Journal of General Virology (2008), 89, 818–828
DOI 10.1099/vir.0.82873-0
Supervirulent pseudorecombination and asymmetric synergism between genomic components of two distinct species of begomovirus associated with severe tomato leaf curl disease in India S. Chakraborty,3 R. Vanitharani,4 B. Chattopadhyay3 and C. M. Fauquet Correspondence C. M. Fauquet
International Laboratory for Tropical Agricultural Biotechnology, Donald Danforth Plant Science Center, 975 N. Warson Road, St Louis, MO 63132, USA
[email protected]
Received 23 January 2007 Accepted 5 November 2007
Isolates of two distinct begomovirus species, the severe strain of the species Tomato leaf curl New Delhi virus (tomato leaf curl New Delhi virus-[India:New Delhi:Severe:1992]; ToLCNDV-[IN:ND:Svr:92], bipartite) and the Varanasi strain of the species Tomato leaf curl Gujarat virus (tomato leaf curl Gujarat virus-[India:Varanasi:2001]; ToLCGV-[IN:Var:01], mono/ bipartite) infect tomato (Lycopersicon esculentum) and cause severe yield losses in northern India. This study investigated the infectivity properties of genomic components of these two species. Both pseudorecombinants were infectious in Nicotiana benthamiana, Nicotiana tabacum and L. esculentum. Enhanced pathogenicity was observed when DNA-A of ToLCNDV-[IN:ND:Svr:92] was trans-complemented with ToLCGV-[IN:Var:01] DNA-B, and was consistently associated with an increase in accumulation of ToLCGV-[IN:Var:01] DNA-B. Mixed infection of ToLCNDV-[IN:ND:Svr:92] and ToLCGV-[IN:Var:01] always showed extremely severe symptoms, suggesting a synergistic interaction between these two viruses. Southern blot analysis of viral DNAs from infected plants showed a significantly higher level of accumulation of both ToLCNDV components and DNA-B of ToLCGV-[IN:Var:01] with no alteration to levels of DNA-A of ToLCGV-[IN:Var:01]. Symptom development and/or higher infectivity of the supervirulent pseudorecombinants correlated with the increased levels of DNA-B accumulation. Protoplast replication assays revealed that enhanced infectivity by the pseudorecombinant occurred at the level of replication, as DNA-A of ToLCNDV-[IN:ND:Svr:92] enhanced ToLCGV-[IN:Var:01] DNA-B replication, whose accumulation was in turn increased by ToLCGV-[IN:Var:01] DNA-A. This is the first report demonstrating a virulent pseudorecombinant between two distinct species of begomoviruses that infect tomato, and is the second report on synergism between begomoviruses. The results revealed that ToLCGV-[IN:Var:01] DNA-B is capable of associating with different DNA-A components, despite having different iteron sequences.
INTRODUCTION Geminiviruses are plant viruses characterized by circular, single-stranded DNA genomes and twinned icosahedral virions (18630 nm) (Stanley et al., 2005). Based on genome organization, host range and insect vector, geminiviruses are divided into four genera: Mastrevirus, Curtovirus, Topocuvirus and Begomovirus (Fauquet & Stanley, 2003; Fauquet et al., 2003). Begomoviruses are transmitted by whiteflies [Bemisia tabaci (Gennadius)] and 3Present address: School of Life Sciences, Jawaharlal Nehru University, New Delhi 110 067, India. 4Present address: University of California at Riverside, Riverside, CA 92521, USA.
818
usually possess a bipartite genome of two approximately 2.7 kb DNA components, designated DNA-A and DNA-B. Monopartite begomoviruses are also known to occur in the Old World (Navot et al., 1991; Dry et al., 1993; Muniyappa et al., 2000; Chatchawankanphanich & Maxwell, 2002; Chakraborty et al., 2003a, b). Tomato leaf curl disease is considered to be one of the major constraints in cultivation of tomato throughout the world (Chakraborty et al., 2003a, b). Tomato leaf curl disease is caused by viruses belonging to the genus Begomovirus (Padidam et al., 1995; Muniyappa et al., 2000; Chakraborty et al., 2003a, b). The DNA-A component encodes six proteins: AV1, the coat protein; AV2, the pre-coat protein; AC1, the essential viral replication-associated protein (Rep) 0008-2873 G 2008 SGM Printed in Great Britain
Virulent pseudorecombination and synergism among ToLCVs
(Laufs et al., 1995); AC2, the transcriptional activator of virion-sense AV1 and BV1 open reading frames (Sunter & Bisaro, 1991, 1992); AC3, the replicational enhancer, which enhances the efficiency of viral replication (Sunter et al., 1990); and AC4, which is putatively responsible for symptom expression (Van Wezel et al., 2002). DNA-B encodes two movement proteins, BV1 and BC1, required for local cell-to-cell movement of the virus and long-distance movement through the phloem (Sanderfoot & Lazarowitz, 1996). The genes on each genomic component are transcribed bidirectionally from a ~200 nt common region (CR) with high sequence identity (90–100 %). The CR contains promoters and sequence elements required for DNA replication and transcription (Eagle et al., 1994; Laufs et al., 1995; Chatterji et al., 1999, 2000). Representatives of several species of tomato leaf curl viruses (ToLCVs) are known to cause tomato leaf curl disease in India. From northern India, isolates of tomato leaf curl New Delhi virus from New Delhi (ToLCNDV[IN:ND:Svr:92]) (Padidam et al., 1995) and from Lucknow (ToLCNDV-[IN:Luc:]) (Srivastava et al., 1995) are reported to be bipartite in nature. Isolates of the species Tomato leaf curl Bangalore virus from Bangalore (Muniyappa et al., 2000) and Tomato leaf curl Karnataka virus from Karnataka (Chatchawankanphanich & Maxwell, 2002) in southern India apparently possess only the DNAA component. Recently, an isolate of a novel species, Tomato leaf curl Gujarat virus, has been reported from Varanasi (ToLCGV-[IN:Var:01]), which is monopartite/ bipartite in nature (Chakraborty et al., 2003b). Under natural conditions, mixed virus infections in a single plant could have biological and epidemiological implications. Harrison et al. (1997) suggested the possibility of synergism between isolates of the two cassava begomovirus species, African cassava mosaic virus and East African cassava mosaic virus. Molecular evidence for synergism between the cassava-infecting begomoviruses African cassava mosaic virus from Cameroon (ACMV-[CM:98]) and East African cassava mosaic virus from Cameroon (EACMV-[CM:98]) has been demonstrated (Fondong et al., 2000; Pita et al., 2001; Vanitharani et al., 2004). Viable pseudorecombinants have been produced in the laboratory by reassortment of the genomic components of closely related virus isolates of ACMV, squash leaf curl virus (SqLCV), tomato golden mosaic virus (TGMV) and bean golden mosaic virus (BGMV) (Stanley et al., 1985; Lazarowitz, 1991; von Arnim & Stanley, 1992; Ingham & Lazarowitz, 1993; Faria et al., 1994) or from isolates of the two species Tomato mottle virus and Bean dwarf mosaic virus (Gilbertson et al., 1993). However, in no case was pseudorecombination between viruses belonging to two different species shown to lead to a more virulent virus. Here, we report for the first time a more virulent pseudorecombinant and an asymmetric synergism between the components of two virus isolates belonging to distinct begomovirus species of the genus Begomovirus, namely the http://vir.sgmjournals.org
severe strain tomato leaf curl New Delhi virus-[India:New Dehi:Severe:1992] (ToLCNDV-[IN:ND:Svr:92]) and the Varanasi strain tomato leaf curl Gujarat virus[India:Varanasi:2001] (ToLCGV-[IN:Var:01]), causing a severe leaf curl disease on tomato in India.
