TY - JOUR
AU - Goldsbrough, Andrew P.
AB - Abstract We have used particle bombardment (biolistics) to deliver replication-competent wheat dwarf virus (WDV)-based constructs, carrying reporter gene sequences fused to the virion sense promoter (Pv) or the CaMV 35S promoter, to suspension culture cells and immature zygotic embryos of wheat. While the replication of WDV double-stranded DNA forms (replicons) was equivalent between wheat suspension culture cells and embryos, GUS reporter gene activity was 20–40 times higher in the embryo cultures. Maximum expression of WDV replicons occurred in the embryonic axis tissue of wheat embryos but their expression in suspension cells was compromised, compared with transiently maintained input plasmid DNA containing the same sequences. From these studies, we propose that WDV replicons are subject to a host cell-controlled competency for virion sense transcription. The term competency is used to distinguish between the phenomenon described here and control of gene expression by specific transcription factors. Control of competency is independent of Pv, the replacement 35S promoter and of the complementary sense control of virion sense expression involving specific sequences in Pv. We propose that factors controlling the competency for replicon expression may be present in cells which, as well as maintaining high rates of DNA synthesis, are totipotent. Cell type control of active chromatin, methylation of specific sequences in WDV minichromosomes and/or interaction of virus-encoded proteins with specific host factors are considered as possible mechanisms. Introduction Wheat dwarf virus (WDV) is a plant DNA virus, classified in the Mastrevirus genus, of the Geminiviridae family (1,2), and has been used previously to study the relationship between DNA replication and transcription (3–5). WDV has a genome size of 2750 bases, which includes four open reading frames (ORFs) potentially encoding for polypeptides of >10 kDa; two on the virion sense (+ve) strand (V1 and V2) and two on the complementary sense (−ve) strand (C1 and C2; 6). Mutational studies of WDV and other Poaceae-infecting geminiviruses have assigned functions to some of these ORF products. The V1 protein may be involved in symptom development and viral spread (7–11), while V2 encodes the coat protein (6,12), required for formation of virus particles and plant infection. Neither V1 nor V2 is required for double-stranded (ds)DNA replication in cereal protoplasts and therefore these two ORFs can be replaced with foreign sequences, typically reporter gene sequences, which facilitate the study of virion sense expression in these viruses (3,4,10,13). The C1 and C2 ORFs encode a putative C1:C2 protein (Rep) which is synthesised from a spliced RNA, resulting in a fusion of these two ORFs (4,14–16). The C1:C2 protein is essential for viral dsDNA replication (4,14) and activation of virion sense gene expression, probably via positive regulation of the virion sense promoter (Pv; 4). Pv is located along with the origin of DNA replication and the promoter of the complementary sense transcription unit (Pc) in the large intergenic region (LIR) of the WDV genome (4,17). It has been recognised that active host DNA synthesis, either in cells that are dividing or undergoing endoreduplication, is a prerequisite for geminiviral DNA replication (10,11,18,19). Digitaria streak virus dsDNA synthesis was optimal in nuclei of S phase cells (20) and tomato golden mosaic virus infection in terminally differentiated cells resulted in accumulation of cellular replication proteins that are not normally switched on in non-dividing cells (21). There is also evidence that WDV is able to produce an appropriate environment in the nucleus to favour both viral replication and transcription, using a similar strategy to simian virus 40 (SV40), adenoviruses and papillomaviruses. RepA has been shown to bind to retinoblastoma (Rb)-like proteins (5,22) and this in turn may induce transcription of genes required for DNA replication and cell cycle progression. As an extension of our previous work (4), we set out to compare levels of virion sense gene expression and DNA replication from WDV-based gene vectors. Particle bombardment technology (biolistics; 23) was used, so that WDV DNA could be delivered to cultured dedifferentiated cells and whole plant tissues such as embryos. Only 1% of the amount of WDV-based vector DNA was required for biolistics (24; Materials and Methods) compared with protoplast electroporation (25,26) or polyethylene glycol (PEG)-mediated procedures (4,27) and, therefore, the biolistic approach was anticipated to more accurately reflect the activity of virion sense gene expression from the replicating WDV DNA molecules (replicons), rather than input plasmid DNA, in wheat suspension cells and embryos. Few transient assays have used replicating vectors to study expression and where WDV-based constructs have been used, studies of replication rather than expression have often been the main aim (17,28,29). Where gene expression has been studied, a transient expression pattern similar to non-replicating constructs was observed (13,30). However, if replicon DNA synthesis was equal to or exceeded replicon DNA degradation, then the period of transient expression may be expected to be prolonged. This maintenance of gene expression in combination with high copy number has been shown in dicot-infecting geminivirus vectors carrying uidA and nptII (31). In cases where replicating, monocot-infecting geminivirus vectors have been employed, the potential benefits have not been conclusively demonstrated, calling into question the competence of particular cell types to maximally express reporter genes included on replicating vectors (13,19,30,32). We set out to investigate the expression of WDV-based constructs in different tissues of wheat and, in doing so, uncovered a host-directed control of virion sense gene expression. Materials and Methods Plasmid constructs pWDV2LUCΔ was constructed by deleting sequences of pWDV2LUC between the KpnI site at the 5′-end of the truncated C1:C2 ORF and the EcoRV site near the 3′-end of the luc gene. The KpnI site was filled in and the plasmid recircularised, forming pWDV2:5′LUC. A blunt-ended 3′-fragment of the luc gene, from the internal SphI site to the XbaI site at the end of the gene, was subcloned into the SmaI site of pBluescript SK+ (33), which had previously had a NotI linker inserted at the EcoRV site of the polylinker. The luc fragment was removed using NotI and ligated into the NotI site of pWDV2:5′LUC, such that the overlapping sequence of the luc gene was a direct repeat, forming pWDV2LUCΔ. The orientation of the luc fragment was confirmed using restriction digests. The AVB365 deletion of pWDV2 was so-called because a 365 bp fragment between the AvaII site (coordinate 142) and the BamHI site (coordinate 507) was removed. The deletion forming pWDV3 