An atypical 3'-controller element mediates low-level transcription of the p6 subgenomic mRNA of Citrus tristeza virus

Membrane association of a nonconserved viral protein confers virus ability to extend its host range

Citrus tristeza virus (CTV), the largest and most complex member of the family Closteroviridae, encodes a unique protein, p33, which shows no homology with other known proteins, however, plays an important role in virus pathogenesis. In this study, we examined some of the characteristics of p33. We show that p33 is a membrane-associated protein that is inserted into the membrane via a transmembrane helix formed by hydrophobic amino acid residues at the C-terminal end of the protein. Removal of this transmembrane domain (TMD) dramatically altered the intracellular localization of p33. Moreover, the TMD alone was sufficient to confer membrane localization of an unrelated protein. Finally, a CTV variant that produced a truncated p33 lacking the TMD was unable to infect sour orange, one of the selected virus hosts, which infection requires p33, suggesting that membrane association of p33 is important for the ability of CTV to extend its host range.

Cis-acting RNA elements in positive-strand RNA plant virus genomes

Positive-strand RNA viruses are the most common type of plant virus. Many aspects of the reproductive cycle of this group of viruses have been studied over the years and this has led to the accumulation of a significant amount of insightful information. In particular, the identification and characterization of cis-acting RNA elements within these viral genomes have revealed important roles in many fundamental viral processes such as virus disassembly, translation, genome replication, subgenomic mRNA transcription, and packaging. These functional cis-acting RNA elements include primary sequences, secondary and tertiary structures, as well as long-range RNA-RNA interactions, and they typically function by interacting with viral or host proteins. This review provides a general overview and update on some of the many roles played by cis-acting RNA elements in positive-strand RNA plant viruses.

A personal history of virus-based vector construction

The ability to express foreign genes or to silence endogenous genes in plants has revolutionized both basic and applied plant biology. Virus-based expression systems, in which the foreign mRNA is greatly amplified by virus replication, can produce very high levels of proteins or peptides in leaves and other tissues. Vectors have been available for about 25 years. They are commonplace as laboratory tools, but their initial commercial expectations have not been met for numerous reasons. Yet, applications of viral vectors are still evolving. This chapter focuses on our laboratory's involvement in developing virus-based vectors in plants. We created the first 'add-a-gene' vectors that were capable of replication and movement throughout plants. These vectors were based on tobacco mosaic virus. Through the evolution of several prototypes, stable vectors were developed that produced relatively large amounts of product in plants. Recently, we created similar vectors for citrus trees based on citrus tristeza virus. Even though the citrus vectors were created as laboratory tools for improving the crop, circumstances have changed the applications to protection and therapy of trees in the field.

Exploring the limits of vector construction based on Citrus tristeza virus

We examined the limits of manipulation of the Citrus tristeza virus (CTV) genome for expressing foreign genes in plants. We previously created a vector with a foreign gene cassette inserted between the major and minor coat protein genes, which is position 6 from the 3' terminus. Yet, this virus has 10 3'-genes with several other potential locations for expression of foreign genes. Since genes positioned closer to the 3' terminus tend to be expressed in greater amounts, there were opportunities for producing greater amounts of foreign protein. We found that the virus tolerated insertions of an extra gene in most positions within the 3' region of the genome with substantially increased levels of gene product produced throughout citrus trees. CTV was amazingly tolerant to manipulation resulting in a suite of stable transient expression vectors, each with advantages for specific uses and sizes of foreign genes in citrus trees.

Virus-based transient expression vectors for woody crops: a new frontier for vector design and use

Virus-based expression vectors are commonplace tools for the production of proteins or the induction of RNA silencing in herbaceous plants. This review considers a completely different set of uses for viral vectors in perennial fruit and nut crops, which can be productive for periods of up to 100 years. Viral vectors could be used in the field to modify existing plants. Furthermore, with continually emerging pathogens and pests, viral vectors could express genes to protect the plants or even to treat plants after they become infected. As technologies develop during the life span of these crops, viral vectors can be used for adding new genes as an alternative to pushing up the crop and replanting with transgenic plants. Another value of virus-based vectors is that they add nothing permanently to the environment. This requires that effective and stable viral vectors be developed for specific crops from endemic viruses. Studies using viruses from perennial hosts suggest that these objectives could be accomplished.