METHODS clones. Full-length infectious clones of ToLCNDV[IN:ND:Svr:92] (DNA-A in pTNDA and DNA-B in pTNDB, hereafter referred to as NA and NB, respectively), ToLCNDV-[IN:ND:Mld:92] (a mild isolate of ToLCNDV, referred to as NAm) and ToLCGV[IN:Var:01] (DNA-A in pTGVA and DNA-B in pTGVB, referred to as VA and VB, respectively) have been described previously (Padidam et al., 1995; Chakraborty et al., 2003b). It should be noted that the severe phenotype is associated with DNA-A of ToLCNDV and not with the DNA-B component (Chatterji et al., 1999), which is why we refer to ToLCNDV-[IN:ND:Svr:92] for the DNA-A component but ToLCNDV-[IN:ND:92] for the DNA-B component. Virus
Sequence analysis. The GenBank accession numbers of DNA-A and DNA-B sequences used in this paper were: ToLCGV-[IN:Var:01] DNAA (AY190290), ToLCGV-[IN:Var:01] DNA-B (AY190291), ToLCNDV[IN:ND:Mld:92] DNA-A (U15016), ToLCNDV-[IN:ND:Svr:92] DNAA (U15015) and ToLCNDV-[IN:ND:92] DNA-B (U15017). Sequences were analysed using the DNASTAR package (version 1.02; DNASTAR) for a Macintosh computer. The multiple sequence alignments MEGALIGN 3.11 program within the DNASTAR package was used to align the CR sequences of the four genomic components. Biolistic inoculation. Cloned DNAs of ToLCNDV-[IN:ND:Svr:92] (NA and NB) and ToLCGV-[IN:Var:01] (VA and VB) were inoculated into 3-week-old seedlings of Nicotiana benthamiana, tomato (Lycopersicon esculentum cv. Organ Spring) and tobacco (Nicotiana tabacum cv. Xanthi) in different combinations (Tables 1 and 2) as described by Chakraborty et al. (2003b). Test plants were maintained under cycles of 16 h light/8 h dark for 6 weeks, scored for symptoms and analysed for viral DNA. DNA preparation. Total DNA extraction was carried out from newly emerging leaves of the test plants (L. esculentum cv. Organ Spring, N. benthamiana and N. tabacum cv. Xanthi) as described by Dellaporta et al. (1983). PCR. DNA from inoculated plants was screened by PCR with Taq DNA polymerase (Invitrogen). The reaction conditions were: denaturation at 94 uC for 2 min, 30 cycles of 1 min at 94 uC, 1 min at 55 uC and 1 min at 72 uC, and a final extension period of 10 min at 72 uC. Primer pairs specific to VA (GAv50: 59-TAACTGACAAAGACATGCGGA-39; GAc1122: 59-AACATTTGTAGACAGTTCAAATAT-39) and VB (GBv74: 59-CTCATCCGATTTGCAACACGT-39; GBc1280: 59-AATTGCATTTTCAAATTCCATGTTCGTACA-39), and primer pairs specific to NA (NDAv41: 59-TCAACCAATGAAATTCACGCTACATGG-39; NDAc1562: 59-CGCGAAAACATCGCTCTCCCAGAAGAAG-39) and NB (NDBv226: 59-TTTCACATAAATCAAATTGCCTTCTTTGCTATGTAT-39; NDBc1287: 59-AATAAGATGTTTCTTCTGCATTTTTATATATTA-39) were used for diagnostic amplification, and PCR products were resolved on 1.2 % agarose gels. Southern blot analysis. Total DNA (4 mg) was fractionated on a
1.2 % agarose gel and transferred to Hybond-N+ membrane (Amersham). Viral DNA was detected by hybridizing blots separately using radiolabelled probes of DNA-A and DNA-B specific to either species. For VA, a SacI–SnaBI fragment (nt 2330–934) and for VB a
819
S. Chakraborty and others
Table 1. Infectivity of genomic DNA-A and DNA-B component combinations of ToLCGV-[IN:Var:01] and ToLCNDV-[IN:ND:Svr:92] by biolistic inoculation on N. benthamiana Inoculum combination
Plants showing symptoms*
First symptom appearance (days p.i.)
Symptom severityD
15/15 15/15 15/15 15/15 11/15 15/15 15/15 15/15 15/15
5 4 4 6 8 6 4 4 3
++ ++++ ++++ ++ ++ ++ +++ ++++ +++++
NA+NB NA+VB NA+NB+VB VA+VB VA+NB VA+NB+VB NA+VA+NB NA+VA+VB NA+VA+NB+VB
*Number of plants showing symptoms/number of plants inoculated. DSeverity of symptoms was scored from mild (+) to severe (+++++).
