Pathogenicity of a Natural Recombinant Associated with Ageratum Yellow Vein Disease: Implications for Geminivirus Evolution and Disease Aetiology

Virology 282, 38–47 (2001) doi:10.1006/viro.2000.0832, available online at http://www.idealibrary.com on

Pathogenicity of a Natural Recombinant Associated with Ageratum Yellow Vein Disease: Implications for Geminivirus Evolution and Disease Aetiology Keith Saunders, Ian D. Bedford, and John Stanley 1 Department of Virus Research, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, United Kingdom Received August 14, 2000; returned to author for revision October 23, 2000; accepted January 11, 2001 Yellow vein disease of Ageratum conyzoides is caused by a viral DNA complex consisting of the genomic component (DNA A) of the monopartite begomovirus Ageratum yellow vein virus (AYVV, family: Geminiviridae) and a small satellite-like DNA ␤ component. AYVV DNA A is unable to induce symptoms in this host alone but can systemically infect A. conyzoides in which it accumulates to low levels. Here, we demonstrate that the yellow vein phenotype can also be produced by co-inoculating A. conyzoides with AYVV DNA A and recDNA-A␤17, a naturally occurring recombinant of approximately the same size as DNA ␤ that contains sequences from both DNA A and DNA ␤. Symptoms induced by DNA A and recDNA-A␤17 in A. conyzoides and Nicotiana glutinosa are qualitatively similar to those associated with DNA A and DNA ␤ although milder. Recombination between DNA A and DNA ␤ to produce a chimera resembling recDNA-A␤17 was observed after whitefly transmission of the disease in A. conyzoides. Hence, such recombination events are likely to occur frequently, implying that recombinants will normally be associated with this type of disease complex in the field. Possible implications of these findings for the evolution of begomoviruses and the aetiology of their diseases are discussed. © 2001 Academic Press

dorecombination) between closely related isolates (Stanley et al., 1985; Aria et al., 1994) and distinct species (Gilbertson et al., 1993; Hou and Gilbertson, 1996; Ho¨fer et al., 1997; Hill et al., 1998; Hou et al., 1998) may also contribute to their diversity. Mixed infections that enable recombination and pseudorecombination to take place have been observed for begomoviruses (Lazarowitz, 1991; Harrison et al., 1997b; Harrison and Robinson, 1999; Roye et al., 1999). Begomovirus sequences can also integrate into the plant genome and enter the germ line, enabling their vertical transmission and thereby providing further opportunities for recombination events to occur (Bejarano et al., 1996; Ashby et al., 1997). Clearly, these nuclear-replicating viruses are highly recombinogenic in nature, and this ability to readily exchange genetic material has undoubtedly facilitated their rapid diversification (Padidam et al., 1999) and emergence as economically important diseases (reviewed by Brown, 1994). The monopartite begomovirus AYVV was isolated from the weed species Ageratum conyzoides exhibiting yellow vein disease (Tan et al., 1995). In addition to a begomovirus component, diseased A. conyzoides plants contain a small (⬃1.4 kb) autonomously replicating satellite-like DNA that is approximately half the size of DNA A (referred to as DNA 1; Saunders and Stanley, 1999). Like its counterpart found in cotton infected with CLCuV (Mansoor et al., 1999), AYVV DNA 1 probably originated from a component of a nanovirus, a multicomponent single-stranded DNA plant virus, by adaptation to whitefly transmission. This implies that mixed infections are

INTRODUCTION The majority of members of the family Geminiviridae belong to the genus Begomovirus (Rybicki et al., 2000). Most begomoviruses have bipartite genomes (DNAs A and B) although several have been isolated that have only a single genomic component resembling DNA A, for example, some isolates of Tomato yellow leaf curl virus (TYLCV) (Kheyr-Pour et al., 1991; Navot et al., 1991) as well as Tomato leaf curl virus (TLCV) (Dry et al., 1993), Ageratum yellow vein virus (AYVV) (Tan et al., 1995), and Cotton leaf curl virus (CLCuV) (Briddon et al., 2000). DNA A encodes a replication-associated protein (Rep) that is essential for viral DNA replication, a replication enhancer protein (REn), the coat protein (CP) and a transcription activator protein (TrAP) that controls late gene expression. DNA B encodes a nuclear shuttle protein (NSP) and a movement protein (MP), both of which are essential for symptomatic infection of plants (reviewed by Bisaro, 1996; Hanley-Bowdoin et al., 1999). Comparative sequence analysis of a large number of begomovirus species, strains, and isolates has demonstrated that their evolution is attributable not only to gradual genetic drift by point mutations but also to frequent recombination events between members of the genus (Padidam et al., 1999) as well as with members of other genera (Stanley et al., 1986; Briddon et al., 1996). The exchange of entire genomic components (pseu1 To whom reprint requests should be addressed. Fax: ⫹44 (0)1603 450045. E-mail: [email protected].

0042-6822/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

38

AGERATUM YELLOW VEIN DISEASE AETIOLOGY

39

possible that the recombinant could also have the capacity to induce disease symptoms. Here, we test this hypothesis by comparing the pathogenic effects of DNA ␤ and recDNA-A␤17 when co-inoculated with AYVV DNA A into the natural host A. conyzoides as well as the highly susceptible hosts Nicotiana glutinosa and N. benthamiana. In addition, we have investigated the propensity of DNA ␤ to undergo recombination with the begomovirus component to produce chimeras resembling recDNAA␤17 during infection in A. conyzoides. RESULTS Recombinant recDNA-A␤17 can induce disease symptoms in A. conyzoides FIG. 1. Begomovirus DNA A and associated satellite-like and recombinant DNAs. DNA A and DNA ␤ sequences are shown in white and gray, respectively. Intergenic region sequences corresponding to the common region of bipartite begomoviruses, also present in recDNAA␤17 (recDNA), are shown in black. The DNA A and DNA ␤ nucleotide coordinates at the recombination points in recDNA-A␤17 are given; nucleotide numbering begins immediately downstream of the nick site within the origin of replication (denoted by the stem–loop structure at the top of each component) according to current convention. The position and orientation of primers V3518 and V3996, used to PCRamplify sequences of DNA ␤, recDNA-A␤17, and de novo recombinants, are indicated by black arrow heads. The positions of SspI sites in DNA ␤ and recDNA-A␤17 that were used to screen for recombination in plants are indicated.

