| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Plant Virology Group, ICGEB Biosafety Outstation, Via Piovega 23, 31056 Ca' Tron di Roncade, Italy
2 Molecular Pathology Group, ICGEB, AREA Science Park, Padriciano 99, 34012 Trieste, Italy
3 INRA, Laboratoire de Biologie Cellulaire, UR501, INRA-Versailles, 78026 Versailles cedex, France
Correspondence
Jeremy R. Thompson
thompson{at}icgeb.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Present address: Department of Plant Pathology & Microbiology, University of California Riverside, Riverside, CA 92521, USA. 
Supplementary material is available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
). It can be classified into two main subgroups, I and II, the former being further subdivided into IA and IB. Its genome comprises three genomic RNAs (gRNAs) named RNA1, 2 and 3, and three subgenomic RNAs (sgRNAs) named 4, 4A and 5. RNA4 is derived from RNA3, and contains the coat protein (CP) gene, and RNA4A is derived from RNA2, and contains the 2b gene. All three gRNAs are capped, as are sgRNA4 and 4A. The promoters of these sgRNAs have been mapped in vitro to nt –28 to +15 and –31 to +13 around their initial nt (+1) on the negative strand, respectively (Chen et al., 2000; Sivakumaran et al., 2002). Both the RNA4 and 4A promoter regions contain stem–loop structures 3' of +1 nt. RNA5, on the other hand, consisting of a mix of the approximately 300 3'-terminal nucleotides of RNAs 2 and 3, contains no ORF, and is uncapped (Blanchard et al., 1996). In addition, it has not been observed in subgroup I CMV strains, but only in subgroup II strains and in the related tomato aspermy virus (TAV) (Palukaitis et al., 1992). A putative promoter-like motif has been identified as a 20–23 nt stretch 5' of the first 5' nucleotide (starting at –1 or +1) (Blanchard et al., 1997; Suzuki et al., 2003), based on two main observations: (i) that there is absolute nucleotide identity between it and a motif at +4 to +24 in the mapped subgenomic promoter of the unrelated benyvirus beet necrotic yellow vein virus (BNYVV) (Balmori et al., 1993), and (ii) that this sequence is conserved in strains of CMV subgroup II and in TAV, but not in subgroup I CMV strains. The function(s) of RNA5 are not known. Symptom attenuation has been proposed based on the observation that the presence of more RNA5 in plants results in less severe symptoms (Shi et al., 2007), and it has been suggested that RNA5 could be directly involved in virus assembly and/or replication (Blanchard et al., 1996; Gould et al., 1978).
The putative promoter-like motif has also been shown to be a hot spot for viral RNA recombination both between CMV RNA3 and TAV RNAs 1 and 2 (Suzuki et al., 2003), TAV RNA3 and CMV RNAs 1 and 2 (Shi et al., 2007) and between the RNAs 3 of CMV and TAV (de Wispelaere et al. 2005). For several other viruses, it has also been demonstrated that recombination occurs preferentially at sites of sgRNA initiation or those recognized by viral replicases (Miller & Koev, 1998; Nagy et al., 1999; Nagy & Simon, 1997). In the case of brome mosaic virus (BMV), also a member of the family Bromoviridae, a functional sgRNA promoter is required for recombination to occur in vitro (Dzianott et al., 2001; Wierzchoslawski et al., 2003). However, it should be noted that the CMV sgRNA4 promoter is not a hot spot for recombination (Aaziz & Tepfer, 1999; de Wispelaere et al., 2005).
The aim of the present study was to analyse the putative promoter-like motif of RNA5 and surrounding 3' untranslated region (3'UTR), both at a structural and biological level. Our approach was to begin by determining in vitro the structural elements of the surrounding 3'UTR, both in the plus and minus viral strands. The region was then mutated in order to elucidate, first, whether RNA5 production was only structurally specific or whether sequence played a role as well, and second, to establish which regions of the CMV genome are involved in RNA5 production. All mutational analyses were carried out by inoculating plants with in vitro transcripts, using RNA3 from either a subgroup I strain (I17F-CMV) or subgroup II strain (R-CMV) in combination with RNAs 1 and 2 from I17F-CMV. This was done to ensure that RNA5 was derived only from RNA3, and not from other gRNAs, as has been reported previously (Blanchard et al., 1996; Gould et al., 1978).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
. Symptom development was recorded up to 12 days post-inoculation (days p.i.), after which no changes were observed. All plants were maintained in greenhouse conditions of 25±2 °C, 75±10 % relative humidity with 16 h day length supplemented when necessary. Fresh leaf tissue (100 mg) was taken from the most apical symptomatic leaf at 12 days p.i., and RNA extracted using the RNeasy Plant Mini kit (Qiagen).