HincII–XcmI fragment (nt 1029–2183) was used as the probe. For NA, a BamHI–EcoRI (nt 133–1701) fragment and for NB a PstI–SapI fragment (nt 2076–1329) was used. The DNA fragments were labelled with [a-32P]dCTP by random oligonucleotide-primed synthesis (Feinberg & Vogelstein, 1983). Viral DNA levels were quantified using a PhosphorImager (Molecular Dynamics).
Protoplast replication assay. Protoplasts were prepared from N.
tabacum BY2 suspension cell cultures as described previously (Fromm et al., 1986). Recombinant plasmids (4 mg) containing the respective viral DNA as a tandem dimeric construct and 20 mg herring sperm DNA were mixed and electroporated in 0.8 ml protoplast suspension. After transfection, protoplasts were maintained at 28 uC, and 48 h
Table 2. Infectivity of pseudorecombinants of ToLCGV-[IN:Var:01] and ToLCNDV-[IN:ND:Svr:92] in N. benthamiana, tomato and tobacco and detection of genomic components by PCR Inoculum/host
VA+VB N. benthamiana
Plants showing symptoms*
24/24
SymptomD
First symptom appearance (days p.i.)
Detection of DNA by PCRd DNA-A
DNA-B
Tomato Tobacco NA+NB N. benthamiana Tomato Tobacco VA+NB N. benthamiana
17/17 11/13
LC, IVCL, CR, LD, Chls, St, Bls, SL LC, CR, LR, YL, MO, VB, Chls
21/21 18/18 13/13
SL, YL, St, LC, Mt LC, CR, YL MO, VB, Chls
5 7 6
21/21 18/18 13/13
21/21 18/18 13/13
8
12/13
9/13
Tomato Tobacco NA+VB N. benthamiana
6/13 7/13
LC, IVCL, CR, Chls, St, Bls, SL Mild LC, YL MiMO, VB
16 9
9/13 9/13
6/13 7/13
22/22
4
22/22
22/22
Tomato Tobacco
13/13 13/13
5 6
13/13 13/13
13/13 13/13
9/13
Mt, MO, YL, St, Chls, Bls, LC, IVCL, LD LC, CR, LR, YL, St, Tw MO, VB, Chls, Mt
6
24/24
24/24
7 6
17/17 12/13
17/17 12/13
*Number of plants showing symptoms/number of plants inoculated. DLC, leaf curling; BV, big veins; MY, mild yellowing; YL, yellow leaf lamina; St, stunting; LD, leaf distortion; VB, vein banding; MO, severe mosaic; LR, leaf rolling; Bls, blistering; SL, small leaves; CR, leaf crinkling; IVCL, interveinal chlorosis; Chls, chlorotic spots; MiMO, mild mosaic; Mt, mottling; SYL, severe yellow leaf lamina; Tw, twisting of petiole. dNumber of plants where viral DNA presence was confirmed. 820
Journal of General Virology 89
Virulent pseudorecombination and synergism among ToLCVs later total DNA was extracted and subjected to Southern blot analysis as described previously (Vanitharani et al., 2003).
RESULTS Synergism between ToLCNDV and ToLCGV Tomato and N. benthamiana plants co-inoculated with the genomic components of ToLCNDV-[IN:ND:Svr:92] and ToLCGV-[IN:Var:01] showed unusually severe symptoms compared with plants infected with either virus alone (Fig. 1a, c). More interestingly, symptom severity was enhanced whenever VA was inoculated with NA irrespective of which DNA-B component was present. For example, when VA was mixed with the NA+NB combination the symptoms were severe (Fig. 1b, Table 1). Plants inoculated with both components of the two virus species exhibited symptoms at least 2 days earlier than those with either virus alone (Table 1). To investigate synergism at the level of viral DNA accumulation, Southern blot analysis was conducted with probes specific for each of the four DNA components (Fig. 2a–d). The level of NA in plants inoculated with NA+NB was 88 % of that detected for the NA+VB combination (Fig. 2a, lanes 1 and 2). However, the NA level increased to 1.8-fold when VA was also inoculated along with NA+NB, suggesting a synergistic interaction between VA and NA (Fig. 2a, lanes 2 and 8). A similar trend was observed when VA was inoculated with NA+VB (2.3-fold increased DNA accumulation) (Fig. 2a, lanes 2 and 7). Also, the NB level increased 1.9-fold when VA was co-inoculated with NA+NB (Fig. 2b, lanes 1 and 7). For VA, however, a different result was observed, in which the VA level was reduced by 58 % when it was co-inoculated with NA compared with the wild-type combination (Fig. 2c, lanes 5 and 7). VB also increased (8-fold) when it was inoculated with VA+NA, rather than VA alone (Fig. 2d, lanes 4 and 8). Together, these results indicate that the DNA-A components of ToLCGV-[IN:Var:01] and ToLCNDV[IN:ND:Svr:92] can interact synergistically with each other, causing an enhanced level of NA and a reduced level of VA in dually infected plants. Infectivity of ToLCNDV-[IN:ND:Svr:92], ToLCGV[IN:Var:01] and their pseudorecombinants in various plant species N. benthamiana, tomato and tobacco plants inoculated with ToLCGV-[IN:Var:01] developed systemic symptoms (Table 2). In N. benthamiana, symptoms appeared as downward leaf curling by 6 days post-inoculation (p.i.), which later developed into typical downward and upward leaf curling, enlarged veins, puckering of the leaves with interveinal chlorosis and stunted growth at 14 days p.i. (Fig. 1d; Table 2). Yellowing of leaf lamina, leaf curling and leaf surface reduction along with stunted growth occurred in tomato (Fig. 1e, Table 2) and severe mosaic and vein banding symptoms occurred in tobacco (Table 2). http://vir.sgmjournals.org
