| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,
1 Centro de Protección Vegetal y Biotecnología, Instituto Valenciano de Investigaciones Agrarias, Moncada, 46113 Valencia, Spain
2 Instituto de Biología Molecular y Celular de Plantas (CSIC-UPV), 46022 Valencia, Spain
Correspondence
Pedro Moreno
moreno_pedgom{at}ivia.gva.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
These authors contributed equally to this work. 
Present address: Instituto de Biología Molecular y Celular de Plantas (CSIC-UPV), Campus UPV CPI 8E, Ingeniero Fausto Elio s/n, 46022 Valencia, Spain.
Present address: Department of Cell and Developmental Biology, John Innes Centre, Norwich NR4 7UH, UK.
The nucleotide sequences reported in this study have been assigned GenBank/EMBL/DDBJ accession numbers FM955890–FM956002.
Supplementary tables of primer sequences and FEL data are available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
97 and 3 % of the virion length, respectively (Febres et al., 1996
; Satyanarayana et al., 2004). The single-stranded, positive-sense CTV genomic RNA (gRNA) is about 20 kb and contains 12 open reading frames (ORFs) potentially encoding at least 19 proteins (Karasev et al., 1995). ORFs 1a and 1b, encoding replication-related proteins, are translated from the gRNA, whereas the 10 3'-proximal ORFs, encoding proteins p33, p6, p65, p61, p27, p25, p18, p13, p20 and p23, are expressed via 3' co-terminal subgenomic RNAs (Hilf et al., 1995). Protein p6 may operate as a membrane anchor (Satyanarayana et al., 2000); proteins p65 (a homologue of the HSP70 heat-shock proteins), p61 and the two coat proteins are involved in virion assembly (Satyanarayana et al., 2000); p20 accumulates in amorphous inclusion bodies (Gowda et al., 2000); and p23, an RNA-binding protein (López et al., 2000), controls asymmetrical accumulation of plus and minus strands during RNA replication (Satyanarayana et al., 2002) and is involved in symptom expression (Ghorbel et al., 2001; Fagoaga et al., 2005). Proteins p23, p20 and p25 act as RNA silencing suppressors (Lu et al., 2004). The functions of p33, p13 and p18 remain unknown.
CTV is primarily dispersed by propagation of infected buds, and is then locally spread by aphids. CTV-induced symptoms include (i) decline of citrus species propagated on sour orange (Citrus aurantium L.) rootstock, (ii) yellowing and growth cessation of sour orange, lemon [C. limon (L.) Burn. f.] or grapefruit (C. paradisi Macf.) seedlings (seedling yellows) or (iii) stunting, stem pitting and poor yield of different citrus varieties regardless of the rootstock used (Moreno et al., 2008). The molecular mechanisms involved in symptom expression are still unknown.
As for other RNA viruses, genetic variation has been observed in CTV isolates resulting from the error-prone nature of RNA polymerases and selection pressures (Domingo & Holland, 1994), superinfection of field trees with divergent CTV variants (Rubio et al., 2001), genetic drift after transmission to new hosts (Albiach-Martí et al., 2000a; D'Urso et al., 2000; Ayllón et al., 2006) or recombination (Rubio et al., 2001; Vives et al., 2005). Characterization of the genetic structure of viral populations and factors contributing to their evolution may help in the understanding of important features like the outbreak of new epidemics or virulence changes in current isolates (Fernández-Cuartero et al., 1994; Escriu et al., 2000). These studies have practical implications in virus control, since durability of host resistance largely depends on the genetic variability of the virus (García-Arenal & McDonald, 2003).
We previously compared the predominant sequence variants of gene p23 from 18 CTV isolates with different origins and pathogenicity characteristics (Sambade et al., 2003). Phylogenetic analyses showed that sequence variants that are predominant in mild isolates {causing mild to moderate symptoms in Mexican lime [C. aurantifolia (Christ.) Swing.] and sometimes decline of sweet orange [C. sinensis (L.) Osb.] propagated on sour orange rootstock} and those predominant in virulent isolates (also inducing seedling yellows and stem pitting in grapefruit or sweet orange) clustered separately. To gain further insight into the mechanisms of CTV evolution, we analysed the genetic variation and phylogenetic relationships in seven gRNA regions of these isolates and investigated recombination between divergent sequence variants. Our analyses showed variable selection pressures along the gRNA and frequent recombination events. They also suggest that CTV variants cluster within at least two evolutionarily divergent lineages.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). These isolates were classified into five biogroups according to the symptoms induced in a panel of indicator hosts (Garnsey et al., 2005). In short, the K isolate is asymptomatic in all hosts (biogroup 0), isolates T32, T55 and T385 induce symptoms in Mexican lime (biogroup 1), T300, T312 and T346 also cause decline of sweet orange grafted on sour orange rootstock (biogroup 2), T36, Galego 50 and C-268-2 additionally induce seedling yellows (biogroup 3), while the remaining isolates, in addition to the latter symptoms, cause stem pitting on grapefruit (Barão B, C-269-6, C-270-3 and VT; biogroup 4) or on grapefruit and sweet orange (T305, T388, Cald-CB and Val-CB; biogroup 5).
cDNA synthesis, cloning and sequencing.
cDNA of regions located in genes p33, p65, p61, p18, p13, p20 and p23 of the CTV gRNA was synthesized by RT-PCR amplification using double-stranded RNA (dsRNA)-rich preparations (Moreno et al., 1990) as template and appropriate primers (Supplementary Table S1, available in JGV Online). Primers amplified 41–44 % of p33, p65 and p61 genes, 76 % of p18 and 87–99 % of p13, p20 and p23. RT-PCR was performed in a 25 µl reaction mix containing 20 mM Tris/HCl, pH 8.4, 50 mM KCl, 500 µg BSA ml–1, 3 mM MgCl2, 4 mM each of dATP, dCTP, dGTP and dTTP, 1 µM each primer, 20 U SuperScript II reverse transcriptase, 1 U RNaseOut and 1 U Taq DNA polymerase (Invitrogen). The reaction was carried out in an air thermal cycler (Idaho Technologies) using the following conditions: 30 min at 46 °C for RT, 2 min at 94 °C, 40 cycles of 5 s at 94 °C, 5 s at 55 °C and 30 s at 72 °C, and a final step of 2 min at 72 °C. The resulting RT-PCR products were cloned in the pGEM-T vector (Promega) (Sambrook et al., 1989).
