Portal de Periódicos da CAPES

Expression of the Oligomerization Domain of the Replication-associated Protein (Rep) of Tomato Leaf Curl New Delhi VirusInterferes with DNA Accumulation of Heterologous Geminiviruses

2001; Elsevier BV; Volume: 276; Issue: 27 Linguagem: Inglês

10.1074/jbc.m100030200

1083-351X

Anju Chatterji, Roger N. Beachy, Claude Fauquet,

Legume Nitrogen Fixing Symbiosis

The minimal DNA binding domain of the replication-associated protein (Rep) of Tomato leaf curl New Delhi viruswas determined by electrophoretic mobility gel shift analysis and co-purification assays. DNA binding activity maps to amino acids 1–160 (Rep-(1–160)) of the Rep protein and overlaps with the protein oligomerization domain. Transient expression of Rep protein (Rep-(1–160)) was found to inhibit homologous viral DNA accumulation by 70–86% in tobacco protoplasts and in Nicotiana benthamiana plants. The results obtained showed that expression of N-terminal sequences of Rep protein could efficiently interfere with DNA binding and oligomerization activities during virus infection. Surprisingly, this protein reduced accumulation of the African cassava mosaic virus, Pepper huasteco yellow vein virusandPotato yellow mosaic virusby 22–48%. electrophoretic mobility shift assays and co-purification studies showed that Rep-(1–160) did not bind with high affinity in vitro to the corresponding common region sequences of heterologous geminiviruses. However, Rep-(1–160) formed oligomers with the Rep proteins of the other geminiviruses. These data suggest that the regulation of virus accumulation may involve binding of the Rep to target DNA sequences and to the other Rep molecules during virus replication. The minimal DNA binding domain of the replication-associated protein (Rep) of Tomato leaf curl New Delhi viruswas determined by electrophoretic mobility gel shift analysis and co-purification assays. DNA binding activity maps to amino acids 1–160 (Rep-(1–160)) of the Rep protein and overlaps with the protein oligomerization domain. Transient expression of Rep protein (Rep-(1–160)) was found to inhibit homologous viral DNA accumulation by 70–86% in tobacco protoplasts and in Nicotiana benthamiana plants. The results obtained showed that expression of N-terminal sequences of Rep protein could efficiently interfere with DNA binding and oligomerization activities during virus infection. Surprisingly, this protein reduced accumulation of the African cassava mosaic virus, Pepper huasteco yellow vein virusandPotato yellow mosaic virusby 22–48%. electrophoretic mobility shift assays and co-purification studies showed that Rep-(1–160) did not bind with high affinity in vitro to the corresponding common region sequences of heterologous geminiviruses. However, Rep-(1–160) formed oligomers with the Rep proteins of the other geminiviruses. These data suggest that the regulation of virus accumulation may involve binding of the Rep to target DNA sequences and to the other Rep molecules during virus replication. common region Tomato golden mosaic virus African cassava mosaic virus Tomato leaf curl New Delhi virus electrophoretic mobility shift assay glutathione S-transferase phosphate-buffered saline Pepper huasteco yellow vein virus Potato yellow mosaic virus Geminiviruses cause economically significant diseases in a wide range of cereal, vegetable, and fiber crops (1Brown J.K. FAO Plant Prot. Bull. 1994; 42: 3-32Google Scholar). These viruses have a single-stranded DNA genome that is replicated in nuclei of infected cells by a rolling circle mechanism (2Saunders K. Lucy A. Stanley J. Nucleic Acids Res. 1991; 19: 2325-2330Crossref PubMed Scopus (167) Google Scholar, 3Stenger D.C. Revington G.N. Stevenson M.C. Bisaro D.M. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 8029-8033Crossref PubMed Scopus (289) Google Scholar). Of the different gene products encoded by the virus, only AC1, the replication-associated protein (Rep), is essential for viral DNA replication. The first step in the replication process involves recognition of specific DNA sequences referred to as iterons, (4Arguello-Astorga G.R. Guevara-Gonzalez R.G. Herrera-Estrella L.R. Rivera-Bustamante R.F. Virology. 1994; 203: 90-100Crossref PubMed Scopus (278) Google Scholar), by the Rep protein in the common region (CR)1 of the virus genome. Most iteron sequences occur as direct repeat motifs of 6–12 base pairs between the TATA box and the start site of transcription of theAC1 gene. The iterons serve as high affinity binding sites of the Rep protein and therefore function as the origin recognition sequences. Specific regions on the N terminus of Rep protein are involved in DNA binding and have been identified for Tomato golden mosaic virus(TGMV) (5Fontes E.P.B. Gladfelter H.J. Schaffer R.L. Petty I.T.D. Hanley-Bowdoin L. Plant Cell. 1994; 6: 405-416Crossref PubMed Scopus (171) Google Scholar, 6Fontes E.P.B. Eagle P.A. Sipe P.S. Luckow V.A. Hanley-Bowdoin L. J. Biol. Chem. 1994; 269: 8459-8465Abstract Full Text PDF PubMed Google Scholar), African cassava mosaic virus(ACMV) (7Hong Y. Stanley J. J. Gen. Virol. 1995; 76: 2415-2422Crossref PubMed Scopus (64) Google Scholar), and Tomato yellow leaf curl virus(8Jupin I. Hericourt F. Benz B. Gronenborn B. FEBS Lett. 1995; 262: 116-120Crossref Scopus (74) Google Scholar). The potential binding site sequences in the common region of theTomato leaf curl New Delhi virus(ToLCNDV) (9Fauquet C.M. Maxwell D.P. Gronenborn B. Stanley J. Arch. Virol. 