not confined to begomoviruses but can also occur between begomoviruses and nanoviruses. Homologues of other nanovirus components have not been isolated from infected A. conyzoides, and DNA 1 plays no essential role in the disease phenotype. However, diseased A. conyzoides plants contain an additional satellite-like DNA, referred to as DNA ␤, of unknown origin (Saunders et al., 2000). AYVV DNA A and DNA ␤ together form the disease complex that is responsible for the yellow vein phenotype. In addition to AYVV DNA A, DNA 1, and DNA ␤, a number of small DNAs have also been isolated from infected A. conyzoides plants. Their sequences indicate that they are chimeras produced by recombination between DNA A and either DNA 1 or DNA ␤. Although originally identified as defective DNAs and designated with the prefix “def”(Stanley et al., 1997), we now refer to them as recombinant (rec) DNAs, either recDNA-A␤ or recDNA-A1, according to their parental DNA components. The DNA ␤ recombinants are all closely related (typified by recDNA-A␤17 (previously def17; Fig. 1), possessing DNA ␤ sequences fused to the AYVV DNA A intergenic region that contains the origin of DNA replication and variable amounts of flanking coding sequences. Because recDNA-A␤17 contains a large part of DNA ␤, including potentially functional open reading frames (ORFs) (Saunders et al., 2000), it was considered

A. conyzoides plants were co-inoculated with AYVV DNA A and DNA ␤, recDNA-A␤17, or a combination of both. Infectivity of DNA A and DNA ␤ in this host was shown to be sporadic when the DNAs were introduced on separate binary vectors (Saunders et al., 2000) and reached an average of only 13% in current experiments (Table 1). In contrast, infection and symptom induction in the highly susceptible host N. benthamiana was typically 100% using this cloning strategy. In an attempt to overcome the infectivity problem in A. conyzoides, DNA A and DNA ␤ (pBin-AYVVA␤), DNA A and recDNA-A␤17 (pBinAYVVArec17), or all three DNAs (pBin-AYVVA␤rec17) were cloned in a single binary vector, thus ensuring that inoculated cells received copies of each component. Routinely, all A. conyzoides plants co-inoculated in this way with DNA A and DNA ␤ became infected and developed typical yellow vein symptoms (Table 1). Plants co-inoculated with DNA A and recDNA-A␤17 also developed yellow vein symptoms although slightly less efficiently (86%), and symptoms were milder than those associated with DNA A and DNA ␤ (Figs. 2A and 2C). However, the onset of symptoms occurred at ⬃9 days postinoculation TABLE 1 Symptom Induction in Plants by Viral, Satellite-like, and Recombinant DNAs Inoculum DNA A DNA A ⫹ DNA ␤ b DNA A ⫹ DNA ␤ c DNA A ⫹ recDNA b DNA A ⫹ recDNA c DNA A ⫹ DNA ␤ ⫹ recDNA c

A. conyzoides a N. benthamiana a N. glutinosa a 0/18 (3) 7/54 (3) 22/22 (3) 4/63 (4) 19/22 (3)

24/25 (5) 23/23 (4) 8/8 (3) 23/23 (4) 8/8 (3)

12/12 (3) not done 12/12 (3) not done 12/12 (3)

19/22 (3)

8/8 (3)

12/12 (3)

a Number of symptomatic plants/number of plants inoculated (number of experiments). b DNA A, DNA ␤, and recDNA-A␤17 partial repeats cloned in separate binary vectors. c DNA A, DNA ␤, and recDNA-A␤17 partial repeats cloned in a single binary vector.

40

SAUNDERS, BEDFORD, AND STANLEY

FIG. 2. Symptoms induced by AYVV DNA A when co-inoculated with either DNA ␤ or recDNA-A␤17. (A) A. conyzoides infected with DNA A and DNA ␤ (left) and DNA A and recDNA-A␤17 (right) photographed 28 days after inoculation. (B) From left to right, an uninfected N. glutinosa plant and plants infected with DNA A alone, DNA A and DNA ␤, and DNA A and recDNA-A␤17, photographed 25 days after inoculation. (C) Upper leaves from an uninfected A. conyzoides plant (centre) and plants infected with DNA A and DNA ␤ (left) and DNA A and recDNA-A␤17 (right). (D) Portions of upper leaves from N. glutinosa plants infected with DNA A alone (left) and DNA A and DNA ␤ (right).

for both inocula. The severe phenotype associated with DNA A and DNA ␤ was maintained in A. conyzoides following whitefly transmission. However, symptoms associated with DNA A and recDNA-A␤17 following whitefly transmission were significantly milder than those in the initially infected plants, and two attempts to whiteflytransmit the disease from these very mildly affected plants have been unsuccessful (data not shown). Nonetheless, the ability of whiteflies to transmit recDNA-A␤17 from the initially infected plant implies that it is encapsidated as has been proposed for DNA ␤ (Saunders et al., 2000). A. conyzoides plants co-inoculated with all three components developed severe symptoms identical to those associated with plants infected with DNA A and DNA ␤. Co-inoculation of DNA A with either DNA ␤ or recDNAA␤17 also induced distinguishable symptoms in other hosts. DNA A alone has been shown to cause a sporadic asymptomatic infection in A. conyzoides (Saunders and Stanley, 1999; Saunders et al., 2000). In contrast, it caused a systemic infection in all inoculated N. glutinosa plants, in which it induced symptoms of stunting and

erect leaves (Fig. 2B; Table 1). When co-inoculated with DNA A, DNA ␤, and recDNA-A␤17, each induced stunting and downward leaf curl in this host, although symptoms were again milder when recDNA-A␤17 was used. Similar to the response in A. conyzoides, N. glutinosa plants co-inoculated with DNA A, DNA ␤, and recDNA-A␤17 displayed symptoms identical to those associated with DNA A and DNA ␤. DNA A and DNA ␤ induced vein greening symptoms in N. glutinosa (Fig. 2D), and a similar though less pronounced phenotype was associated with DNA A and recDNA-A␤17. DNA A alone did not induce this phenotype. Vein greening symptoms are typical of leaf curl disease (probably caused by monopartite begomovirus and DNA ␤ complexes; Saunders et al., 2000) in cotton and hollyhock (Mansoor et al., 1993; Harrison et al., 1997a). Symptoms induced by DNA A and recDNA-A␤17 in N. benthamiana were qualitatively similar to those induced by DNA A and DNA ␤ (Saunders et al., 2000) although slightly milder (data not shown). The difference between phenotypes associated with DNA ␤ and recDNA-A␤17 in this host was less marked than in N. glutinosa.

