ABSTRACT
Rolling-circle replication of single-stranded genomes of plant geminiviruses is initiated by sequence-specific DNA binding of the viral replication-related protein (Rep) to its cognate genome at the replication origin. Monopartite begomovirus-associated betasatellites can be trans replicated by both cognate and some noncognate helper viruses, but the molecular basis of replication promiscuity of betasatellites remains uncharacterized. Earlier studies showed that when tomato yellow leaf curl China virus (TYLCCNV) or tobacco curly shoot virus (TbCSV) is coinoculated with both cognate and noncognate betasatellites, the cognate betasatellite dominates over the noncognate one at the late stages of infection. In this study, we constructed reciprocal chimeric betasatellites between tomato yellow leaf curl China betasatellite and tobacco curly shoot betasatellite and assayed their competitiveness against wild-type betasatellite when coinoculated with TYLCCNV or TbCSV onto plants. We mapped a region immediately upstream of the conserved rolling-circle cruciform structure of betasatellite origin that confers the cognate Rep-mediated replication advantage over the noncognate satellite. DNase I protection and in vitro binding assays further identified a novel sequence element termed Rep-binding motif (RBM), which specifically binds to the cognate Rep protein and to the noncognate Rep, albeit at lower affinity. Furthermore, we showed that RBM-Rep binding affinity is correlated with betasatellite replication efficiency in protoplasts. Our data suggest that although strict specificity of Rep-mediated replication does not exist, betasatellites have adapted to their cognate Reps for efficient replication during coevolution.
IMPORTANCE Begomoviruses are numerous circular DNA viruses that cause devastating diseases of crops worldwide. Monopartite begomoviruses are frequently associated with betasatellites which are essential for induction of typical disease symptoms. Coexistence of two distinct betasatellites with one helper virus is rare in nature. Our previous research showed that begomoviruses can trans replicate cognate betasatellites to higher levels than noncognate ones. However, the molecular mechanisms of betasatellites selective replication remain largely unknown. We investigated the interaction between the begomovirus replication-associated protein and betasatellite DNA. We found that the replication-associated protein specifically binds to a motif in betasatellites, with higher affinity for the cognate motif than the noncognate motif. This preference for cognate motif binding determines the selective replication of betasatellites. We also demonstrated that this motif is essential for betasatellite replication. These findings shed new light on the promiscuous yet selective replication of betasatellites by helper geminiviruses.
INTRODUCTION
Viruses of the family Geminiviridae have small circular single-stranded DNA (ssDNA) genomes encapsidated within geminate particles. They are divided into seven genera based on their genome organization, host range, and insect vector, with the genus Begomovirus being the largest (1). Some begomoviruses in the family Geminiviridae have genomes consisting of two molecules of circular single-stranded DNA, which are referred to as DNA A and DNA B. However, many begomovirus species have only a single genomic component resembling DNA A of the bipartite begomoviruses. Among the monopartite begomoviruses, an increasing number of species, such as ageratum yellow vein virus (AYVV), bhendi yellow vein mosaic virus (BYVMV), cotton leaf curl Multan virus (CLCuMV), and eupatorium yellow vein virus (EpYVV), have been found in association with novel DNA molecules termed betasatellites (formerly known as DNA β) (2–5). Betasatellites are also circular, single-stranded DNAs of approximately 1,350 nucleotides (nt) that have no appreciable sequence identity to their helper begomoviruses. The one exception is a conserved hairpin structure that includes a TAATATTAC loop sequence that is necessary for rolling-circle replication. The betasatellites depend on their helper begomoviruses for replication, and many betasatellites are essential for the induction of typical disease symptoms and viral genome accumulation (6). Betasatellites encode an approximately 13.5-kDa protein known as βC1 (6, 7). Previous studies showed that βC1 interacts with many host proteins to modulate the plant resistance response and facilitates begomovirus infection and transmission (8–13).
Geminiviruses copy their genomes with the aid of host DNA replication machinery, as the viral replication initiation protein (Rep) (also known as AL1, AC1, C1, or L1) is not a DNA polymerase (14–16). Rep is a sequence-specific DNA-binding protein with site-specific nicking and joining activity (17–19). Rep specifically recognizes and binds to repeated sequence motifs (termed iterons) located in the replication origin of the viral genome, a process that is required for initiation of rolling-circle DNA replication (15, 20). For bipartite begomoviruses, sequence-specific Rep iterons' interactions constrain Rep to replicate its cognate DNA A and DNA B components that contain the high-affinity Rep-binding repeated sequences (19, 21). However, in contrast to specific replication of DNA A and DNA B by cognate Rep, betasatellites can interact promiscuously with diverse begomoviruses for trans replication. Cotton leaf curl Multan betasatellite (CLCuMB) was found in association with a complex of distinct monopartite begomoviruses in the field that causes cotton leaf curl disease in south Asia (22). Similarly, ageratum yellow vein betasatellite (AYVB) can be replicated by several monopartite begomoviruses as well as the bipartite Sri Lankan cassava mosaic virus (23). Consistent with these observations, studies with a defective satellite associated with tomato leaf curl virus (ToLCV-sat) showed that although high-affinity binding sites were found in ToLCV-sat, they are dispensable for replication (24). This is in direct contrast with other studies with begomoviruses, where specific interactions between Rep and cis-acting binding sites were prerequisite for replication (16, 19).
Although betasatellite replication is promiscuous, geminiviruses show differential interactions with betasatellites. For example, both AYVB and CLCuMB are trans replicated by AYVV, CLCuMV, and EpYVV but not by honeysuckle yellow vein virus (HYVV) (25). In most cases, cognate helper viruses replicate satellites to higher levels than noncognate viruses. These reassortment assays suggested that replication specificity exists between begomoviruses and betasatellites (25). Previously, we found that both tomato yellow leaf curl China virus (TYLCCNV) and tobacco curly shoot virus (TbCSV) can replicate and stably maintain either of tomato yellow leaf curl China betasatellite (TYLCNB) and tobacco curly shoot betasatellite (TbCSB). However, when both betasatellite molecules were coinoculated with one helper virus, the cognate betasatellite predominated over the noncognate one. At late stages of the infection, the noncognate betasatellite became undetectable in the new leaves of inoculated plants (26). So far, little is known about the molecular mechanisms of betasatellite replication, and the sequence elements conferring promiscuous yet differential replication remain uncharacterized. In the current study, we constructed hybrid satellites of TYLCCNB and TbCSB and used in planta competition assays to locate the cis elements that are responsible for differential replication. Electrophoretic mobility shift assay (EMSA) and DNase I footprinting were used to identify a Rep-binding motif (RBM) upstream of the satellite conserved region (SCR) that is located adjacent to rolling-circle cruciform structure. Our data showed that Reps specifically binds to RBM of betasatellites and with higher affinity for a cognate RBM than a noncognate motif. Rep-RBM binding affinities correlate with betasatellite replication efficiency in protoplasts and with differential interactions between helper viruses and betasatellites in planta.
MATERIALS AND METHODS
Plant material and agroinoculation.
Nicotiana benthamiana plants were grown in an insect-free cabinet at a constant temperature of 25°C with supplementary lighting corresponding to a 16-h day length. Agrobacterium tumefaciens (EHA105) cultures were grown at 28°C for 48 h (optical density at 600 nm [OD600] = 1). In each experiment, 2-week-old N. benthamiana seedlings were agroinoculated with the A. tumefaciens mixtures, which contained equal volumes of the separate bacterial cultures.
Construction of infectious clones and expression plasmids.