Disease symptoms in ToLCNDV-[IN:ND:Svr:92]-infected N. benthamiana plants appeared by 5 days p.i. (Fig. 1h). Later, leaf curling, yellowing of leaf lamina and mottling symptoms were observed (Fig. 1d, Table 2). Typical leaf curl symptoms along with yellowing of leaf lamina and stunting were also observed in tomato infected by ToLCNDV-[IN:ND:Svr:92] (Fig. 1e, Table 2). ToLCNDV[IN:ND:Svr:92] could also infect tobacco and symptoms appeared at 6 days p.i. (Table 2). More severe systemic symptoms on N. benthamiana and tomato appeared 1–2 days earlier upon inoculation with the pseudorecombinant NA+VB. All N. benthamiana plants showed symptoms of leaf curling at 4 days p.i., which finally developed into severe mosaic mottling, leaf distortion, chlorosis and bleaching of leaf lamina, and stunted growth (Fig. 1d, f; Table 2). Similar symptoms occurred in tomato and tobacco (Fig. 1e, Table 2). Unlike NA+VB, VA+NB displayed mild symptoms in N. benthamiana, tomato and tobacco (Table 2). For each virus/host combination, including the one that failed to show distinguishable disease symptoms, all inoculated plants were examined for infection by PCR analysis with specific primers (Table 2). In the case of the pseudorecombinant VA+NB, symptomless infections were detected in all three plant species tested. For example, in N. benthamiana, only 69 % of the inoculated plants showed symptoms but 92 % were infected, based on PCR analysis (Table 2). Both DNA components were detected for ToLCNDV-[IN:ND:Svr:92], ToLCGV-[IN:Var:01] and the virulent pseudorecombinant (NA+VB) in all of the plants (Table 2). However, for the other pseudorecombinant/host combination (VA+NB), which resulted in symptomless infection, only the DNA-A component was detected in some plants. For example, VA was detected in 12/13 N. benthamiana plants inoculated with VA+NB, whereas NB was detected in only nine of these plants. Fig. 1(h) shows the symptom severity in N. benthamiana for wild-type and pseudorecombinant combinations. It was evident that symptoms were severe and appeared earlier when the pseudorecombinant (NA+VB) was inoculated compared with the wild-type virus ToLCNDV[IN:ND:Svr:92] or ToLCGV-[IN:Var:01]. However, inoculation with the VA+NB combination resulted in delayed and reduced symptom induction. Hence, the disease severity associated with each pseudorecombinant combination was distinct, whilst little difference was observed between the wild-type viruses. DNA levels of ToLCNDV-[IN:ND:Svr:92], ToLCGV-[IN:Var:01] and their pseudorecombinants in inoculated plants Both DNA components were detected in N. benthamiana plants infected with ToLCNDV-[IN:ND:Svr:92], ToLCGV[IN:Var:01] and the pseudorecombinants (data not shown). Levels of VA and NA were 14–24 % higher in plants 821
S. Chakraborty and others
Fig. 1. Severe systemic symptoms in N. benthamiana and tomato induced by synergistic interaction (a–c) between ToLCNDV[IN:ND:Svr:92] and ToLCGV-[IN:Var:01] and pseudorecombinants (d–h) between DNA-A of ToLCNDV-[IN:ND:Svr:92] (NA) and DNA-B of ToLCGV-[IN:Var:01] (VB). (a) Synergistic effect in N. benthamiana of a mixture of components of both species (middle plant) compared with the homologous components at 10 days p.i. (b) More severe symptoms induced by VA+NA+NB compared with NA+NB at 10 days p.i. (c) Severe symptoms in tomato at 30 days p.i., induced by (from left to right) ToLCNDV[IN:ND:Svr:92], ToLCNDV-[IN:ND:Svr:92]+ToLCGV-[IN:Var:01] and ToLCGV-[IN:Var:01]. (d) Symptoms in N. benthamiana at 21 days p.i. induced by (from left to right) wild-type ToLCGV-[IN:Var:01], a pseudorecombinant between DNA-A of ToLCNDV[IN:ND:Svr:92] (NA) and DNA-B of ToLCGV-[IN:Var:01] (VB), wild-type ToLCNDV-[IN:ND:Svr:92] and a mock-inoculated control plant. (e) Symptoms in tomato at 25 days p.i. induced by (from left to right) wild-type ToLCGV-[IN:Var:01], a pseudorecombinant of NA and VB, and wild-type ToLCNDV-[IN:ND:Svr:92]. (f) Symptoms of infection induced by the virulent pseudorecombinant in N. benthamiana (close-up view). (g) Symptoms of infection in N. benthamiana caused by the pseudorecombinant between ToLCNDV-[IN:ND:Mld:92] (NAm) DNA-A and VB (right) compared with NAm DNA-A and NB (middle), and a mock-inoculated control (left). (h) Symptom development in N. benthamiana inoculated with wild-type and pseudorecombinant DNAs of ToLCVs.
infected with VB compared with NB (Fig. 2a, lanes 1 and 2; Fig. 2c, lanes 4 and 5). However, accumulation of VB and NB was higher in the presence of NA than VA (Fig. 2b, d). 822
In case of the virulent NA+VB pseudorecombinant, it was observed that accumulation of VB was 4.5-fold higher than the homologous VA+VB combination (Fig. 2d, lanes 2 Journal of General Virology 89
Virulent pseudorecombination and synergism among ToLCVs
Fig. 2. Southern blot analysis showing relative levels of viral DNA accumulation in N. benthamiana plants inoculated with combinations of ToLCV DNA components 21 (a–f) and 30 days (g) p.i. Total DNA (4 mg) isolated from infected plants was loaded in each lane. The blots were hybridized with probes specific to NA (a), NB (b), VB (d, e, g) and VA (c, f). Viral DNA forms are indicated: LIN, double-stranded linear form; OC, open circle; SC, supercoiled; SS, single-stranded. Inoculum combinations are given at the top of each lane. The percentage DNA accumulation specific to each component is indicated below each blot.
