Comparison of the 5' and 3' termini of tomato ringspot virus RNA1 and RNA2: evidence for RNA recombination.

The sequences of the 5' terminal 1140 and 3' terminal 1546 nt of tomato ringspot virus (TomRSV) RNA1 have been determined. These sequences share a high degree of nucleotide sequence similarity with the previously determined TomRSV RNA2 sequence. Eighty-eight percent of the 5' terminal 907 nt of TomRSV RNA1 and RNA2 contain identical nucleotide residues; the first 459 nt are identical at all positions, whereas the next 447 nt are identical at only 75.8% of the nucleotide positions. The region of similarity includes not only the 5' nontranslated leader but also sequence probably encoding polyproteins. The 3' terminal 1533 nt of TomRSV RNA1 and RNA2 are identical and are noncoding. The sequences common to RNA1 and RNA2 account for almost 35% of the total genomic sequence. It is possible that the similar sequences at both ends of TomRSV RNA1 and RNA2 are a result of recombination between these two genomic RNA components.

Nucleotide sequence of tomato ringspot virus RNA-2.

The sequence of tomato ringspot virus (TomRSV) RNA-2 has been determined. It is 7273 nucleotides in length excluding the 3' poly(A) tail and contains a single long open reading frame (ORF) of 5646 nucleotides in the positive sense beginning at position 78 and terminating at position 5723. A second in-frame AUG at position 441 is in a more favourable context for initiation of translation and may act as a site for initiation of translation. The TomRSV RNA-2 3' noncoding region is 1550 nucleotides in length. The coat protein is located in the C-terminal region of the large polypeptide and shows significant but limited amino acid sequence similarity to the putative coat proteins of the nepoviruses tomato black ring (TBRV), Hungarian grapevine chrome mosaic (GCMV) and grapevine fanleaf (GFLV). Comparisons of the coding and non-coding regions of TomRSV RNA-2 and the RNA components of TBRV, GCMV, GFLV and the comovirus cowpea mosaic virus revealed significant similarity for over 300 amino acids between the coding region immediately to the N-terminal side of the putative coat proteins of TomRSV and GFLV; very little similarity could be detected among the non-coding regions of TomRSV and any of these viruses.

Transgenic Nicotiana debneyii expressing viral coat protein are resistant to potato virus S infection.

The coat protein gene from potato virus S (PVS) was introduced into Nicotiana debneyii by leaf disc transformation using Agrobacterium tumefaciens. Transgenic plants expressing the viral coat protein were highly resistant to subsequent infection by the ME strain of PVS as indicated by an absence of symptom development and a lack of accumulation of virus in both the inoculated and upper leaves. As in reported experiments with plants expressing potato virus X coat protein, plants expressing PVS coat protein were also protected from inoculation with PVS RNA. These results provide further evidence that coat protein-mediated protection for these two groups of viruses, which share similar genome organizations, may involve inhibition of some early event in infection other than or in addition to virus uncoating.

Coat protein of melon necrotic spot carmovirus is more similar to those of tombusviruses than those of carmoviruses.

Complementary DNA copies of the genomic RNA of melon necrotic spot virus (MNSV) have been cloned and the region deduced to encode the coat protein has been sequenced. The putative coat protein coding region, located near the 3' end of the genome, consists of 1170 nucleotides and has the potential to encode a 390 amino acid protein of Mr 41,840. Our data show that although MNSV is a carmovirus, its coat protein more closely resembles those of the tombusviruses than those of the carmoviruses sequenced to date, in both the extent of sequence similarity and in the length of the random/arm and protruding domains of the coat protein. Furthermore, dot matrix comparisons revealed sequence similarity between the coat protein protruding domains of MNSV and the cucumber necrosis tombusvirus. This similarity may be involved in one or more of the biological properties these two viruses share, such as the ability to infect cucumbers naturally and to be transmitted by the soil-inhabiting fungus Olpidium radicale.

Organization and interviral homologies of the 3'-terminal portion of potato virus S RNA.

The sequence of 3553 nucleotides corresponding to the 3'-terminal region of potato virus S (PVS) has been determined from cloned cDNA. The sequence obtained contains six open reading frames (ORFs) encoding proteins of Mr 10,734, Mr 32,515, Mr 7,222, Mr 11,802, Mr 25,092 and at least Mr 41,052. The sequence of the 33K ORF has been confirmed to be that of the viral coat protein gene. The nucleotide sequence of this ORF was obtained from plasmids which were isolated by colony hybridization with a specific monoclonal antibody to PVS, and the expression of coat protein fusion products was verified by Western blots of bacterial cell lystates. The deduced amino acid sequence of a 70 amino acid portion from the central region of the PVS coat protein was 59% identical to the analogous region of potato virus X. In addition, the 7K, 12K and 25K ORFs displayed significant sequence homology with the similarly sized ORFs from a number of potexviruses. The partial 41K ORF product was homologous with the C-terminal portion of the viral replicase proteins of potato virus X and white clover mosaic virus.

