Conversion of VP1 to VPg in cells infected by infectious pancreatic necrosis virus.

The RNA polymerase VP1 of IPNV (a bisegmented dsRNA containing virus) is present in the virion both as a free polypeptide and as a genome-linked protein (VPg). Virion VP1 primes viral RNA synthesis in vitro (P. Dobos, 1995, Virology 208, 19-25), and here we present data which suggest that protein-primed RNA synthesis may also take place in infectious pancreatic necrosis virus (IPNV)-infected cells. Anti-VP1 serum immunoprecipitated several polypeptides larger than the 94-kDa VP1 of IPNV from [35S]methionine-labeled infected cell lysates. During denaturing, SDS-polyacrylamide gel electrophoresis these polypeptides formed a characteristic "ladder" which was resistant to alkaline phosphatase but sensitive to RNases, indicating that it consisted of VP1 polypeptides with oligoribonucleotides of various lengths attached to them. Probing the ladder with 5' and 3' end-specific, as well as plus-, or minus-strand-specific oligonucleotides revealed that they represent VP1 linked to 5' terminal sequences of genome segment A- and B-specific plus strands. Pulse-chase experiments in combination with two-dimensional polyacrylamide gel electrophoresis revealed that labeled VP1 could be chased to replicative intermediate, to ssRNA, to dsRNA, and eventually to virion VPg-dsRNA and that VP1 could be released from all these structures by RNase treatment. We suggest that these results are most compatible with the model where a VP1-pN structure acts as a primer for viral RNA synthesis in vivo, a mechanism that has been shown to occur in vitro.

Incomplete dsRNA genomes in purified infectious pancreatic necrosis virus.

The dsRNA containing birnavirus, infectious pancreatic necrosis virus, possesses a virion-associated RNA-dependent RNA polymerase which acts both as primer and as polymerase during in vitro RNA synthesis (P. Dobos, 1995, Virology 208, 19-25). Using [alpha 32P]GTP, we have radiolabeled virion RNA in vitro and found that after deproteinization most of the labeled product comigrated in agarose gels with the 3-kbp viral genome, while the remainder migrated faster than the dsRNA and as a heterogeneous smear. Agarose gel electrophoresis (AGE) of denatured labeled virion RNA showed a radioactive smear ranging from approximately 100 nucleotides to up to 3000 nucleotide the size of genome length single stranded RNA. Hybridization experiments using strand-specific and end-specific obligonucleotides on Northern blots revealed that the radioactivity which migrates with the dsRNA during AGE represents small, 5' end plus RNA molecules of 100-500 nucleotides. The radioactivity in the faster migrating smear denotes incomplete dsRNAs where full-length, unlabeled minus strands are base-paired with labeled plus strands that are 3' truncated to different extents. This was confirmed by reverse transcription-polymerase chain reaction (RT-PCR) using end- and strand-specific obligonucleotide primers. The results indicated that 95% of incomplete dsRNA molecules consisted of full-length minus strands and 3' truncated plus strands. The implications of these findings are discussed in light of RNA replication mechanisms of dsRNA viruses belonging to other families.

Sequence analysis of the bicistronic Drosophila X virus genome segment A and its encoded polypeptides.

Drosophila X virus represents the entomobirnavirus genus of the Family Birnaviridae. Segment A of this bisegmented dsRNA containing virus was cloned and sequenced. The 3360-bp-long nucleotide sequence revealed the presence of two open reading frames (ORFs). A large ORF of 3096 nucleotides, which is flanked by a 107-bp 5' and a 157-bp 3'-untranslated region, and a 711-nucleotide-long small ORF located within the carboxy half of the large ORF but in a different reading frame. The large ORF encodes a 114-kDa polyprotein which is cotranslationally processed by the virus-coded protease VP4 to generate preVP2, VP3, and VP4 (VP1 is encoded by genome segment B). N-terminal amino acid sequencing of VP3 and VP4 established the order NH2-preVP2-VP4-VP3-COOH within the polyprotein. The small ORF straddles the VP4/VP3 junction and is capable of encoding a basic, arginine-rich 27-kDa polypeptide which so far has not been detected in infected cells. The amino acid sequences specified by the two ORFs were compared to those of infectious pancreatic necrosis virus (IPNV) and infectious bursal disease virus (IBDV) that represent the two other genera (aquabirnavirus and avibirnavirus) of the Birnaviridae family. Significant sequence homology among the three viruses was found to be restricted to the amino and carboxy regions of preVP2 and to a small 21-residue-long domain near the carboxy terminal region of VP3. Significant sequence homology is exhibited by the small ORFs of the three viruses.