2000; 145: 1743-1761Crossref PubMed Scopus (70) Google Scholar) genome were identified by site-directed mutagenesis (10Chatterji A. Padidam M. Beachy R.N. Fauquet C.M. J. Virol. 1999; 73: 5481-5489Crossref PubMed Google Scholar). Further analyses using gel shift assays confirmed that the Rep protein specifically binds to the iterated motifs GGTGTCTGGAGTC (nucleotides 2640–2653) in the origin of replication (11Chatterji A. Chatterji U. Beachy R.N. Fauquet C.M. Virology. 2000; 273: 314-350Crossref Scopus (64) Google Scholar). In the present study, our objective was to identify the DNA binding domain of the Rep protein and to determine the nature and contribution of DNA binding and protein oligomerization properties of the Rep protein to limit viral DNA accumulation in plants. In two cases, truncated Rep proteins have been shown to confer resistance to other geminiviruses (7Hong Y. Stanley J. J. Gen. Virol. 1995; 76: 2415-2422Crossref PubMed Scopus (64) Google Scholar, 12Noris E. Accotto G.P. Tavazza R. Brunetti A. Crespi S. Tavazza M. Virology. 1996; 224: 130-138Crossref PubMed Scopus (113) Google Scholar), and the resistance was specific and limited to the homologous virus. We based our choice of truncated Rep protein on the knowledge of overlapping sites for DNA cleavage, domains for DNA binding, and domains for protein oligomerization (13Orozco B.M. Miller A.B. Stellage S.B. Hanley-Bowdoin L. J. Biol. Chem. 1997; 272: 9840-9846Abstract Full Text Full Text PDF PubMed Scopus (98) Google Scholar, 14Heyraud-Nitschke F. Schumacher S. Laufs J. Schaefer S. Schell J. Gronenborn B. Nucleic Acids Res. 1995; 23: 910-916Crossref PubMed Scopus (143) Google Scholar). We hypothesized that a truncated Rep protein that was competent for DNA binding and oligomerization domain might have a greater probability to interfere with the virus replication and might be effective against both homologous and heterologous viruses. In this study, we mapped the minimal binding domain on the Rep protein by electrophoretic mobility shift assays (EMSAs). We also tested the effect of truncated and full-length AC1 sequences on DNA replication of ToLCNDV and other geminiviruses in transient assays using BY2 protoplasts and Nicotiana benthamiana plants. These studies revealed that transient expression of the ToLCNDV-truncated Rep protein encoding the DNA binding and the oligomerization domains could significantly inhibit replication of ToLCNDV viral DNA and to some extent the replication of other geminiviruses having similar iteron sequences. The full-length AC1 genes from the severe and the mild strains of ToLCNDV were amplified by PCR from pMPA1 (DNA-A of the severe strain ToLCNDV) and pMPA2 (DNA-A of the mild strain ToLCNDV) (15Padidam M. Beachy R.N. Fauquet C.M. J. Gen. Virol. 1995; 76: 25-35Crossref PubMed Scopus (216) Google Scholar), cloned in the bacterial expression vector pGEX-4T-3 (Amersham Pharmacia Biotech), and overexpressed inEscherichia coli cells. The recombinant proteins were named according to the number of amino acids at the N or C terminus of the Rep protein. The C-terminal truncations were made by inserting an in-frame stop codon at positions 2436 (pAC1-(1–52)), 2250 (pAC1-(1–114)), and 2110 (pAC1-(1–160)). The truncated AC1sequences were subcloned as a BamHI–XhoI fragment in the pGEX-4T-3 vector, generating pAC1-(1–52), pAC1-(1–114), and pAC1-(1–160), respectively. At the N terminus, the first 21 amino acids of the protein were deleted, and anNheI site was inserted to create an in-frame start codon. The truncated fragment was cloned as a NheI–XhoI fragment in the vector pGEX-4T-3 to produce pAC1-(22–360).The plasmids pAC1-(52–360) and pAC1-(114–360) were produced similarly but had a deletion of the first 51 and 113 amino acids, respectively, from the N terminus of the AC1 gene. The truncated Rep proteins were expressed from plasmids mentioned above in E. colicells. The glutathione S-transferase (GST)-tagged AC1 fusion proteins were purified by glutathione affinity chromatography on glutathione-Sepharose beads according to the manufacturer's recommendations. Briefly, the cells were grown to a density of 0.75–0.8A 600. The cultures were induced by the addition of isopropyl-β-D thiogalactoside at a final concentration of 1 mm and grown further for 2 h. The cells were finally harvested at 4000 rpm (Beckman, JS 10.5 rotor) for 10 min. The pellets were suspended in ice cold 1× PBS (10 mmKH2PO4, 100 mm NaCl) and lysed by sonication. The lysate was clarified at 17,000 × g for 30 min. The resulting supernatant was loaded on a glutathione-Sepharose 4B column (Amersham Pharmacia Biotech) previously equilibrated with 1× PBS. After repeated washing of the column with 1× PBS, the protein was eluted with glutathione elution buffer (Amersham Pharmacia Biotech). The eluted fractions were dialyzed against 1× PBS to remove glutathione and concentrated using centricon filters (Amicon, Centricon). Protein concentrations were estimated using Bradford's reagent (Bio-Rad). Protein extracts from E. coli cells co-expressing the untagged, wild type AC1 and GST fusion of truncated Rep proteins were tested for AC1 oligomerization by co-purification on glutathione-Sepharose. Co-purification of proteins was monitored by resolving the eluted fractions on SDS-PAGE and by immunoblotting. The full-length