Complete nucleotide sequence of the cucumber necrosis virus genome.

The complete nucleotide sequence of the cucumber necrosis virus (CNV) genome has been determined. The genome is 4701 nucleotides in length and contains five long open reading frames (ORF). ORF1 begins at the first AUG codon at the 5' terminus and terminates at an amber codon. The predicted molecular weight of the polyprotein encoded by ORF1 is 33 kilodaltons (kDa). Readthrough of the ORF1 amber codon would yield a protein with a molecular weight of 92 kDa. Comparison of the amino acid sequence of the 92-kDa protein with the putative replicases of carnation mottle virus (CarMV) and barley yellow dwarf virus (BYDV) shows extensive sequence similarity. This suggests that the CNV 92-kDa protein is the viral replicase and, furthermore, suggests a close evolutionary relationship between CNV, CarMV, and BYDV, members of the Tombus-, Carmo-, and Luteovirus groups, respectively. Immediately following the 92-kDa protein is ORF3 which can encode a 40-kDa protein. It is identified as the coat protein based on its similarity in amino acid composition to the previously determined CNV coat protein sequence (J. H. Tremaine, 1972, Virology 48, 582-590) and on its amino acid sequence similarity with the tomato bushy stunt virus coat protein. Two nested ORFs (ORF4 and -5), in different frames, follow the coat protein gene. Although it is not known if both ORFs are expressed, they would encode proteins with predicted molecular weights of 21 and 20 kDa, respectively.

Determination of affinity constants and the number of binding sites of monoclonal antibodies specific for southern bean mosaic virus.

Values of the equilibrium dissociation constant and the number of attachment sites for the binding of radiolabeled F(AB) fragments prepared from four monoclonal antibodies to southern bean mosaic virus have been determined. Monoclonal antibodies B6 and B10 produced against the bean type strain of southern bean mosaic virus (SBMV-B) and 1C4 and 1D6 produced against the cowpea type strain (SBMV-C) were more reactive with the untreated virus than with EDTA-swollen particles. B6 and B10 reacted with 180 and 90 sites, respectively, on the SBMV-B particle and were not reactive with SBMV-C. Antibody 1D6 reacted with 180 sites on both SBMV-B and SBMV-C, whereas 1C4 bound to 180 sites on SBMV-C and 60 sites on SBMV-B. The equilibrium dissociation constants for B10, 1C4, and 1D6 were similar but antibody B6 bound with an 8- to 10-fold lower affinity. Solid phase competitive antibody inhibition analysis showed that 1C4 and 1D6 were mutually inhibitory and that these antibodies also inhibited the binding of B6 and B10. Antibody B6 partially inhibited the binding of 1D6 but did not affect 1C4 or B10 binding. No inhibition of B6, 1C4, or 1D6 binding was observed when B10 was the competing antibody. Possible sites of reaction of these monoclonal antibodies on the SBMV particle are discussed.

Monoclonal antibodies as structural probes of southern bean mosaic virus.

The reactivity of six monoclonal antibodies with native southern bean mosaic virus particles, swollen virus, swollen virus contracted by divalent cations or by pH adjustment to 5.0, and virus coat protein at pH 7.5 and 5.0 were assessed by antigen inhibition enzyme-linked immunosorbent assays. Clones B4, B7, and B11 did not react with native virus but were inhibited by low levels of all nonnative virus antigens. B6 had a high reactivity only with the native virus. B5 and B10 reacted equally with native virus and swollen virus contracted by Ca2+ or by pH adjustment, but reacted only weakly with swollen virus or swollen virus contracted by Mg2+. B5 reacted with viral protein at pH 5.0 and weakly with coat protein at pH 7.5, whereas B10 did not react with protein at pH 7.5 and had only limited reactivity with viral protein at pH 5.0. The reactivity of B5 and B10 with swollen virus increased as the pH was gradually decreased between 7.3 and 6.5. Similarly, their reactivity with swollen-contracted virus decreased as the pH was increased within the same range. Reactivity versus pH plots of these two transition forms showed hysteresis with B10 but did not with B5.

Amino acid sequence of southern bean mosaic virus coat protein and its relation to the three-dimensional structure of the virus.

The amino acid sequence of the coat protein of southern bean mosaic virus has been determined. Chemical techniques were supplemented by a 2.8-A-resolution electron-density map which helped in assigning absolute positions for each peptide. Four amino acids and the placement of one heptapeptide near the amino terminus of the protein were interpolated by comparison with the partially known southern bean mosaic virus RNA sequence. The subunit is heavily lined with basic residues on the side facing the RNA. Subunit-subunit interactions are hydrophobic and ionic. The calcium site on the quasi-threefold axis has three glutamic residues as ligands. There is a disulfide bond, between Cys 168 and 176, within a sequence which is inserted in southern bean mosaic virus relative to the tomato bushy stunt virus structure. This portion is a helix nestling between the "beta-barrel" subunit structure and the RNA interior.

