109

Fast forward genetics based on virus-induced David C Baulcombe

Gene expression in plants can be suppressed in a sequence- Figure 1 specific manner by infection with virus vectors carrying fragments of host genes. Recent developments have revealed that the mechanism of this gene silencing is based on an RNA-mediated defence against viruses. It has also emerged that a related mechanism is involved in the post-transcriptional silencing that accounts for between line variation in transgene expression and cosuppresion of transgenes and endogenous genes. The technology of virus-induced gene silencing is being refined and adapted as a high throughput procedure for functional genomics in plants.

Addresses The Sainsbury Laboratory, , , NR4 7UH, UK; e-mail: [email protected]

Current Opinion in Plant Biology 1999, 2:109–113 http://biomednet.com/elecref/1369526600200109

© Elsevier Science Ltd ISSN 1369-5266

Abbreviations PDS phytoene desaturase PTGS post-transcriptional gene silencing PVX potato virus X RMD RNA-mediated defense TBRV tomato blackring virus TMV tobacco mosaic virus VIGS virus-induced gene silencing

Introduction A limiting step in genetics and molecular biology is the Current Opinion in Plant Biology ability to relate genes to phenotypes and vice versa. In the shadow of genome projects this limitation is more conspic- Nicotiana benthamiana plant exhibiting virus-induced gene silencing of uous than ever. Here, I describe a novel approach to phytoene desaturase. The plant was inoculated three weeks earlier with bridging this gap between genetics and molecular biology. a PVX vector construct carrying a 400 nucleotide fragment of phytoene This approach is based on recent discoveries concerning desaturase cDNA. The bleached symptoms contrast with the normal plant viruses and gene silencing and is still under develop- mild mosaic symptoms associated with PVX infection on this species. ment. The early results, however, provide encouragement that virus-based technology will be a useful complement to ples VIGS has been targeted against a chelatase gene existing functional genomics tools. required for chlorophyll synthesis [3•] and against various transgenes [2•,3•,4••,5]. The most straightforward iteration of this approach is referred to as virus-induced gene silencing (VIGS) and employs virus vectors carrying elements from the exons of The obvious attraction of VIGS for analysis of gene func- plant host genes. Analyses in Nicotiana species have shown tion is speed and ease of adaptation to high throughput that the symptoms on these plants are phenocopies of muta- systems. If the partial sequence of a given gene is tions in the corresponding host genes. For example, with known, a virus vector construct can be assembled within tobacco mosaic virus (TMV) [1] and potato virus X (PVX) two days. This construct may be infectious directly as in vectors [2•] carrying a fragment of phytoene desaturase vitro transcripts of cDNA clones [6] or as DNA, if the (PDS) mRNA, the upper leaves of the infected plant devel- virus has a DNA genome. Alternatively the cDNA of an op a spectacular bleached effect (Figure 1). The cause of RNA virus can be fused to a promoter that is active in these symptoms is a decline in the level of the endogenous plant cells. The cDNA of these constructs is infectious PDS mRNA leading to low levels of phytoene desaturase, a as directly inoculated DNA [7] or by employing an block in carotenoid production and, consequently, the A. tumefaciens Ti plasmid vector to transfer the DNA into absence of protection against photobleaching. In other exam- plant cells [8]. 110 Genome studies and molecular genetics