was made by cutting and filling in the NdeI sites within the C1 ORF, causing a frameshift. Both AVB365 and pWDV3 were described previously (4). A GUS coding sequence (34) was cloned into pJIT30 [a 35S-CaMV poly(A) cassette; 35], generating pJIT58, and the 35S-GUS-CaMV poly(A) fragment was inserted into the BamHI site of AVB365 to give pAVB365:35SGUS, while the same GUS gene used in pWDV2GUS was cloned into the BamHI site of pWDV3 to give pWDV3GUS. pJIT166, a pUC19-based plasmid containing a 35S-GUS cassette, was described previously (36). All plasmids were constructed using standard recombinant DNA procedures (37) and grown in Escherichia coli strain DH5α. Plant material Suspension cultures. Suspension cultures of wheat (Triticum aestivum L.), designated line TaKB1 (38), previously described as Triticum monococcum (4 and references therein) and wheat (cv. Glennson), were maintained in a 35 ml volume, at 25°C on a rotary shaker at 120 r.p.m. with a 3 cm throw, with a 16/8 h light/dark cycle. The cells were subcultured every 4 days, with 10 ml of old culture added to 25 ml of fresh medium. The medium used, called CHS, was Murashige and Skoog (39) medium (Imperial Laboratories, Andover, UK), pH 5.8, supplemented with 3% (w/v) sucrose, 0.2% (w/v) casein enzymatic hydrolysate, 0.02% (w/v) myo-inositol and 1.0 mg l−1 2,4-dichlorophenoxyacetic acid. For bombardment, suspension cells were grown for 5 days after subculturing and gently pelleted by centrifugation at 180 g in an MSE Mistral 2000 swing out rotor (MSE Scientific Instruments, Loughborough, UK). The supernatant was removed and ∼0.5 cm3 packed cell volume was spread in a uniform layer onto a 7 cm filter paper (Whatman No. 1) which had been placed on CHS medium solidified with 0.8% (w/v) Difco BiTek agar (Difco Laboratories, Detroit, MI). Immature embryos. Immature embryos from T.aestivum (cv. Glennson) were dissected from seeds 12–14 days post anthesis. The plants were grown in a greenhouse and supplemented with light to give a 16 h photoperiod and a 20°C day and 14°C night. Harvested seeds were surface sterilised in 10% sodium hypochlorite solution for 10 min and rinsed thoroughly in three changes of sterile distilled water. For bombardment, 20 embryos of between 1 and 2 mm diameter were excised under aseptic conditions and placed in a 2 cm diameter circle on SS2 medium, a modified Murashige and Skoog (39) medium, pH 5.8, supplemented with 20 g sucrose, 150 mg asparagine, 0.5 mg thiamine-HCl, 0.01% (w/v) myo-inositol and 2 mg l−1 2,4-dichlorophenoxyacetic acid (40), with the axis facing up and the scutellar tissue in contact with the medium. Microprojectile bombardment The particle gun apparatus employed for this work was the PDS1000/He device (Bio-Rad). Preparation of 1 µm gold particles, the DNA coating procedure and gun operation were carried out as described (24), except that the coated gold particles were finally resuspended in 40 µl of 100% ethanol and 4 µl aliquots were loaded onto the macrocarriers. The DNA was delivered to suspension cells using 1550 p.s.i. rupture disks, while 1100 p.s.i. rupture disks were selected for wheat embryo bombardment. Nucleic acid isolation and Southern hybridization DNA and RNA from suspension cells and embryos were extracted as previously described (4,6). An aliquot of 5 µg of total genomic DNA was digested with the appropriate restriction enzyme or left undigested and separated on a 1.0% (w/v) agarose, Tris-borate gel (2.4 V cm−1 for 16 h). The DNA was transferred to a nitrocellulose filter by capillary blotting, then subjected to standard prehybridising and hybridising treatments (37). A 1.2 kb HindIII fragment from the C1:C2 region of WDV (see Fig. 1A) was used as a probe. Densitometry Bands on autoradiographs corresponding to WDV DNA were scanned using a Joyce Loebl Chromoscan 3, with a 0.1 mm aperture setting. Absorbence values were compared with values obtained from known quantities of cloned viral DNA, allowing quantification of both input and replicon DNA present in samples extracted from bombarded cells. RNase A/T1 protection assay Luciferase transcript was detected by protection of a luc antisense riboprobe from digestion by RNases A and T1 (41,42). The luc coding sequence (coordinates 1544–1765) inserted into pBluescript SK+ (33) was linearised with HindIII. Approximately 0.25 µg of the linearised DNA was used to make a 32P-labelled probe using the Riboprobe in vitro transcription kit from Promega. The riboprobe was subjected to electrophoresis through 6% polyacrylamide-urea gels (42) and was eluted using 300 µl of 0.5 M ammonium acetate, 1 mM EDTA, 0.2% (w/v) SDS and, after phenol extraction and ethanol precipitation, 0.5 × 106 c.p.m. of probe was co-precipitated with 10–20 µg of total RNA extracted from bombarded plant material. The RNase A/T1-resistant band was detected by autoradiography and the relative amounts at different time points were determined by densitometry. Luciferase assay Bombarded material was pulverised in 200 µl of lysis buffer (100 mM potassium phosphate, pH 7.0, 1 mM DTT and 1 mg ml−1 BSA) using a pestle and mortar. The homogenate was centrifuged for 1 min and 120 µl of supernatant was added to 1 ml of assay buffer (14 mM Gly-Gly buffer, pH 7.8, 14 mM magnesium chloride, 1 mg ml−1 BSA) and 120 µl of 60 mM ATP, pH 7.0. An aliquot of 400 µl of this solution was placed in a 5 ml Rohren tube (Sarstedt, Germany) and the tube placed in a luminometer (LUMAT 9501; Berthold Instruments, Germany). An aliquot of 100 µl of 1 mM d-luciferin (Sigma) was injected into the sample and the relative light units (RLU) emitted over 5 s were measured. Luciferase activity was expressed in RLU mg−1 total protein. Cells bombarded with pUC19 were used as a control and the RLU mg−1 total protein value of the control was subtracted from the RLU mg−1 total protein of the sample. GUS assay The GUS-Light™ chemiluminescent assay kit (Tropix Inc., Bedford, MA) was used for assaying the GUS activity in bombarded samples. The tissue was pulverised in 200 µl of lysis buffer, centrifuged for 1 min and 50 µl of the supernatant was added to 200 µl of reaction buffer. After 1 h incubation in the dark at room temperature, 125 µl of sample was placed in a Rohren tube, in the luminometer. An aliquot of 100 µl of GUS accelerator was injected into the sample and, after a 5 s delay, the RLU emitted over 5 s was measured. As with the luciferase assay, pUC19 bombarded cells were used as a negative control, the control value being subtracted from the value of the sample and the RLU expressed per mg total protein. Protein determination For both the luciferase and GUS assay, the extracted sample was assayed for total protein using the dye binding assay of Bradford (43). Histochemical GUS assay GUS activity was determined histochemically using a 0.5 g l−1 solution of 5-bromo-4-chloro-3-indolyl-β-d-glucuronide (X-Gluc), as described previously (44). Immature embryos were placed