Mutagenesis and transcription.
Point mutations, deletions, insertions and substitutions were generated by PCR-mediated mutagenesis (Liu et al., 2002) using Pfu polymerase (Promega) with primers (62, 262, 272, 254, 256, 258, 51, 52, 181, 182, 19, 20, 179 and 180) shown in Supplementary Table S1 (available with the online version of this paper) in combination with primers M13R and R935+ for R-CMV, and primers M13R and I1178+ for I17F. PCR products generated by the various primer combinations were cloned directly into pUC13 derivatives: R-CMV RNA3 (pR3T7) and I17F-CMV RNA3 (Salanki et al., 1997), using PstI/ApaI and PstI/SalI restriction sites, respectively. An alternative method was used for the R-prom primer (Supplementary Table S1), which made use of the NruI/PstI restriction sites in pR3T7 to ligate directly the PCR product generated by the R-prom primer in combination with R935+. I17R was produced by ligating the NruI/PstI fragment of pR3T7 into the ISLMLR mutant. All resulting mutant clones were sequenced to confirm the presence of the introduced mutations and the absence of any spurious nucleotide changes within the amplified fragment. Capped in vitro transcripts from the full-length PstI- or BamHI-linearized clones pI1T7 (RNA1), pI2T7 (RNA2), pI3T7 (RNA3), pR3T7 (RNA3) and all derived RNA3 mutants were synthesized using the mMessage mMachine high yield capped RNA transcription kit (Ambion), supplemented with 50 U T7 RNA polymerase (Invitrogen) per 20 µl reaction.
Reverse transcription PCR (RT-PCR) and sequence analysis.
Reverse transcription was carried out by mixing 5 µl (1–2 µg) RNeasy extract with 1 µl primer 30 or 32 (Supplementary Table S1) (10 µM) and 6 µl water, followed by heating at 70 °C for 10 min. Reverse transcription with Moloney murine leukemia virus reverse transcriptase (M-MLV; Invitrogen) and PCR with Pfu polymerase were done as described by the manufacturers using primers 184 and –2038 (Supplementary Table S1). PCR products were cleaned and sequenced using primer 186 (Supplementary Table S1) by BMR-genomics, Padua, Italy. All subsequent molecular analyses were carried out using the Vector NTI software (Invitrogen), except for RNA structural predictions which were done with Mfold (http://www.bioinfo.rpi.edu/applications/mfold) (Zuker, 2003) and Pfold (http://www.daimi.au.dk/
compbio/pfold/) (Knudsen & Hein, 2003).
RNA blotting.
All RNA samples were quantified using agarose gel electrophoresis alongside known standards using the Quantity One software (Bio-Rad), and loading volumes adjusted accordingly. The R-CMV-specific DNA probe was produced with primers 31 and 32 (Supplementary Table S1) using 10x DIG labelling mix (Roche). The RNA probe was transcribed (T3 polymerase; Promega) from a PCR product generated using primers 54 and 55 (Supplementary Table S1). Separation of 400 ng total plant RNA by agarose gel electrophoresis was followed by capillary transfer to nitrocellulose membrane, with cross-linking on a standard UV transilluminator for 3 min. Hybridization with DIG Easy Hyb and chemiluminescent detection were done as described by the manufacturer (Roche). Optical density of fragments was determined by analysing trace quantities (intensityxmm) after adjusting with rolling disc background subtraction using the Quantity One software (Bio-Rad). Relative amounts of RNAs 4 and 5 were determined by finding the sum of the intensities of RNA3, 4 and 5 for each sample, and then calculating, as a percentage, the contribution to that sum of each RNA. All mutant and wild-type RNA profiles, once verified by sequencing, were analysed for samples extracted from a minimum of three plants. P-values were calculated from an unpaired type 3 t-test method.
Secondary structure determination.
