- Volume 74, Issue 6, 1993

The glycoprotein H (gH) genes of two avian herpesviruses, Marek’s disease virus and the herpesvirus of turkeys, have been cloned and sequenced and the coding regions found to be of 2439 and 2424 nucleotides respectively. The predicted primary polypeptide products of these open reading frames are 813 and 808 amino acids and correspond to M rs of 90 800 and 91 100. Both amino acid sequences exhibit characteristic glycoprotein features such as hydrophobic signal and anchor sequences and potential sites for N-linked glycosylation. Polypeptide sequence comparison to the other eight available gH sequences revealed more similarity to the alphaherpesvirus subgroup than to either beta- or gammaherpesviruses.

The region of the Orgyia pseudotsugata multinucleocapsid nuclear polyhedrosis virus (OpMNPV) genome containing the ubiquitin and the nuclear matrix-associated protein (39K) genes was sequenced. The first 77 amino acids of the OpMNPV ubiquitin open reading frame (ORF) showed 84.4% and 79% amino acid sequence identity to Autographa californica multinucleocapsid nuclear polyhedrosis virus (AcMNPV) and the Spodoptera frugiperda insect ubiquitin, respectively. The predicted OpMNPV ubiquitin protein contains a 3′ tail of 16 amino acids not present in the AcMNPV or S. frugiperda ubiquitin ORFs. The OpMNPV 39K ORF showed 56% amino acid sequence identity with the AcMNPV 39K ORF. Four additional ORFs from this region were also characterized.

We examined the level of antigenic conservation amongst the haemagglutinins (HAs) of H1 swine influenza viruses, recently isolated from a wide geographical area, in haemagglutination inhibition assays against a panel of four monoclonal antibodies (MAbs). We found a high degree of conservation with a dominant variant (52 of 54 isolates) that reacted with all MAbs. Only two minor variants, each failing to react with one MAb, were found. Using a one-step PCR technique followed by direct sequencing of the products, we examined the HA1 region of the HA RNA of two representative dominant variants. We found no amino acid substitutions relative to a reference strain. The sequences of the HA1 RNA of the two minor variants isolated here and of two other minor variants defined previously were also determined. Each contained inferred amino acid substitutions, all located at different positions on the HA. Finally, we sequenced HA1 RNA obtained from the original pig lung suspensions from which the two dominant and two minor variants had been isolated. Three of the parent viruses were identical to their progeny in eggs whereas the fourth parent virus contained four amino acid differences from its progeny.

Plasmids were prepared that contained cDNA segments of the large genomic segment A of infectious bursal disease virus (IBDV) strain GLS-5. The genes encoding the IBDV structural proteins (VP2, VP3 and VP4) were introduced into the baculovirus transfer vector pBlueBacI to obtain a recombinant baculovirus vIBD-7. When insect cells were infected with recombinant viruses, the result was synthesis of IBDV precursor proteins which were processed by the viral protease. The recombinant IBDV antigens were characterized by immunoprecipitation with monoclonal antibodies (MAbs) and polyclonal antiserum to IBDV, and by antigen-capture ELISA using a panel of IBDV-specific MAbs. The recombinant IBDV antigens closely resembled the native IBDV proteins and reacted with all the GLS-5 strain-specific neutralizing MAbs that recognize only conformational epitopes of IBDV. When susceptible chickens were inoculated with the recombinant IBDV antigens, virus-neutralizing antibodies were induced that conferred up to 79% protection against IBDV challenge.

Levels of insulin mRNA in pancreata from SJL/J male mice susceptible to encephalomyocarditis (EMC)-D virus-induced diabetes started to decrease rapidly 24 h after injection with EMC-D virus and only a trace remained 72 h after injection. In contrast, insulin mRNA in pancreata from C57BL/6J male mice resistant to EMC-D virus-induced diabetes did not show any significant changes 0 to 96 h after injection. EMC-D viral RNA in pancreata from SJL/J mice started to increase rapidly 24 h after injection, reached its peak at 48 h and then decreased gradually. In contrast, EMC-D viral RNA in pancreata from C57BL/6J mice was undetectable except for the 24 and 48 h points after injection. EMC-D virus could bind readily to freshly isolated beta cells from SJL/J mice but scarcely bound to beta cells from C57BL/6J mice. In contrast, there was no significant difference between SJL/J and C57BL/6J mice in binding of EMC-D virus to their cultured beta cells. The rate of EMC-D viral attachment to beta cells from C57BL/6J mice increased significantly during the first 24 h culture period and reached the same rate of attachment as that seen for beta cells from SJL/J mice. This suggests that viral receptors on the beta cells derived from strains of mice resistant to EMC virus-induced diabetes are not expressed in vivo, but are expressed during cell culture, rendering the beta cells susceptible to EMC viral infection. On the basis of our previous and present observations, we conclude that a genetic factor controlling susceptibility to EMC-D virus-induced diabetes may operate by modulating the expression of viral receptors on the beta cells.