and 4; Fig. 2e, lanes 1 and 2). These results showed that VB was efficiently trans-replicated and accumulated more in the presence of NA than in the presence of its cognate DNA-A. Also, accumulation of VA was 10-fold higher in plants in which DNA-B was also detected compared with plants with only DNA-A (Fig. 2f). From these results, it is clear that the severe phenotype is associated with an increased level of DNA-B accumulation in plants. http://vir.sgmjournals.org
Replication of viral DNA components in tobacco protoplasts To investigate whether the virulence of the pseudorecombinant accompanied by higher levels of viral DNA accumulation was due to efficient replication, we performed tobacco protoplast-based replication assays. Protoplasts generated from suspension cultures of N. 823
S. Chakraborty and others
tabacum cell line BY2 were inoculated with various combinations of viral DNA: NA, VA, VA+VB, NA+NB, NA+VA, NA+VB, VA+NB, NA+VA+VB and NA+VA+NB. Replicative viral DNAs were detected as supercoiled circular and single-stranded DNA. Replicative DNA-A of either VA or NA was always detected (Fig. 3a, b), as was DNA-B (NB or VB) when transfected along with either of the DNA-A components (VA or NA) (Fig. 3c, d). The results were consistent with the infectivity data of the pseudorecombinants in plants. The level of NA increased 1.97-fold in the presence of VA (Fig. 3a, lanes 1 and 3). Similarly, the level of NA also increased more than 2-fold in the combinations NA+VA+NB and NA+VA+VB compared with NA+NB and NA+VB, respectively (Fig. 3a, lanes 2, 4, 6 and 7). This clearly indicated the synergistic role of VA in enhancing DNA accumulation in the combinations tested. However, the level of VA remained unchanged in all of the combinations tested, indicating that VA replication is not enhanced (Fig. 3b, lanes 1–6). In the case of NB, replication increased ~1.7fold when VA was transfected along with NA+NB compared with NA+NB alone, whilst NB accumulation
was reduced by 48 % when trans-complemented with VA (Fig. 3c, lanes 1–3). Fig. 3(d) shows that VB was transreplicated more efficiently (~3-fold) with NA than with VA (lanes 1 and 2). Similarly, the combination of VA+NA caused a dramatic increase in VB accumulation (~5.4-fold) compared with VA alone (Fig. 3d, lanes 1 and 3). Comparison of the CRs of ToLCNDV[IN:ND:Svr:92] and ToLCGV-[IN:Var:01] Sequence identity between the CR of VB and the CRs of ToLCNDV are 83 % with NA and 86 % with NB. However, the CRs of VA and VB are only 61 % identical. The lowest sequence identity (52 %) was observed between the CRs of VA and NB. A multiple alignment of the CRs and intercistronic regions of isolates of these two species revealed that the genomic components of ToLCNDV and VB have similar iteron (in bold) sequences (GGTGTCTGGAGT) (Fig. 4). In contrast, the iterons of DNA-A of ToLCGV-[IN:Var:01] and ToLCNDV-[IN:ND:Mld:92] are GGTGTATTGGAGT and GGCGTCTGGCGT, respectively. The only difference between the CR of the DNA-A of the two species is found in the spacer sequence (underlined) between the two iterons, which is 2 nt in the case of ToLCNDV and 3 nt in ToLCGV (Fig. 4). We previously also observed the presence of a third identical iteron at the 59 end of the CRs (Chakraborty et al., 2003b). The CR of VA, when compared with VB, NA and NB, shares more than 85 % identity in the first 80 nt, whereas the region between the TATA box and the hairpin loop is below 50 % (Chakraborty et al., 2003b) (Fig. 4). Inoculation of VB with ToLCNDV-[IN:ND:Mld:92]
Fig. 3. Southern blot analysis of total DNA extracted from tobacco BY2 cells transfected with various combinations of viral DNA components at 48 h post-electroporation. Blots were hybridized with specific probes for DNA-A (NA or VA) or DNA-B (NB or VB). Total DNA (4 mg) isolated from protoplasts was loaded in each lane. Viral DNA forms are indicated as scDNA (supercoiled) and ssDNA (single-stranded). The percentage DNA accumulation specific to each component is indicated below each blot. 824
To investigate whether the pseudorecombinant between ToLCNDV-[IN:ND:Mld:92] (NAm) and VB could infect N. benthamiana and tomato, tandem dimeric constructs of these two genomic components were mixed together and inoculated into both species by biolistic delivery. Systemic symptoms in N. benthamiana plants appeared as small yellow spots on leaf lamina at 16 days p.i., later developing into mosaic mottling, leaf curling and chlorotic spots (Fig. 1g). A total of 6/15 N. benthamiana plants tested showed symptoms, whilst NAm+NB did not induce any symptoms. Southern blot analysis revealed that VB accumulation in plants infected with NAm+VB was 25 % of that for the VA+VB combination, indicating that NAm did not facilitate efficient replication of VB but nevertheless allowed it to occur (Fig. 2g, 30 days p.i.). Replication assays in protoplasts supported the data from plant infections and confirmed that NAm can replicate VB. VB accumulated to 50 % in the presence of NAm compared with VA, as shown by Southern blot analysis (data not shown).
DISCUSSION To examine whether synergism exists between members of two distinct begomovirus species causing tomato leaf curl Journal of General Virology 89
Virulent pseudorecombination and synergism among ToLCVs
Fig. 4. Comparison of the CR sequences of all strains of each of the species of tomato leaf curl virus studied here. The percentage identity within each box is indicated.