AGERATUM YELLOW VEIN DISEASE AETIOLOGY

FIG. 3. Southern blot analysis of viral, satellite-like, and recombinant DNAs associated with systemically infected A. conyzoides plants. Nucleic acids were extracted from upper leaves of plants infected with DNA A and DNA ␤ (lanes 2 and 11–14), DNA A and recDNA-A␤17 (lanes 3 and 15–18), or DNA A, DNA ␤, and recDNA-A␤17 (lanes 4–9). Equal amounts of nucleic acid (5 ␮g) were loaded in each lane. Blots were hybridised with probes to detect either DNA A and recDNA-A␤17 (A) or DNA ␤ (B). Marker DNAs corresponding to partial repeats of DNA A (3.2 kb) and recDNA-A␤17 (1.6 kb) and a dimer of DNA ␤ (2.7 kb) were produced by digestion of 2.5 ng pBin-AYVVA␤rec17 with HindIII, SmaI, and SacI (lanes 1 and 10). The DNA A and DNA ␤ markers contain two copies of their respective probes, and the recDNA-A␤17 marker contains a single copy. The positions of single-stranded forms of DNA A, recDNA-A␤17 (recDNA), and DNA ␤ are indicated.

Symptom severity in A. conyzoides correlates with levels of AYVV DNA A To examine the relative amounts of DNA A, DNA ␤, and recDNA-A␤17 in mixed infections, nucleic acids were extracted from systemically infected symptomatic A. conyzoides plants 21 days postinoculation and analyzed by Southern hybridization using probes to detect either DNA A and recDNA-A␤17 (distinguishable on the basis of size) or DNA ␤ (Fig. 3). Equimolar amounts of marker DNAs corresponding to redundant copies of each DNA component were produced by digestion of pBin-AYVVA␤rec17. Probe specificity was confirmed by hybridization to these markers (lanes 1 and 10) and to DNA isolated from individual symptomatic leaves from plants co-inoculated with either DNA A and DNA ␤ (lane 2) or DNA A and recDNA-A␤17 (lane 3). Predominantly single-stranded DNA forms of all three components accumulated in infected tissues. Double-stranded DNA forms were also detected, although they accumulated to relatively low levels as previously observed in this host (Saunders et al., 2000). The relative accumulation of each DNA component was estimated knowing that the 3.2-kb DNA A marker and the 2.7-kb DNA ␤ marker each contain two copies of their respective probe target, while the

41

1.6-kb recDNA-A␤17 marker contains only a single copy. In an initial experiment, plants were co-inoculated with DNA A, DNA ␤, and recDNA-A␤17 using clone pBinAYVVA␤rec17. The amount of each component varied considerably between samples (lanes 4–9), and either DNA ␤ or recDNA-A␤17 could predominate in individual severely affected leaves. Nonetheless, both components could accumulate to relatively high levels; recDNA-A␤17 accumulated to an average of ⬃125% (range 20–207%) and DNA ␤ to ⬃75% (range 44–129%) of the level of DNA A. In plants co-inoculated with DNA A and DNA ␤ and showing typical severe symptoms, both components accumulated to similar levels (lane 2), while in plants coinoculated with DNA A and recDNA-A␤17 and showing a mild phenotype, the level of DNA A in particular appeared to be greatly reduced (lane 3). To substantiate this observation, A. conyzoides plants were co-inoculated with DNA A and either DNA ␤ (clone pBin-AYVVA␤) or recDNA-A␤17 (pBin-AYVVArec17). On this occasion, DNA ␤ accumulated to an average of ⬃105% (range 85–142%) of the level of DNA A in severely affected leaves (Fig. 3, lanes 11–14). In contrast, DNA A accumulation in leaves showing the mild phenotype typical of co-infection with DNA A and recDNA-A␤17 averaged only ⬃2% (range 0.5–3.9%) of that associated with the DNA A and DNA ␤ infection, while recDNA-A␤17 accumulation remained relatively high at an average of ⬃40% (range 11–62%) of the level of DNA ␤ found in severely affected plants. Overall, these results indicate that recDNA-A␤17 supports a mild symptomatic infection in A. conyzoides associated with extremely low levels of DNA A accumulation. However, even in the presence of high levels of recDNA-A␤17, DNA ␤ can augment DNA A accumulation to levels found in plants infected with only DNA A and DNA ␤. Initial attempts to compare the replication of DNA A, DNA ␤, and recDNA-A␤17 using protoplasts derived from a suspension culture of N. tabacum BY-2 cells (Nagata et al., 1992) were inconclusive because they supported only low levels of replication. Subsequently, replication was compared by infecting N. benthamiana leaf disks with combinations of DNA A, DNA ␤, and recDNA-A␤17. Nucleic acids were extracted at intervals after inoculation and analysed by Southern hybridisation using probes specific for DNA A, DNA ␤, or both DNA A and recDNAA␤17 (Fig. 4). Each component replicated to produce predominantly single-stranded DNA forms. The DNA species that migrates ahead of single-stranded DNA A (Fig. 4A) is probably a subgenomic-sized defective DNA A component. By 6 days postinoculation, DNA ␤ had accumulated to ⬃25% and recDNA-A␤17 to ⬃65% of the level of DNA A (numbers averaged over two experiments) in leaf disks inoculated with pBin-AYVVA␤ (lane 7) and pBin-AYVVArec17 (lane 10), respectively. In leaf disks co-inoculated with DNA A, DNA ␤, and recDNA-A␤17 using clone pBin-AYVVA␤rec17, DNA ␤ had accumulated

42

SAUNDERS, BEDFORD, AND STANLEY

FIG. 4. Southern blot analysis of the accumulation of viral, satellitelike, and recombinant DNAs in N. benthamiana leaf disks. Nucleic acids were extracted from leaf disks inoculated with DNA A (lanes 2–4), DNA A and DNA ␤ (lanes 5–7), DNA A and recDNA-A␤17 (lanes 8–10), or DNA A, recDNA-A␤17, and DNA ␤ (lanes 11–13). Leaf disks were harvested 2 (lanes 2, 5, 8, and 11), 4 (lanes 3, 6, 9, and 12) and 6 (lanes 4, 7, 10, and 13) days postinoculation. Equal amounts of nucleic acid (5 ␮g) were loaded in each lane. Blots were hybridised with probes to detect DNA A (A), DNA A and recDNA-A␤17 (B), or DNA ␤ (C). Lane 1 contains an aliquot (1␮g) of nucleic acid extracted from an A. conyzoides plant co-infected with DNA A, recDNA-A␤17 (recDNA), and DNA ␤, and the positions of their single-stranded DNA forms are indicated. The DNA A, recDNA-A␤17, and DNA ␤ marker fragments (lane 14) are described in the legend to Fig. 3. Note that on this occasion, the 1.6-kb recDNA-A␤17 marker has been lost during the blotting procedure.

