- Volume 73, Issue 4, 1992

The terminal portion of the herpes simplex virus (HSV) genome long repeat region has been shown to contain a neurovirulence gene. Both HSV-1 and HSV-2 mutants deleted in this gene fail to cause central nervous system (CNS) disease in mice. The HSV-1 strain 17 variant 1716, which has a 759 bp deletion encompassing the gene, grows normally in tissue culture but fails to grow following intracerebral inoculation of mice. This paper demonstrates that 1716 is capable of peripheral replication in the footpads of mice. However, no acute replication of virus is detectable in dorsal root ganglia up to 10 days after footpad inoculation. These results imply that the replication defect in 1716 is not host-specific, but is tissue- and/or cell type-specific. Latency reactivation kinetics demonstrate that 1716 is capable of establishing a latent infection, but the kinetics of reactivation are significantly impaired compared to wild-type virus and are dose-dependent. Lack of acute ganglionic replication combined with impaired reactivation kinetics support the conclusion that a proportion of 1716 genomes initiate a lytic infection which then aborts, and a proportion enter the latent state. The results with 1716 imply that its inability to replicate in CNS and peripheral nervous system neurons is specific, and that the block in replication is beyond the stage of adsorption and entry. A prerequisite for any live attenuated HSV vaccine is an inability to initiate CNS involvement following peripheral inoculation. In this respect, 1716 has prototype vaccine potential with the proviso that a direct extrapolation is being made from mouse to man.

In the published DNA sequence of herpes simplex virus type 1 (HSV-1) strain 17 the coding region for the neurovirulence factor ICP34.5 is disrupted by a 2 bp insertion relative to the corresponding sequence in HSV-1 strain F; this difference would render 60% of the ICP34.5 coding sequence out of frame in strain 17. Re-investigation of the strain 17 sequence showed that the plasmid clone used as the major sequencing substrate for the region was atypical in that other clones did not possess the 2 bp insertion. It was concluded that the coding sequence for ICP34.5 is intact in HSV-1 strain 17 and that the HSV-1 DNA sequence should be revised at this locus.

The human embryonal carcinoma cell line NT2/D1 is known to be non-permissive for human cytomegalovirus (HCMV) but becomes permissive after being induced to differentiate by retinoic acid (RA). Because homeobox genes have been reported to be specifically activated in the RA-differentiated NT2/D1 cells, we investigated the possible correlation between expression of homeobox (HOX) 2 genes and expression of the immediate early (IE) genes of HCMV both in NT2/D1 cells and in HCMV permissive human embryonic lung (HEL) cells. HCMV infection did not induce activation of the HOX2A, HOX2E and HOX2I genes in undifferentiated NT2/D1 cells nor affect their activation in the RA-differentiated NT2/D1 cells. By in situ hybridization using a HOX2A RNA probe, HOX2A transcript-positive cells appeared as clusters in RA-differentiated NT2/D1 cells. Viral antigen-positive cells detected by immunofluorescence using an antibody specific for the IE-1 antigen of HCMV appeared as clusters among the population of cells in which the HOX2A transcript was detected. The HOX2A gene only was expressed in HEL cells, however none of the HOX2 genes was expressed in non-permissive HeLa, Raji or mouse embryonic cells. These results suggest that activation of the HOX2A may be necessary for the expression of IE genes. HCMV infection markedly increased the expression of the HOX2E gene in HEL cells in the presence, but not in the absence, of cycloheximide. Ultraviolet-inactivated HCMV also displayed this effect. On the other hand, HCMV infection suppressed expression of the HOX2A gene to some degree at the early and late phases of infection in HEL cells. Activation of the HOX2E gene by HCMV might possibly have a role in virus-induced abnormal embryogenesis.