disease in India, genomic components of the severe strain ToLCNDV-[IN:ND:Svr:92] and the Varanasi isolate ToLCGV-[IN:Var:01] were co-inoculated in N. benthamiana and tomato cv. Organ Spring. To determine whether ToLCNDV-[IN:ND:Svr:92] and ToLCGV-[IN:Var:01] formed infectious pseudorecombinants, we mixed the DNA-A and DNA-B components of these two species (NA and NB, and VA and VB, respectively) and studied their infectivity and DNA accumulation in N. benthamiana and tomato. In addition, we carried out replication analyses in N. tabacum BY2 suspension cell protoplasts. Doubly infected plants inoculated with mixtures of NA+NB and VA+VB produced extremely severe symptoms compared with plants infected with isolates of only one virus species. Southern blot analysis of inoculated plants showed a greater accumulation of NA (2.3-fold) and VB (13-fold) in mixed-infection plants compared with singly infected plants. In contrast, the amount of VA decreased (39 % accumulation), indicating an asymmetric relationship, although the level of NB increased 3-fold. To test whether there was any interaction between ToLCGV[IN:Var:01] and ToLCNDV-[IN:ND:Svr:92], protoplast replication assays were conducted. It was observed that association with VA resulted in an increased level of NA due to enhanced replication of this component. It was also observed that VB was trans-replicated more efficiently (3fold) in combination with NA than with its own DNA-A, VA. In contrast, VA did not support efficient replication of NB (52 % accumulation). The synergistic role of VA and NA, resulting in a much higher level of NA, in turn caused more efficient replication of the DNA-B components, particularly VB, resulting in greater viral DNA accumulation and consequently more severe symptoms in the systemically infected leaves. Synergism between members of two distinct geminivirus species has been demonstrated previously (Harrison et al., 1997; Fondong et al., 2000; Pita et al., 2001). However, no information about pseudorecombination between the components of different isolates of two begomovirus species causing leaf curl disease of tomato is currently available and synergism between such isolates has not been http://vir.sgmjournals.org
demonstrated. Here, we have shown that synergism between two geminiviruses infecting tomato resulted in an increase in viral DNA accumulation and symptom severity. In addition, we demonstrated the occurrence of a more virulent pseudorecombinant between members of the two species, which may explain the sudden breakdown of resistance in tomato cultivars and the development of epidemics in tomato-growing areas in India. Recently, we detected the presence of both ToLCGV-[IN:Var:01] and ToLCNDV-[IN:ND:Svr:92] components in a single severely infected tomato plant under natural conditions (S. Chakraborty and C. M. Fauquet, unpublished data). As ToLCNDV-[IN:ND:Svr:92] and ToLCGV-[IN:Var:01] share the same hosts and are transmitted by the same whitefly vector, they are likely to co-exist in infected plants. The synergism between the two viruses increases the amount of both viruses in the systemically infected leaves and enhances their chance of transmission. As a consequence, doubly infected plants have considerable potential as a source of inoculum for both viruses, and whiteflies feeding on such plants would therefore more easily acquire and transmit both viruses, providing another source of geminivirus biodiversity. The synergism is similar to that observed for potyviruses, which mediate the accumulation of potexviruses, comoviruses and machlomoviruses, although their levels remain relatively unchanged during mixed infections (Damirdagh & Ross, 1967; Calvert & Ghabrial, 1983; Goldberg & Brakke, 1987; Vance, 1991; Vance et al., 1995). Synergism between isolates of two cassava-infecting geminiviruses, ACMV and EACMV, has been observed and the genes mediating synergism have been identified (Fondong et al., 2000; Pita et al., 2001; Vanitharani et al., 2004). Synergism between ACMV and East African cassava mosaic Cameroon virus (EACMCV) is a two-way process, as the presence of the DNA-A component of ACMV or EACMCV enhanced the accumulation of viral DNA of EACMCV and ACMV, respectively, in tobacco BY2 protoplasts (Vanitharani et al., 2004). ACMV AC4 and EACMCV AC2, the putative synergistic genes, are able to suppress post-transcriptional gene silencing (Vanitharani et al., 2004). In the present study, we demonstrated that NA and NB accumulation was 825
S. Chakraborty and others
also increased due to synergism, whilst the level of VA remained unchanged or decreased, indicating a unique asymmetric interaction. All of the pseudorecombinants were infectious in their field host, tomato, as well as in the common laboratory hosts N. benthamiana and tobacco. Symptoms observed by inoculating the pseudorecombinant NA+VB were much more severe in comparison with the wild-type virus isolates. The latent period before the onset of symptoms was also reduced by 2–6 days. We observed a strong positive correlation between symptom severity and viral DNA accumulation, suggesting that VB was trans-replicated more efficiently by NA than by VA. In contrast, symptom severity was less and the test plants took longer to develop symptoms when the VA+NB combination was used. The symptom severity pattern of the supervirulent pseudorecombinant between NA+VB was very different compared with both NA+NB and VA+VB infections. PCR analysis confirmed the presence of the expected DNA-A and DNAB components in each experiment and the absence of potential contamination with homologous components. In some plants inoculated with VA+NB, only DNA-A was detected. It is relevant to mention here that some viruses such as tomato yellow leaf curl Thailand virus have the capacity to induce symptoms with component A alone, whilst others like ToLCGV can replicate and move but do not induce symptoms, as observed here (Fig. 2f). Most pseudorecombinants of closely related isolates of one species have been found to be infectious. For example, mixtures of Kenyan and Nigerian isolates of ACMV (ACMV-[KE:82] and ACMV-[NG], respectively) could infect N. benthamiana and produced wild-type symptoms (Kenyan type) when ACMV-[KE:82] DNA-A was inoculated with ACMV-[NG ] DNA-B (Stanley et al., 1985). For other begomoviruses [e.g. Dominican Republic (DR) and Guatemalan (GA) isolates of BGMV, common and severe strains of TGMV, and SqLCV isolates with broad and restricted host ranges], exchange of genomic components resulted in infectious pseudorecombinants (Lazarowitz, 