The infectious clones of TYLCCNV isolate Y10 (pBinPLUS-Y10-1.7A) and its betasatellite TYLCCNB (pBinPLUS-Y10-2β), TbCSV isolate Y35 (pBinPLUS-Y35-1.7A) and its betasatellite TbCSB (pBinPLUS-Y35-2β), and tomato leaf curl China virus isolate G18 (ToLCCNV, pBinPLUS-G18-1.7A) and its betasatellite ToLCCNB (pBinPLUS-G18-2β) (referred to here as Y10 and Y10β, Y35 and Y35β, and G18 and G18β, respectively) were constructed previously (7, 27, 28). Infectious clones of the hybrid betasatellites, pBinPLUS-Y10β-35βC1, pBinPLUS-Y35β-10βC1, pBinPLUS-Y10β-35AP, and pBinPLUS-Y35β-10AP (referred to here as Y10β-35βC1, Y35β-10βC1, Y10β-35AP, and Y35β-10AP, respectively), were also described previously (29). Other hybrid betasatellite constructs were produced as previously described (29). Briefly, a splicing overlap extension PCR (SOE-PCR) strategy was employed to precisely exchange the LCR (left side of the SCR, a 260-nt fragment upstream of the primer pair β01 and β02 located positions), the RCR (right side of the SCR, a 270-nt fragment downstream of the positions of the primer pair β01 and β02), and the RBM (Rep-binding motif) between Y10β and Y35β or to delete the RBM of the betasatellite (Fig. 1). All PCRs were conducted using Pfu DNA polymerase (Promega, Madison, WI, USA). pBinPLUS-Y10-2β and pBinPLUS-Y35-2β were used as templates for SOE-PCR. For example, to obtain a chimeric satellite containing the TbCSB LCR fragment in the TYLCCNB sequence context (Y10β-35LCR), two independent PCRs were conducted using two pairs of primers. The primers β01 and 35AP-LCR/R were used to amplify a fragment from TYLCCNB (Y10β) containing the RCR, the βC1 open reading frame (ORF), the promoter of βC1 and the A-rich region, from the 5′ terminus to the 3′ terminus, while the primers β02 and 35AP-LCR/F were used to amplify the fragment of the LCR of TbCSB (Y35β). The two PCR products were recovered independently, and each product was mixed in the standard PCR system. After annealing and extension, the flanking primer pair β01/β02 was added to amplify the full-length hybrid betasatellite component. The overlapping PCR products were inserted into a pGEM-T Easy vector (Promega) to produce clones pGEM-Y10β-35LCR (in which the LCR region of TYLCCNB was substituted by the TbCSB LCR region). Accordingly, other hybrid satellites (Fig. 1B and C) were constructed using the same strategy with the primers. All the vectors were sequenced to confirm the fidelity of successful exchanges without mutations introduced by PCR. Dimeric constructs of hybrid betasatellite clones for agroinoculation, pBinPLUS-Y10β-35LCR, pBinPLUS-Y35β-10LCR, pBinPLUS-Y10β-35RCR, pBinPLUS-Y35β-10RCR, pBinPLUS-Y10β-35RBM, pBinPLUS-Y35β-10RBM, pBinPLUS-Y10β-delRBM, pBinPLUS-Y35β-delRBM (referred to here as Y10β-35LCR, Y35β-10LCR, Y10β35-RCR, Y35β10-RCR, Y10β35-RBM, Y35β10-RBM, Y10β-delRBM, and Y35β-delRBM, respectively) were produced using the method described previously (30). All primers used in these experiments are available upon request.
FIG 1.
Genomic organization of the betasatellites used in this study. (A) The betasatellite was partitioned into four parts, the right side of the SCR region (RCR), the βC1 ORF (C1), the promoter and A-rich region (AP), and the left side of the SCR region (LCR). (B and C) Sketches of Y10β (B), Y35β (C), and their hybrid or mutant betasatellites. Pink fragments are from Y10β, while green fragments are from Y35β.
Infectious clones of hybrid viral DNA, pBinPLUS-Y10-35rep and pBinPLUS-Y35-10rep (referred to here as Y10-35rep and Y35-10rep, respectively), were produced by the same strategy of hybrid betasatellite construction. pBinPLUS-Y10-1.7A and pBinPLUS-Y35-1.7A were used as templates for SOE-PCR. For instance, to obtain a chimeric viral DNA containing the TbCSV C1 ORF in the TYLCCNV sequence context (Y10-35rep), three independent PCRs were conducted using three pairs of primers. The primers 35AC1/F1 and 35AC1/R2 were used to amplify the fragment of the C1 ORF of TbCSV, while the primer pairs (the pair Y10/F and 35C1/R1 and the pair 35AC1/F2 and Y10/R) were used to amplify the fragments on the two sides of the C1 ORF from TYLCCNV. The three PCR products were recovered independently, and each product was mixed in the standard PCR system. After annealing and extension were completed, the primer pair Y10/F and Y10/R was added to amplify the full-length hybrid viral DNA component. The overlapping PCR products were inserted into a pGEM-T Easy vector to produce clones pGEM-Y10-35rep (in which the C1 ORF region of TYLCCNV was replaced by the TbCSV C1 ORF) and pGEM-Y35-10rep (in which the C1 ORF region of TbCSV was replaced by the TYLCCNV C1 ORF). Infectious clones of hybrid viral DNA for agroinoculation, pBinPLUS-Y10-35rep and pBinPLUS-Y35-10rep, were produced using the methods described by Cui et al. (7) and Li et al. (27), respectively.
For the construction of the recombinant protein expression vectors, the C1 ORF of TYLCCNV and TbCSV was amplified by PCR using Pfu DNA polymerase (Promega) with the primer pair Y10C1/F and Y10C1/R and the primer pair Y35C1/F and Y35C1/R, respectively, and cloned into a pGEM-T vector (Promega) to obtain pGEM-Y10rep and pGEM-Y35rep. Then, the BamHI/SalI fragments from pGEM-Y10rep and pGEM-Y35rep were inserted into the BamHI/SalI site of the expression vector pMBP-28 (Novagen, Merck KGaA, Darmstadt, Germany) to obtain pMBP-Y10rep and pMBP-Y35rep, respectively. The empty vector pMBP-28 was used to express the MBP tag alone as a control.
Total DNA extraction.
Young leaf tissue from N. benthamiana plants showing typical systemic symptoms was harvested at different days postinoculation (dpi). Viral DNA was extracted as described previously (30), stored at −20°C, and used for PCR and DNA gel blot analysis.
Virus detection by PCR and Southern blot.
Primers for PCR detection were designed on the basis of the nucleotide sequences of Y10, Y35, G18, Y10β, Y35β, and G18β. The primer pairs detY10/F and detDNA A/R, detY35/F and detDNA A/R, detG18/F and detDNA A/R, detY10β/F1 and β02, detY35β/F1 and β02, and detG18β/F and β02 were used for specific detection of Y10-, Y35-, G18-, Y10β-, Y35β-, and G18β-derived molecules, respectively. The primer pair detDNA β/F2 and detY10β/R2 was used for specific detection of Y10β-35RCR and Y10β-35βC1, while the pair detDNA β/F2 and detY35β/R2 was used to detect Y35β-10RCR and Y35β-10βC1 hybrid molecules. PCR amplification was carried out as described by Zhou et al. (30). The PCR products were also cloned, and three independent clones were sequenced to verify that recombination had not occurred and that the PCR results detected betasatellite maintenance.
DNA gel blot analysis was performed as described previously (26). The DNA probes were produced by PCR amplification using specific primer pairs, probe Y10β-AP for Y10β, Y10β-35RCR, Y10β-35βC1, Y10β-35LCR, Y10β-35RBM, and Y10β-delRBM, probe Y35β-AP for Y35β, Y35β-10RCR,Y35β-10βC1, Y35β-10LCR, Y35β-10RBM, and Y35β-delRBM, probe Y10β-βC1 for Y10β and Y10β-35AP, and probe Y35β-βC1 for Y35β and Y35β-10AP. Probes were then labeled with [α-32P]dCTP and used for hybridization. The blots were washed under highly stringent conditions (two 15-min washes in 50 ml of 0.1× SSC [1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate], 0.1% SDS at 60°C) to avoid cross-detection. Hybridization signals were detected by phosphorimaging using a Typhoon 9200 imager (Amersham Pharmacia, Piscataway, NJ, USA).
Recombinant protein expression and purification.
The MBP, MBP-Y10rep, and MBP-Y35rep expression plasmids were transformed separately into Escherichia coli strain BL21(DE3). Protein expression was induced using 0.5 mM isopropyl β-d-thiogalactopyranoside (IPTG) for 24 h at 16°C. The cells were collected by centrifugation (5,000 × g for 5 min) followed by suspension and sonication in MBP column buffer containing 20 mM Tris-Cl (pH 8.0), 200 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol (DTT) and 10% (vol/vol) glycerol. The sonicated extract was then centrifuged at 27,000 × g for 10 min, followed by incubation with amylose resin (NEB) for 2 h at 4°C. After the resin had been washed 5 times with column buffer, the proteins were eluted with column buffer containing 10 mM maltose. The eluted proteins were dialyzed to exchange into column buffer for protein binding assays. Total protein concentrations were measured using a Nanodrop instrument (Nanodrop Technologies, Wilmington, DE, USA). Protein purity was assessed by SDS-PAGE and immunoblot analysis.
DNase I footprinting assay.