We mapped three antigenic domains of continuous epitopes on human cytomegalovirus (CMV) glycoprotein B (gB) by reacting a panel of independently derived monoclonal antibodies with deletion mutants expressed transiently in COS-1 cells. One of these antigenic domains, DC2, maps in the last 75 amino acids of the carboxy terminus. These epitopes are conserved in strains Towne and AD169, as well as in 19 clinical CMV isolates. ELISAs of DC2-reactive antibodies with a set of overlapping synthetic oligopeptides from the carboxy terminus showed that the epitopes of antibodies CH405-1 and CH421-5 map between amino acids 833 and 852 and that the epitope of antibody CH28-2 maps between amino acids 878 and 898. These linear epitopes were grouped into domain DC3. The third antigenic domain, DC1v, maps at the amino-terminal end of CMV strain AD169 gB but is not contained in strain Towne or in 17 of 19 clinical isolates. Epitopes in this domain are likely to map between amino acids 28 and 67, an area where differences occur in the nucleotide sequence of the gB genes from AD169 and Towne. Analysis of CMV-infected cells by flow cytometry with antibodies to the amino- and carboxy-terminal domains revealed that the amino terminus of gB is extracellular and that the carboxy terminus is not exposed on the cell surface.

Computer-assisted analysis of the amino acid sequences of vaccinia virus proteins containing the purine NTP-binding pattern revealed the seven motifs typical of the DNA (RNA) helicase superfamily II in the proteins I8R and A18R. Together with the previously described putative helicases D6R and D11L, the number of putative helicases of this superfamily encoded by the genome of vaccinia virus now reaches four. Aside from the helicase motifs, the sequences of I8R and A18R showed no strong similarity to each other, nor to D6R and D11L. Statistically significant similarity was demonstrated between I8R and the putative RNA helicases involved in pre-mRNA splicing in yeast, PRP2, PRP16 and PRP22, whereas A18R appeared to be related to the putative helicases encoded by the human DNA repair gene ERCC3 and the D10 gene of bacteriophage T5. These findings suggest that I8R may be an RNA helicase. Based on the known properties of the virion NTPases of vaccinia virus, it is possible that the I8R protein may be identical to the previously characterized virion NTPase II. A18R is likely to possess DNA and/or RNA helicase activity. Circumstantial evidence suggests that this activity might be involved in melting duplex structures in late mRNAs. The possibility of independent acquisition of the putative helicases I8R, A18R and a common progenitor to D6R and D11L by an ancestral poxvirus is discussed.

Influenza virus membrane fusion is induced by low pH, which triggers an irreversible conformational change in the viral haemagglutinin (HA). The result of this change is the extrusion of the HA fusion peptide, after which it may act in the fusion of virus and endosomal membranes. Here we describe electron microscopic observations on low pH-treated virus after negative staining or cryo-electron microscopy of virus in the frozen hydrated state. The results indicate a destabilization of the virus membrane at low pH that can be reversed by returning the pH to neutral.

The nucleotide sequence of the mRNA encoding the nucleocapsid (N) protein of bovine respiratory syncytial (BRS) virus, strain 391-2, was determined. Recombinant vectors containing a cDNA of the complete N gene were constructed, and expression of the N protein in eukaryotic cells was demonstrated using two different vector systems. The BRS virus N mRNA was 1197 nucleotides in length, exclusive of poly(A), and had a single major open reading frame that encoded a polypeptide of 391 amino acids with a calculated M r of 42.6K. The nucleotide and amino acid sequences of the BRS virus N gene were compared to those of human respiratory syncytial (HRS) virus strains A2 and 18537, and to BRS virus strain A51908. The level of nucleic acid identity between the N mRNA of BRS virus 391-2 and both HRS virus subtypes was 80 to 81%, whereas the identity between the two BRS virus strains was 97%. A 93 to 94% level of identity was observed between the deduced amino acid sequences of the N protein of BRS virus 391-2 and the corresponding sequences of the two HRS virus strains. The two BRS virus N proteins differed in amino acid sequence at only three positions. Recombinant BRS virus N protein was expressed using two different vector systems: in cells from a plasmid using the vaccinia virus/T7 polymerase expression system or from a recombinant vaccinia virus. N proteins synthesized by the two vector systems migrated with an electrophoretic mobility identical to that of authentic BRS virus N protein, and were precipitated by anti-BRS virus antibodies.