The sequence variants predominant in each CTV isolate were selected by single-strand conformation polymorphism (SSCP) analysis (Rubio et al., 2001). For this purpose, 10 clones from each cDNA product were PCR-amplified as described above and the synthesized DNA was SSCP-analysed in the same gel as the RT-PCR product from which the clones were obtained (Sambade et al., 2003). Clones whose DNA strands co-migrated with the most intense DNA bands of the starting RT-PCR product were sequenced. The nucleotide sequences of the selected cDNA clones were determined in both directions with an ABI PRISM 3100 DNA sequence analyser (Applied Biosystems). The sequence of the virulent CTV isolate NUagA from Japan (GenBank accession no. AB046398
[GenBank]
) was used for comparisons.
Sequence analyses.
Nucleotide sequences were translated to proteins using GeneDoc and multiple protein alignments were performed with the MUSCLE program (Karl & Hugh, 1997; Edgar, 2004). Nucleotide alignments were then obtained by concatenating codons with the amino acid alignment as a guide. Sites containing insertions were removed from all subsequent analyses.
Nucleotide substitution models for different CTV regions were inferred using the model selection tool available at the Datamonkey server (http://www.datamonkey.org) of the HyPhy package (Kosakovsky Pond & Frost, 2005a). Genetic distances and substitution parameters were calculated by maximum-likelihood with the TREE-PUZZLE 5.2 program (Schmidt et al., 2002), assuming that sites had heterogeneous substitution rates described by a gamma distribution with eight classes. The best amino acid substitution model (lowest AIC value among competing models) was inferred with ProtTest (Abascal et al., 2005; available at http://darwin.uvigo.es/software/prottest.html). Codons under selection were detected using the fixed effects maximum-likelihood (FEL) method of the HyPhy package (Kosakovsky Pond & Frost, 2005b). Recombination was detected with the GARD program available at the Datamonkey server using the HKY85 substitution model and a beta–gamma distribution with four classes for rate variation. Recombination events and identification of parental sequences were confirmed further with the RDP3 package (Martin et al., 2005a) that incorporates the recombination-detecting algorithms GENECONV (Padidam et al., 1999), BOOTSCAN (Salminen et al., 1995; Martin et al., 2005b), MAXCHI (Smith, 1992; Posada & Crandall, 2001), CHIMAERA (Posada & Crandall, 2001), SISCAN (Gibbs et al., 2000), 3SEQ (Boni et al., 2007) and RDP (Martin & Rybicki, 2000), using their default parameter values. Average nucleotide distances of CTV isolates were calculated using MEGA 4.0 software (Tamura et al., 2007) after testing homogeneity of pattern substitution among lineages. Evolutionary distances were estimated by the composite maximum-likelihood method, assuming that substitution rates among sites fitted a gamma distribution.
Phylogenetic analysis.
Protein maximum-likelihood trees were inferred with ProtTest (Abascal et al., 2005); significance for the nodes was estimated with 1000 bootstrap replicates using the PHYML program (Guindon & Gascuel, 2003), available at http://phylemon.bioinfo.cipf.es/ (Tárraga et al., 2007). Nucleotide maximum-likelihood trees were constructed by the sequential addition method using HyPhy (Kosakovsky Pond et al., 2005), with the HKY85 substitution model and a gamma distribution with six classes for rate heterogeneity. Topology comparisons were performed by using the Shimodaira & Hasegawa (1999) test implemented in the DNAML program of the PHYLIP 3.67 package (Felsenstein, 2005). Phylogenetic trees were drawn using MEGA 4.0 (Tamura et al., 2007). The ratio of non-synonymous to synonymous substitution rates in different branches of maximum-likelihood trees was estimated using the codon-based genetic algorithm implemented in the GA-BRANCH program (Kosakovsky Pond & Frost, 2005c) available at the Datamonkey server. Split-decomposition analysis of concatenated CTV sequences was performed using the SplitsTree program with default parameters (Huson, 1998).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). At least two clades were observed in most regions. One of these (clade I), comprising the sequence variants of mild isolates T32, T55, T300, T312 and T385, and T346 in p20, was supported by bootstrap values of 86.3 (p33), 67.5 (p65), 96.1 (p61), 65.5 (p18), 75.9 (p13), 99.5 (p20) and 94.4 % (p23). A second clade (clade II), comprising biogroup 5 plus a variable set of isolates of biogroup 4, was supported by bootstrap values of 96.8 (p33), 87.2 (p23), 76.9 (p13) and 56.2 % (p65), although, in regions p33 and p65, a biogroup 3 isolate (Galego 50) was in the same cluster. Although support was less robust for clade II than for clade I, the virulent isolates T388, T305 and NUagA (NUagA-type group) were closely related to each other and distantly related to clade I in all genomic regions, forming a stable nucleus within clade II. Thus, phylogenetic relationships reflect, to some degree, pathogenicity characteristics of the isolates. In p18, all South American isolates except Galego 50 grouped together into a distinct clade (clade III).
|
). The virulent isolates Cald-CB, Barão B and Val-CB clustered together and were closely related to NUagA-type isolates only in p33, p65 and p23. In p33, p65 and p61, isolate K was genetically closer to clade I than to NUagA, but it was located between them in other regions. The two major p13 variants found in Galego 50 were divergent, with the variant Galego 50A grouping with isolates C-270-3 and Barão B and the variant Galego 50B being closer to the NUagA group. Similarly for isolate C-268-2, one of the three p33 variants was close to NUagA and the other two were separated from clade I and NUagA, two of the three p61 variants were close to clade I and the third was separate from both groups (Fig. 1). The different phylogenetic relationships observed for some isolates in different genomic regions suggested that their gRNA might have originated from recombination events between divergent sequences.