The LCR of Y10β or Y35β was amplified by PCR using Y10β or Y35β DNA as the template and 5′ Alexa Fluor 680 (Invitrogen, Carlsbad, CA, USA)-labeled sense and unlabeled antisense primers. For DNase I protection assays of the virus sense strand, the 5′-labeled LCR served as a substrate for Rep binding. The binding reaction was performed in a total volume of 50 μl containing 10 μl of 5× binding buffer (125 mM Tris-Cl [pH 7.5], 25 mM MgCl2, 5 mM DTT, 1 mM EDTA, and 12.5 mM ATP), 0.5 μg poly(dI-dC), 5 ng of the end-labeled LCR DNA fragment, and MBP-Rep protein (0.5 μg to 2.5 μg) or MBP protein (2.5 μg) as a control. The reaction mixture was incubated for 45 min at room temperature. Following incubation, 50 μl of cofactor solution (10 mM MgCl2, 5 mM CaCl2) was added to the reaction mix. The DNase I enzyme dilutions were freshly made and added to each tube (0.005 U). After a 2-min incubation at room temperature, the reaction was stopped by the addition of 100 μl of stop solution (1% SDS, 200 mM NaCl, 20 mM EDTA [pH 8.0], 40 μg/ml tRNA) followed by phenol-chloroform extraction and ethanol precipitation. The pellets were resuspended in 10 μl of sequencing loading dye and loaded on a 20% polyacrylamide gel. The gel was visualized by using an Odyssey infrared imaging system (Li-Cor Biosciences, Lincoln, NE, USA) with the 700-nm channel.
EMSA.
To generate dsDNA probes, Alexa Fluor 680-labeled oligonucleotides were mixed with unlabeled complementary oligonucleotides at equal molar concentrations, incubated for 5 min at 100°C, and allowed to cool slowly to room temperature. For dsDNA binding assays, MBP-Rep proteins (3 μg) were resuspended in binding buffer (25 mM Tris-HCl [pH 7.5], 5 mM MgCl2, 0.2 mM EDTA, 1 mM DTT, 2.5 mM ATP), 100 ng of poly(dI-dC), and 20 fmol of fluorescently labeled oligonucleotide plus increasing amounts of unlabeled competitor oligonucleotide, as indicated in the figure legends. The binding reaction mixtures were incubated for 45 min at room temperature, resolved on a 1% Tris-borate-EDTA (pH 8.5) agarose gel, and visualized by using an Odyssey infrared imaging system in the 700-nm channel.
Transient-replication assays in BY-2 protoplast.
The Rep expression cassettes were constructed by insertion of a 1.1-kb BamHI-SalI Rep ORF fragment of TYLCCNV or TbCSV into the pMD18-T-based vector, which contains the 35S promoter and NOS terminator from pCHF3 vector (31). The SCR of Y10β, Y10β-35RBM, Y10β-delRBM, Y35β, Y35β-10RBM, and Y35β-delRBM were amplified by PCR and cloned into the pMD18-T vector.
Protoplasts were prepared from the N. tabacum suspension cell line BY-2 as described previously (32). The protoplasts were isolated from suspension cells 4 days after subculture by digestion for 1 h with 2% cellulase, 0.2% macerozyme, 0.2% bovine serum albumin (BSA), 1 mM CaCl2, 0.4 M mannitol, 5 mM morpholinepropanesulfonic acid (MES) (pH 5.5). The digested cells were pelleted and washed twice with 0.4 M mannitol, 5 mM MES (pH 5.5) and then resuspended in 70 mM KCl, 0.4 M mannitol, 5 mM MES (pH 5.5). Replication assays were performed by electroporation (250 V) of 5 μg replicon DNA, 10 μg Rep expression cassette DNA, and 20 μg sheared salmon sperm DNA into 2 × 106 protoplasts in a final volume of 0.4 ml. Protoplasts were diluted in 4 ml of BY-2 medium supplemented with 0.4 M mannitol, 5 mM MES (pH 5.5), and 50 μg/μl ampicillin. Total DNA was extracted 48 h after transfection, digested with DpnI, and linearized with EcoRI. Nascent viral DNA accumulation was analyzed by DNA gel blotting with a 32P-labeled DNA corresponding to Y10β or Y35β SCR.
RESULTS
Mapping of the region involved in betasatellite replication selectivity.
Previous studies showed that TYLCCNV-Y10 and TbCSV-Y35 (referred to here as Y10 and Y35, respectively) preferentially maintain their cognate satellites, TYLCCNB and TbCSB (referred to here as Y10β and Y35β, respectively). When Y10 or Y35 was coinoculated with Y10β and Y35β into N. benthamiana plants, the noncognate satellite relative to the helper virus was gradually lost at late stages of infection, while the cognate one was stably maintained (26). To determine the sequence element(s) that mediates maintenance selectivity, we partitioned the betasatellite genome into four regions, the right side of the SCR region (RCR), the βC1 coding region (C1), the promoter and A-rich region (AP), and the left side of the SCR region (LCR) (Fig. 1A). We then constructed several chimeric betasatellites, in which the corresponding parts were exchanged between the two parental satellites Y10β and Y35β (Fig. 1B and C). Four Y10β derivatives, Y10β35RCR, Y10β35C1, Y10β35AP, and Y10β35LCR, contain the RCR, C1 region, AP region, and LCR, respectively, from Y35β, as indicated in the names (Fig. 1B). Similarly, four Y35β derivatives, Y35β10RCR, Y35β10C1, Y35β10AP, and Y35β10LCR, contain the equivalent regions from Y10β (Fig. 1C).
We hypothesized that swapping of satellite regions containing sequence elements that confer replication selectivity would shift the helper virus preference of hybrid satellites. To study their replication competitiveness in relation to wild-type (wt) betasatellite, individual chimeric betasatellites were agro-inoculated together with Y10 and Y10β (Y10+Y10β) or Y35 and Y35β (Y35+Y35β) into N. benthamiana plants. As a control, noncognate wild-type satellite was also included in the in vivo competition assays, generating the inoculum combinations Y10+Y10β+Y35β and Y35+Y35β+Y10β. After inoculation, total DNA was extracted from systemic infected leaves of symptomatic plants, and PCR was used to detect the wild-type and hybrid satellites at 15, 30, and 60 dpi. The data from three independent experiments are summarized in Table 1.
TABLE 1.
PCR detection of trans replication of betasatellites in N. benthamiana plants inoculated with helper virus, together with its cognate wild-type betasatellite and hybrid betasatellite
| Inoculum and time (dpi) | No. of positive plants/totala |
|||
|---|---|---|---|---|
| Y10 | Y35 | Y10β or derivatives | Y35β or derivatives | |
| Y10+Y10β+Y35β | ||||
| 15 | 30/30 | — | 30/30 | 30/30 |
| 30 | 30/30 | — | 30/30 | 7/30 |
| 60 | 30/30 | — | 30/30 | 1/30 |
| Y10+Y10β+Y35β-10RCR | ||||
| 15 | 30/30 | — | 30/30 | 30/30 |
| 30 | 30/30 | — | 30/30 | 10/30 |
| 60 | 30/30 | — | 30/30 | 0/30 |
| Y10+Y10β+Y35β-10C1 | ||||
| 15 | 19/19 | — | 19/19 | 19/19 |
| 30 | 19/19 | — | 19/19 | 6/19 |
| 60 | 19/19 | — | 19/19 | 3/19 |
| Y10+Y10β+Y35β-10AP | ||||
| 15 | 20/20 | — | 20/20 | 20/20 |
| 30 | 20/20 | — | 20/20 | 4/20 |
| 60 | 20/20 | — | 20/20 | 0/20 |
| Y10+Y10β+Y35β-10LCR | ||||
| 15 | 29/29 | — | 29/29 | 29/29 |
| 30 | 29/29 | — | 29/29 | 29/29 |
| 60 | 29/29 | — | 29/29 | 29/29 |
| Y35+Y35β+Y10β | ||||
| 15 | — | 30/30 | 30/30 | 30/30 |
| 30 | — | 30/30 | 15/30 | 30/30 |
| 60 | — | 30/30 | 6/30 | 30/30 |
| Y35+Y35β+Y10β-35RCR | ||||
| 15 | — | 30/30 | 30/30 | 30/30 |
| 30 | — | 30/30 | 15/30 | 30/30 |
| 60 | — | 30/30 | 7/30 | 30/30 |
| Y35+Y35β+Y10β-35C1 | ||||
| 15 | — | 20/20 | 20/20 | 20/20 |
| 30 | — | 20/20 | 8/20 | 20/20 |
| 60 | — | 20/20 | 2/20 | 20/20 |
| Y35+Y35β+Y10β-35AP | ||||
| 15 | — | 19/19 | 19/19 | 19/19 |
| 30 | — | 19/19 | 8/19 | 19/19 |
| 60 | — | 19/19 | 2/19 | 19/19 |
| Y35+Y35β+Y10β-35LCR | ||||
| 15 | — | 28/28 | 28/28 | 28/28 |
| 30 | — | 28/28 | 28/28 | 28/28 |
| 60 | — | 28/28 | 28/28 | 28/28 |
Number of plants containing specified DNA/number of plants inoculated. —, no viral DNA was obtained.
