Genomic organization and expression of Belladonna mottle virus-iowa (Renamed physalis mottle virus)

Abstract

A majority of plant viruses have positive-sense RNA as their genome. The positive-sense RNA viruses show a wide variation in the organization and expression of their genomes. As the present thesis is concerned with the organization and expression of a positive-sense RNA virus, BDMV(I), Chapter 1 describes in some detail the various modes of organization and strategies of expression of different groups of positive-sense RNA plant viruses. These viruses are classified as monopartite, bipartite, and tripartite based on the presence of 1, 2, or 3 RNA segments as their genomes (Fig. 1). In the case of viruses (multipartite) with more than one RNA segment, the simultaneous presence of all the segments is a prerequisite for viral infection.

CPMV was chosen as an example to discuss the genomic organization of a bipartite virus (Fig. 3). This virus adopted strategies of a divided genome and polyprotein processing for the expression of its genes. The genomic organization of CPMV was similar to that of poliovirus, a picornavirus.

BMV, a tripartite virus, on the other hand, used the strategy of a divided genome and synthesis of subgenomic RNA for the expression of its genes (Fig. 4). The genomic and subgenomic RNAs had a 5’ cap structure and a 3' tRNA-like structure. Mutational analysis of the 3' tRNA-like structure enabled the accurate mapping of the promoter region involved in replication.

The monopartite viruses comprise the Sobemo-, Tobamo-, Luteo-, Carmo-, Potex-, Poty-, Tombus-, and Tymovirus groups (Figs. 6–13). Sobemoviruses adopted strategies of polyprotein processing and subgenomic RNA production, while the Tobamo-, Carmo-, and Tombusviruses used subgenomic RNA and read-through of a leaky termination codon. On the other hand, the Potyviruses used only polyprotein processing to generate functional viral proteins, while the Potexviruses adopted the strategy of only subgenomic RNA production. Luteoviruses had a unique way of expressing their replicase gene, in that there was frame shifting during translation.

The genomes of tymoviruses contained three ORFs, two of which overlap at the 5’ terminus in different reading frames (Fig. 13). The tymoviruses followed strategies of polyprotein processing, subgenomic RNA production, read-through, as well as the use of overlapping reading frames for the synthesis of viral proteins. This is the only group of viruses where such an extensive overlap of two genes occurs, the expression of which was confirmed by in vitro and in vivo experiments.

The literature on BDMV(I), a member of the tymovirus group and the subject of this thesis, was discussed in some detail. It was evident from the survey of literature that the genome size of ssRNA plant viruses ranged from 4–9 kb. The genomes were compact, necessitating maximum utilization of the available coding information. The order of arrangement of genes was also common to several positive-sense RNA viruses, in that the RP and CP genes were 5’ and 3’ proximal, respectively, to ensure the correct order of availability of protein products necessary for viral replication. The regulatory sequences were most often located within the short untranslated regions at the 5' and/or 3' regions of the genomic RNA. These regulatory sequences contained "promoter" elements involved in the initiation of the synthesis of +ve and -ve strand RNAs during replication (Fig. 5). The sequences involved in the modulation of translation of viral RNA were generally present at the 5' region. In all these studies, in recent years, site-directed mutagenesis has provided confirmatory evidence for these postulations.

The analysis of the nucleotide sequence of several plant viruses enabled their classification into three superfamilies exhibiting some common features with animal viruses: (i) picorna-like viruses (Fig. 14), (ii) sindbis-like supergroup (Fig. 15), and (iii) Luteo-like virus group (Fig. 16).

The smaller size of the genome and the earlier work in this laboratory prompted the selection of BDMV(I) for a study of its genomic organization and expression. The objectives of the present investigations were: (i) to elucidate the strategies employed by BDMV(I) to express its genomic RNA into CP and non-structural proteins; (ii) to determine the nucleotide sequence at the 3' region of the genomic RNA; (iii) to analyze the sequence information for the presence of ORFs and possible regulatory signals; (iv) to compare the BDMV(I) nucleotide sequence with sequences of other tymoviruses to discern evolutionary relationships; and (v) to understand the role of the 3' untranslated region of the genome in viral replication.