We determined the complete nucleotide sequences of the VP4 genes of five bovine rotavirus strains (A5, 61A, A44, B223 and KK3). The deduced VP4 amino acid sequence of strain A5 with G8 serotype specificity is 91.1% identical to that of strain NCDV of the serotype G6, and strain 61A with G10 serotype specificity has a VP4 amino acid sequence (95.2% identity) similar to that of the UK strain of the G6 serotype. In contrast, the VP4 amino acid sequences of strains A44, B223 and KK3 of the G10 serotype, isolated in Thailand, the U.S.A. and Japan respectively, have very similar sequences to each other, but less similarity (50 to 60%) to other group A rotavirus strains reported so far. Their VP4 genes are 2352 nucleotides in length and encode 772 amino acids, four amino acids fewer than the VP4 proteins of other animal rotaviruses. Thus, the presence of three different VP4 types (P types) and the independent segregation of G serotype and P type were postulated in bovine rotaviruses by VP4 sequence analysis.

We have determined the nucleotide sequence of genes 6 and 10 of porcine rotavirus YM. When the amino acid sequences of VP6 and NS28, the protein products of genes 6 and 10 respectively, were compared with other published sequences it was evident that the proteins of human rotavirus Wa have the highest degree of identity with rotavirus YM. This is in contrast with the observation that when other proteins of these two strains have been compared they have been found to be among the most distantly related pairs of rotavirus strains. This observation is in accordance with the proposed receptor-ligand interaction between NS28 and VP6 during virus morphogenesis, and suggests a specificity in the interaction between these two proteins. In addition, when rotavirus YM VP6, which belongs to subgroup I, was compared with the VP6 proteins of rotavirus strains having different subgroup specificities, it was found to be more closely related to subgroup II rather than subgroup I proteins. This finding allowed us to identify five potential amino acids on VP6 that may contribute to determining the subgroup antigens.

Genomic length cDNAs of papaya mosaic virus (PMV) RNA were generated utilizing reverse transcriptase (RNase H-) for first strand synthesis, Sequenase for second strand synthesis and primers specific for the 5′ and 3′ termini of the viral genome. These cDNAs were cloned into plasmid pUC18 and infectious RNA transcripts were synthesized in vitro from a bacteriophage T7 RNA polymerase promoter incorporated into the 5′ specific primer. The infectivity of transcripts was 16% that of native PMV RNA. Increasing the poly(A) tail length from A24 to A71 produced a 43% increase in infectivity. Transcripts synthesized with or without an m7GpppG cap structure were biologically active although uncapped transcripts were much less infectious. The addition of up to 2434 non-viral nucleotides at the 3′ end of transcripts decreased but did not abolish infectivity. Insertions of two amino acid residues within the polymerase coding region inactivated viral transcripts. A single amino acid deletion within the capsid protein (CP) produced local lesions of a reduced size as compared to native PMV RNA. Viral particles could not be observed in crude extracts from lesions produced by this deletion mutant suggesting that it exists as a naked RNA species within the host. Mutations to the CP suggest that it is required not only for viral assembly but also for some other unidentified function(s) during the replication cycle.