As previously described (26), noncognate, wild-type satellites were initially present in all of the inoculated plants but could not be detected at later stages of infection. For example, among 30 diseased plants inoculated with Y10+Y10β+Y35β, all of the plants contained Y35β at 15 dpi, but only 7 and 1 plants had detectable levels of Y35β at 30 and 60 dpi, respectively (Table 1). Similarly, in plants coinoculated with Y35β-derived chimeras together with Y10+Y10β, all of the hybrid satellites were present in systemic tissues at 15 dpi, indicating that all of the chimeras can replicate and move systemically in the presence of the helper virus Y10. However, at 30 and 60 dpi, hybrid satellites Y35β-10RCR, Y35β-10C1, and Y35β-10AP were outcompeted by Y10β and became less prevalent in the population. Thus, these chimeras maintained Y35β-like replication properties in terms of helper virus preference. In contrast, hybrid satellite Y35β-10LCR was stably maintained by Y10 even at late infection stages, behaving like Y10β. Note that in all cases, the cognate satellite (Y10β) persisted throughout infection. In the reciprocal experiment, Y10-derived hybrid satellites were coinoculated with Y35 and Y35β onto N. benthamiana plants. PCR-based detections showed that Y10β-35LCR was maintained along with Y35β, while the rest of hybrid satellites (Y10β-35RCR, Y10β-35C1, and Y10β-35AP) were gradually lost at late stages of infection (Table 1).
To confirm the PCR results, DNA accumulation of wild-type or chimeric satellites at 60 dpi was analyzed in plants with mixed infections on DNA gel blots. Probes corresponding to divergent sequences in the AP region were used to specifically detect Y10β and its derivatives (Fig. 2A, top) or Y35β and its derivatives (Fig. 2A, bottom). Fragments of the respective βC1 genes were also used as specific probes (Fig. 2B). The blots were washed under highly stringent conditions to avoid cross-detection. The specificities of the probes were verified in lanes 1 to 4 of Fig. 2A, which shows that the Y10β-specific probe does not cross-react with Y35β and vice versa. The DNA gel blots confirmed that noncognate satellites were poorly maintained by helper viruses at 60 dpi when cognate satellites were present (Fig. 2A, lanes 5 to 10). Hybrid satellites containing heterogeneous RCR (Fig. 2A, lanes 15 to 18), C1 (Fig. 2A, lanes 23 to 26), and AP (Fig. 2B, lanes 9 to 12) regions are also poorly maintained by helper viruses. Only when the LCR regions are exchanged, both cognate and hybrid satellites containing cognate LCR are present (Fig. 2A, lanes 31 to 34). Together, these data established that the LCR regions of Y10β and Y35β contain cis elements that confer preferential maintenance by cognate helper viruses.
FIG 2.
DNA gel blot analysis of betasatellite DNA accumulation in N. benthamiana plants coagroinoculated with betasatellites and Y10 or Y35. Total nucleic acids (10 μg) were extracted at 60 dpi from plants showing systemic symptoms. The AP region (A) or βC1 (B) of either Y10β (top) or Y35β (bottom) were used as a probe.
To further confirm the function of LCR region in the selective replication of betasatellites, pairs of reciprocal hybrid betasatellites were coinoculated with either Y10 or Y35 onto N. benthamiana plants. We anticipated that swapping LCR regions would shift the replication predominance of hybrid betasatellites in mixed infections. Indeed, Y10 maintains the cognate LCR-containing satellites regardless of the sequence context (Table 2; Fig. 2A, lanes 27 and 28), as does the Y35 (Table 2; Fig. 2A, lanes 29 and 30). In contrast, exchanging other regions of the two betasatellites does not alter helper virus preference (Table 2; Fig. 2A, lanes 11 to 14 and 19 to 22; Fig. 2B, lanes 5 to 8). These results convincingly showed that the LCR is essential and sufficient to confer cognate helper virus-mediated preferences.
TABLE 2.
PCR detection of trans replication of betasatellites in N. benthamiana plants inoculated with helper virus and two hybrid betasatellites
| Inoculum and time (dpi) | No. of positive plants/totala |
|||
|---|---|---|---|---|
| Y10 | Y35 | Y10β or derivatives | Y35β or derivatives | |
| Y10+Y10β-35RCR+Y35β-10RCR | ||||
| 15 | 29/29 | — | 29/29 | 29/29 |
| 30 | 29/29 | — | 29/29 | 10/30 |
| 60 | 29/29 | — | 29/29 | 5/30 |
| Y35+Y35β-10RCR+Y10β-35RCR | ||||
| 15 | — | 26/26 | 26/26 | 26/26 |
| 30 | — | 26/26 | 11/26 | 26/26 |
| 60 | — | 26/26 | 7/26 | 26/26 |
| Y10+Y10β-35C1+Y35β-10C1 | ||||
| 15 | 20/20 | — | 20/20 | 20/20 |
| 30 | 20/20 | — | 20/20 | 7/20 |
| 60 | 20/20 | — | 20/20 | 2/20 |
| Y35+Y35β-10C1+Y10β-35C1 | ||||
| 15 | — | 19/19 | 19/19 | 19/19 |
| 30 | — | 19/19 | 7/19 | 19/19 |
| 60 | — | 19/19 | 3/19 | 19/19 |
| Y10+Y10β-35AP+Y35β-10AP | ||||
| 15 | 20/20 | — | 20/20 | 20/20 |
| 30 | 20/20 | — | 20/20 | 7/20 |
| 60 | 20/20 | — | 20/20 | 1/20 |
| Y35+Y35β-10AP+Y10β-35AP | ||||
| 15 | — | 21/21 | 21/21 | 21/21 |
| 30 | — | 21/21 | 9/21 | 21/21 |
| 60 | — | 21/21 | 5/21 | 21/21 |
| Y10+Y10β-35LCR+Y35β-10LCR | ||||
| 15 | 25/25 | — | 25/25 | 25/25 |
| 30 | 25/25 | — | 12/25 | 25/25 |
| 60 | 25/25 | — | 5/25 | 25/25 |
| Y35+Y35β-10LCR+Y10β-35LCR | ||||
| 15 | — | 27/27 | 27/27 | 27/27 |
| 30 | — | 27/27 | 27/27 | 9/27 |
| 60 | — | 27/27 | 27/27 | 3/27 |
Number of plants containing the specified DNA/number of plants inoculated. —, no viral DNA was obtained.
The Reps of helper viruses determine the maintenance selectivity.
Rep is the only geminivirus-encoded protein that is essential for viral replication (33). We thus constructed two hybrid helper viruses, Y10-35rep and Y35-10rep, in which the Rep genes were swapped between Y10 and Y35. Each hybrid helper was coinoculated with Y10β and Y35β into N. benthamiana plants. At 7 dpi, most of inoculated plants displayed symptoms characteristic of geminivirus infection, such as leaf curling, indicating that both hybrid viruses are viable (data not shown). Detection of betasatellites by PCR and DNA gel blotting showed that hybrid viruses preferentially maintain the satellite that is cognate to their Reps, i.e., Y10-35rep prefers Y35β and Y35-10rep prefers Y10β (Table 3; Fig. 2A, lanes 3 to 38). Thus, we conclude that the Reps of Y10 and Y35 are involved in the specificity for their cognate betasatellites.
TABLE 3.
PCR detection of trans replication of betasatellites in N. benthamiana plants inoculated with two betasatellites and hybrid helper virus
| Inoculum and time (dpi) | No. of positive plants/totala |
|||
|---|---|---|---|---|
| Y10-35rep | Y35-10rep | Y10β or derivatives | Y35β or derivatives | |
| Y10-35rep+Y10β+Y35β | ||||
| 15 | 20/20 | — | 20/20 | 20/20 |
| 30 | 20/20 | — | 15/20 | 20/20 |
| 60 | 20/20 | — | 6/20 | 20/20 |
| Y35-10rep+Y35β+Y10β | ||||
| 15 | — | 20/20 | 20/20 | 20/20 |
| 30 | — | 20/20 | 20/20 | 12/20 |
| 60 | — | 20/20 | 20/20 | 3/20 |
Number of plants containing the specified DNA/number of plants inoculated. —, no viral DNA was obtained.
Identification of sequence motifs specifically bound to helper-encoded Reps.