The materials and methods used in this study, such as isolation and characterization of BDMV(I) from infected N. glutinosa plants (Figs. 17–19), procedures for in vitro translation, preparation of the cDNA library in pDC18 and using the library to screen for clones generated from the 3' terminal region, sequencing of the clones by direct plasmid sequencing, analysis of the sequence data using GCG software, aminoacylation of BDMV(I) RNA, partial purification of the replicase, and its interaction with a transcript derived from a clone corresponding to the 3' terminal region of the genomic RNA by gel retardation assays, are described in Chapter 2.

BDMV(I) was purified from infected N. glutinosa using solvent extraction, centrifugation, PEG precipitation, and differential centrifugation. The purified virus consisted of two types of particles, which were separated by density gradient centrifugation into B- and T-components, corresponding to intact viral particles and empty capsids, respectively. RNA was isolated from both the B- and T-components by phenol/chloroform extraction. The RNA from the B-component upon agarose gel electrophoresis gave a single band with Mr of 2×10^6, while that from the T-component gave several discrete bands with Mr ranging from 0.2–2×10^6 (Fig. 23).

In vitro translation studies were carried out using the rabbit reticulocyte lysate system on the B- and T-component RNAs to understand the strategies of their expression. The conditions for optimal translation, especially the concentration of K+ and Mg2+, were established (Table 5). The genomic RNA from the B-component yielded proteins with Mr 210, 150, and 78 K (Fig. 25). Probing of these translation products with polyclonal antiserum raised against the CP failed to show the presence of the CP.

The time course of translation of the B-component RNA indicated that the 150 and 78 K proteins could arise from the processing of the 210 K polyprotein (Fig. 26). The 78 K protein could also arise from the translation of an overlapping ORF, in analogy with the observations in other tymoviruses.

The translation of the T-component RNA of BDMV(I) yielded several products, one of which reacted with the specific polyclonal antiserum raised against the CP (Fig. 27), indicating the presence of CP mRNA in the T-component. The translation of this RNA isolated from infected capsicum leaves also showed the presence of CP mRNA (Fig. 28). These results indicated that the subgenomic mRNA for CP was preferentially encapsidated in the T-component of BDMV(I).

A model to explain the possible mechanisms for the expression of BDMV(I) genome was suggested (Fig. 29). The salient features of this model are that the BDMV(I) genomic RNA is translated into a polyprotein, which is further processed to yield two products, one of which could be the putative replicase. The CP is expressed via subgenomic mRNA. In general, tymoviruses seem to follow common strategies of gene expression, such as polyprotein processing, use of subgenomic RNAs, and overlapping reading frames.

A cDNA library of BDMV(I) genomic RNA was constructed in pBC18 (Figs. 30–32). The library was screened with end-labeled and reverse transcribed genomic RNA as probes to pick up BDMV(I)-specific clones (Fig. 33). The positive clones were analyzed by restriction analysis (Figs. 34–36).

Many of the recombinant clones obtained from the library made using EcoRI-linkered cDNA did not release the inserts. An analysis of one such clone, pUT76, indicated the loss of an EcoRI site distal to the MCS (Fig. 36). The sequence of this clone matched the corresponding sequence of the RP gene of EMV, a related tymovirus.

Restriction analysis of the positive clones obtained from the library made using blunt-ended cDNA gave inserts of sizes ranging from 0.2–2.0 kb (Fig. 41). Several of these clones were sequenced by the direct plasmid sequencing method, and a few typical autoradiograms of sequencing gels were presented (Figs. 37 & 43). A contiguous sequence of 1255 nt corresponding to the 3' terminal region of the genomic RNA was obtained, in addition to several stretches of sequences which, upon comparison with the EMV sequence, mapped to the 5’ and internal regions of the RP gene (Figs. 44–46).

The nucleotide sequence of the 3' terminal 1255 nt was analyzed using the GCG software package to identify the presence of ORFs, non-coding regions, and regions that could form secondary structures. This analysis revealed a complete ORF corresponding to the CP and a partial ORF for the C-terminal 178 residues of the RP (Fig. 47). Downstream of the CP gene, a non-coding region was identified which could be folded into a stem-loop structure.