spectively, from DNA extracted from A. conyzoides plants infected with wild-type virus (Fig. 5, lane 1). Digestion of the PCR products with EcoRI, SspI, HindIII, or DraI (lanes 2–5) gave fragments diagnostic of DNA ␤ (most abundant band(s)) and recDNA-A␤17 as predicted from their sequences. As expected, the four fragments resulting from SspI digestion were also amplified from extracts of plants co-inoculated with DNA A and either DNA ␤ (550and 310-bp fragments; lane 7) or recDNA-A␤17 (460- and 370-bp fragments; lane 13). Fragments diagnostic of DNA ␤ were amplified from an extract of plants that had been inoculated with DNA A and DNA ␤ progeny using viruliferous whiteflies (lane 9). However, an additional smaller PCR product that produced novel SspI fragments appeared after the fifth passage using viruliferous whiteflies (lanes 10 and 11). The intact PCR product was cloned, and the sequences of four independent clones were established and shown to be identical. The analysis confirmed that a new recombinant had been produced that was slightly smaller (1190 bp) than both DNA ␤ (1347 bp) and recDNA-A␤17 (1305 bp) as suggested from the size of the PCR fragment. The recombinant differed from recDNA-A␤17 (Fig. 1), having nucleotides 39-937 of DNA ␤ linked to nucleotides 2499-39 of DNA A, but otherwise was similar in overall structure. One recombination point in recDNA-A␤17 has been shown to correspond to the putative nick site within the origin of replication in DNA ␤ (Stanley et al., 1997). The analogous position in the new recombinant occurs downstream of the nick site as previously observed for a number of recDNA-A␤17 homologues (Stanley et al., 1997). On this occasion, the crossover point occurs 39 nucleotides downstream of both the DNA A and DNA ␤ nick sites,

to ⬃25% and recDNA-A␤17 to ⬃20% of the level of DNA A (lane 13). These results indicate that co-infecting DNA ␤ and recDNA-A␤17 accumulate to similar levels but neither outcompetes DNA A replication in N. benthamiana leaf disks over this time period. Recombination occurs between DNA A and DNA ␤ during multiple passaging of clone progeny To examine their potential for recombination, DNA A and DNA ␤ were maintained in A. conyzoides by whitefly transmission. Primers V3518 and V3996 were designed to PCR-amplify a region of either DNA ␤ or a recombinant with a similar structure to recDNA-A␤17 (Fig. 1). The validity of this approach was verified by amplifying DNA ␤ and recDNA-A␤17 fragments of 860 and 830 bp, re-

FIG. 5. PCR-amplification of DNA ␤ and recombinant DNA fragments associated with wild-type and clone progeny infections. Primers V3518 and V3996, specific to DNA ␤ sequences (Fig. 1), were used to PCRamplify a DNA fragment from A. conyzoides extracts of plants infected with wild-type virus (lanes 1–5), DNA A and DNA ␤ (lanes 6 and 7), or DNA A and recDNA-A␤17 (lanes 12 and 13). Fragments were also amplified from extracts of plants infected by whitefly transmission of the progeny of DNA A and DNA ␤, sampled after the first (lanes 8 and 9) and fifth (lanes 10 and 11) passage. Samples were either undigested (lanes 1, 6, 8, 10, and 12) or digested with EcoRI (lane 2), SspI (lanes 3, 7, 9, 11 and 13), HincII (lane 4), and DraI (lane 5). The positions of DNA size markers (kb) in lane M are indicated.

AGERATUM YELLOW VEIN DISEASE AETIOLOGY

suggesting that early events during replication again contribute to recombination. DISCUSSION We have recently demonstrated that yellow vein disease of A. conyzoides is caused by a novel complex comprising AYVV DNA A and satellite-like DNA ␤ (Saunders et al., 2000). DNA ␤ is dependent on DNA A for its replication and encapsidation and is responsible for symptom development and amplification of DNA A to wild-type levels. It was suggested that DNA ␤-mediated amplification of DNA A may involve either systemic movement, conditioning of cells to make them permissive for replication or suppression of a host defense mechanism. Diseased A. conyzoides plants also contain a number of closely related chimeric DNAs that are recombinants between DNA A and DNA ␤ (Stanley et al., 1997). Here, we demonstrate that one such recombinant (recDNA-A␤17) is also capable of inducing the disease phenotype in A. conyzoides when co-inoculated with AYVV DNA A. Thus, the DNA ␤ sequences within recDNA-A␤17, encompassing nucleotides 1–977 and including putative functional ORFs identified by Saunders et al. (2000), are sufficient to induce yellow vein symptoms. As the majority of other recombinants isolated from diseased A. conyzoides closely resemble recDNA-A␤17 (Stanley et al., 1997), it is likely that they represent a population of infectious chimeric DNAs. Our data also suggest that such recombinants may be frequently produced, implying that they will normally be associated with this type of disease complex in the field. This idea is supported by the isolation of a similar recombinant associated with CLCuV-infected cotton in Pakistan (EMBL Accession No. AJ242974; Briddon et al., 2000). Recombinants containing a functional begomovirus origin of replication and of an appropriate size for encapsidation, systemic spread, and whitefly transmission would be expected to be maintained within the population. The proposed ubiquitous nature of the recombinants may be exploited to facilitate cloning of DNA ␤ homologues from other disease complexes by designing diverging primers to the intergenic region of the associated begomovirus with which to PCR-amplify recombinant DNA fragments and, hence, DNA ␤ sequences. Although the onset of symptom development occurred at a similar time when DNA A was co-inoculated with either DNA ␤ or recDNA-A␤17, symptoms induced by recDNA-A␤17 in A. conyzoides were noticeably milder, and this correlated with extremely low levels of DNA A accumulation. This low level of DNA A accumulation is comparable to that observed in asymptomatic A. conyzoides plants infected with DNA A alone (Saunders et al., 2000), indicating that symptoms primarily result from the host response to a DNA ␤ gene product rather than DNA ␤-mediated amplification of a DNA A-encoded