RNA transcripts were generated from a PCR template produced with T7- and T3-tailed primers 96 and 98 (Supplementary Table S1). Transcription mixtures in 100 µl comprised 500 ng DNA template, 20 µl NTPs (10 mM), 20 µl 5x buffer, 10 µl dithiothreitol (0.1M) and either 3 µl T7 polymerase (50 U µl–1; Invitrogen) or 2 µl T3 polymerase (80 U µl–1; Promega). Reaction was incubated for 2 h at 37 °C, verified by gel electrophoresis and stored at –80 °C. Enzymic probing and sequencing of the transcripts was carried out as described previously (Odreman-Macchioli et al., 2000) using primers 97 and 99 (Supplementary Table S1). The RNases used were V1 nuclease (Ambion), T1 nuclease (Ambion) and S1 nuclease (Promega).
Virus quantification.
A virion standard (pseudorecombinant IIR) was purified from infected tobacco by the methods of Lot et al. (1972), and concentrations determined both spectrophotometrically and using the Bradford protein assay (Bio-Rad). Virus quantification was carried out by ELISA according to Jacquemond et al. (1988). Absorbances were measured at 405 nm using a Microplate Reader (model 680) with the Microplate Manager software (Bio-Rad) and a cubic regression method. P-values were calculated from an unpaired type 3 t-test method.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
).
Core putative RNA5 promoter motif contains a conserved stem–loop
A nucleotide alignment of full-length CMV RNA3 sequences from each of subgroups IA, IB and II showed a region that is conserved in subgroup II isolates, though absent in subgroup I isolates, and which contains the 5' terminus of RNA5 at position 1903 (Blanchard et al., 1996) (Fig. 1a and b). Nucleotides identical to the BNYVV promoter (Balmori et al., 1993), and conserved within all subgroup II isolates, defined a ‘core’ 21-mer motif (Box1) in the positive strand of CGUCCGAAGACGUUAAACUAC (nt 1902–1922) (Fig. 1b). Given that sgRNA initiation requires secondary structure (Miller & Koev, 2000), this Box1 sequence was checked in silico for its ability to form secondary structure, and was found to contain a tetramer loop and a stem of 5 or 4 nucleotides in the plus and minus strands, respectively (Fig. 1c). At this stage, equal consideration was taken for both the plus and minus strands, as the mechanism of RNA5 production could involve either internal initiation by the replicase on the negative strand, premature termination on the positive strand (White, 2002), or some other undefined process. In the negative strand, this stem–loop forms a thermodynamically stable UNCG tetraloop, a type that has been shown to be involved in tertiary RNA contacts acting as a nucleation site for folding (Molinaro & Tinoco, 1995; Sarzynska et al., 2003). We then went on to analyse full-length RNA3 sequences on the plus and minus strands using Mfold with standard default settings (Zuker, 2003). Of the 20 most stable structures produced on the plus strand, all contained the stem–loop depicted in Fig. 1(c), while 13 had an extended structure identified subsequently as SLM+. For the minus strand, 13 of the 20 most stable structures obtained had the complementary structure (Fig. 1c).
|
). Structures were identified on both the minus and plus strands from nt 1887 to 1999. For both positive and negative strands, the same digestion patterns were observed in several repetitions of the experiment. The positive strand produced the structure shown in Fig. 2(a), consisting of four stem–loops designated SLJ+, SLK+, SLL+ and SLM+ in accordance with the nomenclature previously applied to the rest of the 3'UTR (Ahlquist et al., 1981). Twenty-eight bases were digested: 12 by V1 nuclease (double-stranded RNA), 4 by T1 nuclease (single-stranded Gs) and 12 by S1 nuclease (single-stranded RNA) (Fig. 2b). All digests were consistent with the predicted structure, except for a weakly T1-digested G in SLM+ at nt 1893, which might be explained by the opposing S1-digested single-stranded U at nt 1920 that resides in an internal bulge region, and a V1-digested U between SLM+ and SLL+ at nt 1925, which could be due to stacking between the stems SLM+ and SLL+. Moving from a 5' to 3' direction (
G values determined in silico): SLM+ is a long stem–loop (G=–10.8) with a 14 bp stem containing large (3 nt) and small (1 nt) asymmetric bulges and a tetramer loop, SLL+ (G=–2.2) has a short (6 nt) stem with a 3 nt asymmetric bulge and a tetramer loop, SLK+ (G=–5.1) contains a 6 nt stem and 10 nt loop, and a 9 nt stem SLJ+ (G=–9.4) has an asymmetric 4 nt bulge and a tetraloop. In SLJ+ both the loop and stem were strongly digested by V1 and T1 nucleases, respectively. RNase probing of the positive strand structure also detected several RTase stops throughout this structure that were conspicuously absent from the structural analysis of the negative strand (see below). RTase-stops are identified by the fact that all lanes (including the control lane) contain a band. These bands, however, are not the results of specific cleavage but are normally generated when the RTase encounters abrupt changes in the RNA secondary structure.