Enhancement or attenuation of disease by deletion of genes from Citrus tristeza virus

Stem pitting is a common virus-induced disease of perennial woody plants induced by a range of different viruses. The phenotype results from sporadic areas of the stem in which normal xylem and phloem development is prevented during growth of stems. These alterations interfere with carbohydrate transport, resulting in reduced plant growth and yield. Citrus tristeza virus (CTV), a phloem-limited closterovirus, induces economically important stem-pitting diseases of citrus. CTV has three nonconserved genes (p33, p18, and p13) that are not related to genes of other viruses and that are not required for systemic infection of some species of citrus, which allowed us to examine the effect of deletions of these genes on symptom phenotypes. In the most susceptible experimental host, Citrus macrophylla, the full-length virus causes only very mild stem-pitting symptoms. Surprisingly, we found that certain deletion combinations (p33 and p18 and/or p13) induced greatly increased stem-pitting symptoms, while other combinations (p13 or p13 plus p18) resulted in reduced stem pitting. These results suggest that the stem-pitting phenotype, which is one of more economically important disease phenotypes, can result not from a specific sequence or protein but from a balance between the expression of different viral genes. Unexpectedly, using green fluorescent protein-tagged full-length virus and deletion mutants (CTV9Δp33 and CTV9Δp33Δp18Δp13), the virus was found at pitted areas in abnormal locations outside the normal ring of phloem. Thus, increased stem pitting was associated not only with a prevention of xylem production but also with a proliferation of cells that supported viral replication, suggesting that at random areas of stems the virus can elicit changes in cellular differentiation and development.

3'-coterminal subgenomic RNAs and putative cis-acting elements of Grapevine leafroll-associated virus 3 reveals 'unique' features of gene expression strategy in the genus Ampelovirus

BACKGROUND

The family Closteroviridae comprises genera with monopartite genomes, Closterovirus and Ampelovirus, and with bipartite and tripartite genomes, Crinivirus. By contrast to closteroviruses in the genera Closterovirus and Crinivirus, much less is known about the molecular biology of viruses in the genus Ampelovirus, although they cause serious diseases in agriculturally important perennial crops like grapevines, pineapple, cherries and plums.

RESULTS

The gene expression and cis-acting elements of Grapevine leafroll-associated virus 3 (GLRaV-3; genus Ampelovirus) was examined and compared to that of other members of the family Closteroviridae. Six putative 3'-coterminal subgenomic (sg) RNAs were abundantly present in grapevine (Vitis vinifera) infected with GLRaV-3. The sgRNAs for coat protein (CP), p21, p20A and p20B were confirmed using gene-specific riboprobes in Northern blot analysis. The 5'-termini of sgRNAs specific to CP, p21, p20A and p20B were mapped in the 18,498 nucleotide (nt) virus genome and their leader sequences determined to be 48, 23, 95 and 125 nt, respectively. No conserved motifs were found around the transcription start site or in the leader sequence of these sgRNAs. The predicted secondary structure analysis of sequences around the start site failed to reveal any conserved motifs among the four sgRNAs. The GLRaV-3 isolate from Washington had a 737 nt long 5' nontranslated region (NTR) with a tandem repeat of 65 nt sequence and differed in sequence and predicted secondary structure with a South Africa isolate. Comparison of the dissimilar sequences of the 5'NTRs did not reveal any common predicted structures. The 3'NTR was shorter and more conserved. The lack of similarity among the cis-acting elements of the diverse viruses in the family Closteroviridae is another measure of the complexity of their evolution.

CONCLUSIONS

The results indicate that transcription regulation of GLRaV-3 sgRNAs appears to be different from members of the genus Closterovirus. An analysis of the genome sequence confirmed that GLRaV-3 has an unusually long 5'NTR of 737 nt compared to other monopartite members of the family Closteroviridae, with distinct differences in the sequence and predicted secondary structure when compared to the corresponding region of the GLRaV-3 isolate from South Africa.

Three genes of Citrus tristeza virus are dispensable for infection and movement throughout some varieties of citrus trees

Citrus tristeza virus (CTV), a member of the Closteroviridae, possesses a 19.3-kb positive-stranded RNA genome that is organized into twelve open reading frames (ORFs). The CTV genome contains two sets of conserved genes, which are characteristic of this virus group, the replication gene block (ORF 1a and 1b) and the quintuple gene block (p6, HSP70 h, p61, CPm, and CP). With the exception of the p6 gene, they are required for replication and virion assembly. CTV contains five additional genes, p33, p18, p13, p20 and p23, in the 3' half of the genome, some of which (p33, p18 and p13) are not conserved among other members of this virus group, and have been proposed to have evolved for specific interactions with the citrus host. In the present study, the requirements for systemic infection of citrus trees of p33, p6, p18, p13 and p20 were examined. Viral mutants with a deletion in the p6 or the p20 ORF failed to infect citrus plants systemically, suggesting their possible roles in virus translocation/systemic infection. However, we found that deletions within the p33, p18 or p13 ORF individually resulted in no significant loss of ability of the virus to infect, multiply, and spread throughout citrus trees. Furthermore, deletions in the p33, p18 and p13 genes in all possible combinations including deletions in all three genes allowed the virus to systemically invade citrus trees. Green fluorescent protein-tagged CTV variants with deletions in the p33 ORF or the p33, p18 and p13 ORFs demonstrated that the movement and distribution of these deletion mutants were similar to that of the wild-type virus.

