural component of the adenovirus virion and suggests that fiber is nonessential for virion assembly. These fiber-deficient particles are poorly adsorbed to cells, consistent with the proposed role of fiber in virus attachment. Further, virion protein precursors, including that of the virion protease, are poorly processed in these particles, suggesting a relationship between the presence of fiber and the maturation of the virus particle.