Genetic variation and selective pressures in different genomic regions
The average nucleotide distance for p33 (0.1641±0.0196) (Table 1) was significantly higher than that of the other regions (ranging from 0.0741±0.0240 to 0.1250±0.0160); the average distance for p13 (0.0741±0.0240) was significantly lower than that of p33 and p61 (model II ANOVA: F6,131=48.4279, P<0.0001; Tukey–Kramer post hoc test at 95 % confidence level). To evaluate the selective constraints operating in each region, codons under selection were detected using the FEL method (Supplementary Table S2, available in JGV Online). Since the rate of non-synonymous substitutions (dN) and the rate of synonymous substitutions (dS) estimates are sensitive to the effect of recombination, we preliminarily screened the different genomic regions with the GARD tool and found significant recombination signals in p61, p20 and p23 genes (positions 283, 252 and 239, respectively). Alignments for these three genes were split in the corresponding non-recombinant regions. A total of 150 codons were subjected to significant purifying selection (Table 1), with ratios of negatively selected codons ranging from 10 % for p20 to 15.87 % for p33. Although these ratios were not significantly different (
2=5.145, 6 d.f., P=0.525), the strength of negative selection estimated by the mean normalized dN–dS values differed among genomic regions (Kruskal–Wallis test: H=53.4090, 6 d.f., P<0.0001), this difference being entirely driven by the less negative dN–dS value estimated for p33 (–2.1098±0.2505) relative to the average value for the other six regions (–7.6334±0.4393) (Dunn's post hoc test, P<0.05). Only codons for amino acids 203 (dN–dS=3.6789) and 244 (dN–dS=6.6412) in p61 showed a significant level of positive selection.
|
) were estimated using a genetic algorithm that identified models with variable numbers of categories per lineage that fitted better to data than the single-ratio (all lineages evolving with equal ) or the fully saturated (each lineage evolves at different ) models (Table 2). Although values and the proportion of associated branches varied in different regions and periods of CTV diversification, most categories had <1 and most branches were assigned to these categories (100 % in p61 and >74 % in the other regions). Fig. 2(b) illustrates a simplified comparison of branch-specific values in phylogenetic trees, grouping these values within four selection intervals: (i) strong negative selection (<0.1), (ii) moderate negative (=0.1–0.4), (iii) weak negative (=0.4–1) and (iv) positive selection (>1). In most cases, internal branches connecting the groups of mild and virulent isolates, particularly clade I and the NUagA group, were associated with <1, indicating that divergence of these genotypes occurred under negative selection pressure. In p65, purifying selection was strong for all internal branches, whereas in p33, p18 and p13, periods of strong and moderate selection alternated. In p33, negative selection was moderate for lineages leading to mild and virulent groups and strong in the internal branch leading to the out-groups of C-270-3 and the ancestor of T36 and C-268-2B; however, the opposite was true in p18, with strong selection for lineages leading to the mild and virulent groups and moderate in the lineage leading to C-270-3 and other South American isolates. In p61, p20 and p23, selection was moderate for lineages leading to virulent or mild groups and to C-270-3, and in p23 it was weak or even positive after diversification of the cluster formed by K and clade I. Positive selection occurred during limited periods in all regions but p61 and, except for p13, it was observed only in terminal branches, with the highest frequency being observed in p33. For some isolates, a characteristic positive or weak negative selection was detected in most regions, i.e. T385 (p33, p61, p13, p20 and p23), T312 (p33, p65, p61, p18 and p23) or VT (p33, p65, p61 and p18).
|
|
test that compares the goodness of a set of competing phylogenetic trees to describe the evolution of a given alignment. Maximum-likelihood estimates of the substitution parameters (ratio of transversion to transition rates, shape parameter of the gamma distribution of substitution rates per site and relative substitution rates) were inferred using TREE-PUZZLE. This analysis confirmed the differences in tree topology inferred for different genomic regions, although, in p61, p20 and p23, the trees inferred from the entire region were not significantly worse than those inferred from each partition defined by GARD, suggesting that partitions established by GARD were mostly due to differences in branch length (P<0.05).