The complete nucleotide sequence of the large (L) protein gene of simian virus 5 (SV5) was determined from cDNA of the genomic RNA and mRNA, and found to be 6804 bases in length, exclusive of a poly(A) tract. The sequence contained an open reading frame of 6765 nucleotides encoding 2255 amino acids. Results of dot matrix comparisons of the L protein of SV5 with those of human parainfluenza type 3 virus and Sendai virus indicated that there are five conserved domains, and that each domain contains characteristic sequence(s). The L protein of SV5 was detected in purified virions using antiserum directed against an oligopeptide corresponding to the N-terminal region.

Viral haemorrhagic septicaemia is the most important viral disease of trout in Europe. The causative agent, viral haemorrhagic septicaemia virus (VHSV), a member of the lyssavirus genus of the rhabdoviridae family, was formerly believed to be confined to portions of the European continent; however in 1988, VHSV was isolated from adult chinook (Oncorhynchus tshawytscha) and coho (O. kisutch) salmon returning to two hatcheries in the northwestern part of the State of Washington, U.S.A. Initial fears were that the virus had been imported into North America, perhaps by aquaculture activities. The nucleotide and deduced amino acid sequences of the nucleocapsid (N) gene of one of the North American (Makah) isolates of VHSV were determined and compared with published sequences of a European reference strain of VHSV (07-71) and the Round Butte strain of infectious haematopoietic necrosis virus (IHNV), another salmonid fish rhabdovirus that is enzootic in western North America. The N gene of the Makah isolate of VHSV shared a similarity of 88.433% at the nucleotide level and 94.802% at the amino acid level with the N gene of the European strain of VHSV, and 62.121% amino acid similarity with the N protein of IHNV. Like the European reference isolate, the North American isolate of VHSV showed three domains in the N protein, the central one being the most conserved and the likely site of interaction with genomic RNA. This was also the region of highest similarity with the amino acid sequence of IHNV. The sequence data suggested that the Makah and 07-71 isolates were of independent origin.

Chimpanzees were inoculated intravenously with the H strain of hepatitis C virus (HCV), and analysed for viraemia using the polymerase chain reaction and for a humoral immune response using first and second generation anti-HCV ELISAs and an immunoblot assay (4-RIBA). In all seven chimpanzees studied, viraemia occurred several weeks before a significant increase in serum alanine transferase (ALT) activity, whereas the first circulating anti-HCV antibodies became detectable at the time of significant increase in ALT levels, provided the second generation ELISA or 4-RIBA was used. On the basis of the duration of viraemia the chimpanzees studied could be assigned to two different groups: those in which viraemia disappeared in conjunction with or shortly after seroconversion, and those remaining viraemic for many weeks after the appearance of antibodies. The clearance of HCV from the circulation did not correlate with the antibody pattern determined using 4-RIBA, i.e. the HCV-specific assays currently available do not enable us to predict whether an infected chimpanzee will develop persistent viraemia. Only two of the seven chimpanzees analysed developed anti-core protein (c-22) antibodies, which appeared at the same time as the first ALT peak, whereas all animals developed antibodies to the non-structural protein, c-33, and these antibodies persisted.

Picorna-like virus particles, 29 nm in diameter, were purified from apparently healthy Laodelphax striatellus Fallen. The virus particles had a buoyant density of 1.352 g/ml in CsCl and a sedimentation coefficient of 161 s. The virus capsid proteins consisted of three major polypeptides of M rs 36500, 33000 and 28000, and three minor polypeptides. The virus contained a major ssRNA of M r 2.8 × 106 and was also frequently associated with a minor dsRNA of M r 4×106. The 3′ end of the ssRNA had a poly(A) tract of about 60 adenine residues. The virus has been provisionally named himetobi P virus.