and truncated Rep proteins were detected using the polyclonal anti-AC1 antibody. A similar procedure was used to assess the oligomerization of full-length Rep proteins from other geminiviruses with the truncated Rep protein of ToLCNDV (pAC1-(1–160)). Protein extracts from E. coli cells co-expressing wild type AC1 of Pepper huasteco yellow vein virus(PHYVV), Potato yellow mosaic virus(PYMV), and ACMV and the GST-tagged pAC1-(1–160) were incubated with GST-Sepharose beads, washed thoroughly with 1× PBS, and eluted with glutathione elution buffer (Amersham Pharmacia Biotech). The eluate was resolved on SDS-PAGE gels and then transferred to nitrocellulose membranes and detected by immunoblotting using polyclonal anti-AC1 antibody and anti-GST antibody. For expression of the truncated AC1 gene in plant cells, the mutants described above were subcloned as BamHI fragments in the plant expression vector pILTAB 350. This placed the DNA fragment 3′ of the cassava vein mosaic virus promoter (16Verdaguer B. de Kochko A. Fux C. Beachy R.N. Fauquet C.M. Plant Mol. Biol. 1998; 37: 1055-1067Crossref PubMed Scopus (74) Google Scholar) upstream of the AC1gene sequences to produce the gene expression cassettes, pILTAB 401 (encoding AC1-(1–52)), pILTAB 402 (encoding AC1-(1–114)), and pILTAB 403 (encoding AC1-(1–160)), respectively. Constructions of infectious clones of plasmids containing full-length DNA of ToLCNDV are named pMPA1 and pMPB1 and were previously described (15Padidam M. Beachy R.N. Fauquet C.M. J. Gen. Virol. 1995; 76: 25-35Crossref PubMed Scopus (216) Google Scholar). John Stanley (John Innes Institute, Norwich, United Kingdom) generously provided full-length infectious dimers of ACMV-Kenya, pCLV 1.3A, and pCLV 2B (17Stanley J. Nature. 1983; 305: 643-645Crossref Scopus (132) Google Scholar). Infectious monomers of PHYVV (18Bonilla-Ramirez G.M. Guevara-Gonzalez R.G. Garzon-Tiznado J.A. Ascencio-Ibanez J.T. Torres-Pacheo I. Rivera-Bustamante R.F. J. Gen. Virol. 1997; 78: 947-951Crossref PubMed Scopus (19) Google Scholar) were kindly provided by Riviera Bustamante (CINVESTAV, Irapuato, Mexico). The PYMV clones have been described (19Umaharan P. Padidam M. Phelps R.H. Beachy R.N. Fauquet C.M. Phytopathology. 1998; 88: 1262-1268Crossref PubMed Scopus (66) Google Scholar). The sequences of the synthetic oligonucleotides used as probes or competitors in EMSAs are given in Table III. In the case of the severe and mild strains of ToLCNDV, the 18-mer oligonucleotides corresponding to the binding sites of the Rep protein were used as probe (11Chatterji A. Chatterji U. Beachy R.N. Fauquet C.M. Virology. 2000; 273: 314-350Crossref Scopus (64) Google Scholar). For the geminiviruses, ACMV, PHYVV, and PYMV, fragments of their CR sequences were synthesized and used as competitors in EMSAs. All oligonucleotides were synthesized commercially by Life Technologies, Inc.Table IIIThe sequence of synthetic oligonucleotides used in EMSAsVirusSequence in the CR of the viral genome3-aThe numbers within parenthesis indicate the coordinates of the CR sequences used as competitor in EMSAs for the respective virus.GenBank™ accession no.ToLCNDV (s)ATTGGTGTCTGGAGTCCC (2632–2653)U15015, U15017ToLCNDV (m)ATTGGCGTCTGGCGTCCC (2632–2653)U15016PHYVVATCGGTGTATTGGTAGCCAAT (2554–2574)PHV-mex, X70418, 70419PYMV/TTATCGGTGTATTGGGGTACTAT (2508–2528)PYMV/TT, AF039031ACMVATTGGAGA-(40 bp)-GGAGACAT (2625–2682)ACMV-K, j02057, j020583-a The numbers within parenthesis indicate the coordinates of the CR sequences used as competitor in EMSAs for the respective virus. Open table in a new tab The single-stranded 18-mer oligonucleotides containing the potential binding sites of the Rep protein of ToLCNDV were annealed to their complementary strands. The oligonucleotides were end-labeled with [γ-32P]ATP using T4 polynucleotide kinase and purified on polyacrylamide gels. The final concentration of the probes was 500 pm (30,000 cpm). The concentration of competitor DNA used was 50 nm per reaction. Both the probe and the competitor DNAs were purified on Sephadex G-25 columns, quantified by scintillation counting, and diluted to 30,000 cpm for each binding reaction. The binding assays were performed using purified Rep protein. Typically, the binding reactions contained 500 ng of pure protein, 1 ng of labeled DNA, and 0.2 μg of poly(dI-dC). Binding buffer contained 20 mm HEPES, pH 7.5, 60 mm KCl, 1 mm dithiothreitol, and 15% glycerol. Reactions were incubated at 25 °C for 30 min, and the complexes were resolved on 4% polyacrylamide gels in 0.25× TBE buffer. The gels were dried on Whatman paper and exposed to x-ray film. Comparative efficiency of binding was analyzed by quantifying the amount of radioactivity in the retarded bands using a PhosphorImager (Molecular Dynamics). Protoplasts derived from Nicotiana tabacum BY-2 suspension cultures were used for transfection with viral DNA (20Watanabe Y. Meshi T. Okada Y. FEBS Lett. 1987; 219: 65-69Crossref Scopus (98) Google Scholar). Protoplasts were collected from cultures 48 h postinoculation for DNA isolation and analysis. One million protoplasts were inoculated by electroporation (250 V, 500 microfarads) with 2 μg each of A and B component DNAs and 40 μg of sheared herring sperm DNA (10Chatterji A. Padidam M. Beachy R.N. Fauquet C.M. J. Virol. 