To assess the frequency and extension of recombination during CTV diversification, the seven genomic regions were concatenated and recombination events were identified using the RDP3 program that implements several recombination-detecting methods (Martin et al., 2005a), using default setting parameters for the subset of fast detection methods GENECONV, BOOTSCAN, MaxChi, CHIMAERA, Sister-Scan, 3SEQ and RDP. Isolate C-268-2 was excluded to avoid arbitrary assembling of its p33 and p61 variants. A total of 14 recombination events were detected by at least one of the methods, but only those predicted by at least four different methods were accepted (Fig. 2c) and assignment of parental and daughter sequences was confirmed by constructing maximum-likelihood trees (Fig. 2b). For example, isolate C-269-6 grouped with NUagA-type isolates in regions p33, p65, p23 and p20, but with C-270-3 in p61 and p18 and as an outgroup in p13 (Fig. 2b). A recombination involving a NUagA ancestor as major parental, and the p61 and p18 regions from a C-270-3 ancestor, was identified by the seven methods used (Fig. 2c). Recombination events involving the same parental strains were also detected for Cald-CB and Val-CB in p18, Barão B in p18 and p13 and Galego 50A in p13. Isolates Barão B and Val-CB, which grouped together and distant from other isolates in p20, presented an additional recombination in this region between a NUagA-type and an unknown ancestor not represented in the alignment. Galego 50 was closely related to VT in most regions (p33, p65, p61, p18 and p20) but not in p23 and p13. In this latter region, the variant Galego 50B grouped with VT and the variant Galego 50A with C-270-3, as a result of a recombination between VT and C-270-3 ancestors (Fig. 2b and c). In p23, Galego 50 was close to isolates C-270-3 and C-268-2, and RDP3 predicted a recombination between a Cald-CB ancestor as a major parental and an unknown isolate providing the 3' end of p20 and p23. The non-assignment of C-270-3 as a minor parental was due to the use of UPGMA trees in RDP3 default analysis (not shown), but inspection of maximum-likelihood trees indicated that VT and C-270-3 are likely to be the major and minor parental strains, respectively (Fig. 2b and c).
Isolate T346 was close to clade I in regions p33, p61 and p23, to C-270-3 in p65 and was in intermediate positions in p20, p18 and p13, an incongruence that was compatible with a recombination event in the p65 region between T312 and C-270-3 ancestors as major and minor parentals, respectively. The phylogenetic relationships of isolates T346 and T36 widely varied among regions, while in p33, p65 and p23, both isolates were divergent (genetic distances from 0.1180 to 0.2447), in p61, p18 and p13 they were closely related (genetic distances from 0.0090 to 0.0535) and two recombination events involving regions p61 and p13 were detected between ancestors of these isolates. In p13 and p18, T346 and K clustered in the same group. These results and the genetic distances observed in these regions are compatible with a recombination between K and T36 in p18 and p13 and later recombination between T346 and T36 in p13, and with other possibly older recombination events between T346 and T36 in p61.
Finally, isolate C-268-2 contained three major diverged variants in p33 and p61, but it was monomorphic for the other regions. Furthermore, it clustered with C-270-3 in p65, p18, p20 and p23 regions, but close to clade I in p13, suggesting that this isolate may be the result of several recombination events.
A phylogenetic network for CTV isolates
Due to the recombinant nature of CTV genomes, bifurcating phylogenetic trees do not properly reflect the actual evolutionary history of different isolates, since one isolate may be directly linked to more than one ancestral sequence. To provide a more accurate representation of those relationships, a phylogenetic network was constructed from the concatenated alignment of the seven regions by the split-decomposition method implemented in SplitsTree (Huson, 1998) (Fig. 3). Isolate C-268-2 was excluded for the reasons given above.
|
), were interconnected in the network in a complex pattern (Fig. 3), supporting recombination between them [as detected by RDP3 (Fig. 2c)]. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2001) and (ii) in citrus plants successively co-inoculated with a mild and a virulent CTV isolate, symptom onset was associated with predominance of the sequence variant characteristics of the virulent isolate (Sambade et al., 2002, 2007). Analysis of selective pressures acting on different codons suggests that all regions examined are mostly subjected to purifying selection, with only two codons in p61 being positively selected. Purifying selection measured as normalized dN–dS showed a similar proportion of selected sites and selection intensity among regions, except for p33 which had less intense selection. p33 had less negative dN–dS values, indicating that more non-synonymous substitutions are allowed in this region; this region also had higher evolutionary distances. Although net selection pressure was similar in the other regions, branch-specific analysis showed that the strength of selection varied depending on the genomic region and period of CTV diversification, i.e. this pressure was strong in the diversification of clade I and NUagA-type isolates in p65 but was moderate or weak in p23.
Data available on the functional domains of CTV proteins are still limited, and the function of proteins p33, p13 and p18, which are dispensable for CTV infection and movement (Tatineni et al., 2008), is unknown. Although selective pressures are less intense in p33, the fraction of selected sites is similar in all regions, indicating selective constraints to amino acid changes and providing candidate positions to test in functional studies. Negative selection was expected in p65, p61, p20 and p23, considering their role in CTV biology. Genes p65 and p61 are part of a conserved five-gene block encoding proteins involved in virion assembly and movement (Satyanarayana et al., 2000; Dolja et al., 2006). The p65 region analysed here encodes five of the eight motifs conserved among HSP70 proteins (Pappu et al., 1994). Proteins p20 and p23 act as silencing suppressors (Lu et al., 2004). Within the p20 region, amino acids I38, Y113, R130, L137, S141 and L159 are strictly conserved among silencing suppressors of closteroviruses (Reed et al., 2003). Gene p23, which has been completely sequenced, contains the RNA-binding domain and putative zinc finger required for asymmetrical accumulation of positive and negative RNA strands, with conserved residues C68, C71, H75 and C85 being involved in this activity (López et al., 2000; Satyanarayana et al., 2002). These amino acids were encoded by invariable codons in all CTV isolates, with the exception of L137 in p20 and C71 in p23, which were found to be under significant purifying selection using cut-off values of P=0.5 and P=0.1, respectively. The observation that p23 is a pathogenicity determinant in Citrus spp. (Ghorbel et al., 2001; Fagoaga et al., 2005) is consistent with separation of the mild and virulent CTV isolates in phylogenetic analysis.