A stable RNA virus-based vector for citrus trees

Virus-based vectors are important tools in plant molecular biology and plant genomics. A number of vectors based on viruses that infect herbaceous plants are in use for expression or silencing of genes in plants as well as screening unknown sequences for function. Yet there is a need for useful virus-based vectors for woody plants, which demand much greater stability because of the longer time required for systemic infection and analysis. We examined several strategies to develop a Citrus tristeza virus (CTV)-based vector for transient expression of foreign genes in citrus trees using a green fluorescent protein (GFP) as a reporter. These strategies included substitution of the p13 open reading frame (ORF) by the ORF of GFP, construction of a self-processing fusion of GFP in-frame with the major coat protein (CP), or expression of the GFP ORF as an extra gene from a subgenomic (sg) mRNA controlled either by a duplicated CTV CP sgRNA controller element (CE) or an introduced heterologous CE of Beet yellows virus. Engineered vector constructs were examined for replication, encapsidation, GFP expression during multiple passages in protoplasts, and for their ability to infect, move, express GFP, and be maintained in citrus plants. The most successful vectors based on the 'add-a-gene' strategy have been unusually stable, continuing to produce GFP fluorescence after more than 4 years in citrus trees.

Completion of the mapping of transcription start sites for the five-gene block subgenomic RNAs of Beet yellows Closterovirus and identification of putative subgenomic promoters

In the positive-sense RNA genome of Beet yellows Closterovirus (BYV), the 3'-terminal open reading frames (ORFs) 2-8 are expressed as a nested set of subgenomic (sg) RNAs. ORFs 2-6, coding for the structural and movement proteins, form a 'five-gene block' conserved in closteroviruses. We mapped the 5'-end of the ORF 4 sgRNA, which encodes the p64 protein, at adenosine-11169 in the BYV genome. This completes the mapping of the transcription start sites for the five-gene block sgRNAs of BYV. Computer-assisted analysis of the sequences upstream of BYV ORFs 2, 3, 4, 5, and 6 revealed two conserved motifs, which might constitute the subgenomic promoter elements. These motifs are conserved in the equivalent positions upstream of three orthologous genes of Citrus tristeza Closterovirus and two orthologous genes of Beet yellow stunt Closterovirus.

Evolutionary analysis of genetic variation observed in citrus tristeza virus (CTV) after host passage

We have studied the genetic variability in two genes (p18 and p20) from two groups of Citrus tristeza virus (CTV) isolates. One group (isolates T385, T317, T318, and T305) was derived from a Spanish source by successive host passages while the other (isolates T388 and T390) was obtained after aphid transmission from a Japanese source. A total of 274 sequences were obtained for gene p18 and 451 for p20. In the corresponding phylogenetic trees, sequences derived from the severe isolates (T318, T305, and T388) clustered together and separately from those derived from mild or moderate isolates (T385, T317, and T390), regardless of their geographic origin. Hierarchical analyses of molecular variance showed that up to 53% of the total genetic variability in p18 and up to 87% of the variation in p20 could be explained by differences in the pathogenicity features of the isolates. Neutrality tests revealed that different selection forces had been acting between isolates and between genes, with purifying selection being suggested for p18 from isolates T385 and T390 and for p20 from isolates T385, T317, and T388, and balancing selection for p18 from isolates T318, T305, and T388 and for p20 from isolates T318 and T390. Furthermore, several models of codon selection were observed, with purifying selection being the most notable one, compatible with low effective population size of the virus populations resulting from transmission bottlenecks. We found no evidence of recombination playing a significant role during p18 and p20 evolution in these isolates. These results suggest that hosts can be an important evolutionary factor for CTV isolates.

Variations in two gene sequences of Citrus tristeza virus after host passage

We estimated genetic variation in two groups of Citrus tristeza virus (CTV) isolates: one of them (isolates T385, T317, T318 and T305) derived from a Spanish source by successive host passages, and the other (isolates T388 and T390), obtained after aphid transmission of a Japanese source. The population structure of these isolates had been characterized by single-strand conformation polymorphism analysis of genes p18 and p20. The nucleotide sequences of representative haplotypes of each isolate and gene were used to estimate genetic diversity within and between isolates and to evaluate genetic differentiation between populations. Phylogenetic analysis of p18 and p20 sequence variants showed two main groups: one them included variants predominant in the severe isolates (T318, T305 and T388), and the other comprised variants present in both mild (T385, T317) and severe isolates. Most sequence variants of isolate T390 were not associated to these groups. In some isolates, within-isolate diversity was higher than diversity with other isolates because their population contained distantly related sequence variants, some of which were genetically close to variants predominant in the second isolate. Isolates T388 and T390 were genetically different for the two genes, as estimated by the F statistic. Furthermore, genetic differentiation between T385 and T317, T318 and T305 increased after each host passage. Our results suggest that aphid transmission and host passage may significantly alter the composition of CTV populations and thus be an important factor in their evolution.