43

symptom determinant. Unlike DNA ␤, recDNA-A␤17 contains DNA A nucleotides 2472-56 (Fig. 1) encompassing the authentic begomovirus replication origin and adjacent virion-sense and complementary-sense promoters. Hence, DNA A iterons involved in Rep binding during the initiation of replication (Argu¨ello-Astorga et al., 1994) are present in recDNA-A␤17 although identical motifs do not occur in DNA ␤ (Saunders et al., 2000). Despite these differences, DNA ␤ and recDNA-A␤17 replicate to similar levels in leaf disks of the permissive host N. benthamiana. However, it remains to be seen if the authentic begomovirus origin confers a significant replicative advantage on recDNA-A␤17 in the natural host A. conyzoides. Although recDNA-A␤17 may contain the same gene(s) as DNA ␤, it is possible that gene expression is adversely affected by the presence of begomovirus regulatory elements that participate in TrAP-mediated transactivation of the virion-sense promoter (Sunter and Bisaro, 1992) and Rep-mediated downregulation of the complementary-sense promoter (Sunter et al., 1993). Indeed, the control of virion-sense expression by TrAP has been proposed for TLCV (Dry et al., 2000), and the TLCV Rep binding site has been mapped between its promoter and coding sequences, consistent with a regulatory role (Behjatnia et al., 1998), suggesting that comparable regulatory elements also occur in monopartite begomoviruses. As a consequence, recDNA-A␤17 may be less well suited to amplify DNA A during systemic infection of A. conyzoides, resulting in only mild symptoms. However, our co-infection experiments using this host demonstrate that the proposed defect in recDNA-A␤17 is overcome by DNA ␤. While it is not yet known why recDNA-A␤17 accumulates at the expense of the begomovirus component in A. conyzoides (in the absence of DNA ␤), this is clearly a host-specific phenomenon as both components accumulate efficiently in N. benthamiana plants (Stanley et al., 1997) and DNA A accumulates to higher levels than recDNA-A␤17 in N. benthamiana leaf disks (this paper). It is likely that the drastic reduction in DNA A levels is responsible for the rapid loss in the ability to transmit the disease by whiteflies. This indicates that DNA ␤ has a selective advantage over recDNA-A␤17 and suggests that, although recombinants like recDNA-A␤17 have the potential to co-exist with DNA A, they are unlikely to occur in the field in the absence of DNA ␤. However, the possibility that recombinants associated with uncharacterised begomovirus diseases have gradually evolved to overcome such defects cannot be ruled out at the present time. The evolutionary origin of DNA ␤ remains to be determined. It shares no extensive sequence homology with bipartite begomoviruses although it has a number of features in common with DNA B that probably reflects convergent evolution of these components. Both DNA ␤ and DNA B are replicated in trans by DNA-A-encoded

44

SAUNDERS, BEDFORD, AND STANLEY

Rep and encapsidated in DNA-A-encoded CP. DNA B is generally considered to play an essential role in bipartite begomovirus infection, although systemic movement of DNA A alone can occur under certain conditions (Klinkenberg and Stanley, 1990; Evans and Jeske, 1993). DNA ␤ is not essential for systemic movement of AYVV DNA A although it is required for the amplification of DNA A to wild-type levels in systemically infected tissues and so may be considered as an essential component of the disease (Saunders and Stanley, 1999; Saunders et al., 2000). Amplification of DNA A in plants by DNA B is mediated by NSP and MP, which function cooperatively to facilitate cell-to-cell movement (Noueiry et al., 1994; Sanderfoot and Lazarowitz, 1995). MP is encoded on the complementary-sense strand and has been shown to be an important symptom determinant (von Arnim and Stanley, 1992; Pascal et al., 1993; Duan et al., 1997). Similarly, we have identified an ORF encoded on the complementary-sense strand of DNA ␤, currently designated ␤C1, that contributes to symptom production (Saunders et al., 2000 and unpublished data). Having acquired DNA A sequences, the recombinant recDNA-A␤17 shares structural features with DNA B not seen for DNA ␤ (Fig.1). DNA A and DNA B contain a region of ⬃200 nucleotides (designated the common region) that is highly conserved between genomic components of individual begomovirus species, located mainly within the intergenic region. The common region contains the Rep binding site and a conserved stem– loop motif that includes the nick site for the initiation of rolling circle replication (reviewed by Hanley-Bowdoin et al., 1999). Recombination between common region sequences of DNA A and DNA B plays an important role in maintaining these essential cis-acting motifs on both components. For example, in the case of African cassava mosaic virus, lethal mutations within the stem–loop motif of DNA B, including deletion of the entire sequence, are rapidly corrected by recombination with DNA A (Roberts and Stanley, 1994). This may occur either by homologous recombination between common region sequences or by a replicative mechanism mediated by the nick site within the origin of replication that is known to be a recombinational hot-spot (Etessami et al., 1989; Stenger et al., 1991; Stanley, 1995). Similarly, we have demonstrated that DNA ␤ can acquire sequences from DNA A by a process that, at least for recDNA-A␤17, involves the putative nick site of DNA ␤ (Fig. 1). The resulting recombinants essentially contain a common region together with the adjacent DNA A intergenic region and the 5⬘ terminus of the Rep ORF. Interestingly, the common region sequences of some bipartite begomoviruses, for example Tomato golden mosaic virus and Bean golden mosaic virus (Hamilton et al., 1984; Howarth et al., 1985), also extend into the 5⬘ terminus of the Rep ORF, suggesting that DNA B similarly acquired this region from DNA A.

Although recDNA-A␤17 is clearly a derivative of DNA ␤, having been produced by recombination with the begomovirus component, it can induce the disease phenotype, has the potential to exist in nature in the absence of DNA ␤, and shows structural and functional similarities to DNA B. Indeed, without prior knowledge of the DNA ␤ progenitor, recDNA-A␤17 may have been considered to be a novel type of begomovirus component on the basis of these criteria. Based on the conservation of their sequences, it was suggested that DNA B-encoded NSP originated from CP by gene duplication (Kikuno et al., 1984), and this idea is supported by the fact that both are sequence non-specific single-stranded DNA binding proteins that participate in nuclear trafficking (Pascal et al., 1994; Kunik et al., 1998; Palanichelvam et al., 1998) and that they show some functional redundancy (Pooma et al., 1996; Ingham et al., 1995). Hence, DNA B may have evolved from a monopartite geminivirus DNA A component by functional adaptation of the virion-sense CP gene for nuclear trafficking and exchange of existing complementary-sense genes for a MP gene. However, while DNA B and DNA ␤ are probably evolutionarily unrelated and simply represent functionally convergent disease components (vis-a`-vis DNA A amplification), it is not inconceivable that DNA B evolved from a satellitelike DNA progenitor encoding a complementary-sense MP (symptom determinant) gene that acquired a common region from DNA A (as has recDNA-A␤17) together with the virion-sense CP gene and, in doing so, increased in size to that of a begomovirus component. In conclusion, our findings provide direct evidence that recombinational events that result in significant structural changes can still produce a viable viral DNA component. This serves to illustrate the immense potential of recombination to rapidly produce novel viruses and diseases. MATERIALS AND METHODS Clone construction and plant inoculation The construction of pHNBin419 and pBinAYVrec17 (previously pBinAYVdef17) containing partial repeats of AYVV DNA A and recDNA-A␤17, respectively, in the binary vector pBin19 has been described (Tan et al., 1995; Stanley et al., 1997). The construction of a tandem repeat of DNA ␤ in pBinPlus (pBin-AYVV␤) and its cloning together with the partial repeat of AYVV DNA A on the same binary vector (pBin-AYVVA␤) has been described (Saunders et al., 2000). DNA ␤ sequences in pBinAYVVA␤ were removed by SacI digestion, and the resulting plasmid was self-ligated to produce pBin-AYVVA. The partial repeat of recDNA-A␤17 in pAYVrec17/1.2 (previously pAYVdef17/1.2; Stanley et al., 1997) was subcloned as a HindIII fragment into pBin-AYVVA and into pBinAYVVA␤ to produce pBin-AYVVArec17 and pBinAYVVA␤rec17, respectively. Agrobacterium tumefaciens