1999; 73: 5481-5489Crossref PubMed Google Scholar). For co-inoculation experiments, 2 μg of the plasmid DNA containing the expression cassettes with truncated AC1 gene sequences were used. Total DNA from the protoplasts was extracted 48 h after transfection (21Dellaporta S.L. Wood J. Hicks J.B. Plant Mol. Biol. 1983; 1: 19-21Crossref Scopus (6371) Google Scholar, 22Mettler I.J. Plant Mol. Biol. Rep. 1987; 5: 346-349Crossref Scopus (148) Google Scholar). Viral DNA accumulation was analyzed by Southern blotting (10Chatterji A. Padidam M. Beachy R.N. Fauquet C.M. J. Virol. 1999; 73: 5481-5489Crossref PubMed Google Scholar). Total proteins were extracted from the protoplasts 48 h after transfection by sonication of the cell pellets in ice cold 1× PBS (10 mm KH2PO4, 100 mmNaCl). The lysate was clarified at 17,000 × g for 15 min, and the resulting supernatant was used for immunoprecipitations. Immunoprecipitations were done by incubating 50 μg of total protein extracts with polyclonal anti-AC1 antiserum (1 mg) overnight at 4 °C. Protein-antibody complexes were incubated with protein A-agarose for 2 h at 4 °C and then washed with 1× PBS. Bound proteins were eluted from the agarose beads in SDS-PAGE sample buffer by boiling at 100 °C for 5 min. Proteins resolved on the gel were transferred on nitrocellulose membranes and analyzed by immunoblotting with polyclonal anti AC1 antibody using 3,3′-diaminobenzidine tetrahydrochloride for colorimetric quantitation of the expressed Rep levels in the cell extracts. Two-week-old seedlings of N. benthamiana were grown in magenta boxes and inoculated with partial tandem dimers of viral DNA using a Bio-Rad helium-driven particle gun (10Chatterji A. Padidam M. Beachy R.N. Fauquet C.M. J. Virol. 1999; 73: 5481-5489Crossref PubMed Google Scholar). Ten plants were inoculated with each mutant using 0.5 μg each of DNA-A and DNA-B genomic components per plant. Plants were observed for symptom development, and newly emerging leaves were harvested for Southern blot analysis 4 weeks postinoculation. DNA extractions from systemically infected leaf samples were completed as described by Dellaportaet al. (21Dellaporta S.L. Wood J. Hicks J.B. Plant Mol. Biol. 1983; 1: 19-21Crossref Scopus (6371) Google Scholar) and from protoplasts by following the procedure of Mettler (22Mettler I.J. Plant Mol. Biol. Rep. 1987; 5: 346-349Crossref Scopus (148) Google Scholar). Total DNA (4 μg) was fractionated on 1% agarose gels without ethidium bromide and transferred to nylon membranes. Viral DNA was detected by using a 900-base pairAflII–PstI fragment of the A component containing sequences from the open reading frames of theAC1, AC2, and AC3 genes or a probe specific for the B component (878-base pair PCR-amplifiedBC1 gene). The amount of viral DNA was quantified as previously described (10Chatterji A. Padidam M. Beachy R.N. Fauquet C.M. J. Virol. 1999; 73: 5481-5489Crossref PubMed Google Scholar). In the case of geminiviruses other than ToLCNDV, fragments of their AC1 and BC1 genes were amplified and used as probes to analyze the replication levels of viral DNA. The Rep protein binds specifically to a directly repeated DNA sequence motif in the common region of the ToLCNDV genome (11Chatterji A. Chatterji U. Beachy R.N. Fauquet C.M. Virology. 2000; 273: 314-350Crossref Scopus (64) Google Scholar). Purified Rep proteins truncated at amino acids 160, 114, and 52 were used to map the C-terminal boundary of the Rep DNA binding domainin vitro. As a control, full-length Rep protein (amino acids 1–360) was used in all assays. The truncated and full-length Rep proteins were expressed in E. coli with a GST tag and affinity-purified on a glutathione-Sepharose 4B column. The affinity-purified proteins were highly enriched as determined by Coomassie staining following electrophoresis on SDS-PAGE gels. The proteins were detected in immunoblots using anti-GST antibody (data not shown). The purified Rep proteins were tested for their ability to bind a radiolabeled 18-mer (nucleotides 2632–2653) that contains the Rep binding site sequence, 5′-GGTGTCTGGAGTC-3′. DNA-protein complexes that contained Rep-(1–360) and Rep-(1–160) were detected. No binding was observed for Rep-(1–52) or Rep-(1–114) (Fig.1A, lanes 1–4). These results located the C-terminal boundary of the DNA binding domain of the Rep protein between amino acids 115 and 160. The N-terminal boundary of the DNA binding domain was determinedin vitro by comparing the binding of full-length Rep-(1–360) and Rep-(22–360), Rep-(52–360), and Rep-(114–360) to the 18-base pair iteron sequence, 5′-GGTGTCTGGAGTC-3′ in EMSAs. The DNA-protein complexes were observed in case of full-length Rep protein, while no DNA-protein complexes were detected for the Rep-(22–360), Rep-(52–360), or Rep-(114–360) (Fig. 1B, lanes 2–4). These results demonstrated that the sequences within the first 21 amino acids of the Rep protein are essential for protein-DNA interactions. Together, these results placed the DNA binding domain of ToLCNDV Rep protein between amino acids 1 and 160. To determine if truncations at the N and the C termini of Rep protein affect its ability to oligomerize, GST-tagged truncated Rep proteins were co-expressed with untagged wild type full-length Rep protein in bacterial cells and co-purified on glutathione-Sepharose beads. The bound fractions were eluted and analyzed in immunoblots using polyclonal anti-AC1 antiserum. The wild type Rep-(1–360) co-purified with GST-tagged truncated proteins