Mapping of a serotype specific epitope of the major capsid protein VP2 of infectious pancreatic necrosis virus.

A serotype-specific, conformational epitope of VP2 of infectious pancreatic necrosis virus (IPNV) (Jasper), has been mapped by restriction enzyme site-specific deletions of cloned viral cDNA. Subclones encoding fragments of VP2 were expressed in Escherichia coli followed by polyacrylamide gel electrophoresis and Western blot analysis. The epitope (between amino acid residues 242 and 325) reacted with monoclonal anti-VP2 antibodies that neutralized the homologous (Jasper), but not the heterologous (Sp) serotype of IPNV.

Identification and analysis of a putative origin of DNA replication in the Choristoneura fumiferana multinucleocapsid nuclear polyhedrosis virus genome.

A recombinant plasmid containing the Choristoneura fumiferana multinucleocapsid nuclear polyhedrosis virus (CfMNPV) HindIII R fragment (m.u. 2.2-3.9) was shown to undergo CfMNPV infection-dependent DNA replication in Cf-124T cells. Replication of this DNA sequence was detectable by 24 hr p.i. and did not appear to have resulted as a consequence of recombination with the virus genome. Replication was inhibited by mimosine, an inhibitor of eukaryotic DNA replication. These data suggest that HindIII R of CfMNPV DNA contains an origin of DNA replication which we call ori1. HindIII R contains five GC-rich and three AT-rich regions and a 0.9-kb homologous repeat region 1 (hr1). Two short 440- and 740-bp contiguous sequences at the right end of the HindIII R fragment separately exhibited limited ori function. HindIII R subfragments with optimal ori activity contained a cluster of repeated and inverted sequences including nine copies of a 50-bp homologous repeat sequence (hr1a to hr1i) within hr1. The CfMNPV hr1 sequence was somewhat homologous with the homologous repeat (hr) of the putative Autographa californica MNPV (AcMNPV) replication origins. HindIII Y, another CfMNPV DNA fragment containing an hr sequence, hr3, also supported infection-dependent DNA replication, suggesting that it too contains an ori. Although replication of a putative AcMNPV origin (HindIII Q) was detectable in CfMNPV-infected Cf-124T cells, replication of CfMNPV HindIII R was not detectable in AcMNPV-infected Spodoptera frugiperda cells.

Protein-primed RNA synthesis in vitro by the virion-associated RNA polymerase of infectious pancreatic necrosis virus.

The 94-kDa virion-associated RNA-dependent RNA polymerase (RdRp) is present in infectious pancreatic necrosis virus in two forms: (i) as a free polypeptide (VP1) and (ii) as a genome linked protein (VPg) (J. G. Calvert et al., 1991, J. Gen. Virol. 72, 2563-2567). VP1 was guanylylated in vitro by incubating purified virus in the presence of [alpha2P]GTP. During further incubation in an in vitro RNA polymerase reaction mixture (in the presence of unlabeled GTP), the radiolabeled VP1-pG complex was "chased" via nascent RNA strands and replicative intermediates to a VP1-dsRNA complex. Labeled VP1-pG was recovered from the intermediate as well as from the final reaction products by digestion with RNase A and RNase V1, a dsRNA-specific nuclease. Analysis of the reaction products indicated that only the plus strands of the two genome segments were being synthesized in vitro which remained base-paired to their templates. The results suggest that in vitro transcription by the virion RdRp is primed by VP1 and then proceeds via an asymmetric, semiconservative, strand-displacement mechanism.

Evidence for the detection of the infectious pancreatic necrosis virus polyprotein and the 17-kDa polypeptide in infected cells and of the NS protease in purified virus.

The larger genome segment A of infectious pancreatic necrosis virus (IPNV) contains two open reading frames (ORFs): (i) the large ORF encodes a 106-kDa polyprotein (PP) (NH2-pVP2-NS-VP3-COOH) which is cotranslationally cleaved by the NS protease to generate the major capsid polypeptides VP2 and VP3; (ii) the second small ORF, which overlaps the 5' end of the PP ORF but in a different reading frame, encodes a 17-kDa arginine-rich polypeptide. Hitherto, neither the PP nor the 17-kDa polypeptide have been identified in infected cells, and the NS (nonstructural) polypeptide was thought not to be part of the virion. The smaller genome segment B of IPNV encodes a 94-kDa minor polypeptide VP1. Using recombinant baculoviruses expressing VP1 and PP as markers, the PP could be unambiguously identified in Western blots of infected fish-cell lysates and in purified IPNV. Anti-17-kDa and anti-NS serum was produced by injecting rabbits with bacterially expressed fusion proteins containing these polypeptides. Labeled 17-kDa polypeptide was immune-precipitated from infected cell lysates using the anti-17-kDa serum, whereas the NS and NSt (a truncated form of NS) polypeptides were identified in infected cells by immune precipitation and Western blotting using the anti-NS serum. Western blots of purified virus revealed two forms of truncated NS: (i) the NSt found in infected cells and (ii) a smaller polypeptide NSta. The identity of the virion NSt/NSta was also demonstrated by peptide mapping.