Phylogenetic analysis showed that the mild isolates T32, T55, T300, T312 and T385 form a clade (clade I) supported by bootstrap values >70 % in six regions, thus defining a CTV lineage which also includes isolate T30 from Florida and probably others from Colombia and Taiwan (Ruiz-Ruiz et al., 2006; Albiach-Martí et al., 2000b). A second clade (clade II) including a variable set of virulent isolates was supported by bootstrap values >75 % in three regions, suggesting a higher recombination frequency for those isolates. Within clade II, isolates T305, T388 and NUagA were closely related to each other and distantly related to clade I in all regions and, together with other virulent isolates, showed a star-like evolution pattern from a common ancestor in the phylogenetic network. These data suggest that virulent isolates could represent a second CTV lineage distantly related to clade I that would also include isolate T318A from Spain (Ruiz-Ruiz et al., 2006). A third clade, including Brazilian and Argentinean isolates genetically related to isolate C-270-3, was observed in several genomic regions. Sequence comparisons showed that several clones of two Colombian isolates released in GenBank were closely related to C-270-3, while others were closely related to virulent isolates in p23. However, the latter were related to clade I or to virulent isolates but not to C-270-3 in p33 (not shown). Although a complete gRNA sequence from South American isolates is currently not available, these findings are compatible with the diversification of a third CTV lineage and frequent recombinations in this area. All lineages included closely related CTV variants from distant locations, a finding that, together with the radiation pattern observed for diversification of some CTV isolates, provide further support to a previous suggestion that genetic flow has occurred (Rubio et al., 2001).
Most isolates showed incongruent phylogenetic relationships in different regions, suggesting frequent recombination events, a possibility that was supported by recombination-detecting methods and by a split-decomposition phylogenetic network. Nine of the 19 isolates compared were recombinant, including most isolates from Brazil (between ancestors of NUagA or VT and C-270-3) and Argentina (C-268-2 appears as a mosaic of mild, severe and C-270-3-type sequence variants), in agreement with the results obtained for these and other Argentinean isolates (Iglesias et al., 2008). Recombination involving p18 was so frequent that, in this region, all South American isolates except Galego 50 formed a monophyletic group diverged from the other isolates. A similar grouping was described for genes p25 and p27, located upstream of p18, for Argentinean isolates (Iglesias et al., 2008). The variable position of K, T346 and T36 in phylogenetic trees, their lack of association with other isolates and the network topology suggest that they might represent CTV lineages with a more complex history involving recombination events among their ancestors, and possibly with genotypes unrelated to those analysed here.
The phylogenetic network grouped CTV isolates in two major clusters separated by long splits: one of them comprises isolates of biogroups 0, 1 and 2 (except for isolate T346) that induce mild to moderate symptoms in the most sensitive hosts and cause symptomless infections in grapefruit and sweet orange seedlings, and the other includes isolates of biogroups 3, 4 and 5 (except for isolate T36) that induce severe symptoms in sensitive hosts and are also pathogenic on grapefruit or sweet orange. Sequence separation between mild and virulent CTV isolates is consistent with a different host response after infection, as indicated by specific changes induced in the citrus transcriptome by both types of isolates (Gandía et al., 2007). This separation suggests that virulence might be an important evolutionary factor shaping CTV populations.
Homologous recombination seems a common process in some plant RNA viruses, particularly potyviruses (Chare & Holmes, 2005) and bromoviruses (Codoñer & Elena, 2008). It has been postulated that recombination might prevent accumulation of deleterious mutations in small populations and/or allow a faster adaptation to changing environments (Lai, 1992; Roossinck, 1997; García-Arenal et al., 2001). On the other hand, simulation studies showed that recombination in RNA viruses is more likely to create combinations of deleterious mutations than purge them from genomes, thus causing fitness reductions (Holmes, 2003). Our results and previous analyses (Hilf, 2009) suggest that RNA recombination is a major factor in CTV variation and is likely to play a role in its evolution. Potential factors contributing to frequent recombination in CTV include: (i) dispersal of diverged CTV genotypes in the same area by movement of infected buds, (ii) the long life of citrus trees providing many opportunities for repeated infections with divergent sequence variants (Rubio et al., 2001; Weng et al., 2007; Gomes et al., 2008), (iii) the presence in infected cells of several viral RNA species produced during CTV replication that probably facilitates recombination events (Hilf et al., 1995; Yang et al., 1997; Ayllón et al., 1999; Gowda et al., 2003) and (iv) the large size of the CTV genome that may accumulate 2–3 mutations per genome and replication round. Recombination might help in maintaining functional genomes, even if many non-functional recombinants were produced (Allison et al., 1990).
In summary, analyses of genetic variation in the 3' half of the CTV genome suggest that at least two different lineages might have evolved and that selective pressure against amino acid changes, gene flow, homologous recombination and perhaps virulence may be important factors in CTV evolution and in shaping CTV populations.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Albiach-Martí, M. R., Guerri, J., Hermoso de Mendoza, A., Laigret, F., Ballester-Olmos, J. F. & Moreno, P. (2000a). Aphid transmission alters the genomic and defective RNA populations of Citrus tristeza virus isolates. Phytopathology 90, 134–138.[CrossRef][Medline]
Albiach-Martí, M. R., Mawassi, M., Gowda, S., Satyanarayana, T., Hilf, S., Shanker, M. E., Almira, E. C., Vives, M. C., López, C. & other authors (2000b). Sequences of Citrus tristeza virus separated in time and space are essentially identical. J Virol 74, 6856–6865.
Allison, R., Thompson, C. & Ahlquist, P. (1990). Regeneration of a functional RNA virus genome by recombination between deletion mutants and requirement for Cowpea chlorotic mottle virus 3a and coat genes for systemic infection. Proc Natl Acad Sci U S A 87, 1820–1824.