|
). One digest was not consistent with the predicted structure: V1 at A (nt 1890), but can be explained by the fact that this base forms a 1 nt bulge in a well-defined stem–loop. The overall structures identified are similar to those obtained for the plus strand, except for parts of SLM– and SLL–. SLJ– (G=–5.7) has an 8 nt loop (three of which were strongly digested by S1) separated from an asymmetric 3 nt UUA bulge by a 3 bp stem. The base of SLJ– comprises four base pairings, three of which are G-C. Moving in the 3' direction, SLK– (G=–1.3) comprises a weak 5 bp stem and 10 nt loop. SLL– (G=–7.1) is a longer, more stable structure than SLL+, with an asymmetric 2 nt bulge, an 8 nt loop and a 10 bp stem. The stem proximal to the loop contains nucleotides the complement of which in SLL+ is contained within the loop. The length of SLL– means that the configuration of SLM– is not that of a long stem–loop, as in SLM+, but a double stem–loop, the 5' structure (SLMa– G=–3.2) being that of the stem–loop containing Box1 identified in Fig. 1(b)
.
|
Further analysis using the Pfold software with the 3'UTR of cucumoviruses R-CMV, I17F-CMV, peanut stunt virus (PSV) and P-TAV on the plus strand was able to clearly identify structures SLJ+ and SLK+, including compensatory mutations (Supplementary Fig. S1). SLK+ in R-CMV and P-TAV were identical, except for one nucleotide consisting of a 6 nt stem and 10 nt loop, whereas I17F-CMV and PSV SLK+ contained a 7 nt stem (Supplementary Fig. S1b). For SLJ+ all four structures had common features: a four base-paired stem, a UAAA asymmetric bulge and a U-rich tetraloop (Supplementary Fig. S1c). Recently, Shi et al. (2007) identified a recombination hot spot in TAV (corresponding to a G at nt 1968 in R-CMV), which in the in silico structure presented lies in a long 34 nt single-stranded stretch. In the structure determined in our study, this hot spot is found between SLK– and SLJ– (Fig. 3a).
These results, in combination with those previously obtained (Ahlquist et al., 1981; Felden et al., 1994; Joshi et al., 1983; Rietveld et al., 1983), now provide an almost complete map of the putative secondary structure of the plus-strand 3'UTR of CMV RNA3 (Supplementary Fig. S2). There are two main domains: the tRNA-like pseudoknotted structure consisting of the 132 3'-terminal nucleotides (2075–2206), and a linear structure of eight stem–loops (SLF to SLM) incorporating the 184 nucleotides 5' to the tRNA-like structure. Of the linear portion, only SLI has not been demonstrated experimentally. It consists of a 6 nt stem and a hexaloop, the sequence of which is in the most conserved region of the 3'UTR between subgroup I and II strains. Interestingly, all structures can also be predicted for CMV subgroup I strain I17F, except Box1-containing SLM.
RNA5 production has sequence specificity
To determine whether the production of RNA5 was dependent on structure alone or sequence, three mutants were engineered (Fig. 4a). SLMloopIn contained an inverted tetraloop and closing base pair, which was predicted to maintain the SLM hairpin structure. SLM*1 contained two mutations that disrupt the stem. SLMrest contained an additional two mutated nucleotides that restored the disrupted SLM*1 structure, though not the sequence, to wild-type SLM. Plants were inoculated with each mutant, and were screened 12 days p.i. for the engineered mutant sequences by RT-PCR and sequencing; all samples were faithful to the original transcripts inoculated. Sequence conformity of the progeny virus was similarly confirmed in all mutants described in the following sections. The RNA of three plants was then analysed by RNA blotting for each mutant. The results (Fig. 4b and c) showed that, in comparing the relative amounts of each RNA (the proportion of the intensities of RNA4 and 5 relative to total RNA derived from RNA3 – i.e. the sum intensities of RNA3, 4 and 5), RNA4 levels remained statistically constant, while RNA5 levels were significantly depleted in all mutants when compared with the wild-type stoichiometry (P=0.015, 0.025 and 0.024 for SLMloopIn, SLM*1 and SLMrest, respectively).