AGERATUM YELLOW VEIN DISEASE AETIOLOGY

possessing the Ti plasmid pGV3850 (Zambryski et al., 1983) was transformed with pBinAYVVA, pBinAYVVArec17, and pBin-AYVVA␤rec17 by triparental mating (Ditta et al., 1980). A. conyzoides, N. benthamiana, and N. glutinosa plants were agroinoculated as described by Tan et al. (1995). Plants were screened daily for the development of symptoms. Whitefly transmission studies were carried out as previously described (Saunders et al., 2000). Analysis of viral DNAs in infected plants Nucleic acids were isolated from infected A. conyzoides plants using the procedure of Covey and Hull (1981) and were fractionated by agarose gel electrophoresis and transferred to membranes as previously described (Saunders and Stanley, 1999). Detection of specific DNA species was achieved using probes generated by PCR amplification performed using the appropriate DNA template, which were subsequently oligolabelled as described by Feinberg and Vogelstein (1983). A probe to detect both DNA A and recDNA-A␤17 was generated using virion-sense primer V4616 (corresponding to DNA A and recDNA-A␤17 nucleotides 2482–2505 and 1046–1069, respectively) and complementary-sense primer V4617 (corresponding to DNA A and recDNAA␤17 nucleotides 2689–2716 and 1253–1280, respectively). A DNA-␤-specific probe was generated using virionsense primer V4614 (corresponding to DNA ␤ nucleotides 1091–1112) and complementary-sense primer V4615 (corresponding to DNA ␤ nucleotides 1302–1326). DNA A, DNA ␤, and recDNA-A␤17 levels were quantitated by phosphorimaging using a Fujifilm BAS-MP imaging plate, BAS 1000 imager and Mac Bas v2.5 software. Analysis of viral DNA replication in leaf disks N. benthamiana leaf disks were prepared and maintained as described by Klinkenberg et al. (1989). Leaf disks were inoculated with combinations of components using clones pBin-AYVVA, pBin-AYVVA␤, pBinAYVVArec17, and pBin-AYVVA␤rec17. A probe to specifically detect DNA A was generated using the MluI(579)– KpnI(1570) fragment encompassing the 3⬘ termini of the CP and Rep genes, a region not found in any of the recombinants characterised to date. Probes to detect DNA ␤ and recDNA-A␤17 were as described above. Cloning of recombinant viral DNA The sequence across the intergenic region of either DNA ␤ or recombinant DNA was PCR-amplified from DNA isolated from infected plants using the DNeasy Plant Mini Kit (Qiagen) together with virion-sense primer V3996 (corresponding to DNA ␤ nucleotides 795–821) and complementary-sense primer V3518 (corresponding to DNA ␤ nucleotides 286–310) (Fig. 1). The amplified

45

fragments were purified by a Wizard PCR Preps DNA purification system (Promega) and cloned into pGEM-T Easy (Promega). The sequences of both strands of cloned recombinant DNAs were determined using an ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin–Elmer) with SP6 and T7 sequencing primers. Sequencing products were resolved using an ABI 373 automated sequencer. Sequences were analysed using version 8 of the program library of the Genetics Computer Group (Devereux et al., 1984). Restriction endonuclease digestions were performed on purified PCR-amplified fragments using conditions recommended by the supplier. ACKNOWLEDGMENTS We thank Professor J.W. Davies for his encouragement and interest throughout this study and Margaret Boulton for her constructive criticism of the manuscript. This work was funded by a grant-in-aid to the John Innes Centre from the Biotechnology and Biological Sciences Research Council. AYVV was maintained and manipulated with the authorisation of the Ministry of Agriculture, Fisheries, and Food under the Plant Health (Great Britian) Order 1993, licences PHL 11/2300(6/ 1998) and PHL 11B/3069(6/1999).

REFERENCES Argu¨ello-Astorga, G.R., Guevara-Gonza´lez, R.G., Herrera-Estrella, L.R. and Rivera-Bustamante, R.F. (1994). Geminivirus replication origins have a group-specific organization of iterative elements: A model for replication. Virology 203, 90–100. Aria, J. C., Gilbertson, R. L., Hanson, S. F., Morales, S. F., Ahlquist, P., Loniello, O., and Maxwell, D. P. (1994). Bean golden mosaic geminivirus type II isolates from the Dominican Republic and Guatemala: Nucleotide sequences, infectious pseudorecombinants, and phylogenetic relationships. Phytopthology 84, 321–329. Ashby, M. K., Warry, A., Bejarano, E. R., Khashoggi, A., Burrell, M., and Lichtenstein, C. P. (1997). Analysis of multiple copies of geminiviral DNA in the genome of four closely related Nicotiana species suggests a unique integration event. Plant Mol. Biol. 35, 313–321. Behjatnia, S. A. A., Dry, I. B., and Rezaian, M. A. (1998). Identification of the replication-associated protein binding domain within the intergenic region of tomato leaf curl geminivirus. Nucleic Acids Res. 26, 925–931. Bejarano, E. R., Khashoggi, A., Witty, M., and Lichtenstein, C. P. (1996). Integration of multiple repeats of geminiviral DNA into the nuclear genome of tobacco during evolution. Proc. Natl. Acad. Sci. USA 93, 759–764. Bisaro, D. (1996). Geminivirus DNA replication. In “DNA Replication in Eukaryotic Cells” (M. DePamphilis, Ed.), pp. 833–854. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Briddon, R. W., Bedford, I. D., Tsai, J. H., and Markham, P. G. (1996). Analysis of the nucleotide sequence of the treehopper-transmitted geminivirus, tomato pseudo-curly top virus, suggests a recombinant origin. Virology 219, 387–394. Briddon, R. W., Mansoor, S., Bedford, I. D., Pinner, M. S., and Markham, P. G. (2000). Clones of cotton leaf curl geminivirus induce symptoms atypical of cotton leaf curl disease. Virus Genes 20, 19–26. Brown, J. K. (1994). Current status of Bemisia tabaci as a plant pest and virus vector in agroecosystems woldwide. FAO Plant Prot. Bull. 42, 332. Covey, S. N., and Hull, R. (1981). Transcription of cauliflower mosaic virus DNA. Detection of transcripts, properties and location of the