Begomovirus Rep binds to viral dsDNA intermediates in a sequence-specific manner, and this activity is essential for viral replication (21, 34). To investigate specific binding between Rep and betasatellite DNA sequences, we expressed MBP-tagged Y10 Rep (MBP-Y10Rep) and Y35 Rep (MBP-Y35Rep) in E. coli and purified the recombinant proteins. DNase I footprinting assays were carried out to identify the specific Rep-binding site(s) in the LCR region. The Y10β LCR (276 nucleotides) (Fig. 3A) and the Y35β LCR (272 nucleotides) (Fig. 3B) DNA fragments labeled at the 5′ ends of the sense strand were used as substrates for Y10 Rep or Y35 Rep binding in the DNase I protection assay. Labeled LCR dsDNAs were incubated with the MBP tag alone (as a control) or with increasing amounts of the Rep proteins followed by DNase I digestion. The locations of the protected sites were determined by comparing digestion products with a labeled DNA sequencing ladder run in parallel on denaturing gels. The results revealed that both Y10 Rep and Y35 Rep proteins could bind to an approximately 30-nucleotide (nt) region of the 5′ terminus (located at positions 1050 to 1080 in the Y10β genome and 1068 to 1098 in the Y35β genome) of either cognate or noncognate LCR (Fig. 3A and B).
FIG 3.
Footprinting analysis of Rep-binding sites in the LCR of Y10β (A) or Y35β (B). 5′ Alexa Fluor 680-labeled, sense-strand LCR dsDNA was used as a substrate. The labeled LCR DNA was incubated with increasing amounts (0.5 μg, 1 μg, and 2.5 μg) of MBP-Rep protein or MBP protein (2.5 μg) as a control followed by DNase I digestion. The size markers, run in parallel, are shown on the left. (C) The sequences of the protected regions from Y10β and Y35β are compared. The nucleotides in red indicate the direct repeats in the Rep-binding motifs.
We named the 30-nt region the Rep-binding motif (RBM). Analyses of RBM sequence showed that there are two direct repeats in the 30-nt region (GGGACACC in Y10β and GAGGACC in Y35β), separated by a 10-nt (in Y10β) or 12-nt (in Y35β) interval sequence (Fig. 3C). Next, we used EMSA to confirm that the 30-nt RBMs specifically bind to the Reps. Fluorescently labeled RBM oligonucleotides were chemically synthesized and annealed to produce the dsDNA probes. As shown in Fig. 4A, the recombinant MBP-Y10-Rep fusion protein bound efficiently to its cognate RBM probes, resulting in two retarded bands compared to the free probes (Fig. 4A, lanes 1 to 3). In competition binding assays, the addition of 10-, 100-, and 500-fold molar excesses of GFP dsDNA only marginally affected the amount of upper shifted bands (Fig. 4A, lanes 4 to 6). Note that the lower shifted bands were dissociated following addition of the GFP competitor (Fig. 4A, lanes 4 to 6), suggesting that this Rep-RBM complex is less stable. However, the addition of increasing amounts of unlabeled Y10-RMB competitors effectively dissociated the Y10 Rep-probe binding (Fig. 4A, lanes 7 to 9), indicating that the binding of Y10 Rep to Y10 RBM is specific. Similarly, Y35 Rep also bound to its cognate RMB sequence efficiently and specifically (Fig. 4B, lanes 1 to 9). These data confirmed that the RBMs in Y10β and Y35β were sequence-specific binding sites for the cognate Reps.
FIG 4.
EMSA of viral Rep protein and betasatellite RBM probe. Binding reaction mixtures contained the pair Y10 Rep and Y10β RBM probe (A), Y35 Rep and Y35β RBM probe (B), Y10 Rep and Y35β RBM probe (C), or Y35 Rep and Y10β RBM probe (D). All reactions were resolved on 1% agarose gels and visualized using an Odyssey infrared imaging system to compare relative binding efficiencies. Increasing amounts of dsDNA competitor were 10, 100, and 500 times the amounts of the probes (molar ratio). For quantification, we set the background of the reaction without protein as 0 (lane 1) and the binding band of the reaction without any competitors as 100% (lane 3) and then standardize the binding bands in all reactions. Results are representative of at least three independent experiments.
Since Y10 and Y35 can also replicate the noncognate betasatellites although their cognate ones were favored during mixed infections (Fig. 2A) (21), we then analyzed whether these two Reps could also bind efficiently to noncognate RBM. Indeed, EMSA showed that Y10 Rep and Y35 Rep bound efficiently to their noncognate Y35β (Fig. 4C, lanes 1 to 3) and Y10 β (Fig. 4D, lanes 1 to 3) RBMs, respectively. Again, excess GFP DNA competed poorly for binding, suggesting that Rep–noncognate-RMB binding is also specific (Fig. 4C and D, lanes 4 to 6).
Rep binds more strongly to its cognate betasatellite RBM than to a noncognate RBM.
To investigate whether Rep-mediated preferential replication of a cognate betasatellite in vivo correlates with differential Rep-RBM binding in vitro, we compared the binding affinities of Rep to cognate and noncognate RBMs. For this purpose, unlabeled Y35 RBM or Y10 RBM dsDNAs were used as competitors in binding assays to evaluate their ability to compete for Y10 Rep bound to the Y35β RBM probe. Both competitors efficiently dissociate Y10 Rep-probe binding when present in excess (10-, 100- and 500-fold molar excess over Y35β RBM probe) (Fig. 4C, lanes 7 to 12), consistent with the finding that Y10 Rep binds specifically to both RBMs. Quantitative analyses showed that Y10β RBM competes for Y10 Rep to a greater degree than Y35β RBM (Fig. 4C, bottom, lanes 10 to 12 versus 7 to 9). In contrast, Y35β RBM could not compete for Y10 Rep binding to its cognate (Y10β) RBM (Fig. 4A, lanes 10 to 12). Surprisingly, we consistently observed that addition of excess of Y35β facilitated specific Y10 Rep-Y10β RBM binding (Fig. 4A, lanes 10 to 12). One possibility is that Y35β RBM promotes the formation of stable complexes (upper shifted bands) by dissociating unstable Y10 Rep-Y10β RBM complexes (lower shifted bands). Independent of the mechanism, our data indicate that the binding affinity of Y10 Rep for its cognate (Y10β) RBM is higher than its affinity for the noncognate (Y35β) RMB. A similar conclusion can be drawn for Y35 Rep, which displayed stronger binding to the Y35β RBM than the Y10β RBM in competition assays (Fig. 4B and D, lanes 7 to 12). Thus, these data show that Reps bind to their cognate RBM more strongly than to a noncognate RBM.
The betasatellite RBMs confer preferential maintenance in vivo.
To determine whether the RBM contributes to satellite competitiveness in vivo, we constructed two chimeric betasatellite clones, Y10β-35RBM and Y35β-10RBM, in which the RBM was exchanged between Y10β and Y35β (Fig. 1B and C). N. benthamiana plants were agroinoculated with helper virus together with its cognate wild-type betasatellite and the hybrid betasatellite containing cognate RBM. In the Y10β and Y35β-10RBM coinoculated plants; both betasatellites were maintained by the Y10 helper virus at all time points tested (15, 30, and 60 dpi) (Table 4; Fig. 5, lanes 11 to 13). Similarly, Y10β-35RBM also copersists with Y35β throughout infection in the presence of the Y35 helper virus (Table 4; Fig. 5, lanes 14 to 16). When two RBM hybrid betasatellites (Y10β-35RBM and Y35β-10RBM) were coinoculated with either Y10 or Y35 onto N. benthamiana plants, in both cases, only the hybrid betasatellite with the cognate RBM relative to the given helper virus was maintained at a late stage of infection (Table 4; Fig. 5, lanes 5 to 10). These results indicate that switching the RBM sequence is sufficient to reverse the helper virus preference during infections.
TABLE 4.
PCR detection of trans replication of betasatellites in N. benthamiana plants inoculated with helper virus together with its cognate wild-type betasatellite and RBM hybrid betasatellite or with two RBM hybrid betasatellites
| Inoculum and time (dpi) | No. of positive plants/totala |
|||
|---|---|---|---|---|
| Y10 | Y35 | Y10β or derivatives | Y35β or derivatives | |
| Y10+Y10β+Y35β-10RBM | ||||
| 15 | 25/25 | — | 25/25 | 25/25 |
| 30 | 25/25 | — | 25/25 | 25/25 |
| 60 | 25/25 | — | 25/25 | 25/25 |
| Y35+Y35β+Y10β-35RBM | ||||
| 15 | — | 25/25 | 25/25 | 25/25 |
| 30 | — | 25/25 | 25/25 | 25/25 |
| 60 | — | 25/25 | 25/25 | 25/25 |
| Y10+Y10β-35RBM+Y35β-10RBM | ||||
| 15 | 25/25 | — | 25/25 | 25/25 |
| 30 | 25/25 | — | 10/25 | 25/25 |
| 60 | 25/25 | — | 5/25 | 25/25 |
| Y35+Y35β-10RBM+Y10β-35RBM | ||||
| 15 | — | 25/25 | 25/25 | 25/25 |
| 30 | — | 25/25 | 25/25 | 13/25 |
| 60 | — | 25/25 | 25/25 | 8/25 |
Number of plants containing the specified DNA/number of plants inoculated. —, no viral DNA was obtained.