Figure 2 There are two lines of supporting evidence for this model. First there is direct evidence that both RNA nepoviruses [11••] and DNA caulimoviruses can induce RMD [12,13••]. Activation of RMD Initially the plants infected with these viruses exhibit symptoms and the viruses accumulate to a high level. However, the upper leaves of these plants emerging after systemic infection have recovered. They are symptom free and have only low levels of viral RNA. This recovery can be attributed to RMD because, in the nepovirus infected plants, there is RNA sequence-dependent resistance against viruses inoculated to these upper leaves [11••]. In the caulimovirus infected plants the recovery is probably due to RMD as there is post-transcriptional suppression of Virus accumulation (particles/cell) Virus viral RNA in the symptom-free leaves. Time Current Opinion in Plant Biology The second supporting evidence for RMD is indirect and Typical time course of virus accumulation in an infected cell. The graph is based on analyses of plants exhibiting post-transcrip- illustrates schematically the kinetics of virus accumulation in an tional gene silencing (PTGS) of transgenes. PTGS infected cell. There is an initial phase of rapid replication and virus involves sequence-specific turnover of RNA and is one of accumulation that progresses through to the final period when the reasons for between-line variation in transgene expres- accumulation has been arrested. The phases shown would take place over a period of a few days, depending on the type of virus. sion levels [14]. Lines in which a transgene is expressed at a low level include those in which PTGS is active and tar- geted against the transgene RNA. High level expressors After inoculation of the vector there is a delay of only 7–21 are those in which PTGS is either weak or inactive. days before VIGS-related symptoms develop in the upper leaves of the infected plant. The relationship from gene to The link with RMD is from the finding that PTGS is sup- phenotype, therefore, can be established within a few weeks. pressed by virus-encoded proteins that are also suppressors of In genomics jargon this strategy is referred to as reverse anti-viral defense in the host [15••–18••]. Presumably the genetics. VIGS can also be used in a forward genetics strate- suppressed anti-viral defense is the putative RMD and the gy although the resources required are substantial — it suppressed PTGS was activated when the transgene or its would be necessary to construct cDNA libraries in virus vec- RNA was detected by the RMD surveillance system. The tor constructs and to systematically analyse large numbers of suppressors of PTGS would be part of a counter defense sys- infected plants. tem that allows viruses to accumulate after activation of RMD [19•]. With some viruses the final steady state level of virus In the later sections of this article I assess the potential, accumulation would be determined by the effectiveness of the limitations and the practicalities of VIGS both for the suppressors. In other instances the virus may use evasion forward and for reverse genetics. Because the develop- instead of suppression as a strategy to counteract RMD. ment of VIGS technology is influenced by knowledge, however, I first review the current views about the Analysis of PTGS has revealed that there is a signal of underlying mechanism. gene silencing that can move systemically in plants [20••,21••,22•,23•]. This signal can be rationalised in terms The mechanism of VIGS of RMD as it would allow the plant to prevent spread of It is likely that VIGS is caused by an RNA-mediated the virus. If the signal moves at the same time or ahead of defense (RMD) mechanism in plants against viruses the virus it would prime RMD in the cells at, or just [9,10]. According to this idea there is an as yet uncharac- beyond the infection front. Consequently the RMD would terised surveillance system in plants that can specifically be activated more rapidly than in the initially infected cells recognise viral nucleic acids and give sequence specificity and the spread of the virus would be delayed. to RMD. Normally, with wild-type isolates of plant virus- es, it is thought that this mechanism is activated as the A final point about PTGS concerns DNA methylation. virus begins to accumulate in the plant (Figure 2). There is an association of DNA methylation and PTGS in Eventually, as a result of RMD, virus accumulation slows several experimental systems [4••,24–27] and evidence down and eventually stops. that the methylation may be targeted to transgenes by RNA [4••,28]. These observations are difficult to reconcile In the situations when a genetically modified virus has with the proposed similarity of PTGS and RMD as most similarity to a gene in the host plant, the RMD would tar- plant viruses have RNA genomes. Perhaps the DNA get both the viral RNA and the corresponding endogenous methylation data are an indicator that the underlying mRNA. VIGS would result from the targeting of the mechanism of PTGS is involved not only in anti-viral endogenous mRNA defense. It could be, for example, that antiviral defense is Fast forward genetics based on virus-induced gene silencing Baulcombe 111