|
). The substitution mutants SLM*3, SLK* and SLJ* all contained three to four base substitutions that were designed to disrupt the designated stem structure. The substituting sequence was based exactly on the complementary strand of the stem, except for SLM*3 where GAC (nt 1910–1912) were substituted by AUU instead of CUG, because the latter was assessed in silico to form a stable stem with part of the SLM bulge two nucleotides below. Being in the coding region, the mutation in mutant CPSt* was designed to disrupt a stem predicted in silico by introducing silent mutations at nt 1864, 1867 and 1870 (Supplementary Fig. S3). The deletion mutant Rbox1 had a 22 nt deletion from 1900 to 1921 including almost all of Box1, whereas in the SLMintoR mutant, a larger Box1 deletion (nt 1886–1922) was substituted with the equivalent I17F-CMV sequence (nt 1915–1940). The latter mutant was originally engineered because of the assumed deleterious effects of removing 23 nt from the virus, yet surprisingly both infected similarly to wild-type. Results with 12 days p.i. samples showed that all mutants, except SLJ*, significantly affected RNA5 levels relative to all RNA3-derived RNAs, ranging from 12 (SLK*) to 17 % (CPSt*) when compared with pseudorecombinant IIR RNA5 level (51 %) (Fig. 5b and c). Mean RNA4 levels were also affected, though not in a statistically significant manner. For the deletion mutants Rbox1 and SLMintoR, RNA5 was practically eliminated at 1 (P=0.010) and 2 % (P=0.010), respectively. These results therefore identify the 3' limit of those nucleotides required for optimal RNA5 production as occurring between +49 and +80. Clearly, if downstream adjacent structural elements are required for efficient RNA5 production, SLK, but not SLJ, is involved. Upstream of the 5' terminal nucleotide of RNA5, however, the picture is not as clear, with mutant CPSt* yielding significantly reduced levels of RNA5, its most 3' mutated nucleotide being at position –34. Further mutation analyses into the CP region would be required to clarify this point.
|
, ISLMLR in which nt 1929–1961 of I17F-CMV were substituted by nt 1902–1943 of R-CMV, and finally, I17R in which nt 1929–2216 of I17F-CMV were substituted by nt 1902–2206 of R-CMV 3'UTR (Fig. 6a). RNA blot analysis of samples extracted 12 days p.i. from infected plants showed that the Box1 sequence alone (mutant I+box1) was not sufficient to produce detectable amounts of RNA5 (Fig. 6b and c), whereas an extra 21 nt of R-CMV sequence downstream of the Box1 insertion (mutant ISLMLR) was sufficient to produce low, though detectable, amounts of RNA5 (6 % of total RNA3-derived RNA). The additional 21 nt do not fully restore SLL+ and only partially restore SLL–, indicating again that sequence has an important role in RNA5 production. The amount of RNA5 produced could be increased further still with the almost complete 3'UTR of R-CMV (I17R), although it did not reach the same levels as for pseudorecombinant IIR (20 % compared with 47 %, respectively). Interestingly, in the RNA blots, one quarter (100 ng) of the set amount of total RNA was loaded to the well in order to obtain a similar chemiluminescent signal for RNAs 3 and 4 between I17R and the rest of the samples. Significantly, levels of RNA 1 and 2 in both the pseudorecombinant IIR and I17R were reduced to the point of being undetectable in comparison with I17F-CMV, where levels were similar to those of RNA3 and 4. Nevertheless, when the same RNA probe was used on those mutants analysed in Fig. 5 appreciable levels of RNAs 1 and 2, irrespective of the levels of RNA5, were not detected (results not shown).
|
). These variations in viral RNA stoichiometry, therefore, provide a possible function for RNA5 that should be studied further, namely the downregulation of RNAs1 and 2, and to a lesser extent RNA4. For BMV in yeast, blocking sgRNA4 transcription increased replication of RNA3 by up to 350 %, therefore implying that RNA3 replication is somehow inhibited by sgRNA4, and that RNA3 and sgRNA4 mutually interfere with each other (Grdzelishvili et al., 2005).