Although increasingly unlikely, the possibility of a scrapie-specific nucleic acid carried by infectious prion particles is still unresolved. Return refocusing gel electrophoresis was developed to detect homogeneous and heterogeneous nucleic acids extracted from highly purified scrapie prion preparations. This method was improved with respect to the size range from 13 to 1100 nucleotides (nt) over which analyses could be performed. The yield of nucleic acid, particularly of small DNA oligonucleotides and polyadenylated RNA, was determined after deproteinization and two-phase extraction. Despite extensive nuclease digestions some small polynucleotides remained. Although a scrapie-specific nucleic acid cannot be excluded, the results further define the possible characteristics of a hypothetical molecule. If homogeneous in size, such a molecule would be <80 nt in length at a particle-to-infectivity ratio near unity, if heterogeneous, scrapie-specific nucleic acids would have to include molecules smaller than 240 nt.

Double-stranded RNAs were isolated from plants infected with five different isolates of apple chlorotic leaf spot virus (ACLSV). Analysis by PAGE and by Northern blot hybridization showed that six major species of viral dsRNA of approximately 7.5, 6.4, 5.4, 2.2, 1.1 and 1.0 kbp can be detected in infected plants, irrespective of the ACLSV isolate used. In addition to the dsRNA of 7.5 kbp corresponding to the full-length genome, the size and position on the genome of the 2.2 and 1.1 kbp species indicate that these are very probably double-stranded forms of subgenomic RNAs allowing the expression of the internal open reading frames coding respectively for the ACLSV 50K and coat proteins. The subgenomic messenger for the coat protein was indeed detected in total RNA preparations from infected plants. Surprisingly, the two most abundant dsRNA species, of 6.4 and 5.4 kbp, were found to be 5′-coterminal with the genomic RNA. A model for the expression of the genome of ACLSV and for the production of the molecules 5′-coterminal with the genomic RNA is presented.

The small nuclear inclusion (NIa) protein of the tobacco etch virus (TEV) is synthesized initially as part of a genome-derived high M r precursor. The NIa protein releases itself from this genome-derived precursor by self-cleavage, or an autocatalytic processing event. Cleavage between specific glutamine-glycine dipeptides at the N and C termini generates the 430 amino acid or 49000 M r (49K) NIa protein. The requirements of this autocatalytic release, or cis cleavage, were examined by constructing gene cassettes encoding the TEV NIa protein which could be ligated into particular locations in cDNA of the TEV genome and also into foreign gene DNA sequences. Using cell-free transcription and translation systems, polyproteins containing TEV NIa sequences were synthesized and assayed for (i) autocatalysis and (ii) the ability of a functional NIa proteinase, purified from plant tissue, to cleave in bimolecular or trans reactions various artificial polyproteins which contained an inactive form of the NIa proteinase. The NIa self-cleavage events required an active proteinase sequence and a consensus TEV cleavage site sequence at the N and C termini. These results were consistent for NIa protein sequences placed at a foreign TEV cleavage site or in unrelated proteins. Differences were noted in the trans cleavage of these sites.

The 5′- and 3′-terminal nucleotide sequences of the dsRNA genome segments of rice ragged stunt virus (RRSV), a member of the plant Reoviridae, were determined and compared with those published for other viruses in this family. The 5′- and 3′-terminal regions of the RRSV plus strand RNA from all genome segments were found to have the same conserved hexanucleotide (5′ GAUAAA---) and tetranucleotide (---GUGC 3′) sequences, respectively. These conserved terminal sequences were different from those found in viruses in the Phytoreovirus and Fijivirus genera. This result confirms that, as already suggested, RRSV should be placed in a third genus.