1991; von Arnim & Stanley, 1992; Ingham & Lazarowitz, 1993; Faria et al., 1994) and induced symptoms resembling wild-type infections, except for BGMV-[DR] DNA-A and BGMV-[GA] DNA-B, which produced delayed and attenuated symptoms (Faria et al., 1994). In the latter case, the heterologous combinations may not interact as efficiently as the homologous combinations. However, in no case was a more virulent pseudorecombinant demonstrated. The viable nature of these pseudorecombinants was attributed to the highly conserved nature of AC1 and the CR, and also to identical iteron sequences. A highly specific interaction between the Rep proteins of mild and severe strains of ToLCNDV and their cognate iteron sequences also demonstrated the intimate relationship between these elements and the consequence in terms of DNA accumulation and symptoms (Chatterji et al., 1999, 2000). This lead to the concept that matching N-terminal amino acid sequences of the Rep protein and iteron sequences are 826
required for efficient replication and consequently a severe symptomatology. It also supported the species concept, with the exception of recombinants in these regions, where these two elements differ between species. In this respect, it is surprising to find a field isolate of a very severe virus such as ToLCGV-[IN:Var:01] having a CR sharing only ~60 % identity between the DNA-A and DNA-B components, although their iterons are identical, differing only in the numbers of spacing nucleotides. An important result described in this paper is the evidence of a more virulent pseudorecombinant provided by VB when co-inoculated with NA. This DNA-B component is highly similar to NB (they share 80 % overall nucleotide identity and their CRs are 86 % identical) (Chakraborty et al., 2003b). One of the characteristics of plants infected with the NA+VB pseudorecombinant was an increased level of VB accumulation. The replication assays conducted in tobacco BY2 cells supported the hypothesis that VB replication was more efficient for this pseudorecombinant. This is presumably due to a more efficient interaction of the Rep protein of ToLCNDV-[IN:ND:Svr:92] with the DNA-B CR of ToLCGV-[IN:Var:01] than with its own DNA-B CR. It is relevant here to mention that AC1 of VA shares only 71 % nucleotide identity with NA and that the Rep proteins are 78 % similar in their predicted amino acid sequence (Chakraborty et al., 2003b). We have shown that the pseudorecombinant VA+NB is poorly infectious in N. benthamiana, tomato and tobacco plants, presumably due to the low sequence identity in their CRs (52 %), which may result in inefficient interaction of the ToLCGV[IN:Var:01] Rep protein with the heterologous DNA-B CR. However, it is interesting to note that the iterons of both DNA-B components are identical, and that there are only single-nucleotide differences between the two CRs. In conclusion, we have reported for the first time synergism between two tomato leaf curl geminiviruses occurring in India. We conclude that exchange of genomic components of the members of two distinct species of begomoviruses causing tomato leaf curl disease in India can form infectious pseudorecombinants, and furthermore that the heterologous pseudorecombinant NA+VB is more virulent than the original homologous virus composition. In addition, NA is able to replicate and move VB very efficiently, despite the non-matching iterons (Chatterji et al., 1999, 2000). This clearly indicates that a perfect match between the N-terminal Rep sequence and the iteron sequences is not the determining factor in this case. This suggests that there are other factors that compensate a nonmatching iteron interaction between the Rep of ToLCNDV-[IN:ND:Mld:92] and ToLCGV-[IN:Var:01] DNA-B CR. In addition, all the experiments in protoplasts and in planta showed that the replication and accumulation of the ToLCGV-[IN:Var:01] DNA-B component was much higher and may consequently translate into a better movement of both viral DNAs, thereby inducing more severe symptoms. Further experiments on the exchange of open reading frames and CRs of DNA-B among the two Journal of General Virology 89
Virulent pseudorecombination and synergism among ToLCVs
species and their effect on plant phenotype and replication in protoplasts may lead to a better understanding of the molecular mechanisms involved in asymmetric synergism.
Fondong, V. N., Pita, J. S., Rey, M. E. C., de Kochko, A., Beachy, R. N. & Fauquet, C. M. (2000). Evidence of synergism between African
cassava mosaic virus and a new double-recombinant geminivirus infecting cassava in Cameroon. J Gen Virol 81, 287–297. Fromm, M. E., Taylor, L. P. & Walbot, V. (1986). Stable transformation
ACKNOWLEDGEMENTS
of maize after gene transfer by electroporation. Nature 319, 791–793.
The study was supported by the Danforth Plant Science Center and a BOYSCAST Fellowship from India for S. C. The authors would like to thank the greenhouse personnel for their good care of the plants.
Gilbertson, R. L., Hidayat, S. H., Paplomatas, E. J., Rojas, M. R., Hou, Y.-M. & Maxwell, D. P. (1993). Pseudorecombination between
infectious cloned DNA components of tomato mottle and bean dwarf mosaic geminiviruses. J Gen Virol 74, 23–31. Goldberg, K. B. & Brakke, M. K. (1987). Concentration of maize
REFERENCES
chlorotic mottle virus increased in mixed infections with maize dwarf mosaic virus, strain B. Phytopathology 77, 162–167.
Calvert, L. A. & Ghabrial, S. A. (1983). Enhancement by soybean
Harrison, B. D., Zhou, X., Otim-Nape, G. W., Liu, Y. & Robinson, D. J. (1997). Role of a novel type double infection in the geminivirus-
mosaic virus of bean pod mottle virus titer in doubly infected soybean. Phytopathology 73, 992–997. Chakraborty, S., Pandey, P. K., Banerjee, M. K., Kalloo, G. & Fauquet, C. M. (2003a). A new begomovirus species causing tomato leaf curl
disease in Varanasi, India. Plant Dis 87, 313. Chakraborty, S., Pandey, P. K., Banerjee, M. K., Kalloo, G. & Fauquet, C. M. (2003b). Tomato leaf Gujarat virus, a new begomovirus species
causing a severe leaf curl disease of tomato in Varanasi, India. Phytopathology 93, 1485–1496. Chatchawankanphanich, O. & Maxwell, D. P. (2002). Tomato leaf
curl Karnataka virus from Bangalore, India, appears to be a recombinant begomovirus. Phytopathology 92, 637–645. Chatterji, A., Padidam, M., Beachy, R. N. & Fauquet, C. M. (1999).
Identification of replication specificity determinance in tomato leaf curl virus from New Delhi. J Virol 73, 5481–5489. Chatterji, A., Chatterji, U., Beachy, R. N. & Fauquet, C. M. (2000).