FIG 5.
DNA gel blot analysis of betasatellite DNA accumulation in N. benthamiana plants coagroinoculated with betasatellites and Y10 or Y35. Total nucleic acids (10 μg) were extracted at 60 dpi from plants showing systemic symptoms. The AP region of Y10β (top) or Y35β (bottom) was used as the probe.
We previously reported a different tomato-infecting begomovirus isolate, tomato leaf curl China virus (ToLCCNV-G18), in association with the betasatellite (G18β). Sequence alignment showed that G18β shared only 60.3% overall nucleotide sequence homology to Y10β (data not shown), while the postulated RBM sequence of G18β is similar with that of Y10β. G18β shares an identical RBM sequence (GGGACACC) with Y10β in the first repeat, but the second repeat contains one substitution (GAGACACC). To further test the role of RBM in determining helper virus preference, G18 or Y10 was used as the helper to support the mixed infection of G18β and Y10β in N. benthamiana plants. As expected, both helper viruses can stably maintain the cognate and noncognate betasatellites simultaneously (Table 5), indicating that their Reps cannot discriminate between G18β and Y10β. In contrast, when mixtures of G18β and Y35β were coinoculated with G18 or Y35 onto plants, only the cognate betasatellite was maintained by a given helper virus at late stages of infection (Table 5). These data further highlighted the important role of the RBM in conferring selective maintenance of begomovirus-associated betasatellites.
TABLE 5.
PCR detection of trans replication of betasatellites in N. benthamiana plants inoculated with two satellites and one helper virus
| Inoculum and time (dpi) | No. of positive plants/totala |
|||
|---|---|---|---|---|
| Y10 | G18A | Y10β | G18β | |
| Y10+Y10β+G18β | ||||
| 15 | 15/15 | — | 15/15 | 15/15 |
| 30 | 15/15 | — | 15/15 | 15/15 |
| 60 | 15/15 | — | 15/15 | 15/15 |
| G18+ G18β+Y10β | ||||
| 15 | — | 15/15 | 15/15 | 15/15 |
| 30 | — | 15/15 | 15/15 | 15/15 |
| 60 | — | 15/15 | 15/15 | 15/15 |
| Y35+Y35β+G18β | Y35 | G18A | Y35β | G18β |
| 15 | 15/15 | — | 15/15 | 15/15 |
| 30 | 15/15 | — | 15/15 | 6/15 |
| 60 | 15/15 | — | 15/15 | 3/15 |
| G18+G18β+Y35β | ||||
| 15 | — | 15/15 | 15/15 | 15/15 |
| 30 | — | 15/15 | 3/15 | 15/15 |
| 60 | — | 15/15 | 0/15 | 15/15 |
Number of plants containing the specified DNA/number of plants inoculated. —, no viral DNA was obtained.
Strong Rep binding to a cognate RBM is required for efficient betasatellite replication in protoplasts.
The correlation between Rep-RBM interactions and betasatellite maintenance suggests that differential replication mediates the selective loss of a noncognate versus a cognate betasatellite in planta. However, infection studies in whole plants, like those described above, do not separate replication from encapsidation or movement, all of which could impact betasatellite maintenance. Thus, we used transient replication assays in BY-2 protoplasts to examine the relationship between Rep-RBM binding and replication without the complications of other viral processes. We produced plasmid-based replicon DNAs, in which an ∼500-fragment consisting of the satellite conserved region (SCR) was amplified from wild-type or hybrid satellites and inserted into the bacterial cloning vector pMD18-T. The SCR encompasses the LCR and RCR and presumably contains all the sequence elements necessary for trans replication (25). The replicon DNAs were coelectroporated into BY-2 protoplasts with a binary Agrobacterium plasmid that expresses Y10 Rep or Y35 Rep driven by the 35S promoter of cauliflower mosaic virus. Total DNA was isolated from the transfected cells and digested with DpnI, to distinguish nascent DNA from input bacterial DNA. No hybridization signal corresponding to nascent DNA was detected in the absence of Rep (Fig. 6A and B, lanes 1 and 4), indicating that Rep is required for betasatellite DNA replication. Both Y10 Rep and Y35 Rep supported replication of Y10β SCR and Y35β SCR-containing replicons. Comparison of the levels of nascent products indicated that both Y10 Rep and Y35 Rep replicated the cognate CR-containing replicon more efficiently than the noncognate one (Fig. 6A and B, lanes 2 and 3). Exchanging the RBMs' sequences reversed the Rep preference (Fig. 6A and B, lanes 5 and 6), demonstrating that the RBM determines the differential replication efficiency. Hence, the differential interactions of begomovirus to cognate and noncognate betasatellites likely occur at the level of replication. The Rep-RMB binding affinity correlates with satellite replication efficiency in protoplast and preferential interaction with helper viruses in vivo.
FIG 6.
Transient replication of Y10β (A) and Y35β (B) in tobacco BY-2 protoplasts. The protoplasts were coelectroporated with a wild-type or RBM mutant DNA β and Rep protein. Total DNA was isolated 48 h postelectroporation, digested with DpnI, and linearized with EcoRI. Nascent betasatellite DNA accumulation was analyzed on DNA gel blots using a radiolabeled Y10β or Y35β SCR probe. Results are representative of at least three independent experiments.
The involvement of the RMB in betasatellite replication was also investigated by means of transient-replication assays using RMB deletion constructs. Deletion of RMB sequence rendered the replicons nonfunctional in replication (Fig. 6A and B, lanes 7 to 9). Infectious clones of betasatellites with the RMB deleted also failed to accumulate in planta or to induce the typical betasatellite-associated symptoms when coinoculated with helper viruses (Table 6). In conclusion, our data showed that high-affinity binding of Rep to RBM is required for efficient betasatellite replication.
TABLE 6.
PCR detection of trans replication of betasatellites in N. benthamiana plants coinoculated with helper virus and RBM deletion betasatellites
| Inoculum and time (dpi) | No. of positive plants/totala |
|||
|---|---|---|---|---|
| Y10 | Y35 | Y10βdelRBM | Y35βdelRBM | |
| Y10+ Y10β-delRBM | ||||
| 15 | 10/10 | — | 0/10 | — |
| 30 | 10/10 | — | 0/10 | — |
| Y35+ Y10β-delRBM | ||||
| 15 | — | 10/10 | 0/10 | — |
| 30 | — | 10/10 | 0/10 | — |
| Y35+ Y35β-delRBM | ||||
| 15 | — | 10/10 | — | 0/10 |
| 30 | — | 10/10 | — | 0/10 |
| Y10+ Y35β-delRBM | ||||
| 15 | 10/10 | — | — | 0/10 |
| 30 | 10/10 | — | — | 0/10 |
Number of plants containing the specified DNA/number of plants inoculated. —, no viral DNA was obtained.
DISCUSSION
Although betasatellites completely rely on their helper virus-encoded Rep for replication, they differ from DNA B in that they display a less-stringent requirement for interaction with Rep, which is encoded on DNA A (35–38). For example, the AYVV betasatellite and CLCuMV betasatellite can be replicated by CLCuMV and AYVV, respectively (25, 39). However, the molecular basis of promiscuous replication of betasatellites remains uncharacterized. In this study, we mapped a region upstream of the conserved rolling-circle cruciform structure in the betasatellite origin (an approximately 260-nt fragment) that confers a cognate Rep-mediated replication advantage over noncognate satellites. This advantage was determined by the begomovirus-encoded Rep (Table 3), which specifically binds to a sequence motif (RBM) within the SCR and contributes to differential replication of cognate over noncognate betasatellites.