Rep-(22–360), Rep-(52–360), and Rep-(114–360) (Fig. 1C, lanes 1–5), suggesting that truncations made at the N terminus in the Rep did not affect the ability of the Rep protein to oligomerize with itself, although each of the truncated proteins was deficient for DNA binding. The effect of Rep protein on viral DNA replication was investigated by co-inoculating N. tabacum BY2 protoplasts with DNA-A and various cassettes that express truncated AC1gene sequences from the CsVMV promoter. ToLCNDV DNA-A replicated in BY-2 cells and accumulated high levels of single-stranded (ss) and supercoiled (sc) DNA (Fig. 2A,lane 1). In contrast, there was a significant decrease in the level of viral DNA replication (78% drop) in the presence of Rep-(1–160) (Fig. 2A, lane 4, Table I). Reduction in replication was estimated by quantifying the amount of radioactivity using a PhosphorImager (Storm 860; Molecular Dynamics). The reduction in virus replication was not as dramatic in the presence of Rep-(1–52) (Fig. 2A, lane 2) or Rep-(1–114) (Fig. 2A, lane 3) when compared with Rep-(1–160) (Fig. 2A, lane 4). EMSAs showed that the Rep-(1–52) and Rep-(1–114) do not bind DNA (Fig.1A, lanes 2 and 3), implying that an intact DNA binding domain and/or a protein oligomerization domain is essential for inhibition of replication.Table IVirus replication in BY2 protoplasts and N. benthamiana plants co-inoculated with truncated Rep protein gene constructs and the viral DNA of the severe strain of ToLCNDVVirus constructSymptom expression1-aA total of 10 plants were inoculated per experiment, and each experiment was repeated three times. Shown are the number of plants infected/number of plants inoculated. Plants were scored for symptom expression 3 weeks postinoculation.Replication1-bThe numbers refer to the amount (in percentage) of viral DNA replication in protoplasts electroporated with similar constructs. The viral DNA was quantified using a PhosphorImager (Molecular Dynamics).ProtoplastsPlantsDNA-ADNA-B%%%A1+BSevere, 10/10100100100A2+BMild, 10/105548–5010–15A1+B+Rep-(1–52)/A1Severe, 10/1010092–9892–110A1+B+Rep-(1–114)/A1Severe, 10/1090–9289–9290–98A1+B+Rep-(1–160)/A1Very mild,1-cAbout 55% of plants were asymptomatic, 30% showed mild chlorosis, and only 15% of plants expressed mild symptoms of leaf curl (Table I). None of the plants showed severe infection or stunted growth found in wild type infection. Most of the plants inoculated with AC1–1-(1–52) and Ac1–2-(1–114) developed severe symptoms 7 days post inoculation (Table II). 10/1022–2814–305–14A1+B+Rep-(1–52)/A2Severe, 10/1010094–96100A1+B+Rep-(1–114)/A2Severe, 10/1010089–9398–100A1+B+Rep-(1–160)/A2Severe, 10/1078–8070–7494–98A2+B+Rep-(1–160)/A2No symptoms, 10/10505612–141-a A total of 10 plants were inoculated per experiment, and each experiment was repeated three times. Shown are the number of plants infected/number of plants inoculated. Plants were scored for symptom expression 3 weeks postinoculation.1-b The numbers refer to the amount (in percentage) of viral DNA replication in protoplasts electroporated with similar constructs. The viral DNA was quantified using a PhosphorImager (Molecular Dynamics).1-c About 55% of plants were asymptomatic, 30% showed mild chlorosis, and only 15% of plants expressed mild symptoms of leaf curl (Table I). None of the plants showed severe infection or stunted growth found in wild type infection. Most of the plants inoculated with AC1–1-(1–52) and Ac1–2-(1–114) developed severe symptoms 7 days post inoculation (Table II). Open table in a new tab Similar experiments were conducted with the mild strain of ToLCNDV in which analogous truncated mutations of the Rep gene were co-introduced in tobacco protoplasts with DNA-A from the mild strain. In these studies, an analogous inhibition of viral DNA accumulation in BY2 protoplasts was detected (Fig. 2A, lanes 5–8, and Table II).Table IIVirus replication in BY2 protoplasts and N. benthamiana plants co-inoculated with truncated Rep protein constructs and viral DNA of the mild strain of ToLCNDVVirus constructSymptom expression2-aA total of 10 plants were inoculated per experiment, and each experiment was repeated three times. Shown are the number of plants infected/number of plants inoculated. Plants were scored for symptom expression 3 weeks postinoculation.Replication2-bThe numbers refer to the amount (in percentage) of viral DNA replication in protoplasts electroporated with similar constructs. The viral DNA was quantified using a PhosphorImager (Molecular Dynamics).ProtoplastsPlantsDNA-ADNA-BA1+BSevere, 10/10100100100A2+BMild, 10/10555210–12A2+B+Rep-(1–52)/A2Mild, 10/10555610–13A2+B+Rep-(1–114)/A2Mild, 10/104542–458–10A2+B+Rep-(1–160)/A2No symptoms, 10/10122010–14A2+B+Rep-(1–52)/A1Mild, 10/10506010–12A2+B+Rep-(1–114)/A1Mild, 10/10525810–13A2+B+Rep-(1–160)/A1Mild, 10/1028–3036–4010–122-a A total of 10 plants were inoculated per experiment, and each experiment was repeated three times. Shown are the number of plants infected/number of plants inoculated. Plants were scored for symptom expression 3 weeks postinoculation.2-b The numbers refer to the amount (in percentage) of viral DNA replication in protoplasts electroporated with similar constructs. The viral DNA was quantified using a PhosphorImager (Molecular Dynamics). Open table in a new tab To determine the relative expression levels of the three truncated Rep proteins in transfected tobacco protoplasts, total proteins were extracted from the tobacco protoplasts 48 h after infection and immunoprecipitated with the anti-AC1 antibody, and the protein-antibody complexes were resolved on SDS-PAGE gels. All of the three truncated proteins could be detected in immunoblots from the transfected protoplasts when developed using 3,3′-diaminobenzidine tetrahydrochloride. 