Expression of infectious pancreatic necrosis virus polyprotein and VP1 in insect cells and the detection of the polyprotein in purified virus.

In order to study the molecular biology of infectious pancreatic necrosis virus (IPNV) replication, six different recombinant baculoviruses were constructed. The following four recombinants contained genome segment A-specific sequences; (i) AcPP contained the complete polyprotein coding region and Spodoptera frugiperda (Sf) cells infected by these recombinants synthesized the 106-kDa polyprotein (NH2-preVP2-NS protease-VP3-COOH), which was only partially processed by the protease to yield preVP2 and VP3 and unprocessed polyprotein; (ii) AcPP(S) and AcPP(Ss) represented 3' truncated sequences of the segment A cDNA where the VP3 coding region and that coding for 30 and 98 carboxy terminal amino acids of NS in the two constructs, respectively, were deleted. AcPP(S) demonstrated partial, and that of AcPP(Ss), complete loss of proteolytic activity, demonstrating that the carboxy one-third of the 29-kDa NS protease is necessary for the formation of the active enzyme; and (iii) AcPP(B/B) contained all but the first 180 nt of the pVP2 gene, the complete NS coding region, and the amino end of VP3. Analysis of cells coinfected with AcPP(Ss) and AcPP(B/B) showed either that the protease did not work in trans or that the alteration of the structure of the substrate prevented cleavage. Recombinant baculoviruses AcVP1VL and AcVP1ETL contained IPNV genome segment B cDNA encoding the 94-kDa VP1 which is the viral RNA-dependent RNA polymerase. AcVP1VL contained the whole segment B cDNA, whereas in AcVP1ETL, the 5' non-coding sequences were deleted resulting in the production of large amounts of VP1 when Sf cells were infected with this recombinant. The use of recombinants AcPP and AcVP1ETL as well as monoclonal antibodies and VP1-specific sera allowed the unambiguous identification of the high molecular weight minor polypeptides present in purified IPNV demonstrating the presence of both VP1 and the polyprotein in purified virus preparations.

Adenine DNA methyltransferase M.CviRI expression accelerates apoptosis in baculovirus-infected insect cells.

The adenine DNA methyltransferase M.CviRI (TGCmA) gene from chlorella virus XZ-6E was cloned into the Autographa californica nuclear polyhedrosis virus (AcMNPV) genome and expressed in Spodoptera frugiperda insect cells under the control of two tandemly arranged viral promoters, the early ETL promoter and the late polyhedrin promoter. M.CviRI activity was first detected at 10 hr p.i and reached a maximum at 48 hr p.i. Viral DNA synthesized in insect cells infected with M.CviRI expressing virus (AcMTRI) was methylated at all TGCA sites. Unexpectedly, AcMTRI-infected cells lysed 48 hr earlier than wild-type AcMNPV-infected cells. Moreover, cellular DNA, but not viral DNA, from AcMTRI-infected cells was degraded to fragment sizes characteristic of apoptosis. These results suggest that M.CviRI methylation influences the onset of viral cytopathic effects and induces an apoptosis-like response.

In vitro guanylylation of infectious pancreatic necrosis virus polypeptide VP1.

Incubation of purified infectious pancreatic necrosis virus (IPNV) in the presence of [alpha 32P]GTP resulted in the formation of VP1-GMP. The GMP is linked to VP1 by a phosphodiester bond and its formation does not require the presence of divalent cations. In contrast to reovirus guanylyl transferase, the formation of IPNV VP1-GMP is not reversible and the guanylylation reaction is not inhibited by inorganic pyrophosphate. Furthermore, the IPNV VP1-GMP cannot transfer the GMP to an acceptor molecule (such as GTP) indicating that VP1 is not a capping enzyme. Time-course experiments revealed that after the initial guanylylation of VP1 to form VP1-pG, a second GMP is added to form VP1-pGpG, the formation of which is template-dependent. Since VP1 is present in the virion both as a free polypeptide and in a genome-linked form as VPg, and it is also the virion-associated RNA polymerase, the results suggest that VP1 may function as a primer during in vitro RNA synthesis.