Ayllón, M. A., López, C., Navas-Castillo, J., Mawassi, M., Dawson, W. O., Guerri, J., Flores, R. & Moreno, P. (1999). New defective RNAs from citrus tristeza virus: evidence for a replicase-driven template switching mechanism in their generation. J Gen Virol 80, 817–821.[Abstract]
Ayllón, M. A., Rubio, L., Sentandreu, V., Moya, A., Guerri, J. & Moreno, P. (2006). Variations in two gene sequences of citrus tristeza virus after host passage. Virus Genes 32, 119–128.[CrossRef][Medline]
Boni, M. F., Posada, D. & Feldman, M. W. (2007). An exact nonparametric method for inferring mosaic structure in sequence triplets. Genetics 176, 1035–1047.
Chare, E. R. & Holmes, E. C. (2005). A phylogenetic survey of recombination frequency in plant RNA viruses. Arch Virol 151, 933–946.[CrossRef][Medline]
Codoñer, F. M. & Elena, S. F. (2008). The promiscuous evolutionary history of the Bromoviridae family. J Gen Virol 89, 1739–1747.
Dolja, V. V., Kreuze, J. F. & Valkonen, J. P. T. (2006). Comparative and functional genomics of closteroviruses. Virus Res 117, 38–51.[CrossRef][Medline]
Domingo, E. & Holland, J. J. (1994). Mutation rates and rapid evolution of RNA viruses. In The Evolutionary Biology of Viruses, pp. 161–184. Edited by S. S. Morse. New York: Raven Press.
D'Urso, F., Ayllón, M. A., Rubio, L., Sambade, A., Hermoso de Mendoza, A., Guerri, J. & Moreno, P. (2000). Contribution of uneven distribution of genomic RNA variants of Citrus tristeza virus (CTV) within the plant to changes in the viral population following aphid transmission. Plant Pathol 49, 288–294.[CrossRef]
Edgar, R. C. (2004). MUSCLE: a multiple sequence alignment method with reduced time and space complexity. Bioinformatics 5. doi: 10.1186/1471-2105-5-113
Escriu, F., Fraile, A. & García-Arenal, F. (2000). Evolution of virulence in natural populations of the satellite RNA of Cucumber mosaic virus. Phytopathology 90, 480–485.[CrossRef][Medline]
Fagoaga, C., López, C., Moreno, P., Navarro, L., Flores, R. & Peña, L. (2005). Viral-like symptoms induced by the ectopic expression of the p23 gene of Citrus tristeza virus are citrus specific and do not correlate with the pathogenicity of the virus strain. Mol Plant Microbe Interact 18, 435–445.[CrossRef][Medline]
Febres, V. J., Ashoulin, L., Mawassi, M., Frank, A., Bar-Joseph, M., Manjunath, K. L., Lee, R. F. & Niblett, C. L. (1996). The p27 protein is present at one end of Citrus tristeza virus particles. Phytopathology 86, 1331–1335.
Felsenstein, J. (2005). PHYLIP (Phylogeny Inference Package) version 3.6. Distributed by the author. Department of Genome Sciences, University of Washington, Seattle, USA.
Fernández-Cuartero, B., Burgyan, J., Aranda, M. A., Salanki, K., Moriones, E. & García-Arenal, F. (1994). Increase in the relative fitness of a plant virus RNA associated with its recombinant nature. Virology 203, 373–377.[CrossRef][Medline]
Gandía, M., Conesa, A., Ancillo, G., Gadea, J., Forment, J., Pallás, V., Flores, R., Duran-Vila, N., Moreno, P. & Guerri, J. (2007). Transcriptional response of Citrus aurantifolia to infection by Citrus tristeza virus. Virology 367, 298–306.[CrossRef][Medline]
García-Arenal, F. & McDonald, B. A. (2003). An analysis of the durability of resistance to plant viruses. Phytopathology 93, 941–952.[CrossRef][Medline]
García-Arenal, F., Fraile, A. & Malpica, J. M. (2001). Variability and genetic structure of plant virus populations. Annu Rev Phytopathol 39, 157–186.[CrossRef][Medline]
Garnsey, S. M., Civerolo, E. L., Gumpf, D. J., Paul, C., Hilf, M. E., Lee, R. F., Brlansky, R. H., Yokomi, R. K. & Hartung, J. S. (2005). Biological characterization of an international collection of Citrus tristeza virus (CTV) isolates. In Proceedings of the 16th Conference of the International Organization of Citrus Virologists, pp. 75–93, Edited by M. E. Hilf, N. Duran-Vila & M. A. Rocha-Peña. Riverside, CA: International Organisation of Citrus Virologists. http://www.ivia.es/iocv/
Ghorbel, R., López, C., Fagoaga, C., Moreno, P., Navarro, L., Flores, R. & Peña, L. (2001). Transgenic citrus plants expressing the Citrus tristeza virus p23 protein exhibit viral-like symptoms. Mol Plant Pathol 2, 27–36.[CrossRef]
Gibbs, M. J., Armstrong, J. S. & Gibbs, A. J. (2000). Sister-Scanning: a Monte Carlo procedure for assessing signals in recombinant sequences. Bioinformatics 16, 573–582.
Gomes, C. P. C., Nagata, T., de Jesus, W. C., Jr, Neto, C. R. B., Pappas, G. J., Jr & Martin, D. P. (2008). Genetic variation and recombination of RdRp and HSP 70h genes of Citrus tristeza virus isolates from orange trees showing symptoms of citrus sudden death disease. Virol J 5. doi:10.1186/1743-422x-5-9
Gowda, S., Satyanarayana, T., Davis, C. L., Navas-Castillo, J., Albiach-Martí, M. R., Mawassi, M., Valkov, N., Bar-Joseph, M., Moreno, P. & Dawson, W. O. (2000). The p20 gene product of Citrus tristeza virus accumulates in the amorphous inclusion bodies. Virology 274, 246–254.[CrossRef][Medline]
Gowda, S., Ayllón, M. A., Satyanarayana, T., Bar-Joseph, M. & Dawson, W. O. (2003). Transcription strategy in a Closterovirus: a novel 5'-proximal controller element of Citrus tristeza virus produces 5'- and 3'- terminal subgenomic RNAs and differs from 3' open reading frame controller elements. J Virol 77, 340–352.[CrossRef][Medline]
Guindon, S. & Gascuel, O. (2003). A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 52, 696–704.