3,3′-Diaminobenzidine tetrahydrochloride produces a brown precipitate with the peroxidase and thereby provided a direct measure of the amount of antibody bound to the expressed protein in the samples, revealing that all three truncated Rep proteins were expressed stably and in equivalent amounts in the protoplasts (Fig.2B, lanes 2–6). Two-week-old seedlings ofN. benthamiana plants were co-bombarded with 2 μg each of infectious dimers of ToLCNDV DNA-A and DNA-B in the presence or absence of genes encoding Rep-(1–160). The plants were observed daily for symptom development. All of the plants inoculated only with the wild type virus DNAs developed severe symptoms 5 days after inoculation. In contrast, plants co-inoculated with the virus and the genes encoding Rep-(1–160) developed milder symptoms of ToLCNDV infection (Table I). About 55% of the plants were asymptomatic, 30% showed mild chlorosis, and 15% expressed mild leaf curl symptoms (Table I). None of the plants showed severe infection or stunted growth as in plants infected only with ToLCNDV. Most of the plants co-inoculated with Rep-(1–52) and Rep-(1–114) developed severe symptoms by 7 days postinoculation (Tables I and II). The levels of viral DNA in ToLCNDV-infected plants were analyzed by Southern blot analysis of young leaves sampled 28 days postinoculation using probes that detected DNA-A and DNA-B (see “Materials and Methods” and Fig. 3, A andB, respectively). The amount of viral DNA ranged from undetectable to very low (an average of 15% of the wild type levels) in asymptomatic plants, and the accumulation of both genomic components increased with increasing severity of symptom expression. Plants co-inoculated with expression cassettes Rep-(1–52) and Rep-(1–114) developed intermediate to severe symptoms in most of the plants and accumulated viral DNA between 85 and 92% of wild type infection. To investigate the potential of truncated Rep protein to inhibit the replication of other geminiviruses, we selected examples of viruses that belonged to the Old World (ACMV) and New World geminiviruses (PHYVV and PYMV-TT). We reasoned that for the Rep to be able to interfere in replication of heterologous geminiviruses, it must (a) bind to the origin sequences of these viruses and (b) oligomerize with their Rep proteins. For the EMSA studies, fragments of the intergenic region sequences of the selected heterologous geminiviruses close to the TATA box were chosen. The coordinates of these sequences are given in TableIII. To determine if the putative iteron sequences of the other geminiviruses could compete with the cognate iteron sequences of ToLCNDV for binding to ToLCNDV Rep protein, synthetic oligonucleotides encoding the CR sequences from each virus were synthesized and used as competitors in EMSAs. None of the CR sequences were effective competitors in EMSA with the ToLCNDV Rep protein (Fig. 4A,lanes 3–6) and did not affect binding of the Rep protein with its cognate 13-mer iteron sequences to a significant degree. The crude lysates of E. coli cells co-expressing wild type Rep proteins from ACMV, PHYVV, or PYMV and the GST-tagged ToLCNDV Rep-(1–160) were tested for the ability to bind to each other. Crude protein extracts from bacterial cells co-expressing the target proteins were loaded on a GST-Sepharose column, washed extensively with 1× PBS, and eluted with glutathione elution buffer. The resulting fractions were detected in immunoblots using polyclonal anti-AC1 and anti-GST antibodies. In immunoblots, the anti-AC1 antibody detected wild type untagged Rep proteins from ACMV, PHYVV, and PYMV-TT that co-purified with the ToLCNDV Rep-(1–160) (Fig. 4B, lanes 2–6). When the same blot was washed and reprobed with anti-GST antibody, only the truncated Rep protein of ToLCNDV (Rep-(1–160)) in each of the samples was detected (Fig. 4C,lanes 2–6). To determine if Rep-(1–160) could reduce accumulation of other geminiviruses, in vivo replication assays were conducted by co-bombarding the N. benthamiana plants with partial/tandem dimers of full-length A and B components of ACMV, PHYVV, and PYMV-TT with genes encoding Rep-(1–160) of ToLCNDV. Most plants developed typical symptoms of virus infection within 21–27 days postinoculation as opposed to 7–10 days required for the symptoms on control plants to develop infection. Southern blot analysis of viral DNA extracted from the plants harvested 28 days postinoculation showed a minor reduction in the levels of virus accumulation as compared with the control plants (TableIV and Fig.5, A and B). However, the decrease in replication was not as significant as the inhibition observed in the case of ToLCNDV DNA.Table IVRegulation of virus DNA replication in BY2 protoplasts by the N-terminal sequences of AC1 gene of ToLCNDVVirusPutative iteron4-aThe sequences indicate putative iteron sequences.N-Rep sequenceEMSAReplication4-bOnly Rep-(1–160) was tested in competition experiments. The numbers indicate relative inhibition of virus replication levels as compared with a wild type ToLCNDV