Identification of bent DNA and ARS fragments in the genome of Choristoneura fumiferana nuclear polyhedrosis virus.

We identified four CfMNPV DNA fragments with autonomously replicating sequences (ARS) functional in Saccharomyces cerevisiae. A 0.9-kb fragment which, mapped to 54.5 to 55.3 map units within EcoRI HI of the CfMNPV genome, showed the strongest ARS activity of the four. Sequence analysis of this 0.9-kb DNA segment revealed an A+T-rich region separated from a G+C-rich region by 320 bp. Although no sequence matched exactly the ARS core-consensus sequence, 13 near-matches differing by only one or two nucleotides from the core-consensus sequence, were identified. Ten near-matches were clustered within a 105-bp A+T-rich region, and were arranged as inverted repeats. A section of bent DNA structure was predicted within this region. The bent DNA, which showed temperature-dependent retardation during polyacrylamide gel electrophoresis, was unique as its sequence was arranged as a symmetrical 'tilde' (approximately) structure. The second (1.0 kb) and third (1.6 kb) ARS-bearing fragments mapped within EcoRI-E and -B fragments which contain homologous repeat sequences. The fourth (1.5 kb) fragment had the weakest ARS activity and mapped to the EcoRI-D or -B regions of the genome.

Antigenic and genomic differences of two Jasper strains of infectious pancreatic necrosis virus.

Strains of infectious pancreatic necrosis virus (IPNV-Jasper) obtained from two different laboratories were compared serologically with polyclonal and monoclonal antibodies. Nucleotide sequence and restriction endonuclease patterns of 359-bp fragment of genome segment A cDNA were also compared. Substantial differences were found in both analyses that will support the fact that the two Jasper strains are not identical.

Characterization of the VPg-dsRNA linkage of infectious pancreatic necrosis virus.

By the use of strong denaturing agents, a genome-linked protein (VPg)-RNA complex was purified from infectious pancreatic necrosis virus. Ribonuclease treatment of 125I-labelled VPg-RNA released a 90K polypeptide identical to the minor structural polypeptide VP1 (the putative RNA polymerase), as determined by peptide mapping. The polypeptide is linked to the RNA by a serine-5' GMP phosphodiester bond. The results identify birnaviruses as the only dsRNA viruses with a VPg, the size of which is the largest of the VPgs of RNA viruses.

Sequence analysis of infectious pancreatic necrosis virus genome segment B and its encoded VP1 protein: a putative RNA-dependent RNA polymerase lacking the Gly-Asp-Asp motif.

The genome segment B sequence of infectious pancreatic necrosis virus was determined for both the Jasper and Sp serotypes. The sequences are 2784 and 2630 bp long, respectively, and contain a single large open reading frame encoding the VP1 protein, the putative RNA-dependent RNA polymerase (RdRp) of IPNV. The proteins exhibit an 88% homology with each other, but only 41% with infectious bursal disease virus (IBDV) VP1, another member of the Birnaviridae. Despite the low overall homology between the IPNV and IBDV VP1 proteins, homologous regions were detected within the central portion of the proteins. The carboxy-proximal regions of the VP1, which contain very low amino acid homology, displayed evidence of conservation in structural features such as a hydrophilic, highly basic domain. Consensus sequences associated with GTP-binding proteins and RdRps were also detected in VP1. However, unlike the RdRps associated with single-stranded plus RNA viruses, the birnavirus RdRp lacks the Gly-Asp-Asp motif characteristic of this enzyme family.

Cloning and expression of NDV hemagglutinin-neuraminidase cDNA in a baculovirus expression vector system.

The hemagglutinin-neuraminidase (HN) gene of the Hitchner B1 strain of Newcastle disease virus (NDV) was cloned as a cDNA and inserted into a baculovirus expression vector. The recombinant HN (recHN) expressed in Spodoptera frugiperda cells had both hemagglutinating and neuraminidase activities both of which were inhibited by polyclonal anti-NDV sera or a monoclonal antibody (MAb) against HN. Infected insect cells could hemadsorb chicken red blood cells suggesting that the recHN is properly glycosylated and transported to the cell surface. A 67-kDa recHN precursor and a 74-kDa, presumably mature, recHN from infected cells were detected by Western blot analysis and were found to comigrate with similar proteins from NDV-infected chick embryo fibroblast cells. The kinetics of synthesis of recHN was similar to that for polyhedrin and some HN appeared in the extracellular medium. HN was copurified with extracellular virus (ECV) from the extracellular medium and was used to immunize chickens. The anti recHN serum was specific to NDV in both ELISA and Western blot analysis.