Previously, several sequence motifs have been implicated in betasatellite replication. Eini et al. found that a G-box motif located 143 nt upstream of the βC1 start codon is required for efficient replication of CLCuMB (40). Studies of tomato leaf curl virus (TLCV) defective satellite DNA showed that an approximately 330-nt region, including the conserved nonanucleotide sequence TAATATTAC, is essential for replication (41). Notably, this region aligns well with the SCR of AYVB, which has been shown to be essential for betasatellite replication (25). The involvement of the SCR in betasatellite replication was also confirmed in our transient-replication assay in BY-2 protoplasts (Fig. 6). According to the proposed model of geminivirus rolling-circle replication, binding of viral Rep to the replication origin of the double-stranded (dsDNA) genome is a key step in the initiation of viral DNA replication. Rep specifically binds to iterative sequences, or “iterons,” located upstream of potential stem-loop structures of its cognate genome (21, 34, 42, 43), catalyzes a cleavage reaction in the conserved nonanucleotide sequence within the origin (18, 21), and ligates DNA at the conserved hairpin structure (18, 19). However, unlike DNA B, betasatellites generally lack significant sequence homology with their helper virus, and helper virus-related iteron sequences are frequently not found in betasatellites. It is unclear how Rep mediates origin recognition and trans replication of betasatellite DNA. Studies with ToLCV-sat identified two high-affinity Rep-binding sites (GGTGTCT) upstream of the conserved rolling circle cruciform structure which are identical to the ToLCV iteron sequence but occur in an inverted orientation (24, 44, 45). Surprisingly, mutation of both motifs does not completely abolish the replication of ToLCV or of ToLCV-sat (24). These data suggest that high-affinity Rep replication origin binding is not required for ToLCV-sat replication, consistent with its promiscuous replication by diverse helper viruses, such as African cassava mosaic virus and beet curly top virus, that recognize different iteron sequences (44). However, it should be noted that both ToLCV and ToLCV-sat Rep binding site mutants accumulated at greatly reduced levels in infected plants compared to wild-type clones (24), suggesting that high-affinity Rep binding is essential for efficient replication. We identified a specific Rep-binding motif (RBM) in TYLCCNB and TbCSB that are essential for betasatellite replication in protoplasts. The RBM contains two direct repeats and is located in a position analogous to the putative iterons of ToLCV-sat and AYVB (24, 25). Interestingly, the RBM binds specifically to both its cognate Rep and to a noncognate Rep, albeit with lower affinity (Fig. 4). The binding affinities for Reps correlate with the replication efficiencies in protoplasts (Fig. 6) and preferential maintenance of the betasatellites in mixed infection (Fig. 2; Tables 1 to 4). Our data are inconsistent with a model in which high-affinity Rep binding is essential for efficient replication and stable maintenance of the betasatellite in plants. Hence, the association of betasatellites with monopartite begomoviruses, although with less stringent interaction, is similar to that of DNA A and DNA B of bipartite begomoviruses.
Bipartite begomovirus-encoded Reps usually are highly specific for their cognate DNAs, and reassortants resulting from mixed infection are often not viable or less pathogenic than the parental viruses (46, 47). In contrast, betasatellites can be replicated by diverse helper viruses. However, despite the promiscuous replication, coexistence of two distinct betasatellites with one helper virus is rarely reported in nature, indicating that competition exist between coinfected satellites. Phylogenetic analyses have shown a high correlation between the variability of the betasatellite molecules and their cognate helper viruses (30), indicating betasatellites have coevolved with their helper virus, and reassortment events are not prevalent. Our in vitro Rep binding assays and in vivo competition experiments suggested that betasatellites have adapted to their cognate Rep for high-affinity binding and efficient replication, thereby gaining maintenance advantage over noncognate satellites.
ACKNOWLEDGMENTS
We thank Ruiqiang Ye from National Institute of Biological Science, Beijing, People's Republic of China, for technical support in BY-2 cell culture.
REFERENCES
- 1.Varsani A, Navas-Castillo J, Moriones E, Hernández-Zepeda C, Idris A, Brown J, Murilo Zerbini F, Martin D. 2014. Establishment of three new genera in the family Geminiviridae: Becurtovirus, Eragrovirus and Turncurtovirus. Arch Virol 159:2193–2203. doi: 10.1007/s00705-014-2050-2. [DOI] [PubMed] [Google Scholar]
- 2.Briddon RW, Mansoor S, Bedford ID, Pinner MS, Saunders K, Stanley J, Zafar Y, Malik KA, Markham PG. 2001. Identification of DNA components required for induction of cotton leaf curl disease. Virology 285:234–243. doi: 10.1006/viro.2001.0949. [DOI] [PubMed] [Google Scholar]
- 3.Saunders K, Bedford ID, Yahara T, Stanley J. 2003. Aetiology: the earliest recorded plant virus disease. Nature 422:831. doi: 10.1038/422831a. [DOI] [PubMed] [Google Scholar]
- 4.Saunders K, Bedford ID, Briddon RW, Markham PG, Wong SM, Stanley J. 2000. A unique virus complex causes Ageratum yellow vein disease. Proc Natl Acad Sci U S A 97:6890–6895. doi: 10.1073/pnas.97.12.6890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jose J, Usha R. 2003. Bhendi yellow vein mosaic disease in India is caused by association of a DNA beta satellite with a begomovirus. Virology 305:310–317. doi: 10.1006/viro.2002.1768. [DOI] [PubMed] [Google Scholar]
- 6.Zhou X. 2013. Advances in understanding begomovirus satellites. Annu Rev Phytopathol 51:357–381. doi: 10.1146/annurev-phyto-082712-102234. [DOI] [PubMed] [Google Scholar]
- 7.Cui X, Tao X, Xie Y, Fauquet CM, Zhou X. 2004. A DNAβ associated with Tomato yellow leaf curl China virus is required for symptom induction. J Virol 78:13966–13974. doi: 10.1128/JVI.78.24.13966-13974.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yang JY, Iwasaki M, Machida C, Machida Y, Zhou X, Chua NH. 2008. βC1, the pathogenicity factor of TYLCCNV, interacts with AS1 to alter leaf development and suppress selective jasmonic acid responses. Gene Dev 22:2564–2577. doi: 10.1101/gad.1682208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhang T, Luan JB, Qi JF, Huang CJ, Li M, Zhou XP, Liu SS. 2012. Begomovirus-whitefly mutualism is achieved through repression of plant defences by a virus pathogenicity factor. Mol Ecol 21:1294–1304. doi: 10.1111/j.1365-294X.2012.05457.x. [DOI] [PubMed] [Google Scholar]
- 10.Shen Q, Liu Z, Song F, Xie Q, Hanley-Bowdoin L, Zhou X. 2011. Tomato SlSnRK1 protein interacts with and phosphorylates betaC1, a pathogenesis protein encoded by a geminivirus beta-satellite. Plant Physiol 157:1394–1406. doi: 10.1104/pp.111.184648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Li R, Weldegergis BT, Li J, Jung C, Qu J, Sun Y, Qian H, Tee C, van Loon JJA, Dicke M, Chua N-H, Liu S-S, Ye J. 2014. Virulence factors of geminivirus interact with MYC2 to subvert plant resistance and promote vector performance. Plant Cell 26:4991–5008. doi: 10.1105/tpc.114.133181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yang X, Xie Y, Raja P, Li S, Wolf JN, Shen Q, Bisaro DM, Zhou X. 2011. Suppression of methylation-mediated transcriptional gene silencing by βC1-SAHH protein interaction during geminivirus-betasatellite infection. PLoS Pathog 7:e1002329. doi: 10.1371/journal.ppat.1002329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Li F, Huang C, Li Z, Zhou X. 2014. Suppression of RNA silencing by a plant DNA virus satellite requires a host calmodulin-like protein to repress RDR6 expression. PLoS Pathog 10:e1003921. doi: 10.1371/journal.ppat.1003921. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gutierrez C, Ramirez-Parra E, Mar Castellano M, Sanz-Burgos AP, Luque A, Missich R. 2004. Geminivirus DNA replication and cell cycle interactions. Vet Microbiol 98:111–119. doi: 10.1016/j.vetmic.2003.10.012. [DOI] [PubMed] [Google Scholar]
- 15.Hanley-Bowdoin L, Settlage SB, Robertson D. 2004. Reprogramming plant gene expression: a prerequisite to geminivirus DNA replication. Mol Plant Pathol 5:149–156. doi: 10.1111/j.1364-3703.2004.00214.x. [DOI] [PubMed] [Google Scholar]