Intermediates of the in vitro assembly and disassembly of southern bean mosaic virus.

Southern bean mosaic virus (SBMV) was swollen by treatment with EDTA at pH 7.5 and dissociated into RNA and protein in 1 M NaCl. Aliquots of this preparation were diluted with appropriate buffers to obtain samples in varying concentrations of NaCl, and components of these samples were sedimented through sucrose solutions and dissolved in 0.01 M Tris-HCI buffer, pH 7.5. The protein content and sedimentation properties of components in these preparations were determined. When the NaCl molarity in the treatment exceeded 0.6 M the preparations contained RNA with approximately six protein subunits per SBMV RNA molecule. The protein content of the preparations increased from 30 protein subunits per RNA molecule to 145 protein subunits per RNA molecule as the NaCl molarity used in the treatment was decreased from 0.5 to 0.1 M. The positions of sedimentation of components in these preparations in density gradient centrifugation were intermediate between those of RNA and EDTA-swollen virus. The sedimentation rate of these assembled components increased as the NaCl molarity used in the treatment was decreased. Similar components were assembled when preparations of RNA and protein dissociated from SBMV by dialysis in neutral buffers containing EDTA and 1 M NaCl were diluted to lower NaCl molarities. When SBMV was swollen by treatment with EDTA and dissociated in various concentrations of NaCl, the components formed were similar to those obtained by assembly in the same NaCl molarities. Preparations in the pH 7.5 buffer contained single components which sedimented at 56 S, 55 S, 54 S, 51 S, 46 S, 38 S, 33 S, and 24 S. With the exception of the 24 S component, components formed by disassembly in the same NaCl molarities and dissolved in pH 5.0 buffer sedimented faster.

A highly basic cyanogen bromide peptide from sowbane mosaic virus protein.

The protein of sowbane mosaic virus was cleaved with cyanogen bromide (CNBr) and a highly basic peptide, sCB-1, was isolated by Sephadex and ion-exchange chromatography. The amino acid composition and sodium dodecyl sulfate-polyacrylamide gel electrophoresis indicated that sCB-1 had 49 amino acid residues. Eighteen of these residues are basic, fifteen are lysine, and three are arginine. sCB-1 moved faster in polyacrylamide gel electrophoresis at pH 2.2 than a similar sized CNBr peptide from the bean strain of southern bean mosaic virus with twelve basic residues. The highly basic nature of sCB-1 suggests that this portion of the SoMV protein sequence is involved in protein-RNA binding in the virus particles.

Chemical and serological properties of a cyanogen bromide peptide of southern bean mosaic virus protein.

Southern bean mosaic virus (SBMV) protein was cleaved with cyanogen bromide and a highly basic peptide, CB-1, was isolated by ion exclusion and ion-exchange chromatography. Twelve peptides were separated from a tryptic digest of CB-1 by ion-exchange chromatography and the composition of these peptides was similar to that of peptides released from EDTA-swollen virus particles by limited tryptic digestion. The composition and N-termini of the tryptic peptides indicated CB-1 was from the N-terminus of SBMV protein and contained 48 amino acid residues. The CB-1 peptide moved rapidly to the cathode in polyacrylamide gel electrophoresis at pH 3.9 and contained nine arginine residues, three lysine residues, and no acidic amino acid residues. It was shown to interact with purified viral RNA, sodium dextran sulfate, and calf thymus DNA. Antiserum to sodium dodecyl sulfate (SDS)-dissociated virus gave a reaction of partial identity between the CB-1 peptide and the SDS-dissociated virus in SDS gel diffusion tests. The CB-1 peptide did not react with antiserum to SDS-dissociated, trypsin-treated virus. Gel diffusion tests conducted in saline agar gels between trypsin-treated virus and SBMV, with SBMV antiserum, did not show differences in their serological properties. Antiserum to the CB-1 peptide conjugated to tomato bushy shunt virus reacted with SBMV but SBMV antiserum did not react with CB-1 or the CB-1-tomato bushy shunt virus conjugate.

A dendrogram of plant viruses.

To facilitate the recognition of plant viruses with similar characteristics a dendrogram of characterized viruses was constructed. The sequence of criteria included: type of nucleic acid; single or double stranded; presence or absence of lipid envelope; helical or nonhelical symmetry; and divided or single genome. Nonhelical RNA viruses with divided genomes were further divided into viruses with one or more than one capsid size. Those with one capsid size were subdivided into viruses with one or more than one sedimenting component. Nonhelical RNA viruses with a single genome were divided according to their RNA size, and their sensitivity to sodium dodecyl sulfate and ethylenediaminetetraacetic acid.