Expression in Escherichia coli of the major outer capsid protein of infectious pancreatic necrosis virus.

The outer capsid polypeptide, VP2, represents the major neutralizing antigen of infectious pancreatic necrosis virus (IPNV). A 926-bp viral cDNA, encoding an N-terminal truncated VP2, was cloned into the pWR590 expression plasmid family resulting in a C-terminal extension of a truncated Escherichia coli beta-galactosidase (beta Gal) under the control of the lac promoter. When cells transformed by in-phase hybrid plasmids were induced by isopropylthiogalactoside, high levels of the 100-kDa beta Gal-VP2 fusion protein accumulated within 4 h after induction. The fusion protein reacted in Western blots both with rabbit anti-beta Gal and with neutralizing mouse anti-VP2 monoclonal antibody. Sera of rabbits immunized with semipurified fusion protein reacted with the VP2 polypeptide in Western blots and with intact purified virus in ELISA and also neutralized IPNV infectivity in a plaque-reduction assay. Out-of-phase hybrid plasmids did not produce the fusion protein but expressed a small amount of structurally discrete VP2-specific sequences probably by internal initiation of translation at an in-phase AUG codon near the 5' end of the VP2 gene.

Synthesis of the infectious pancreatic necrosis virus polyprotein, detection of a virus-encoded protease, and fine structure mapping of genome segment A coding regions.

Full-length and truncated genome segment A-specific infectious pancreatic necrosis virus cDNA was subcloned into plasmid transcription vectors, and runoff transcripts were produced in vitro. These transcripts were translated in cell-free rabbit reticulocyte lysates and the translation products were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Virus-specific polypeptides were gel purified and mapped by partial proteolysis with N-chlorosuccinimide and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Peptide profiles were compared with those of the corresponding polypeptides purified from infectious pancreatic necrosis virus-infected cells or prepared by in vitro translation of denatured genomic RNA. The cDNA directed the synthesis of authentic pVP2, VP3, and NS polypeptides as well as a number of previously undescribed polypeptides. A 101,000-molecular-weight polypeptide was isolated and shown to be the unprocessed infectious pancreatic necrosis virus polyprotein. The NS polypeptide appears to be a virus-encoded protease responsible for the cleavage of pVP2 from the polyprotein. The carboxy terminus of NS was mapped to within three or four amino acids on the polyprotein. The most likely internal translation start sites responsible for NS and VP3 production in vitro were also mapped.

Mapping of the large RNA genome segment of infectious pancreatic necrosis virus by hybrid arrested translation.

Segment A, the larger dsRNA segment of IPNV which encodes three of the four virus-coded polypeptides (preVP2, VP3, and NS) was cloned and physically mapped. The plus and minus RNA strands of the virus genome were separated and the A+ and B+ RNA strands were identified. A nested set of cDNA subclones, coterminal with the 5' end of A+ RNA, were used in hybrid arrested translation experiments. Hybrid arrest conditions which blocked the 5' two-thirds of A+ RNA allowed the in vitro synthesis of only VP3, while hybridization of the RNA to cDNA representing the 5' half of A+ RNA allowed the synthesis of both NS and VP3 but not of preVP2. In vitro translation of A+ RNA yielded all three polypeptides. It is, therefore, concluded that the order of the three polypeptides on A+ RNA is 5'-preVP2-NS-VP3-3'. These results imply that internal initiation of translation could take place on the RNA at least in vitro at sites located hundreds of nucleotides downstream from the first in-phase AUG codon near the 5' end.

Experimental induction of the carrier state in yearling brook trout: a model challenge protocol for IPNV immunization.

Brook trout fry (Salvelinus fontinalis) were not protected from infectious pancreatic necrosis virus (IPNV) challenge by immersion vaccination with inactivated, purified virus at concentrations of 10(7) to 10(9) pfu/ml. Mortalities in vaccinated groups were higher than for the unvaccinated control group and appeared to be dose-dependent. A challenge protocol for adult brook trout was developed for future vaccine trials. A single intraperitoneal injection of virulent, purified virus was sufficient to make long-lasting carriers of 16 month-old trout. Fish underwent a transient viremia, identified by virus isolation from plasma and leucocytes. Feces were the most reliable samples for identification of IPNV carriers by non-sacrificial testing. Many fish in the remaining infected group were still carriers 12 and 27 weeks post-infection.