- 16.Hanley-Bowdoin L, Bejarano ER, Robertson D, Mansoor S. 2013. Geminiviruses: masters at redirecting and reprogramming plant processes. Nat Rev Microbiol 11:777–788. doi: 10.1038/nrmicro3117. [DOI] [PubMed] [Google Scholar]
- 17.Heyraud-Nitschke F, Schumacher S, Laufs J, Schaefer S, Schell J, Gronenborn B. 1995. Determination of the origin cleavage and joining domain of geminivirus Rep proteins. Nucleic Acids Res 23:910–916. doi: 10.1093/nar/23.6.910. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Laufs J, Traut W, Heyraud F, Matzeit V, Rogers SG, Schell J, Gronenborn B. 1995. In vitro cleavage and joining at the viral origin of replication by the replication initiator protein of tomato yellow leaf curl virus. Proc Natl Acad Sci U S A 92:3879–3883. doi: 10.1073/pnas.92.9.3879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Orozco BM, Hanley-Bowdoin L. 1996. A DNA structure is required for geminivirus replication origin function. J Virol 70:148–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Gutierrez C. 1999. Geminivirus DNA replication. Cell Mol Life Sci 56:313–329. doi: 10.1007/s000180050433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Fontes EP, Eagle PA, Sipe PS, Luckow VA, Hanley-Bowdoin L. 1994. Interaction between a geminivirus replication protein and origin DNA is essential for viral replication. J Biol Chem 269:8459–8465. [PubMed] [Google Scholar]
- 22.Mansoor S, Briddon RW, Bull SE, Bedford ID, Bashir A, Hussain M, Saeed M, Zafar Y, Malik KA, Fauquet C, Markham PG. 2003. Cotton leaf curl disease is associated with multiple monopartite begomoviruses supported by single DNA beta. Arch Virol 148:1969–1986. doi: 10.1007/s00705-003-0149-y. [DOI] [PubMed] [Google Scholar]
- 23.Saunders K, Salim N, Mali VR, Malathi VG, Briddon R, Markham PG, Stanley J. 2002. Characterisation of Sri Lankan cassava mosaic virus and Indian cassava mosaic virus: evidence for acquisition of a DNA B component by a monopartite begomovirus. Virology 293:63–74. doi: 10.1006/viro.2001.1251. [DOI] [PubMed] [Google Scholar]
- 24.Lin B, Akbar Behjatnia SA, Dry IB, Randles JW, Rezaian MA. 2003. High-affinity Rep-binding is not required for the replication of a geminivirus DNA and its satellite. Virology 305:353–363. doi: 10.1006/viro.2002.1671. [DOI] [PubMed] [Google Scholar]
- 25.Saunders K, Briddon RW, Stanley J. 2008. Replication promiscuity of DNA-beta satellites associated with monopartite begomoviruses; deletion mutagenesis of the Ageratum yellow vein virus DNA-beta satellite localizes sequences involved in replication. J Gen Virol 89:3165–3172. doi: 10.1099/vir.0.2008/003848-0. [DOI] [PubMed] [Google Scholar]
- 26.Qing L, Zhou X. 2009. Trans-replication of, and competition between, DNAβ satellites in plants inoculated with Tomato yellow leaf curl China virus and Tobacco curly shoot virus. Phytopathology 99:716–720. doi: 10.1094/PHYTO-99-6-0716. [DOI] [PubMed] [Google Scholar]
- 27.Li Z, Xie Y, Zhou X. 2005. Tobacco curly shoot virus DNAbeta is not necessary for infection but intensifies symptoms in a host-dependent manner. Phytopathology 95:902–908. doi: 10.1094/PHYTO-95-0902. [DOI] [PubMed] [Google Scholar]
- 28.Yang X, Guo W, Ma X, An Q, Zhou X. 2011. Molecular characterization of tomato leaf curl China virus, infecting tomato plants in China, and functional analyses of its associated betasatellite. Appl Environ Microbiol 77:3092–3101. doi: 10.1128/AEM.00017-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ding C, Qing L, Li Z, Liu Y, Qian Y, Zhou X. 2009. Genetic determinants of symptoms on viral DNA satellites. Appl Environ Microbiol 75:5380–5389. doi: 10.1128/AEM.01193-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zhou X, Xie Y, Tao X, Zhang Z, Li Z, Fauquet CM. 2003. Characterization of DNA beta associated with begomoviruses in China and evidence for co-evolution with their cognate viral DNA-A. J Gen Virol 84:237–247. doi: 10.1099/vir.0.18608-0. [DOI] [PubMed] [Google Scholar]
- 31.Cai XZ, Zhou X, Xu YP, Joosten MH, de Wit PJ. 2007. Cladosporium fulvum CfHNNI1 induces hypersensitive necrosis, defence gene expression and disease resistance in both host and nonhost plants. Plant Mol Biol 64:89–101. doi: 10.1007/s11103-007-9136-0. [DOI] [PubMed] [Google Scholar]
- 32.Qi Y, Ding B. 2002. Replication of Potato spindle tuber viroid in cultured cells of tobacco and Nicotiana benthamiana: the role of specific nucleotides in determining replication levels for host adaptation. Virology 302:445–456. doi: 10.1006/viro.2002.1662. [DOI] [PubMed] [Google Scholar]
- 33.Nash TE, Dallas MB, Reyes MI, Buhrman GK, Ascencio-Ibanez JT, Hanley-Bowdoin L. 2011. Functional analysis of a novel motif conserved across geminivirus Rep proteins. J Virol 85:1182–1192. doi: 10.1128/JVI.02143-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Fontes E, Luckow VA, Hanley-Bowdoin L. 1992. A geminivirus replication protein is a sequence-specific DNA binding protein. Plant Cell 4:597–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Briddon RW, Stanley J. 2006. Subviral agents associated with plant single-stranded DNA viruses. Virology 344:198–210. doi: 10.1016/j.virol.2005.09.042. [DOI] [PubMed] [Google Scholar]
- 36.Mansoor S, Briddon RW, Zafar Y, Stanley J. 2003. Geminivirus disease complexes: an emerging threat. Trends Plant Sci 8:128–134. doi: 10.1016/S1360-1385(03)00007-4. [DOI] [PubMed] [Google Scholar]
- 37.Mansoor S, Zafar Y, Briddon RW. 2006. Geminivirus disease complexes: the threat is spreading. Trends Plant Sci 11:209–212. doi: 10.1016/j.tplants.2006.03.003. [DOI] [PubMed] [Google Scholar]
- 38.Stanley J. 2004. Subviral DNAs associated with geminivirus disease complexes. Vet Microbiol 98:121–129. doi: 10.1016/j.vetmic.2003.10.005. [DOI] [PubMed] [Google Scholar]
- 39.Briddon RW, Bull SE, Amin I, Idris AM, Mansoor S, Bedford ID, Dhawan P, Rishi N, Siwatch SS, Abdel-Salam AM, Brown JK, Zafar Y, Markham PG. 2003. Diversity of DNA beta, a satellite molecule associated with some monopartite begomoviruses. Virology 312:106–121. doi: 10.1016/S0042-6822(03)00200-9. [DOI] [PubMed] [Google Scholar]
- 40.Eini O, Behjatnia SA, Dogra S, Dry IB, Randles JW, Rezaian MA. 2009. Identification of sequence elements regulating promoter activity and replication of a monopartite begomovirus-associated DNA beta satellite. J Gen Virol 90:253–260. doi: 10.1099/vir.0.002980-0. [DOI] [PubMed] [Google Scholar]
- 41.Li D, Behjatnia SA, Dry IB, Randles JW, Eini O, Rezaian MA. 2007. Genomic regions of tomato leaf curl virus DNA satellite required for replication and for satellite-mediated delivery of heterologous DNAs. J Gen Virol 88:2073–2077. doi: 10.1099/vir.0.82853-0. [DOI] [PubMed] [Google Scholar]
- 42.Argüello-Astorga GR, Guevara-González RG, Herrera-Estrella LR, Rivera-Bustamante RF. 1994. Geminivirus replication origins have a group-specific organization of iterative elements: a model for replication. Virology 203:90–100. doi: 10.1006/viro.1994.1458. [DOI] [PubMed] [Google Scholar]
- 43.Argüello-Astorga GR, Ruiz-Medrano R. 2001. An iteron-related domain is associated to motif 1 in the replication proteins of geminiviruses: identification of potential interacting amino acid-base pairs by a comparative approach. Arch Virol 146:1465–1485. doi: 10.1007/s007050170072. [DOI] [PubMed] [Google Scholar]
- 44.Dry IB, Krake LR, Rigden JE, Rezaian MA. 1997. A novel subviral agent associated with a geminivirus: the first report of a DNA satellite. Proc Natl Acad Sci U S A 94:7088–7093. doi: 10.1073/pnas.94.13.7088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Akbar Behjatnia SA, Dry IB, Ali Rezaian M. 1998. Identification of the replication-associated protein binding domain within the intergenic region of tomato leaf curl geminivirus. Nucleic Acids Res 26:925–931. doi: 10.1093/nar/26.4.925. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Hou YM, Gilbertson RL. 1996. Increased pathogenicity in a pseudorecombinant bipartite geminivirus correlates with intermolecular recombination. J Virol 70:5430–5436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Gilbertson RL, Hidayat SH, Paplomatas EJ, Rojas MR, Hou YM, Maxwell DP. 1993. Pseudorecombination between infectious cloned DNA components of tomato mottle and bean dwarf mosaic geminiviruses. J Gen Virol 74:23–31. doi: 10.1099/0022-1317-74-1-23. [DOI] [PubMed] [Google Scholar]