Absence of RNA5 does not alter virus levels or symptom development
To address the hypothesis that RNA5 is involved in symptom attenuation, two mutants, namely RBox1 and SLM*3, were compared with the pseudorecombinant IIR in an infectivity assay. RBox1 was chosen because of its almost wild-type stoichiometry in the absence of RNA5, while SLM*3 was found in pilot studies to display delayed, often attenuated, symptoms and reduced virus accumulation in the plant. RNA blot analysis also showed that SLM*3 RNA accumulation in the infected leaf was lower than wild-type (Fig. 5b). Five plants were inoculated in three independent assays with each of the three virus types and then scored for appearance of symptoms, after which coat protein levels were measured by ELISA at 12 days p.i. (Fig. 7). Results indicated that, for RBox1, infection was comparable with pseudorecombinant IIR, with mean calculated values of 295 and 332 µg g–1 leaf tissue, respectively. There was also no visible difference between the leaf mosaic symptoms they caused, therefore suggesting that RNA5 is not directly involved in symptom attenuation, contrary to what has been postulated previously by Shi et al. (2007). For SLM*3, on the other hand, appearance of symptoms was delayed and virus accumulation diminished.
|
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Ahlquist, P., Dasgupta, R. & Kaesberg, P. (1981). Near identity of 3' RNA secondary structure in bromoviruses and cucumber mosaic virus. Cell 23, 183–189.[CrossRef][Medline]
Balmori, E., Gilmer, D., Richards, K., Guilley, H. & Jonard, G. (1993). Mapping the promoter for subgenomic RNA synthesis on beet necrotic yellow vein virus RNA 3. Biochimie 75, 517–521.[Medline]
Blanchard, C. L., Boyce, P. M. & Anderson, B. J. (1996). Cucumber mosaic virus RNA 5 is a mixed population derived from the conserved 3'-terminal regions of genomic RNAs 2 and 3. Virology 217, 598–601.[CrossRef][Medline]
Blanchard, C. L., Higgins, T. J. & Anderson, B. J. (1997). RNAs 4A and 5 are present in tomato aspermy virus and both subgroups of cucumber mosaic virus. Arch Virol 142, 1273–1283.[CrossRef][Medline]
Chen, M. H., Roossinck, M. J. & Kao, C. C. (2000). Efficient and specific initiation of subgenomic RNA synthesis by cucumber mosaic virus replicase in vitro requires an upstream RNA stem-loop. J Virol 74, 11201–11209.
de Wispelaere, M., Gaubert, S., Trouilloud, S., Belin, C. & Tepfer, M. (2005). A map of the diversity of RNA3 recombinants appearing in plants infected with cucumber mosaic virus and tomato aspermy virus. Virology 331, 117–127.[CrossRef][Medline]
Dzianott, A., Rauffer-Bruyere, N. & Bujarski, J. J. (2001). Studies on functional interaction between brome mosaic virus replicase proteins during RNA recombination, using combined mutants in vivo and in vitro. Virology 289, 137–149.[CrossRef][Medline]
Eperon, L. P., Graham, I. R., Griffiths, A. D. & Eperon, I. C. (1988). Effects of RNA secondary structure on alternative splicing of pre-mRNA: is folding limited to a region behind the transcribing RNA polymerase? Cell 54, 393–401.[CrossRef][Medline]
Felden, B., Florentz, C., Giege, R. & Westhof, E. (1994). Solution structure of the 3'-end of brome mosaic virus genomic RNAs. Conformational mimicry with canonical tRNAs. J Mol Biol 235, 508–531.[CrossRef][Medline]
Gould, A. R., Palukaitis, P., Symons, R. H. & Mossop, D. W. (1978). Characterization of a satellite RNA associated with cucumber mosaic virus. Virology 84, 443–455.[CrossRef][Medline]
Grdzelishvili, V. Z., Garcia-Ruiz, H., Watanabe, T. & Ahlquist, P. (2005). Mutual interference between genomic RNA replication and subgenomic mRNA transcription in brome mosaic virus. J Virol 79, 1438–1451.
Jacquemond, M., Amselem, J. & Tepfer, M. (1988). A gene coding for a monomeric form of cucumber mosaic virus satellite RNA confers tolerance to CMV. Mol Plant Microbe Interact 1, 311–316.[Medline]
Joshi, R. L., Joshi, S., Chapeville, F. & Haenni, A. L. (1983). tRNA-like structures of plant viral RNAs: conformational requirements for adenylation and aminoacylation. EMBO J 2, 1123–1127.[Medline]
Knudsen, B. & Hein, J. (2003). Pfold: RNA secondary structure prediction using stochastic context-free grammars. Nucleic Acids Res 31, 3423–3428.