in individual wells of a Falcon Microtest III flexible assay plate (Becton Dickinson Labware, Oxnard, CA) containing 500 µl of X-Gluc solution. Plates were incubated for 16 h at 37°C. Results Plasmid constructs The WDV-based vectors used for particle bombardment were similar to those described in our previous work (4), except that either the luciferase (luc; 45) or β-glucuronidase (GUS; 34) coding sequences were inserted between the BstEII site (coordinate 500) and the MluI site (coordinate 1246) of WDV (6), replacing the V2 ORF in a transcriptional fusion to the virion sense promoter, generating pWDV2LUC (Fig. 1A) and pWDV2GUS, respectively. The WDV-based plasmids were designed to permit escape of a viral genome by recombination between repeat segments of viral DNA (Fig. 1A and B), followed by replication to high copy number in the nucleus of the host cell (4). Replication of WDV DNA in wheat suspension cultures and immature zygotic embryos Cell suspensions of T.aestivum TaKB1 and wheat (cv. Glennson) immature zygotic embryos were bombarded with pWDV2LUC (Fig. 1A) and total DNA was harvested after 2, 5 and 8 days. Southern blot analysis of the total DNA extracted from both suspension cells and embryos revealed the presence of the supercoiled, linear and open circular dsDNA forms of the WDV2LUC replicon (Fig. 1C and data not shown). Using the isoschizomers DpnI and MboI allowed distinction between WDV DNAs present in our samples as replicon or input plasmid (Figs 1C and 2). Input DNA (detected only in total DNA samples extracted from bombarded suspension cells and not from bombarded embryos), prepared from a dam+E.coli strain, is methylated at the sequence 5′-GATC-3′ (46) and can be digested with DpnI, yielding five fragments which cross-hybridised with the C1:C2 probe (Fig. 1A), whereas the 3.8 kb replicon molecule (WDV2LUC; Fig. 1B), synthesised in planta, is not adenine methylated (47) and was susceptible to digestion with the restriction enzyme MboI, releasing three fragments which cross-hybridised with a C1:C2 probe (Fig. 1A–C). Band a (Fig. 1C) was common to both MboI- and DpnI-sensitive DNAs, generated from replicon and input plasmid DNA, respectively, and was therefore chosen for densitometry. It should be noted that when input DNA was digested with DpnI, there was always a background of partially digested products (Fig. 1C). The three DpnI sites in the complementary sense ORFs of pWDV2-type constructs, giving rise to bands b and e (Fig. 1A and C), are always cut very slowly. However, longer exposures of the autoradiographs caused these bands to become more visible (48). Although partial digestion occurred in independent DNA samples and with different batches of DpnI, the background of partial products was always <10% of the total DNA and, given the magnitude of the relative changes in input DNA levels over time, we considered that this did not substantially interfere with our observations. A similar amount of replicon DNA was detected in both suspension cells and embryos at equivalent time points. In suspension culture samples, input plasmid DNA always declined rapidly over the time course to almost undetectable levels by 8 days post-bombardment (d.p.b.). Input plasmid DNA was never detected in total DNA samples extracted from bombarded embryos (Fig. 2A). Over the experimental period, suspension cells typically doubled in mass, while the fresh weight of the embryo material increased 7-fold (Fig. 2B). Although pWDV2LUC contained an intact C1 and C2 ORF and an LIR, no direct replication of the plasmid, resulting in a 7.9 kb MboI-sensitive molecule, was observed on Southern blots (Fig. 1C) and this is consistent with other reports (30). Figure 1 View largeDownload slide Organization of the WDV-based plasmid vector pWDV2LUC, showing relevant restriction sites, the expected replicon molecule generated by homologous recombination (WDV2LUC) and restriction pattern of the two DNA molecules. (A) Circular map of pWDV2LUC, showing the relative positions of the ORFs C1, C2 and V1, the small intergenic regions (SIRs) and LIR containing the virion and complementary sense promoters (Pv, filled arrow; Pc, empty arrow). The dotted outline represents the region of pBluescript SK+. The LUC reporter and ampicillin resistance genes are marked, along with restriction sites (E, H, K, N, S and X stand for EcoRV, HindIII, KpnI, NotI, SphI and XbaI, respectively). The sequence ‘gatc’ represents sites in input plasmid DNA that can be cleaved with DpnI. The bold arc defines the 1.2 kb fragment used as a probe for hybridisation. The bands expected to hybridise to the probe after digestion with DpnI are also marked (a, b, c, d and e). (B) Circular map of WDV2LUC, the expected replicon molecule generated by homologous recombination of pWDV2LUC in wheat cells. The C1, C2 and V1 ORFs are shown, along with the position of the SIR, LIR and LUC reporter gene. The sequence ‘gatc’ represents the sites which can be cleaved with MboI on the in planta-generated replicon. The three fragments generated by digestion with MboI which hybridise to the probe are indicated (a, b and e). (C) Autoradiograph of a Southern blot of input plasmid DNA (pWDV2LUC) and DNA extracted from TaKB1 cells 5 days after bombardment with pWDV2LUC. The lanes marked Input DNA represent 5 µg TaKB1 genomic DNA mixed with 100 pg of pWDV2LUC plasmid, while tracks labelled Bombarded Cells contain 5 µg total genomic DNA from TaKB1 cells inoculated with pWDV2LUC. The DNA was digested with either DpnI (D), MboI (M) or left undigested (U). Open circular (oc), linear (lin) and supercoiled (sc) forms of the replicon molecule (WDV2LUC) are indicated, along with the source of the digestion products. A 1.2 kb DNA fragment comprising the WDV complementary sense ORFs and small intergenic region was used as a probe. Lane L contains a mixture of λ DNA digested with HindIII and HindIII+EcoRI. Band a was used for quantification by densitometry. Figure 1 View largeDownload slide Organization of the WDV-based plasmid vector pWDV2LUC, showing relevant restriction sites, the expected replicon molecule generated by homologous recombination (WDV2LUC) and restriction pattern of the two DNA molecules. (A) Circular map of pWDV2LUC, showing the relative positions of the ORFs C1, C2 and V1, the small intergenic regions (SIRs) and LIR containing the virion and complementary sense promoters (Pv, filled arrow; Pc, empty arrow). The dotted outline represents the region of pBluescript SK+. The LUC reporter and ampicillin resistance genes are marked, along with restriction sites (E, H, K, N, S and X stand for EcoRV, HindIII, KpnI, NotI, SphI and XbaI, respectively). The sequence ‘gatc’ represents sites in input plasmid DNA that can be cleaved with DpnI. The bold arc defines the 1.2 kb fragment used as a probe for hybridisation. The bands expected to hybridise to the probe after digestion with DpnI are also marked (a, b, c, d and e). (B) Circular