Sequence parameters that determine specificity of binding of the replication-associated protein to its cognate in two strains of Tomato leaf curl virus-New Delhi. Virology 273, 341–350. Damirdagh, I. S. & Ross, A. F. (1967). A marked synergistic
interaction of potato viruses X and Y in inoculated leaves of tobacco. Virology 31, 296–307. Dellaporta, S. L., Woods, J. & Hicks, J. B. (1983). A plant DNA
induced epidemic of severe cassava mosaic in Uganda. Ann Appl Biol 131, 437–448. Ingham, D. J. & Lazarowitz, S. G. (1993). A single missense mutation
in the BR1 movement protein alters the host range of squash leaf curl virus. Virology 196, 694–770. Laufs, J., Jupin, I., David, C., Schumacher, S., Heyraud-Nitschke, F. & Gronenborn, B. (1995). Geminivirus replication: genetic and
biochemical characterization of Rep protein function, a review. Biochimie 77, 765–773. 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. Muniyappa, V., Venkatesh, H. M., Ramappa, H. K., Kulkarni, R. S., Zeidan, M., Tarba, C. Y., Ghanim, M. & Czosnek, H. (2000). Tomato
leaf curl virus from Bangalore (ToLCV-Ban4): sequence comparison with Indian ToLCV isolates, detection in plants and insects, and vector relationships. Arch Virol 145, 1583–1598. Navot, N., Pichersky, E., Zeidan, M., Zamir, D. & Czosnek, H. (1991).
Tomato yellow leaf virus: a whitefly-transmitted geminivirus with a single genomic component. Virology 185, 151–161. Padidam, M., Beachy, R. N. & Fauquet, C. M. (1995). Tomato leaf curl
minipreparation: version II. Plant Mol Biol Rep 1, 19–21.
geminivirus from India has a bipartite genome and coat protein is not essential for infectivity. J Gen Virol 76, 25–35.
Dry, I. B., Rigden, J. E., Krake, L. R., Mullineaux, P. M. & Rezaian, M. A. (1993). Nucleotide sequence and genome organization of tomato leaf
Pita, J. S., Fondong, V. N., Sangare, A., Otim-Nape, G. W., Ogwal, S. & Fauquet, C. M. (2001). Recombination, pseudorecombination and
curl geminivirus. J Gen Virol 74, 147–151. Eagle, P. A., Orozco, B. M. & Hanley-Bowdoin, L. (1994). A DNA
sequence required for geminivirus replication also mediates transcriptional regulation. Plant Cell 6, 1157–1170.
synergism of geminiviruses are the determinant keys to the epidemic of severe cassava mosaic disease in Uganda. J Gen Virol 82, 655–665. Sanderfoot, A. A. & Lazarowitz, S. G. (1996). Getting it together in
Faria, J. C., Gilbertson, R. L., Hanson, S. F., Morales, F. J., Ahlquist, P., Loniello, O. & Maxwell, D. P. (1994). Bean golden mosaic geminivirus
plant virus movement: cooperative interactions between bipartite geminivirus movement proteins. Trends Cell Biol 6, 353–358.
type III isolates from the Dominican Republic and Guatemala: nucleotide sequences, infectious pseudorecombinants, and phylogenetic relationships. Phytopathology 84, 321–329.
Srivastava, K. M., Hallan, V., Raizada, R. K., Chandra, G., Singh, B. P. & Sane, P. V. (1995). Molecular cloning of Indian tomato leaf
Fauquet, C. M. & Stanley, J. (2003). Geminivirus classification
and nomenclature: progress and problems. Ann Appl Biol 142, 165–189.
curl virus genome following a simple method of concentrating the supercoiled replicative form of viral DNA. J Virol Methods 51, 297–304. Stanley, J., Townsend, R. & Curson, S. J. (1985). Pseudore-
Fauquet, C. M., Bisaro, D. M., Briddon, R. W., Brown, J. K., Harrison, B. D., Rybicki, E. P., Stenger, D. C. & Stanley, J. (2003). Revision of
combinants between cloned DNAs of two isolates of cassava latent virus. J Gen Virol 66, 1055–1061.
taxonomic criteria for species demarcation in the family Geminiviridae, and an updated list of begomovirus species. Arch Virol 148, 405–421.
Stanley, J., Bisaro, D. M., Briddon, R. W., Brown, J. K., Fauquet, C. M., Harrison, B. D., Rybicki, E. P. & Stenger, D. C. (2005). Geminiviridae.
Feinberg, A. P. & Vogelstein, B. (1983). A technique for radiolabelling
DNA restriction endonuclease fragments to high specific activity. Ann Biochem 132, 6–13. http://vir.sgmjournals.org
In Virus Taxonomy. Eighth Report of the International Committee on Taxonomy of Viruses, pp. 301–326. Edited by C. M. Fauquet, M. A. Mayo, J. Maniloff, U. Desselberger & L. A. Ball. London: Elsevier/ Academic Press. 827
S. Chakraborty and others Sunter, G. & Bisaro, D. M. (1991). Transactivation in a geminivirus:
AL2 gene product is needed for coat protein expression. Virology 180, 416–419. Sunter, G. & 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., Harititz, M. D., Hormuzdi, S. G., Brough, C. L. & Bisaro, D. M. (1990). Genetic analysis of tomato golden mosaic virus: ORF
AL2 is required for coat protein accumulation while ORF AL3 is necessary for efficient DNA replication. Virology 179, 69–77. Van Wezel, R., Dong, X., Blake, P., Stanley, J. & Hong, Y. (2002).
Differential roles of geminivirus Rep and AC4 (C4) in the induction of necrosis in Nicotiana benthamiana. Mol Plant Pathol 3, 461–471. Vance, V. B. (1991). Replication of potato virus X RNA is altered in
coinfections with potato virus Y. Virology 182, 486–494.
828
Vance, V. B., Berger, P. H., Carrington, J. C., Hunt, A. G. & Shi, X. M. (1995). 59 promximal potyviral sequences mediate potato virus X/
potyviral synergistic disease in transgenic tobacco. Virology 206, 583–590. Vanitharani, R., Chellappan, P. & Fauquet, C. M. (2003). Short
interfering RNA-mediated interference of gene expression and viral DNA accumulation in cultured plant cells. Proc Natl Acad Sci U S A 100, 9632–9636. Vanitharani, R., Chellappan, P., Pita, J. S. & Fauquet, C. M. (2004).
Differential roles of AC2 and AC4 of cassava geminiviruses in mediating synergism and suppression of posttranscriptional gene silencing. J Virol 78, 9487–9493. von Arnim, A. & Stanley, J. (1992). Determinants of tomato golden
mosaic virus symptom development located on DNA-B. Virology 186, 286–293.
Journal of General Virology 89