46

SAUNDERS, BEDFORD, AND STANLEY

gene encoding the virus inclusion body protein. Virology 111, 463– 474. Devereux, J., Haeberli, P., and Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12, 387–395. Ditta, G., Stanfield, S., Corbin, D., and Helinski, D. R. (1980). Broad host range DNA cloning system for gram-negative bacteria: Construction of gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 77, 7347–7351. Dry, I. B., Rigden, J. E., Krake, L. R., Mullineaux, P. M., and Rezaian, M. A. (1993). Nucleotide sequence and genome organization of tomato leaf curl geminivirus. J. Gen. Virol. 74, 147–151. Dry, I., Mullineaux, P., and Rezaian, A. (2000). Regulation of tomato leaf curl viral gene expression in host tissues. Mol. Plant-Microbe Interact. 13, 529–537. Duan, Y-P., Powell, C. A., Purcifull, D. E. Broglio, P., and Hiebert, E. (1997). Phenotypic variation in transgenic tobacco expressing mutated geminivirus movement/pathogenicity (BC1) proteins. Mol. Plant-Microbe Interact. 10, 1065–1074. Etessami, P., Watts, J., and Stanley, J. (1989). Size reversion of African cassava mosaic virus coat protein gene deletion mutants during infection of Nicotiana benthamiana. J. Gen. Virol. 70, 277–289. Evans, D., and Jeske, H. (1993). DNA B facilitates, but is not essential for, the spread of abutilon mosaic virus in agroinoculated Nicotiana benthamiana. Virology 194, 752–757. Feinberg, A. P., and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction fragments to high specific activity. Anal. Biochem. 132, 6–13. Gilbertson, R. L., Hidayat, S. H., Paplomatas, E. J., Rojas, M. R., Hou, Y-M, and Maxwell, D. P. (1993). Pseudorecombination between infectious cloned DNA components of tomato mottle and bean dwarf mosaic geminiviruses. J. Gen. Virol. 74, 23–31. Hamilton, W. D. O., Stein, V. E., Coutts, R. H. A., and Buck, K. W. (1984). Complete nucleotide sequence of the infectious cloned DNA components of tomato golden mosaic virus: Potential coding regions and regulatory sequences. EMBO J. 3, 2197–2205. Hanley-Bowdoin, L., Settlage, S. B., Orozco, B. M., Nagar, S., and Robertson, D. (1999). Geminiviruses: Models for plant DNA replication, transcription, and cell cycle regulation. Crit. Rev. Plant Sci. 18, 71–106. Harrison, B. D., Liu, Y. L., Khalid, S., Hameed, S., Otim-Nape, G. W., and Robinson, D. J. (1997a). Detection and relationships of cotton leaf curl virus and allied whitefly-transmitted geminiviruses occurring in Pakistan. Ann. Appl. Biol. 130, 61–75. Harrison, B. D., and Robinson, D. J. (1999). Natural genomic and antigenic variation in whitefly-transmitted geminiviruses (begomoviruses). Annu. Rev. Phytopathol. 37, 369–398. Harrison, B. D., Zhou, X., Otim-Nape, G. W., Liu, Y., and Robinson, D. J. (1997b). Role of a novel type of double infection in the geminivirusinduced epidemic of severe cassava mosaic in Uganda. Ann. Appl. Biol. 131, 437–438. Hill, J. E., Strandberg, J. O., Hiebert, E., and Lazarowitz, S. G. (1998). Asymmetric infectivity of pseudorecombinants of cabbage leaf curl virus and squash leaf curl virus: Implications for bipartite geminivirus evolution and movement. Virology 250, 283–292. Ho¨fer, P., Engel, M., Jeske, H., and Frischmuth, T. (1997). Host range limitation of a pseudorecombinant virus produced by two distinct bipartite geminiviruses. Mol. Plant-Microbe Interact. 10, 1019–1022. Hou, Y.-M., and Gilbertson, R. L. (1996). Increased pathogenicity in a pseudorecombinant bipartite geminivirus correlates with intermolecular recombination. J. Virol. 70, 5430–5436. Hou, Y.-M. Paplomatas, E. J., and Gilbertson, R. L. (1998). Host adaptation and replication properties of two bipartite geminiviruses and their pseudorecombinants. Mol. Plant-Microbe Interact. 11, 208–217. Howarth, A. J., Caton, J., Bossert, M., and Goodman, R. M. (1985). Nucleotide sequence of bean golden mosaic virus and a model for