related to the mechanisms used to protect against retro- this RNA would include an element from a host sequence posons and other forms of mobile DNA in plants [29]. so that the RMD silences the corresponding gene. This RNA would not need to be infectious, however, because it Development of VIGS technology would be delivered into cells by expression of a transgene. This relationship of VIGS, PTGS and RMD has important implications for further development of VIGS technology. It An advantage of transgenic VIGS over ‘conventional’ means, for example, that VIGS would be weak if the virus PTGS using sense or antisense transgenes should be con- vector encodes a suppressor of PTGS and that the most sistency. The transgenic VIGS constructs would be effective VIGS vectors will be those that use evasion rather designed to produce an RNA that is detected by the RMD than suppression as a strategy to overcome RMD. Vectors surveillance system and it would be expected that all, or at based on potyviruses, therefore, would be poor VIGS vectors least most, of the lines would exhibit silencing of the tar- because they produce strong suppressors of PTGS get gene. In PTGS it is likely that the transgene RNA is [15••,16••,18••]. Cucumber mosaic virus would be unsuitable only recognised by the RMD surveillance system as a for the same reason [16••,17••]. Conversely, PVX and tomato result of sequence rearrangements, DNA methylation of blackring virus (TBRV) would be suitable vectors for VIGS. chromatin structures that are associated with the transgene Neither of these viruses is able to reverse PTGS in trans- after its introduction into the plant genome. As a result, the genic plants [16••,17••] and it can be assumed that they do degree of PTGS varies greatly between lines. not produce suppressors or that their suppressors are weak. Transgenic VIGS is less amenable than infectious VIGS to TBRV is a seed and pollen transmitted virus as a result of its high throughput applications because of the transformation ability to infect the progenitor cells of the pollen and eggs in step. In plants that can be transformed with high efficiency, the meristems [30]. This ability to penetrate the meristem however, it is quite feasible to produce the tens of thou- means that TBRV would be a particularly effective vector sands of transformed lines that would be necessary to for VIGS. Most plant viruses, including PVX, are excluded survey the gene function in a typical plant. Once produced, from the meristematic zone of infected plants [31] and the seed from these lines would be a permanent resource would not be able to silence genes required for meristem that could be used in many different functional screens. identity and early in leaf and flower development. Unfortunately, as yet, there are no full length cDNAs of The version of transgenic VIGS corresponding most close- TBRV or related viruses that could be developed as VIGS ly to infectious VIGS employs amplicons. Amplicons are vectors with the ability to target these important genes. transgenes comprising a promoter and terminator directing transcription of a modified viral vector RNA in the plant There are several other ways in which understanding of cell [39]. The modifications may include deletion or muta- PTGS could facilitate development of VIGS technology. tion of viral genes required for spread of the virus in the For example, it may be possible to enhance effectiveness infected plant and of any other functions that are secondary of VIGS vectors by engineering increased potential to pro- to replication. Expression of the amplicon in the plant cell duce double stranded RNA. Double stranded RNA has could activate RMD and, if there are host sequences in the been implicated as an initiator of PTGS in plant systems amplicon construct, PTGS of the corresponding host gene. [32•,33••,34•] and of a similar gene silencing phenomenon in nematodes [35,36•]. There are several instances of amplicon transgenes although they are not always referred to as such. A PVX amplicon pro- It may also be possible to enhance VIGS by modification duced consistent silencing as expected [39]. Similarly, a of the host plant. For example, it might be expected that petunia chalcone synthase gene could be targeted by a gem- VIGS would be stronger in the background of host gene ini virus amplicon with a chalcone synthase insert [40] and a mutations that enhance gene silencing (egs) [37]. The brome mosaic virus amplicon conferred virus resistance that products of these genes could be repressors of RMD. It was probably due to RMD [41]. The phenotype of ampli- might also be expected that VIGS would be stronger in cons derived from cucumber mosaic virus [42] and a severe plants that over-express genes required for RMD than in a strain of TMV, however, did not appear to involve gene wild-type host. The genes that are required for gene silencing [43]. The differences between these amplicons silencing can be recognised by analysis of the suppressed may be due to the suppressors of RMD encoded in the cor- gene silencing (sgs) [38•] phenotype of mutant alleles. responding viral genomes — genomes producing little or no silencing may produce strong suppressors of RMD, whereas Transgenic VIGS good silencers may be those that do not encode suppressors In the applications of VIGS, as described above, the acti- or from which the suppressor genes have been deleted. vator of gene silencing is the viral RNA in the infected plant cells. The silencing effect is systemic because the Some alternative approaches to PTGS in transgenic plants virus vector is able to spread through the infected plant. In may also be considered as transgenic ‘pseudoVIGS’. The transgenic VIGS the aim is to produce an RNA that can be success of these approaches depends on the ability of non recognised by the RMD surveillance system. As in VIGS, viral transgenes to produce the virus-like that activate 112 Genome studies and molecular genetics