Rep by Rep-(1–160) as determined by Southern blotting and PhosphorImager analysis.AB%%%ACMVGGAGAMRTPPRFRIQANKYFLTYPKC48–6225PHVGGTGAMPLPKRFLNAKNYFLTYPQC26–5219PYMVGGTGTMP-PKRFRINANKYFLTYPKC43–2227ToLCNDV (s)GGTGTMAPPRFRVNANKYFLTYPKC10070–8686–95ToLCNDV (m)GGCGTMASPRFRIDANKYFLTYPKC36–4012–144-a The sequences indicate putative iteron sequences.4-b Only Rep-(1–160) was tested in competition experiments. The numbers indicate relative inhibition of virus replication levels as compared with a wild type ToLCNDV Rep by Rep-(1–160) as determined by Southern blotting and PhosphorImager analysis. Open table in a new tab We determined the nature and the significance of the DNA binding and protein oligomerization functions of a truncated Rep-(1–160) protein to interfere in DNA accumulation of homologous and heterologous geminiviruses. Our studies show that while both activities of Rep-(1–160) contribute to interference in DNA accumulation, protein oligomerization, unlike DNA binding, is nonspecific and can occur between the Rep proteins of two unrelated geminiviruses. We mapped the DNA binding domain on the Rep protein of ToLCNDV to amino acids 1–160 and showed that the transient expression of this Rep sequence significantly inhibits ToLCNDV DNA accumulation in inoculated tobacco protoplasts and plants. Of the three C-terminal truncations made in the AC1 gene, only Rep-(1–160) bound the iteron DNA sequences in vitro, and the two truncations Rep-(1–52) and Rep-(1–114) were not competent to bind viral DNAin vitro. None of the N-terminal truncations tested (i.e. Rep-(22–360), Rep-(52–360), or Rep-(114–360)) bound viral DNA, indicating that an intact N terminus is required for the Rep protein to bind the origin sequences. In co-purification assays, each of the three N-terminal truncated proteins bound with the wild type Rep protein as detected by immunoblotting of the bound fractions. The co-immunoprecipitation assays indirectly suggested that the oligomerization domain might overlap the DNA binding domain. In TGMV, it is known that the oligomerization domain overlaps the DNA binding domain (13Orozco B.M. Miller A.B. Stellage S.B. Hanley-Bowdoin L. J. Biol. Chem. 1997; 272: 9840-9846Abstract Full Text Full Text PDF PubMed Scopus (98) Google Scholar). In co-infection studies, the sequences comprising the Rep-(1–52) or Rep-(1–114) amino acids of the Rep protein did not cause a significant reduction in ToLCNDV accumulation. However, Rep-(1–160) bound with high affinity to the iteron sequences and reduced viral replication in protoplasts and plants. Colorimetric quantification of the truncated proteins revealed that all three proteins were expressed in equivalent amounts, ruling out the possibility that poor expression and/or instability of Rep-(1–52) and Rep-(1–114) in tobacco protoplasts may have compromised their ability to inhibit virus replication. In related studies, we observed that N. benthamiana plants co-bombarded with plasmids that produced Rep-(1–160) and infectious ToLCNDV produced a range of symptoms from asymptomatic to a mild leaf curl. None of the plants developed the severe puckering and blistering associated with wild type virus infection. Southern blot analysis of the infected plants showed that accumulation of viral DNA in plants with mild or no symptoms was much less than in plants showing severe symptoms. More importantly, the degree of inhibition in plants was similar to those observed in BY-2 protoplasts, indicating that the impact is probably on virus replication. Rep-(1–160) contains the DNA binding domain of the ToLCNDV-Rep. By analogy with the Rep proteins of TGMV and Tomato yellow leaf curl virus, this fragment is expected to contain the domains for DNA cleavage and ligation, as well as protein oligomerization domain (13Orozco B.M. Miller A.B. Stellage S.B. Hanley-Bowdoin L. J. Biol. Chem. 1997; 272: 9840-9846Abstract Full Text Full Text PDF PubMed Scopus (98) Google Scholar,23Laufs J. Traut W. Heyraud F. Matzeit V. Rogers S.G. Schell J. Gronenborn B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 3879-3883Crossref PubMed Scopus (259) Google Scholar). This region of the Rep protein of ToLCNDV is involved in the specificity of origin recognition and binding (10Chatterji A. Padidam M. Beachy R.N. Fauquet C.M. J. Virol. 1999; 73: 5481-5489Crossref PubMed Google Scholar). Considering the various activities that are associated with Rep-(1–160), it is possible to suggest the mode of action of Rep-(1–160) in limiting virus DNA accumulation. One possibility is that the Rep-(1–160) protein reduces replication by competing with the viral Rep protein for binding the iteron sequences in the origin. The truncated Rep protein may therefore behave as a dominant negative mutant (24Herskowitz I. Nature. 1987; 329: 219-222Crossref PubMed Scopus (869) Google Scholar) and block virus replication. Another possibility is that the truncated Rep protein does not contain the NTP binding domain present on the C terminus of the Rep protein. The NTP binding domain is required for replication (25Desbeiz C. David C. Mettouchi A. Laufs J. Gronenborn B. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5640-5644Crossref PubMed Scopus (106) Google Scholar), and the lack of this region may interfere with the normal replication process of the virus. The fact that the Rep protein represses its own transcription may be yet another explanation for the inhibition of virus replication. Presumably, binding by the Rep protein to the origin is responsible for the repression of AC1 gene transcription in TGMV (26Sunter G. Hartitz M.D. Bisaro D.M. Virology. 