Hilf, M. E. (2009). Two distinct evolutionary pathways for Citrus tristeza virus: recombination defines two gene modules and provides for increased genetic diversity in a narrow host range plant virus. In Proceedings of the 17th Conference of the International Organization of Citrus Virologists (in press).
Hilf, M. E., Karasev, A. V., Pappu, H. R., Gumpf, D. J., Niblett, C. L. & Garnsey, S. M. (1995). Characterization of Citrus tristeza virus subgenomic RNAs in infected tissue. Virology 208, 576–582.[CrossRef][Medline]
Holmes, E. C. (2003). Error thresholds and the constraints to RNA virus evolution. Trends Microbiol 11, 543–546.[CrossRef][Medline]
Huson, D. H. (1998). SplitsTree: analyzing and visualizing evolutionary data. Bioinformatics 14, 68–73.
Iglesias, N. G., Gago-Zachert, S. P., Robledo, G., Costa, N., Plata, M. I., Vera, O., Grau, O. & Semorile, L. C. (2008). Population structure of Citrus tristeza virus from field Argentinean isolates. Virus Genes 36, 199–207.[CrossRef][Medline]
Karasev, A. V., Boyko, V. P., Gowda, S., Nikolaeva, O. V., Hilf, M. E., Koonin, E. V., Niblett, C. L., Cline, K., Gumpf, D. J. & other authors (1995). Complete sequence of the Citrus tristeza virus RNA genome. Virology 208, 511–520.[CrossRef][Medline]
Karl, B. N. & Hugh, B. N. (1997). GeneDoc: a tool for editing and annotating multiple sequence alignments. www.nrbsc.org/gfx/genedoc/ebinet.htm.
Kosakovsky Pond, S. L. & Frost, S. D. W. (2005a). Datamonkey: rapid detection of selective pressure on individual sites of codon alignments. Bioinformatics 21, 2531–2533.
Kosakovsky Pond, S. L. & Frost, S. D. W. (2005b). Not so different after all: a comparison of methods for detecting amino acid sites under selection. Mol Biol Evol 22, 1208–1222.
Kosakovsky Pond, S. L. & Frost, S. D. W. (2005c). A genetic algorithm approach to detecting lineage-specific variation in selection pressure. Mol Biol Evol 22, 478–485.
Kosakovsky Pond, S. L., Frost, S. D. W. & Muse, S. V. (2005). HyPhy: hypothesis testing using phylogenies. Bioinformatics 21, 676–679.
Lai, M. M. (1992). RNA recombination in animal and plant viruses. Microbiol Rev 56, 61–79.
López, C., Navas-Castillo, J., Gowda, S., Moreno, P. & Flores, R. (2000). The 23-kDa protein coded by the 3'-terminal gene of Citrus tristeza virus is an RNA-binding protein. Virology 269, 462–470.[CrossRef][Medline]
Lu, R., Folimonov, A., Shintaku, M., Li, W. X., Falk, B. W., Dawson, W. O. & Ding, S. W. (2004). Three distinct suppressors of RNA silencing encoded by a 20-kb viral RNA genome. Proc Natl Acad Sci U S A 101, 15742–15747.
Martin, D. P. & Rybicki, E. (2000). RDP: detection of recombination amongst aligned sequences. Bioinformatics 16, 562–563.
Martin, D. P., Williamson, C. & Posada, D. (2005a). RDP2: recombination detection and analysis from sequence alignments. Bioinformatics 21, 260–262.
Martin, D. P., Posada, D., Crandall, K. A. & Williamson, C. (2005b). A modified bootscan algorithm for automated identification of recombinant sequences and recombination breakpoints. AIDS Res Hum Retroviruses 21, 98–102.[CrossRef][Medline]
Moreno, P., Guerri, J. & Muñoz, N. (1990). Identification of Spanish strains of Citrus tristeza virus by analysis of double-stranded RNA. Phytopathology 80, 477–482.[CrossRef]
Moreno, P., Ambrós, S., Albiach-Martí, M. R., Guerri, J. & Peña, L. (2008). Citrus tristeza virus: a pathogen that changed the course of the citrus industry. Mol Plant Pathol 9, 251–268.[CrossRef][Medline]
Padidam, M., Sawyer, S. & Fauquet, C. M. (1999). Possible emergence of new geminiviruses by frequent recombination. Virology 265, 218–225.[CrossRef][Medline]
Pappu, H. R., Karasev, A. V., Anderson, E. J., Pappu, S. S., Hilf, M. E., Febres, V. J., Eckloff, R. M., McCaffery, M., Boyko, V. & Gowda, S. (1994). Nucleotide sequence and organization of eight 3' open reading frames of the Citrus tristeza closterovirus genome. Virology 199, 35–46.[CrossRef][Medline]
Posada, D. & Crandall, K. A. (2001). Evaluation of methods for detecting recombination from DNA sequences: computer simulations. Proc Natl Acad Sci U S A 98, 13757–13762.