map of WDV2LUC, the expected replicon molecule generated by homologous recombination of pWDV2LUC in wheat cells. The C1, C2 and V1 ORFs are shown, along with the position of the SIR, LIR and LUC reporter gene. The sequence ‘gatc’ represents the sites which can be cleaved with MboI on the in planta-generated replicon. The three fragments generated by digestion with MboI which hybridise to the probe are indicated (a, b and e). (C) Autoradiograph of a Southern blot of input plasmid DNA (pWDV2LUC) and DNA extracted from TaKB1 cells 5 days after bombardment with pWDV2LUC. The lanes marked Input DNA represent 5 µg TaKB1 genomic DNA mixed with 100 pg of pWDV2LUC plasmid, while tracks labelled Bombarded Cells contain 5 µg total genomic DNA from TaKB1 cells inoculated with pWDV2LUC. The DNA was digested with either DpnI (D), MboI (M) or left undigested (U). Open circular (oc), linear (lin) and supercoiled (sc) forms of the replicon molecule (WDV2LUC) are indicated, along with the source of the digestion products. A 1.2 kb DNA fragment comprising the WDV complementary sense ORFs and small intergenic region was used as a probe. Lane L contains a mixture of λ DNA digested with HindIII and HindIII+EcoRI. Band a was used for quantification by densitometry. Reporter gene expression from pWDV2LUC in wheat suspension cells and immature zygotic embryos Samples of the same pWDV2LUC-bombarded suspension cells and embryos were assayed for luciferase activity 2, 5 and 8 d.p.b. (Fig. 2B). In six independent experiments, luciferase activity from bombarded suspension cells and embryos was always highest after 2 days and declined steadily to <10% of this value by day 8. The decline in luciferase expression was matched by a similar decline in the steady-state levels of luc transcript, measured by RNase A/T1 protection, indicating that the decline in activity was at the level of transcription or RNA stability (Fig. 2B). However, luc activity from embryo material was more than double that observed in suspension culture cells (Fig. 2B). Replacement of Pv with the CaMV 35S promoter confers the same patterns of gene expression Pv in pWDV2GUS was replaced with the cauliflower mosaic virus (CaMV) 35S promoter in the deletion AVB365 (4), generating pAVB365:35SGUS. This deletion did not interfere with WDV dsDNA replication (4,48), but the pattern of decline in GUS expression from pAVB365:35SGUS in suspension cells over 8 days was identical to that of pWDV2GUS and to pJIT166, a non-replicating plasmid containing the GUS reporter gene fused to the 35S promoter (Fig. 3A). pWDV3GUS, which contained a frameshift mutation in C2, did not generate replicons (4,48), produced no detectable GUS activity when bombarded into suspension cells (Fig. 3A) and, therefore, acted as a negative control for background levels of GUS activity that is sometimes associated with plant cell extracts. Figure 2 View largeDownload slide Replication and virion sense expression of pWDV2LUC in bombarded wheat suspension cells and immature zygotic embryos, along with fresh weight gain of the tissue, over an 8 day time course. (A) Graphs showing amounts of input and replicon DNA in bombarded suspension cells and embryos 2, 5 and 8 d.p.b. At each time point, cells from three bombarded dishes were pooled, total DNA was extracted and used for the quantification of WDV DNA. Data shown are mean values estimated from analysis of three individual Southern blots. Black bars represent standard errors. (B) Graphs showing luciferase activity, luc mRNA levels and fresh weight gain of wheat suspension cells and immature zygotic embryos at 2, 5 and 8 d.p.b. Eighteen independently inoculated dishes of suspension cells and embryos (20 embryos per dish) from six bombardment experiments were used to calculate the mean luc activity and fresh weight gain. Black bars represent standard errors. Total RNA was extracted from three dishes of bombarded suspension cells or embryos and used for RNase A/T1 protection. The protected band was quantified by densitometry and data are expressed as a percentage of the day 2 value. Figure 2 View largeDownload slide Replication and virion sense expression of pWDV2LUC in bombarded wheat suspension cells and immature zygotic embryos, along with fresh weight gain of the tissue, over an 8 day time course. (A) Graphs showing amounts of input and replicon DNA in bombarded suspension cells and embryos 2, 5 and 8 d.p.b. At each time point, cells from three bombarded dishes were pooled, total DNA was extracted and used for the quantification of WDV DNA. Data shown are mean values estimated from analysis of three individual Southern blots. Black bars represent standard errors. (B) Graphs showing luciferase activity, luc mRNA levels and fresh weight gain of wheat suspension cells and immature zygotic embryos at 2, 5 and 8 d.p.b. Eighteen independently inoculated dishes of suspension cells and embryos (20 embryos per dish) from six bombardment experiments were used to calculate the mean luc activity and fresh weight gain. Black bars represent standard errors. Total RNA was extracted from three dishes of bombarded suspension cells or embryos and used for RNase A/T1 protection. The protected band was quantified by densitometry and data are expressed as a percentage of the day 2 value. When embryos were bombarded with replicating WDV-based plasmids (pWDV2GUS and pAVB365:35SGUS), reporter gene activity decreased over time, but the mean values of GUS activity were very large; 50 times higher than embryos inoculated with pJIT166 2 d.p.b. (Fig. 3B), between 20 and 40 times greater than in suspension cells at equivalent time points. Despite the decrease in activity over time, GUS activity from embryos bombarded with the replicating constructs was 10-fold higher after 8 days than values in suspension cells after 2 days (Fig. 3A and B). In contrast, GUS expression from pJIT166 was approximately equal in both suspension cells and immature zygotic embryos (Fig. 3A and B) over 8 days. No GUS expression was detected in embryos bombarded with pWDV3GUS (Fig. 3B). Figure 3 View largeDownload slide Comparison of GUS activities from four different plasmid constructs in wheat suspension cells and immature zygotic embryos. Mean GUS activity (RLU mg−1 total protein) was measured after 2, 5 and 8 days using 16 independently inoculated dishes of material at each time point, taken from four experiments. Black bars represent standard errors. (A) GUS activity in bombarded wheat suspension cells. (B) GUS activity in bombarded wheat immature zygotic embryos. Figure 3 View largeDownload slide Comparison of GUS activities from four different plasmid constructs in wheat suspension cells and immature zygotic embryos. Mean GUS activity (RLU mg−1 total protein) was measured after 2, 5 and 8 days using 16 independently inoculated dishes of material at each time point, taken from four experiments. Black bars represent standard errors. (A) GUS activity