RMD. The gene silencing from simultaneous expression of Acknowledgements sense and antisense RNAs [33••] or from transgene con- I am grateful to the Gatsby Charitable Foundation for supporting work in • the Sainsbury Laboratory and to my colleagues for many stimulating structs with inverted repeats in the transcribed regions [32 ] discussions about the role and mechanisms of gene silencing in plants. could both be considered in this ‘pseudoVIGS’ category . References and recommended reading VIGS versus other techniques of functional Papers of particular interest, published within the annual period of review, genomics have been highlighted as: Conventional approaches to forward genetics generally • of special interest •• of outstanding interest involve insertional mutagenesis either by T-DNA [44] or 1. Kumagai MH, Donson J, Della-Cioppa G, Harvey D, Hanley K, Grill LK: transposable elements [45,46]. These are powerful Cytoplasmic inhibition of carotenoid biosynthesis with virus- approaches that have led to many advances in gene discov- derived RNA. Proc Natl Acad Sci USA 1995, 92:1679-1683. ery; however, there are several complications with 2. Ruiz MT, Voinnet O, Baulcombe DC: Initiation and maintenance of insertional mutagenesis of genes required for basic cell func- • virus-induced gene silencing. Plant Cell 1998, 10:937-946. Compares VIGS of a transgene and an endogenous gene, as in [3•]. The tion or early development. In these instances, because the analysis reveals separate initiation and maintenance stages of VIGS and a insertional mutation often results in complete inactivation of surprising difference in the properties of VIGS targetted against transgenes and endogenous genes. the target gene, the mutant phenotype will be embryo lethal and difficult to interpret in terms of the cellular or biochem- 3. Kjemtrup S, Sampson KS, Peele CG, Nguyen LV, Conkling MA, • Thompson WF, Robertson D: Gene silencing from plant DNA ical role of the gene product. Conversely, if the mutant gene carried by a geminivirus. Plant J 1998, 14:91-100. is one of a family it is likely that there may be no phenotype Shows that a DNA geminivirus can initiate VIGS although does not rule out that viral RNA is the initiator of gene silencing. As with [2•] compares VIGS of an insertional mutant because the functional homologues of endogenous genes and transgenes. may compensate for the loss of function at the mutant locus. 4. Jones AL, Thomas CL, Maule AJ: De novo methylation and co- •• suppression induced by a cytoplasmically replicating plant RNA VIGS and related PTGS-based approaches may circumvent virus. EMBO J 1998, 17:6385-6393. Extends and establishes general importance of RNA directed methylation of these problems. For example, with multigene families, the DNA as reported originally in [28]. This paper used a viral system instead of a gene silencing mechanism will target not only the precise viroid as in the previous work. The paper is also important because it demon- homologue of the gene in the VIGS vector; it will also target strates VIGS in a legume. Most of the other examples are in Nicotiana species. partial homologues that vary by up to 10 or 20%. The chlorot- 5. Lindbo JA, Silva-Rosales L, Proebsting WM, Dougherty WG: Induction of a highly specific antiviral state in transgenic plants: ic and dwarf phenotypes that result from PTGS of tobacco implications for regulation of and virus ribulose bisphosphate carboxylase (rubisco) is a good illustra- resistance. Plant Cell 1993, 5:1749-1759. tion of how gene silencing can target a multigene family 6. Donson J, Kearney CM, Hilf ME, Dawson WO: Systemic expression [47,48]. It is unlikely that a phenotype would result from of a bacterial gene by a tobacco mosaic virus-based vector. Proc Natl Acad Sci USA 1991, 88:7204-7208. insertional mutagenesis of rubisco because it is encoded in a 7. Sablowski RWM, Baulcombe DC, Bevan MW: Expression of a multigene family. The precise mismatch between different flower-specific Myb protein in leaf cells using a viral vector tobacco rubisco genes is not known but, by comparison with causes ectopic activation of a target promoter. Proc Natl Acad Sci other solanaceous plants, it is likely to be up to 14% [49]. USA 1995, 92:6901-6905. 