Liu, S., He, X., Park, G., Josefsson, C. & Perry, K. L. (2002). A conserved capsid protein surface domain of cucumber mosaic virus is essential for efficient aphid vector transmission. J Virol 76, 9756–9762.
Lot, H., Marrou, J., Quiot, J. B. & Esvan, C. (1972). Contribution à l'étude du virus de la mosaïque du concombre (CMV). I. Méthode de purification rapide du virus. Ann Phytopath 4, 25–38. in French
Miller, W. A. & Koev, G. (1998). Getting a handle on RNA virus recombination. Trends Microbiol 6, 421–423.[CrossRef][Medline]
Miller, W. A. & Koev, G. (2000). Synthesis of subgenomic RNAs by positive-strand RNA viruses. Virology 273, 1–8.[CrossRef][Medline]
Molinaro, M. & Tinoco, I., Jr (1995). Use of ultra stable UNCG tetraloop hairpins to fold RNA structures: thermodynamic and spectroscopic applications. Nucleic Acids Res 23, 3056–3063.
Nagy, P. D. & Simon, A. E. (1997). New insights into the mechanisms of RNA recombination. Virology 235, 1–9.[CrossRef][Medline]
Nagy, P. D., Pogany, J. & Simon, A. E. (1999). RNA elements required for RNA recombination function as replication enhancers in vitro and in vivo in a plus-strand RNA virus. EMBO J 18, 5653–5665.[CrossRef][Medline]
Odreman-Macchioli, F. E., Tisminetzky, S. G., Zotti, M., Baralle, F. E. & Buratti, E. (2000). Influence of correct secondary and tertiary RNA folding on the binding of cellular factors to the HCV IRES. Nucleic Acids Res 28, 875–885.
Palukaitis, P. & Garcia-Arenal, F. (2003). Cucumoviruses. Adv Virus Res 62, 241–323.[Medline]
Palukaitis, P., Roossinck, M. J., Dietzgen, R. G. & Francki, R. I. (1992). Cucumber mosaic virus. Adv Virus Res 41, 281–348.[Medline]
Rietveld, K., Pleij, C. W. & Bosch, L. (1983). Three-dimensional models of the tRNA-like 3' termini of some plant viral RNAs. EMBO J 2, 1079–1085.[Medline]
Salanki, K., Carrere, I., Jacquemond, M., Balazs, E. & Tepfer, M. (1997). Biological properties of pseudorecombinant and recombinant strains created with cucumber mosaic virus and tomato aspermy virus. J Virol 71, 3597–3602.[Abstract]
Sarzynska, J., Nilsson, L. & Kulinski, T. (2003). Effects of base substitutions in an RNA hairpin from molecular dynamics and free energy simulations. Biophys J 85, 3445–3459.[Medline]
Shi, B. J., Symons, R. H. & Palukaitis, P. (2007). The cucumovirus 2b gene drives selection of inter-viral recombinants affecting the crossover site, the acceptor RNA and the rate of selection. Nucleic Acids Res 36, 1057–1071.[CrossRef][Medline]
Sivakumaran, K., Chen, M.-H., Roossinck, M. J. & Kao, C. C. (2002). Core promoter for initiation of Cucumber mosaic virus subgenomic RNA4A. Mol Plant Pathol 3, 43–52.[CrossRef]
Suzuki, M., Hibi, T. & Masuta, C. (2003). RNA recombination between cucumoviruses: possible role of predicted stem-loop structures and an internal subgenomic promoter-like motif. Virology 306, 77–86.[CrossRef][Medline]
White, K. A. (2002). The premature termination model: a possible third mechanism for subgenomic mRNA transcription in (+)-strand RNA viruses. Virology 304, 147–154.[CrossRef][Medline]
Wierzchoslawski, R., Dzianott, A., Kunimalayan, S. & Bujarski, J. J. (2003). A transcriptionally active subgenomic promoter supports homologous crossovers in a plus-strand RNA virus. J Virol 77, 6769–6776.
Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31, 3406–3415.
Received 4 February 2008;
accepted 8 March 2008.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