gene regulation in geminiviruses. Proc. Natl. Acad. Sci. USA 82, 3572–3576. Ingham, D. J., Pascal, E., and Lazarowitz, S. G. (1995). Both bipartite geminivirus movement proteins define viral host range, but only BL1 determines viral pathogenicity. Virology 207, 191–204. Kheyr-Pour, A., Bendahmane, M., Matzeit, V., Accotto, G-P. M., Crespi, S., and Gronenborn, B. (1991). Tomato yellow leaf curl virus from Sardinia is a whitefly-transmitted monopartite geminivirus. Nucleic Acids Res. 19, 6763–6769. Kikuno, R., Toh, H., Hayashida, H., and Miyata, T. (1984). Sequence similarity between putative gene products of geminiviral DNAs. Nature 308, 562. Klinkenberg, F. A., Ellwood, S., and Stanley, J. (1989). Fate of African cassava mosaic virus coat protein deletion mutants after agroinoculation. J. Gen. Virol. 70, 1837–1844. Klinkenberg, F. A., and Stanley, J. (1990). Encapsidation and spread of African cassava mosaic virus DNA A in the absence of DNA B when agroinoculated to Nicotiana benthamiana. J. Gen. Virol. 71, 1409– 1412. Kunik, T., Palanichelvam, K., Czosnek, H., Citovsky, V., and Gafni, Y. (1998). Nuclear import of the capsid protein of tomato yellow leaf curl virus (TYLCV) in plant and insect cells. Plant J. 13, 393–399. Lazarowitz, S. G. (1991). Molecular characterization of two bipartite geminiviruses causing squash leaf curl disease: Role of viral replication and movement functions in determining host range. Virology 180, 70–80. Mansoor, S., Bedford, I., Pinner, M. S., Stanley, J., and Markham, P. G. (1993). A whitefly-transmitted geminivirus associated with cotton leaf curl disease in Pakistan. Pak. J. Bot. 25, 105–107. Mansoor, S., Khan, S. H., Bashir, A., Saeed, M., Zafar, Y., Malik, K. A., Briddon, R., Stanley, J., and Markham, P. G. (1999). Identification of a novel circular single-stranded DNA associated with cotton leaf curl disease in Pakistan. Virology 259, 190–199. Nagata, T., Nemoto, Y., and Hasezawa, S. (1992). Tobacco BY-2 cell line as the “Hela” cell in the cell biology of higher plants. Int. Rev. Cytol. 132, 1–30. Navot, N., Pichersky, E., Zeidan, M., Zamir, D., and Czosnek, H. (1991). Tomato yellow leaf curl virus: A whitefly-transmitted geminivirus with a single genomic component. Virology 185, 151–161. Noueiry, A. O., Lucas, W. J., and Gilbertson, R. L. (1994). Two proteins of a plant DNA virus coordinate nuclear and plasmodesmal transport. Cell 76, 925–932. Padidam, M., Sawyer, S., and Fauquet, C. M. (1999). Possible emergence of new geminiviruses by frequent recombination. Virology 265, 218–225. Palanichelvam, K., Kunik, T., Citovsky, V., and Gafni, Y. (1998). The capsid protein of tomato yellow leaf curl virus binds cooperatively to singlestranded DNA. J. Gen. Virol. 79, 2829–2833. Pascal, E., Goodlove, P. E., Wu, L. C., and Lazarowitz, S. G. (1993). Transgenic tobacco plants expressing the geminivirus BL1 protein exhibit symptoms of viral disease. Plant Cell 5, 795–807. Pascal, E., Sanderfoot, A. A., Ward, B. M., Medville, R., Turgeon, R., and Lazarowitz, S. G. (1994). The geminivirus BR1 movement protein binds single-stranded DNA and localizes to the cell nucleus. Plant Cell 6, 995–1006. Pooma, W., Gillette, W. K., Jeffrey, J. L., and Petty, I. T. D. (1996). Host and viral factors determine the dispensability of coat protein for bipartite geminivirus systemic movement. Virology 218, 264–268. Roberts, S., and Stanley, J. (1994). Lethal mutations within the conserved stem-loop of African cassava mosaic virus DNA are rapidly corrected by genomic recombination. J. Gen. Virol. 75, 3203–3209. Roye, M. E., Wernecke, M. E., McLaughlin, W. A., Nakhla, M. K., and Maxwell, D. P. (1999). Tomato dwarf leaf curl virus, a new bipartite geminivirus associated with tomatoes and peppers in Jamaica and mixed infection with tomato yellow leaf curl virus. Plant Pathol. 48, 370–378. Rybicki, E. B., Briddon, R. W., Brown, J. K., Fauquet, C. M., Maxwell, D. P.,

AGERATUM YELLOW VEIN DISEASE AETIOLOGY Harrison, B. D., Markham, P. G., Bisaro, D. M., Robinson, D., and Stanley, J. (2000). Geminiviridae. In “Virus Taxonomy. Seventh Report of the International Committee on Taxonomy of Viruses” (M.H.V. Regenmortel, C.M. Fauquet, D. H. L. Bishop, E.B. Carstens, M.K. Estes, S.M. Lemon, J. Maniloff, M.A. Mayo, D.J. McGeoch, C.R. Pringle, and R.B. Wickner, Eds.), pp. 285–297. Academic Press, San Diego. Sanderfoot, A. A., and Lazarowitz, S. G. (1995). Cooperation in viral movement: The geminivirus BL1 movement protein interacts with BR1 and redirects it from the nucleus to the cell periphery. Plant Cell 7, 1185–1194. Saunders, K., Bedford, I. D., Briddon, R. W., Markham, P. G., Wong, S. M., and Stanley, J. (2000). A unique virus complex causes Ageratum yellow vein disease. Proc. Natl. Acad. Sci. USA 97, 6890–6895. Saunders, K., and Stanley, J. (1999). A nanovirus-like DNA component associated with yellow vein disease of Ageratum conyzoides: Evidence for interfamilial recombination between plant DNA viruses. Virology 264, 142–152. Stanley, J. (1995). Analysis of African cassava mosaic virus recombinants suggest strand nicking occurs within the conserved nonanucleotide motif during the initiation of rolling circle DNA replication. Virology 206, 707–712. Stanley, J., Markham, P. G., Callis, R. J., and Pinner, M. S. (1986). The nucleotide sequence of an infectious clone of the geminivirus beet curly top virus. EMBO J. 5, 1761–1767. Stanley, J., Saunders, K., Pinner, M. P., and Wong, S. M. (1997). Novel

47

defective interfering DNAs associated with ageratum yellow vein geminivirus infection of Ageratum conyzoides. Virology 239, 87–96. Stanley, J., Townsend, R., and Curson, S. J. (1985). Pseudorecombinants between cloned DNAs of two isolates of cassava latent virus. J. Gen. Virol. 66, 1055–1061. Stenger, D. C., Revington, G. N., Stevenson, M. C., and Bisaro D. M. (1991). Replicational release of geminivirus genomes from tandemly repeated copies: Evidence for rolling circle replication of a plant viral DNA. Proc. Natl. Acad. Sci. USA 88, 8029–8033. Sunter, G., and Bisaro, D. M. (1992). Transactivation of geminivirus AR1 and BR1 gene expression by the viral AL2 gene product occurs at the level of transcription. Plant Cell 4, 1321–1331. Sunter, G., Hartitz, M. D., and Bisaro, D. M. (1993). Tomato golden mosaic virus leftward gene expression: Autoregulation of geminivirus replication protein. Virology 195, 275–280. Tan, P. H. N., Wong, S. M., Wu, M., Bedford, I. D., Saunders, K., and Stanley, J. (1995). Genome organization of ageratum yellow vein virus, a monopartite whitefly-transmitted geminivirus isolated from a common weed. J. Gen. Virol. 76, 2915–2922. von Arnim, A., and Stanley, J. (1992). Determinants of tomato golden mosaic virus symptom development located on DNA B. Virology 186, 286–293. Zambryski, P., Joos, H., Genetello, C., Leemans, J., van Montagu, M., and Schell, J. (1983). Ti plasmid vector for the introduction of DNA into plant cells without alteration of their normal regeneration capacity. EMBO J. 2, 2143–2150.