1993; 195: 275-280Crossref PubMed Scopus (100) Google Scholar, 27Eagle P.A. Orozco B.M. Hanley-Bowdoin L. Plant Cell. 1994; 6: 1157-1170PubMed Google Scholar) and ACMV (7Hong Y. Stanley J. J. Gen. Virol. 1995; 76: 2415-2422Crossref PubMed Scopus (64) Google Scholar). Constitutive expression of the truncated viral Rep protein could repress the transcription of the wild type AC1gene by binding to the origin, thereby influencing viral accumulation levels. Alternately, as suggested in the case of wheat dwarf virus, constitutively expressed Rep may adversely affect the integrity of the viral DNA by introducing nicks at cryptic motifs (28Heyraud F. Matzeit V. Kammann M. Schaefer S. Schell J. Gronenborn B. EMBO J. 1993; 12: 4445-4452Crossref PubMed Scopus (90) Google Scholar). Comparison of co-infection experiments performed using truncated Rep proteins of the mild and the severe strains of ToLCNDV in protoplasts and plants suggested that only homologous Rep sequences could reduce virus accumulation with high efficiency. The Rep-(1–160) from the severe strain did not significantly restrict the virus replication of the mild strain and vice versa. Our previous work showed that the mild and the severe strains of ToLCNDV exhibit selectivity in binding to their cognate iteron sequences (11Chatterji A. Chatterji U. Beachy R.N. Fauquet C.M. Virology. 2000; 273: 314-350Crossref Scopus (64) Google Scholar). Hence, the inability of Rep-(1–160) of one strain to limit virus DNA accumulation of the heterologous strain may reflect specificity of interaction by the Rep protein for its cognate origin DNA sequences. Since the Rep-(1–160) can bind to DNA and form oligomers, these results support the hypothesis that DNA binding and protein oligomerization are important in inhibition of virus replication by Rep-(1–160). Notwithstanding that Rep-(1–160) of the severe strain of ToLCNDV had a modest effect on the replication of the mild strain, we were interested to know if Rep-(1–160) can interfere with the replication of related geminiviruses that have similar sequences in their origins of replication. The reduction in virus accumulation in the case of unrelated geminiviruses was rather surprising, considering that none of the CR sequences from these viruses were effective competitors for Rep-(1–160) in EMSAs. However, the results were not unexpected because the co-purification experiments revealed the ability of Rep-(1–160) to interact with the Rep proteins of heterologous geminiviruses. These data suggest that since the Rep-(1–160) protein does not bind to the heterologous CR sequences, reduction in virus accumulation may result from oligomerization of Rep-(1–160) with the Rep proteins of the unrelated geminiviruses. This level of virus reduction was similar to the reduction of mild strain ToLCNDV accumulation by the Rep-(1–160). We suggest that oligomers of Rep-(1–160) could potentially interfere with the replication complexes formed during infection by PHYVV, PYMV, ACMV, or the mild strain of ToLCNDV. Formation of heteromultimers via protein-protein interactions (29Settlage S.B. Miller B. Hanley-Bowdoin L. J. Virol. 1996; 70: 6790-6795Crossref PubMed Google Scholar) has been reported between TGMV and BGMV. The formation of heteromultimeric complexes might sequester the wild type Rep oligomers that otherwise would participate in the formation of a replication complex or might prevent recognition of origin sequences. Several approaches to control replication of geminiviruses have been developed. Transgenic N. benthamiana plants that accumulate defective interfering DNA (30Stanley J. Frischmuth T. Ellwood S. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 6291-6295Crossref PubMed Scopus (126) Google Scholar, 31Frischmuth T. Stanley J. Virology. 1991; 183: 539-544Crossref PubMed Scopus (53) Google Scholar) of ACMV were less susceptible to ACMV infection, but resistance was confined to closely related strains of ACMV. Transgenic N. tabacum expressing antisense RNA targeted against TGMV AL1 (32Day A.G. Bejarano E.R. Buck K.W. Burrell M. Lichtenstein C.P. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 6721-6725Crossref PubMed Scopus (131) Google Scholar) or Tomato yellow leaf curl virus(33Bendahmane M. Gronenborn B. Plant Mol. Biol. 1997; 33: 351-357Crossref PubMed Scopus (94) Google Scholar) showed that specificity of resistance depended on the level of homology between the antisense RNA and the target virus sequence. Finally, the possibility of expressing a full-length AC1transgene in ACMV (34Hong Y. Stanley J. Mol. Plant Microbe Interact. 1996; 9: 219-225Crossref Scopus (76) Google Scholar) and the N-terminal sequences of Tomato yellow leaf curl virusRep (12Noris E. Accotto G.P. Tavazza R. Brunetti A. Crespi S. Tavazza M. Virology. 1996; 224: 130-138Crossref PubMed Scopus (113) Google Scholar) in virus resistance has also been documented. Our studies demonstrate the potential of using Rep proteins that are mutated in the oligomerization and DNA binding domain to interfere with viral DNA replication. Experiments are in progress to test the stable expression and efficiency of Rep-(1–160) in transgenic tobacco and tomato plants for resistance to ToLCNDV.