The sequence corresponding to the C-terminal 178 residues of RP was compared with the corresponding regions in the RPs of other tymoviruses (Fig. 48). This comparison revealed a high degree of sequence homology among the RP sequences of tymoviruses. Within the coding region of the RP, upstream of the CP ORF, a highly conserved stretch of 17 nucleotides (Tymobox), observed in other tymoviruses, was also identified in BDMV(I) (Fig. 49). The functional importance of this region was discussed.

The deduced CP sequence of BDMV(I) was identical to that obtained earlier by direct protein sequencing, except for an extra dipeptide in the latter. A comparison of the CP sequences of nine tymoviruses, including BDMV(I), revealed that the overall homology was less than that observed when the RPs were compared (Fig. 50). Another unusual feature noticed in this comparison was the invariant presence of glycines at several positions in the sequence, although this was not an abundant amino acid in the protein (Fig. 50). These glycines might be present at interstrand regions and hence important for maintaining the tertiary structure.

It was observed that BDMV(I) CP lacks a basic amino terminal arm, which was invoked for stabilizing the protein-RNA interactions in the case of several other viruses. Polyamines, which altered the spectral properties of the RNA, probably replaced the function of the basic amino terminal arm (Fig. 51).

An expressing immunoscreening clone of BDMV(I) was identified by Western blotting and IPTG induction experiments (Figs. 52–54). Sequencing of this clone showed that the protein was truncated at the N-terminus by 21 amino acids and was in-frame with the lacZ gene of the vector. This truncated protein could be used in future studies aimed at establishing the role of the N-terminal region of CP in the assembly of BDMV(I).

The analysis of the 3' terminal noncoding region showed that the terminal 80 nt could be folded into a characteristic tRNA-like structure (Fig. 55). Within this structure, a sequence capable of forming a "pseudoknot" was seen as part of the aminoacyl acceptor arm (Fig. 56). The alignment of this region with corresponding regions of other tymoviruses revealed that the loops were better conserved than the base-paired regions. One possible reason for the high degree of conservation of the loops could be their involvement in interactions with viral/host proteins participating in viral replication.

A phylogenetic tree to position BDMV(I) among tymoviruses was constructed based on aligned sequences of CPs and the C-terminal 178 residues of RPs. This construct was made using a comparison matrix procedure (Table 8). The trees obtained using either CP or RP sequences had similar topology. The tree (Fig. 58) showed that BDMV(I) was closer to EMV than BDMV(E), regarded as a related strain of the same virus. In view of this observation, it is suggested that BDMV(I) be renamed Physalis mottle virus (PhMV).

The functional importance of the 3' terminal region of the genomic RNA was studied by carrying out aminoacylation experiments (Table 9) and the interaction of a partially purified replicase with a labeled transcript obtained from a clone corresponding to the 3' terminal 240 nt of the genomic RNA.

The aminoacylation experiments showed that the tRNA-like structure was capable of accepting valine in preference to other amino acids (Table 9). The physiological significance of this structure capable of acylation by valine is not yet clear.

The viral replicase was identified and solubilized from infected N. glutinosa leaves (Table 10 and Fig. 59). The product of the replicase reaction had an Mr identical to the expected size of the genomic RNA (Fig. 59), indicating that the replicase was capable of synthesizing a full-length genomic RNA.

The interaction of the replicase with a transcript corresponding to the 3' terminal portion of the genomic RNA was examined by first constructing a subclone of TA42 in pBluescript (Figs. 60, 61). The radiolabeled transcript was characterized (Fig. 62) and its binding to the replicase was monitored by gel retardation assays (Figs. 63, 64). The ability of this transcript to inhibit replicase activity was additional evidence for the specific nature of the interactions (Table 11).

The results described in this thesis showed that BDMV(I) followed the strategy of polyprotein processing and subgenomic RNA production for the expression of its proteins. The 3’ terminal genomic RNA sequence determined in this study revealed the presence of a non-coding region at the 3' end, possibly involved in regulation of replication, the CP gene, and the C-terminal 178 residues of the RP gene. A quantitative comparison of this sequence with those of other tymoviruses indicated that BDMV(I) is not a strain of BDMV(E) but a distinct tymovirus, now renamed Physalis mottle virus (PhMV).