in bombarded wheat suspension cells. (B) GUS activity in bombarded wheat immature zygotic embryos. Histochemical staining reveals microscopic areas of GUS expression in suspension cells, but macroscopic regions of GUS expression associated with the embryonic axis On dishes of bombarded suspension cells, small GUS-positive cell clumps were revealed by X-Gluc treatment 2, 5 and 8 d.p.b. Both replicating and non-replicating constructs gave similar numbers of GUS-positive cells and the number of blue spots always declined over the 8 day time period (Fig. 4A–I). Figure 4 View largeDownload slide Localisation of GUS reporter gene expression in wheat suspension cells and immature zygotic embryos. Photographs of wheat suspension cells (A–I) and immature zygotic embryos (J–R) stained with X-Gluc solution after bombardment with either pWDV2GUS (A, D, G, J, M and P), pAVB365:35SGUS (B, E, H, K, N and Q) or pJIT166 (C, F, I, L, O and R). (A–C) and (J–L) show samples stained at 2 d.p.b., while (D–F) and (M–O), and (G–I) and (P–R) show samples stained at 5 and 8 d.p.b., respectively. Figure 4 View largeDownload slide Localisation of GUS reporter gene expression in wheat suspension cells and immature zygotic embryos. Photographs of wheat suspension cells (A–I) and immature zygotic embryos (J–R) stained with X-Gluc solution after bombardment with either pWDV2GUS (A, D, G, J, M and P), pAVB365:35SGUS (B, E, H, K, N and Q) or pJIT166 (C, F, I, L, O and R). (A–C) and (J–L) show samples stained at 2 d.p.b., while (D–F) and (M–O), and (G–I) and (P–R) show samples stained at 5 and 8 d.p.b., respectively. In contrast, the intensity, size and persistence of areas of staining on embryos bombarded with either pWDV2GUS or pAVB365:35SGUS (Fig. 4J–K, M–N and P–Q) were much greater than on the dishes of bombarded suspension cells (Fig. 4A–I), whereas embryos inoculated with the non-replicating vector pJIT166 showed only small, discrete blue GUS staining spots, an observation more similar to the situation found in bombarded suspension cells (Fig. 4L, O and R). Out of a total of 1080 embryos bombarded over three separate experiments, 97.5% of those inoculated with the WDV-based constructs were positive for GUS activity 2 d.p.b. and 82.5% were positive 8 d.p.b. This can be compared with 85% GUS-positive embryos 2 days and 32.5% positive embryos 8 days after inoculation with pJIT166. Embryos bombarded with WDV-based vectors typically displayed large, readily visible, GUS-positive areas on the embryonic axis at all the time points of the experiment (Fig. 4J–K, M–N and P–Q). Much smaller discrete spots were also occasionally observed on callusing scutellar tissue and these were of a similar size to the single type of GUS-positive spots found on pJIT166-inoculated embryos (Fig. 4L, O and R). It was noticeable that staining was never detected on the green, developing leaf primordia (Fig. 4P–R). Contribution of input plasmid and replicon DNA to gene expression in different cell types To quantify the contribution of replicon DNA to the level of reporter gene expression, the vector pWDV2LUCΔ was constructed (Materials and Methods), in which the luc coding sequence provided the overlapping sequences for recombination-mediated escape of the WDV2LUC replicon, prior to its amplification by dsDNA replication. As described above, pWDV2LUC (Fig. 1A) relied upon homologous recombination between repeat sequences of the C1 and C2 ORFs for escape of the replicon (Fig. 1B). However, the pWDV2LUCΔ plasmid did not include a complete luciferase coding sequence and therefore was not expected to produce luc activity, but replicons generated from this plasmid via homologous recombination between the overlapping section of the two incomplete luc coding regions could contain an intact luc coding sequence. Therefore, any luciferase activity observed in pWDV2LUCΔ-inoculated cells would reflect virion sense expression from replicon DNA alone. In contrast, pWDV2LUC could generate luciferase activity from both input plasmid and replicon DNA. We took advantage of the activity of Pv in E.coli (32) by performing a luciferase assay on pulverised bacterial cells containing either pWDV2LUC or pWDV2LUCΔ, to confirm that while luciferase activity was readily detectable in bacteria harbouring pWDV2LUC (1.82 × 107 RLU mg−1 total protein), <0.02% of that activity was detected in those containing pWDV2LUCΔ. Both pWDV2LUC and pWDV2LUCΔ were compared directly by bombardment into wheat suspension cells and immature zygotic embryos, followed by assaying for replication and luc activity after 2, 5 and 8 days. The ability to generate WDV2LUCΔ replicons from pWDV2LUCΔ was not impaired in suspension cells when compared with pWDV2LUC-bombarded cells over an 8 day time course, but luciferase activity from pWDV2LUCΔ detected in suspension cells was <20% of the level detected from pWDV2LUC-inoculated cells after 2 days (Fig. 5A). Luciferase activity from both plasmids declined between 2 and 8 d.p.b., but the values of luc expression suggested that only a small proportion of the reporter gene activity detected in suspension cells was encoded by replicon molecules. In bombarded embryos, both pWDV2LUC and pWDV2LUCΔ replicated equally well, but no difference was observed in luc activity from the two plasmids (Fig. 5B) and luc activity from pWDV2LUC was considerably higher in bombarded embryos than in suspension cells (Fig. 5A and B). These data imply that in immature zygotic embryos, the majority of reporter gene expression must have been generated from the replicon molecules, as pWDV2LUCΔ cannot generate a functional luc coding sequence. Figure 5 View largeDownload slide Comparison of replication and virion sense expression from pWDV2LUC and pWDV2LUCΔ in wheat suspension cells and immature zygotic embryos. At each of the three time points, bombarded material from three dishes were pooled, total DNA was extracted and used to determine the quantity of WDV DNA. Mean amounts of replicon DNA were estimated from densitometry of three separate Southern blots. Mean luc activities were calculated 2, 5 and 8 d.p.b. from 18 dishes of independently inoculated material taken from six independent experiments. Black bars represent standard errors. (A) Bombarded wheat suspension cells. (B) Bombarded wheat immature zygotic embryos. Figure 5 View largeDownload slide Comparison of replication and virion sense expression from pWDV2LUC and pWDV2LUCΔ in wheat suspension cells and immature zygotic embryos. At each of the three time points, bombarded material from three dishes were pooled, total DNA was extracted and used to determine the quantity of WDV DNA. Mean amounts of replicon DNA were estimated from densitometry of three separate Southern blots. Mean luc activities were calculated 2, 5 and 8 d.p.b. from 18 dishes of independently inoculated material taken from six independent experiments. Black bars represent standard errors. (A) Bombarded wheat suspension