Alkali-denatured cauliflower mosaic virus (CaMV) virions incorporated ssDNA, added exogenously, into multimolecular complexes during dialysis against a neutral buffer. CaMV coat protein binding to tracer DNA, assessed by gel electrophoresis and autoradiography, was highly cooperative as judged by the absence of intermediate-sized complexes. The incorporation of labelled BglII fragments of a plasmid containing CaMV DNA into complexes was prevented by the presence of 0.16 g/l unlabelled calf thymus DNA. Lower concentrations of competitor DNA allowed binding of some BglII fragments although preventing the binding of others. The self-annealing poly(dI-dC) was much less efficient than calf thymus DNA in preventing the incorporation of fragments into complexes, suggesting a binding preference for ss- over dsDNA. In addition, dsDNA, minimally cross-linked to prevent strand separation, was bound only weakly. End-labelled ssDNA fragments in complexes were partially protected against DNase I. The nucleic acid-binding activity of CaMV coat protein may be responsible for the organization of replication complexes, the precursors to virion particles.

Transgenic plants producing the 30K temperature-sensitive transport protein (TP) of tobacco mosaic virus (TMV) mutant Ni2519 (affecting cell-to-cell transport) were found to: (i) be susceptible to wild-type TMV U1 at 24 °C (a permissive temperature for Ni2519 TP), (ii) acquire a certain level of resistance to TMV U1 accumulation when maintained at 33 °C (a non-permissive temperature for Ni2519 TP) and (iii) lose the resistance to wild-type TMV after their transfer from 33 °C to 24 °C. It is suggested that reversible temperature-dependent conformational changes in Ni2519 TP are responsible for these phenomena and that production of a TP which is only partially functional in transgenic plants confers on these plants a resistance to the virus owing to reduction of the level of cell-to-cell transport. Transgenic tobacco plants producing the 32K TP of brome mosaic virus (BMV) acquired resistance to TMV U1 suggesting that BMV TP is partially functional in tobacco plants.

The first predicted polypeptide encoded by the potyvirus tobacco vein mottling virus (TVMV) is a highly positively charged protein of predicted M r 29K that functions as a protease to perform the first predicted cleavage in the potyvirus polyprotein. We expressed this protein (P1pro) fused with glutathione S-transferase (GST) and purified the fusion protein from engineered Escherichia coli. We found that the intact fusion protein, as well as samples in which the P1pro portion was liberated from GST by pretreatment with thrombin, was able to bind RNA. Binding activity was optimal at relatively high KCl concentrations, suggesting an interaction dependent on a specific protein structure and not just on the binding of the negatively charged phosphate backbone by the positively charged P1pro polypeptide. The TVMV P1pro preferred ssRNA over DNA or dsRNA, and showed a possible preference for sequences containing oligo(G) tracts. Like other potyvirus-encoded proteins, the TVMV P1pro therefore possesses more than one demonstrable biochemical activity and probably plays multiple roles in the TVMV life cycle.

The nucleotide sequence of a satellite RNA associated with a strawberry isolate (H) of strawberry latent ringspot nepovirus (SLRSV) was determined from cDNA copies and the 5′ end sequence was deduced from directly sequenced virion RNA. At the 3′ end a poly(A) sequence was identified. A long open reading frame encoding a polypeptide of 331 amino acids (M r 36488) was determined. Sequence comparisons showed that SLRSV satellite RNA has no extensive homology with other sequences in the GenEmbl and Swiss-Prot databases.

The 16.7 kbp dsRNA specific to the ‘447’ cytoplasmic male sterility (CMS) line of Vicia faba was labelled in vitro with [α-32P]ATP and poly(A) polymerase, and by T4 RNA ligase-mediated addition of [32P]pCp. Analysis of the reaction products under denaturing conditions revealed in both cases extensive labelling of a 4.5 kb ssRNA, already detected in previous experiments in which the RNA-dependent RNA polymerase associated with the dsRNA was allowed to pursue RNA synthesis on preinitiated complexes. Mobility shift analysis of total pCp-labelled dsRNA revealed not two but three different 3′ termini. The most prominent sequencing pattern corresponded to the 4.5 kb ssRNA, indicating that this RNA species has a preferentially accessible, free 3′ OH extremity. Northern blot analysis of the denatured dsRNA confirmed that the 4.5 kb ssRNA is a subgenomic mRNA and detected its counterpart of about 12 kb. Nearly all 16.7 kbp dsRNA molecules featured an interrupted positive-sense strand, indicating a marked prevalence of transcription over replication complexes. This unusual strategy of transcription by a strand displacement mechanism, following initiation at an internal discontinuity, is compared with that of other dsRNA viruses or defective viruses, and is discussed in relation to the expression of the CMS trait.