Reed, J. C., Kasschau, K. D., Prokhnevsky, A. I., Gopinath, K., Pogue, G. P., Carrington, J. C. & Dolja, V. V. (2003). Suppressor of RNA silencing encoded by Beet yellows virus. Virology 306, 203–209.[CrossRef][Medline]
Roossinck, M. J. (1997). Mechanisms of plant virus evolution. Annu Rev Phytopathol 35, 191–209.[CrossRef][Medline]
Rubio, L., Ayllón, M. A., Kong, P., Fernández, A., Polek, M., Guerri, J., Moreno, P. & Falk, B. W. (2001). Genetic variation of Citrus tristeza virus isolates from California and Spain, evidence for mixed infections and recombination. J Virol 75, 8054–8062.
Ruiz-Ruiz, S., Moreno, P., Guerri, J. & Ambrós, S. (2006). The complete nucleotide sequence of a severe stem pitting isolate of Citrus tristeza virus from Spain, comparison with isolates from different origins. Arch Virol 151, 387–398.[CrossRef][Medline]
Salminen, M. O., Carr, J. K., Burke, D. S. & McCutchan, F. E. (1995). Identification of breakpoints in intergenotypic recombinants of HIV type 1 by Bootscanning. AIDS Res Hum Retroviruses 11, 1423–1425.[Medline]
Sambade, A., Rubio, L., Garnsey, S. M., Costa, N., Müller, G. W., Peyrou, M., Guerri, J. & Moreno, P. (2002). Comparison of viral RNA populations of pathogenically distinct isolates of Citrus tristeza virus, application to monitoring cross-protection. Plant Pathol 51, 257–265.[CrossRef]
Sambade, A., López, C., Rubio, L., Flores, R., Guerri, J. & Moreno, P. (2003). Polymorphism of a specific region in gene p23 of Citrus tristeza virus allows discrimination between mild and severe isolates. Arch Virol 148, 2325–2340.[CrossRef][Medline]
Sambade, A., Ambrós, S., López, C., Ruiz-Ruiz, S., Hermoso de Mendoza, A., Flores, R., Guerri, J. & Moreno, P. (2007). Preferential accumulation of severe variants of Citrus tristeza virus in plants co-inoculated with mild and severe variants. Arch Virol 152, 1115–1126.[CrossRef][Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Satyanarayana, T., Gowda, S., Boyko, V. P., Albiach-Martí, M. R., Mawassi, M., Navas-Castillo, J., Karasev, A. V., Dolja, V., Hilf, M. E. & other authors (1999). An engineered closterovirus RNA replicon and analysis of heterologous terminal sequences for replication. Proc Natl Acad Sci U S A 96, 7433–7438.
Satyanarayana, T., Gowda, S., Mawassi, M., Albiach-Martí, M. R., Ayllón, M. A., Robertson, C., Garnsey, S. M. & Dawson, W. O. (2000). Closterovirus encoded HSP70 homolog and p61 in addition to both coat proteins function in efficient virion assembly. Virology 278, 253–265.[CrossRef][Medline]
Satyanarayana, T., Bar-Joseph, M., Mawassi, M., Albiach-Martí, M. R., Ayllón, M. A., Gowda, S., Hilf, M. E., Moreno, P., Garnsey, S. M. & Dawson, W. O. (2001). Amplification of Citrus tristeza virus from a cDNA clone and infection of citrus trees. Virology 280, 87–96.[CrossRef][Medline]
Satyanarayana, T., Gowda, S., Ayllón, M. A., Albiach-Martí, M. R., Rabindran, S. & Dawson, W. O. (2002). The p23 protein of Citrus tristeza virus controls asymmetrical RNA accumulation. J Virol 76, 473–483.
Satyanarayana, T., Gowda, S., Ayllón, M. A. & Dawson, W. O. (2004). Closterovirus bipolar virion: evidence for initiation of assembly by minor coat protein and its restriction to the genomic RNA 5' region. Proc Natl Acad Sci U S A 101, 799–804.
Schmidt, H. A., Strimmer, K., Vingron, M. & von Haeseler, A. (2002). TREE-PUZZLE: maximum likelihood phylogenetic analysis using quartets and parallel computing. Bioinformatics 18, 502–504.
Shimodaira, H. & Hasegawa, M. (1999). Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Mol Biol Evol 16, 1114–1116.
Smith, J. M. (1992). Analyzing the mosaic structure of genes. J Mol Evol 34, 126–129.[Medline]
Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007). MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24, 1596–1599.
Tárraga, J., Medina, I., Arbiza, L., Huerta-Cepas, J., Gabaldón, T., Dopazo, J. & Dopazo, H. (2007). Phylemon: a suite of web tools for molecular evolution, phylogenetics and phylogenomics. Nucleic Acids Res 35, W38–W42
Tatineni, S., Robertson, C. J., Garnsey, S. M., Bar-Joseph, M., Gowda, S. & Dawson, W. O. (2008). Three genes of Citrus tristeza virus are dispensable for infection and movement throughout some varieties of citrus trees. Virology 376, 297–307.[CrossRef][Medline]
Vives, M. C., Rubio, L., Sambade, A., Mirkov, T. E., Moreno, P. & Guerri, J. (2005). Evidence of multiple recombination events between two RNA sequence variants within a Citrus tristeza virus isolate. Virology 331, 232–237.[CrossRef][Medline]
Weng, Z., Barthelson, R., Gowda, S., Hilf, M. E., Dawson, W. O., Galbraith, D. W. & Xiong, Z. (2007). Persistent infection and promiscuous recombination of multiple genotypes of an RNA virus within a single host generate extensive diversity. PLoS One 2, e917[CrossRef][Medline]
Yang, G. A., Mawassi, M., Gofman, R., Gafny, R. & Bar-Joseph, M. (1997). Involvement of a subgenomic mRNA in the generation of a variable population of defective Citrus tristeza virus molecules. J Virol 71, 9800–9802.[Abstract]
Received 22 October 2008;
accepted 12 February 2009.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