cells. (B) Bombarded wheat immature zygotic embryos. Discussion Relationship between levels of WDV DNA and gene expression in wheat suspension cultures and immature zygotic embryos From our experiments, we concluded that the full potential for virion sense expression from WDV-based replicating vectors was somehow compromised in suspension culture cells and that reporter gene activity in suspension cells correlated more closely with the decline in levels of input plasmid DNA than with the maintenance of WDV replicon DNA, raising doubts about the ability of WDV replicons to fully express virion sense genes in these cells. Over a number of experiments, levels of replicon DNA generated from pWDV-based plasmids were equivalent in both suspension cultures and embryos up to 8 d.p.b. (e.g. pWDV2LUC in Figs 2A and 5A and B), but input plasmid DNA was only detected in suspension culture cells (Figs 1C and 2A), not in bombarded embryos (Fig. 2B). An active mechanism for removal of m6A DNA (i.e. input plasmid DNA) may exist in these embryonic cells, as has been reported in meristematic barley cells (49), but if so, would be absent in suspension cells. It was possible to achieve levels of WDV replicon which represented the majority of WDV DNA present in the cell but this was not reflected in high levels of replicon-directed virion sense expression and this decline in virion sense reporter gene activity was matched by a corresponding decline in transcript (Fig. 2B). Levels of reporter gene expression were always greater in cells of the immature zygotic embryo than in suspension cultures, despite there being approximately equal amounts of replicon DNA and, in the latter, extra copies of reporter genes on remaining input DNA. In embryos, luc expression driven by Pv was consistently greater than double that observed in suspension cultures (Fig. 2A and B), while GUS expression, driven by either the Pv or CaMV 35S promoter, was 20–40 times higher in embryos compared with suspensions (Fig. 3A and B). Furthermore, in embryos, replicating constructs produced 50 times more GUS activity than a non-replicating 35S-GUS plasmid (pJIT166), but in suspension cells, pJIT166 yielded similar amounts of GUS activity as replicating WDV-based constructs (Fig. 3A and B). These results appear more extreme when using the GUS reporter gene, rather than luc, and this may be due to an accumulation of GUS enzyme in the cell, compared with the more rapid turnover of luciferase (50). It should also be emphasised that, despite equivalent levels of replicon DNA generated from the WDV-based plasmid containing 35S-GUS (pAVB365:35SGUS) in suspension cells and embryos, the huge difference in expression levels between the two experimental systems was maintained as observed for pWDV2GUS, in which the reporter gene is fused to Pv. The 35S promoter has been shown to be unaffected by flanking WDV sequences (4). High levels of reporter gene expression from replicating WDV-based vectors can only be realised in immature zygotic embryos Replicating viral vectors showed no advantage in terms of reporter gene expression in suspension cells, but high levels of GUS activity from replicating WDV-based constructs was detected in embryos, appearing as mass stained areas in and adjacent to the embryonic axis (Fig. 4J–K, M–N and P–Q). Since only replicon DNA was detected in these embryos (Fig. 2B), this staining was interpreted as being derived solely from replicon DNA. Large regions of GUS staining were limited to the embryonic axis and were never observed in cells of the scutellum or developing leaf primordium, implying that the passive spread of replicating WDV-based molecules occurred between dividing cells within the embryonic axis, but not to adjacent tissues. Constructs incapable of replicating produced only localised gene expression in bombarded embryonic cells and their immediate descendants (e.g. pJIT166 in Fig. 4L, O and R), as did replicating vectors inoculated into callusing scutellar tissue (e.g. Fig. 4N) or dedifferentiated suspension cells (Fig. 4A–B, D–E and G–H), where input plasmid DNA was responsible for most of the expression. The virion sense proteins have been implicated in the systemic spread of WDV and other Poaceae-infecting geminiviruses through their hosts (7,8,10,11), but V2 was deleted in the WDV plasmid vectors (Fig. 1A; Materials and Methods). During active cell division, high copy replicon molecules may be maintained in both daughter cells, but as cells differentiate (e.g. into leaf mesophyll cells) and enter G0, replication of viral dsDNA becomes reduced in line with diminished host cell DNA synthesis (20) and this could account for the lack of large sectors of GUS staining in suspension cells and on scutellar-derived callus tissue. What is the relative contribution of input plasmid DNA and replicon DNA to total reporter gene expression? Comparisons between pWDV2LUC and pWDV2LUCΔ allowed us to estimate that virion sense expression from WDVLUC replicon DNA in suspension cells accounted for <20% of the total activity, but all of the luciferase activity in the immature zygotic embryos (Fig. 5A and B). These data suggest that WDV replicons did not achieve maximal rates of virion sense expression in wheat suspension cells compared with the same sequences on input plasmid DNA. Therefore, in suspension cells, Pv must be more active on the input plasmid DNA than on the replicon. Our earlier data (Fig. 3A and B) also imply that this promoter displayed differential activity on input plasmid and replicon, respectively. Comparison with previous studies in cereal protoplasts and embryos Expression of WDV-based vectors in protoplasts of cereal cell suspension cultures has generally been related to the appearance of replicating WDV DNA (4,13,19,32). However, with the exception of one published report (4), input DNA has not been seen to persist for long periods in protoplast cultures (13) and therefore no comments have been made about the possible contribution of input DNA to total levels of expression. Our data do not disagree with previous reports, as some expression was observed from the replicon molecules (Fig. 5A). Comparisons with previously published work are further complicated by the use of replication-defective mutants of WDV-based plasmids, which have parts of C1 and C2 disrupted, as non-replicating controls (13,19,32). Pv-directed expression is controlled by C1 and C2, independent of replication, and, as a consequence, the contribution of high copy number replicons to virion sense reporter gene expression may have been overestimated (4). Our observations that pWDV2GUS gave no greater GUS activity than either replicating or non-replicating 35S-based constructs in suspension cultures are consistent with the same comparison made in T.monococcum protoplasts and maize suspension culture cells (13,19). Wild-type WDV DNA inoculated into protoplasts of cereal suspension cultures produced dsDNA replicons, but no single-stranded (ss)DNA was observed (10,13,32), suggesting that expression of V2 may not occur in these cells. Any ssDNA intermediates (arising from rolling circle replication of WDV; 17) not sequestered into virus particles would be converted to dsDNA forms (19,51). This explanation is consistent with the low levels of replicon-associated virion sense gene expression reported here. However, in barley anther culture-derived embryos (undergoing rapid growth and differentiation) bombarded with wild-type WDV DNA, viral ssDNA was predominant (52), suggesting that expression of V2 was functional. Comparison of non-replicating 35S-reporter gene fusions with their WDV- or MSV-based counterparts, imbibed into desiccated cereal embryos or agro-infected into maize seedlings, respectively (53,54), revealed elevations of reporter gene activity of up to 20-fold in geminivirus-based constructs over non-replicating 35S plasmid controls, further supporting the observations reported here. Requirements for high level replicon-associated expression From all the above considerations, we propose that determinants for virion sense expression are present in cells that are undergoing or are about to undergo differentiation. The suspension cells we have used are dedifferentiated, have lost totipotency and may have less of the factors required to support replicon-associated virion sense expression. In contrast to this host cell specificity for maximum virion sense expression, dsDNA replication requires only that the host cell is actively carrying out DNA synthesis, a view consistent with many previous reports (3 and references therein). Possible mechanisms The tissue specificity of replicon-associated virion sense expression is independent of the control of the C1 or C1:C2 protein and is not mediated by Pv, since maximum expression of WDV-based, 35S-containing constructs were also tissue dependent. Virion sense expression from input plasmid DNA in suspension culture cells was better than that from the replicon DNA, so we propose that replicon DNA, but not input plasmid DNA, is subject to a host cell determination of competence to transcribe virion sense genes. The term competence is used to distinguish between this form of control exerted irrespective of the promoter residing in place of Pv and the control of gene expression mediated by cellular or viral-encoded transcription factors exerted upon specific promoter sequences. Similar observations made in maize suspension cell protoplasts, in which WDV-GUS replicons attained 2500 copies per cell but GUS expression did not undergo such a corresponding increase when compared with a non-replicating 35S-GUS plasmid, was attributed to a titration out of factors required for gene expression (19), but our data are not explained by this suggestion, as we would expect that a limitation of host factors would apply equally to promoters on both replicon and input plasmid DNA. The ability of WDV to generate replicons, capable or incapable of virion sense expression, may be analogous to SV40 and CaMV, where two subsets of minichromosomes may exist in infected cells, competent either for transcription or replication (55–57). Thus, the WDV replicon as a minichromosome may have cell type-dependent activation states of its chromatin, controlling access of transcription factors to any promoter carried on the replicon. In bombarded suspension culture cells, chromatin associated with introduced plasmid DNA may well be different from that associated with the in planta-generated replicon. We have demonstrated that the methylation state of the input and replicon DNAs are different (Fig. 1C), which may be the cause of the observed difference in their expression in suspension cells. Using m5C-specific restriction enzymes, we found no differences in the methylation pattern between replicons from embryos and suspension cells (48). However, this assay would account for only a small fraction of all potential C-methylated sites (58). Small regions of methylated DNA can act as foci for inactive chromatin assembly, leading to the inactivation of distant promoter sequences, irrespective of their specific modes of control (59). Therefore, altered states of methylation and chromatin as modes of control of replicon-directed virion sense expression need not be mutually exclusive. Both complementary sense WDV proteins (C1 and C1:C2) contain an Rb-binding motif (like those found in SV40 and human adenoviruses), allowing the potential activation of Rb or Rb-like host factors (5,22). C1, for which no function has been previously demonstrated, is required for maximum virion sense gene expression (5,48) but not viral DNA replication, suggesting that C1 alone may be responsible for interactions with host transcription factors. If such factors were lacking in terminally dedifferentiated cells, this could explain tissue-specific gene expression patterns observed in these experiments. Acknowledgements We gratefully acknowledge the skilled technical assistance of Sandra Robinson for the preparation of wheat immature zygotic embryos and Dr Margaret Boulton for critical reading of the manuscript. P.G. was supported by a CASE studentship from the Biotechnology and Biological Sciences Research Council and Plant Breeding International plc. This work was conducted under licence no. PHF 1185B/17/111 from the Ministry of Agriculture, Fisheries and Food. References 1 Murphy F.A.,  Fauquet C.M.,  Bishop D.H.L.,  Ghabrial S.A.,  Jarvis A.W.,  Martelli G.P.,  Mayo M.A.,  Summers M.D.. ,  Virus Taxonomy. Sixth Report of the International Committee on Taxonomy of Viruses ,  1995 New York, NY Springer-Verlag 2 Padidam M.,  Beachy R.N.,  Fauquet C.M.. ,  J. Gen. Virol. ,  1995, vol.  76 (pg.  249- 263) CrossRef Search ADS PubMed  3 Mullineaux P.M.,  Davies J.W.,  Woolston C.J..  Wilson T.M.A.,  Davies J.W.. ,  Genetic Engineering with Plant Viruses ,  1992 Boca Raton, FL CRC Press(pg.  187- 215) 4 Hofer J.M.I.,  Dekker E.L.,  Reynolds H.V.,  Woolston C.J.,  Cox B.S.,  Mullineaux P.M.. ,  Plant Cell ,  1992, vol.  4 (pg.  213- 223) CrossRef Search ADS PubMed  5 Collin S.,  Fernandez-Lobato M.,  Gooding P.S.,  Mullineaux P.M.,  Fenoll C.. ,  Virology ,  1996, vol.  219 (pg.  324- 329) CrossRef Search ADS PubMed  6 Woolston C.J.,  Barker R.,  Gunn H.,  Boulton M.I.,  Mullineaux P.M.. ,  Plant Mol. Biol. ,  1988, vol.  11 (pg.  35- 43) CrossRef Search ADS PubMed  7 Mullineaux P.M.,  Boulton M.I.,  Bowyer P.,  Vlugt R.V.D.,  Marks M.,  Donson J.,  Davies J.W.. ,  Plant Mol. Biol. ,  1988, vol.  11 (pg.  57- 66) CrossRef Search ADS PubMed  8 Boulton M.I.,  Steinkellner H.,  Donson J.,  Markham P.G.,  King D.I.,  Davies J.W.. ,  J. Gen. Virol. ,  1989, vol.  70 (pg.  2309- 2323) CrossRef Search ADS PubMed  9 Lazarowitz S.G.,  Pinder A.J.,  Damsteegt V.D.,  Rogers S.G.. ,  EMBO J. ,  1989, vol.  8 (pg.  1023- 1032) PubMed  10 Woolston C.J.,  Reynolds H.V.,  Stacey N.J.,  Mullineaux P.M.. ,  Nucleic Acids Res. ,  1989, vol.  17 (pg.  6029- 6041) CrossRef Search ADS PubMed  11 Boulton M.I.,  Pallaghy C.K.,  Chatani M.,  MacFarlane S.,  Davies J.W.. ,  Virology ,  1993, vol.  192 (pg.  85- 93) CrossRef Search ADS PubMed  12 Morris-Krsinich B.A.M.,  Mullineaux P.M.,  Donson J.,  Boulton M.I.,  Markham P.G.,  Short M.N.,  Davies J.W.. ,  Nucleic Acids Res. ,  1985, vol.  13 (pg.  7237- 7256) CrossRef Search ADS PubMed  13 Matzeit V.,  Schaefer S.,  Kammann M.,  Schalk H.J.,  Schell J.,  Gronenborn B.. ,  Plant Cell ,  1991, vol.  3 (pg.  247- 258) CrossRef Search ADS PubMed  14 Schalk H.J.,  Matzeit V.,  Schiller B.,  Schell J.,  Gronenborn B.. ,  EMBO J. ,  1989, vol.  8 (pg.  359- 364) PubMed  15 Accotto G.P.,  Donson J.,  Mullineaux P.M.. ,  EMBO J. ,  1989, vol.  8 (pg.  1033- 1039) PubMed  16 Dekker E.L.,  Woolston C.J.,  Xue Y.,  Cox B.,  Mullineaux P.M.. ,  Nucleic Acids Res. ,  1991, vol.  19 (pg.  4075- 4081) CrossRef Search ADS PubMed  17 Heyraud F.,  Matzeit V.,  Kammann M.,  Schaefer S.,  Schell J.,  Gronenborn B.. ,  EMBO J. ,  1993, vol.  12 (pg.  4445- 4452) PubMed  18 Townsend R.,  Watts J.,  Stanley J.. ,  Nucleic Acids Res. ,  1986, vol.  14 (pg.  1253- 1265) CrossRef Search ADS PubMed  19 Timmermans M.C.P.,  Das O.P.,  Messing J.. ,  Nucleic Acids Res. ,  1992, vol.  20 (pg.  4047- 4054) CrossRef Search ADS PubMed  20 Accotto G.P.,  Mullineaux P.M.,  Brown S.,  Marie D.. ,  Virology ,  1993, vol.  195 (pg.  257- 259) CrossRef Search ADS PubMed  21 Nagar S.,  Pedersen T.J.,  Carrick K.M.,  Hanley-Bowdoin L.,  Robertson D.. ,  Plant Cell ,  1995, vol.  7 (pg.  705- 719) CrossRef Search ADS PubMed  22 Xie Q.,  Suarez-Lopez P.,  Gutierrez C.. ,  EMBO J. ,  1995, vol.  14 (pg.  4073- 4082) PubMed  23 Klein T.M.,  Wolf E.D.,  Wu R.,  Sanford J.C.. ,  Nature ,  1987, vol.  327 (pg.  776- 777) CrossRef Search ADS   24 Kikkert J.R.. ,  Plant Cell Tissue Organ Culture ,  1993, vol.  33 (pg.  221- 226) CrossRef Search ADS   25 Dekeyser R.A.,  Claes B.,  De Rycke R.M.U.,  Habets M.E.,  Van Montagu M.C.,  Caplan A.B.. ,  Plant Cell ,  1990, vol.  2 (pg.  591- 602) CrossRef Search ADS PubMed  26 D'Halluin K.,  Bonne E.,  Bossut M.,  De Beuckeleer M.,  Leemans J.. ,  Plant Cell ,  1992, vol.  4 (pg.  1495- 1505) CrossRef Search ADS PubMed  27 Datta S.K.,  Peterhans A.,  Datta K.,  Potrykus I.. ,  BioTechnology ,  1990, vol.  8 (pg.  736- 740) CrossRef Search ADS   28 Kammann M.,  Schalk H.J.,  Matzeit V.,  Schaefer S.,  Schell J.,  Gronenborn B.. ,  Virology ,  1991, vol.  184 (pg.  786- 790) CrossRef Search ADS PubMed  29 Kammann M.,  Matzeit V.,  Schmidt B.,  Schell J.,  Walden R.,  Gronenborn B.. ,  Gene ,  1991, vol.  104 (pg.  247- 252) CrossRef Search ADS PubMed  30 Suarez-Lopez P.,  Gutierrez C.. ,  Virology ,  1997, vol.  227 (pg.  389- 399) CrossRef Search ADS PubMed  31 Hayes R.J.,  Petty I.T.D.,  Coutts R.H.A.,  Buck K.W.. ,  Nature ,  1988, vol.  334 (pg.  179- 182) CrossRef Search ADS   32 Ugaki M.,  Ueda T.,  Timmermans M.C.P.,  Vieira J.,  Elliston K.O.,  Messing J.. ,  Nucleic Acids Res. ,  1991, vol.  19 (pg.  371- 377) CrossRef Search ADS PubMed  33 Alting-Mees M.A.,  Short J.M.. ,  Nucleic Acids Res. ,  1989, vol.  17 pg.  9494  CrossRef Search ADS PubMed  34 Jefferson R.A.,  Kavanagh T.A.,  Bevan M.W.. ,  EMBO J. ,  1987, vol.  6 (pg.  3901- 3907) PubMed  35 Mullineaux P.M.,  Guerineau F.,  Accotto G.P.. ,  Nucleic Acids Res. ,  1990, vol.  18 (pg.  7259- 7265) CrossRef Search ADS PubMed  36 Guerineau F.,  Lucy A.,  Mullineaux P.M.. ,  Plant Mol. Biol. ,  1992, vol.  18 (pg.  815- 818) CrossRef Search ADS PubMed  37 Sambrook J.,  Fritsch E.F.,  Maniatis T.. ,  Molecular Cloning: A Laboratory Manual ,  1989 2nd Edn. Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press 38 Leitch S.A.R.,  Schwarzacher T.,  Wang M.L.,  Leitch I.J.,  Surlan-Momirovich G.,  Moore G.,  Heslop-Harrison J.S.. ,  Plant Cell Tissue Organ Culture ,  1993, vol.  33 (pg.  287- 296) CrossRef Search ADS   39 Murashige T.,  Skoog F.. ,  Physiol. Plant. ,  1962, vol.  15 (pg.  473- 497) CrossRef Search ADS   40 Sears R.G.,  Deckard E.L.. ,  Crop Sci. ,  1982, vol.  22 (pg.  546- 550) CrossRef Search ADS   41 Melton D.A.,  Krieg P.A.,  Rebagliati M.R.,  Maniatis T.,  Zinn K.,  Green M.R.. ,  Nucleic Acids Res. ,  1984, vol.  12 (pg.  7035- 7056) CrossRef Search ADS PubMed  42 Mullineaux P.M.,  Rigden J.E.,  Dry I.B.,  Krake L.R.,  Rezaian M.A.. ,  Virology ,  1993, vol.  193 (pg.  414- 423) CrossRef Search ADS PubMed  43 Bradford M.M.. ,  Anal. Biochem. ,  1976, vol.  72 (pg.  248- 254) CrossRef Search ADS PubMed  44 McCabe D.E.,  Swain W.F.,  Martinell B.J.,  Christou P.. ,  BioTechnology ,  1988, vol.  6 (pg.  923- 926) CrossRef Search ADS   45 Ow D.W.,  Wood K.V.,  DeLuca M.,  De Wet J.R.,  Helinski D.R.,  Howell S.H.. ,  Science ,  1986, vol.  234 (pg.  856- 859) CrossRef Search ADS PubMed  46 Nelson M.,  McClelland M.. ,  Nucleic Acids Res. ,  1989, vol.  17 (pg.  389- 415) CrossRef Search ADS PubMed  47 Chen D.F.,  Dale P.J.,  Heslop-Harrison J.S.,  Snape J.W.,  Harwood W.,  Bean S.,  Mullineaux P.M.. ,  Plant J. ,  1994, vol.  5 (pg.  429- 436) CrossRef Search ADS   48 Gooding P.S.. ,  PhD thesis ,  1995 University of East Anglia 49 Rogers S.W.,  Rogers J.C.. ,  Plant Mol. Biol. ,  1992, vol.  18 (pg.  945- 961) CrossRef Search ADS PubMed  50 Martin T.,  Wohner R.V.,  Hummel S.,  Willmitzer L.,  Frommer W.B..  Gallegher S.R.. ,  GUS Protocols ,  1992 New York, NY Academic Press(pg.  23- 43) 51 Donson J.,  Morris-Krsinich B.A.M.,  Mullineaux P.M.,  Boulton M.I.,  Davies J.W.. ,  EMBO J. ,  1984, vol.  3 (pg.  3069- 3073) PubMed  52 Creissen G.,  Smith C.,  Francis R.,  Reynolds H.,  Mullineaux P.. ,  Plant Cell Rep. ,  1990, vol.  8 (pg.  680- 683) CrossRef Search ADS PubMed  53 Topfer R.,  Gronenborn B.,  Schell J.,  Steinbiss H.H.. ,  Plant Cell ,  1989, vol.  1 (pg.  133- 139) CrossRef Search ADS PubMed  54 Shen W.H.,  Hohn B.. ,  Plant J. ,  1994, vol.  5 (pg.  227- 236) CrossRef Search ADS   55 Edenberg H.J.. ,  Nature ,  1980, vol.  286 (pg.  529- 531) CrossRef Search ADS PubMed  56 Covey S.N.. ,  Semin. Virol. ,  1991, vol.  2 (pg.  151- 159) 57 Covey S.N.,  Turner D.S.,  Stratford R.,  Saunders K.,  Lucy A.,  Riseborough S.,  Ray P.. ,  Plant Mol. Biol. ,  1991, vol.  2 (pg.  1- 10) 58 Pfeifer G.P.,  Steigerwald S.D.,  Mueller P.R.,  Wold B.,  Riggs A.D.. ,  Science ,  1989, vol.  246 (pg.  810- 813) CrossRef Search ADS PubMed  59 Kass S.U.,  Goddard J.P.,  Adams R.L.P.. ,  Mol. Cell. Biol. ,  1993, vol.  13 (pg.  7372- 7379) CrossRef Search ADS PubMed  © 1999 Oxford University Press
TI - Plant cell-directed control of virion sense gene expression in wheat dwarf virus
JF - Nucleic Acids Research
DO - 10.1093/nar/27.7.1709
DA - 1999-04-01
UR - https://www.deepdyve.com/lp/oxford-university-press/plant-cell-directed-control-of-virion-sense-gene-expression-in-wheat-mtEEYLt5dw
SP - 1709
EP - 1718
VL - 27
IS - 7
DP - DeepDyve
ER -