8. Turpen TH, Turpen AM, Weinzettl N, Kumagai MH, Dawson WO: When VIGS is targeted against lethal genes the phenotype Transfection of whole plants from wounds inoculated with Agrobacterium tumefaciens containing cDNA of tobacco mosaic is death, as in insertional mutagenesis, but the dying virus. J Virol Methods 1993, 42:227-240. process may be more informative than the embryo lethal 9. van Kammen A: Virus-induced gene silencing in infected and phenotype of insertional mutants. Because the VIGS is transgenic plants. Trends Plant Sci 1997, 2:409-411. applied to mature plants and because there may be only 10. Dawson WO: Gene silencing and virus resistance: a common partial suppression of the target gene, the dying process mechanism. Trends Plant Sci 1996, 1:107-108. may be informative about the role of the corresponding 11. Ratcliff F, Harrison BD, Baulcombe DC: A similarity between viral gene products. It might be expected, for example, that •• defense and gene silencing in plants. Science 1997, 276:1558- 1560. death due to a block in lipid biosynthesis would be clearly With [13••] provides an indication, in non-transgenic plants and with a wild- different from death due to suppression of DNA synthesis type virus, for an RMD that resembles PTGS. or proteins required for respiration 12. Al-Kaff NS, Covey SN, Kreike MM, Page AM, Pinder R, Dale PJ: Transcriptional and post-transcriptional plant gene silencing in Conclusions response to a pathogen. Science 1998, 279:2113-2115. VIGS vectors have been developed from the genomes of 13. Covey SN, Al-Kaff NS, Langara A, Turner DS: Plants combat •• infection by gene silencing. Nature 1997, 385:781-782. TMV, PVX and TGMV for applications in Nicotiana With [11••] provides an indication, in non-transgenic plants and with a wild- species. To develop these vectors for use in other plant type virus, for an RMD that resembles PTGS. species it will be possible to build on recent discoveries that 14. Depicker A, Van Montagu M: Post-transcriptional gene silencing in VIGS and PTGS are related to antiviral defense in plants plants. Curr Opin Cell Biol 1997, 9:372-382. and that viruses produce suppressors of this antiviral 15. Kasschau KD, Carrington JC: A counter-defensive strategy of plant •• viruses: suppression of post-transcriptional gene silencing. Cell defense. When these vectors are available for Arabidopsis, 1998, 95:461-470. maize and other species it will be possible to combine inser- This paper and the others appearing within a few weeks of each other •• •• tional mutagenesis and VIGS to develop a powerful [16 –18 ] provide indirect but compelling evidence that RMD is a general defense mechanism against plant viruses and that viruses have evolved combined approach to functional surveys of plant genomes. counter-defense strategies. Fast forward genetics based on virus-induced gene silencing Baulcombe 113

16. Brigneti G, Voinnet O, Wan-Xiang L, Liang-Hui J, Ding S-W, 31. Matthews REF: Plant , edn 3. San Diego: Academic Press; •• Baulcombe DC: Viral pathogenicity determinants are suppressors 1991. of transgene silencing in Nicotiana benthamiana. EMBO J 1998, 17 32. Hamilton AJ, Brown S, Han YH, Ishizuka M, Lowe A, Solis AGA, :6739-6746. • This paper and the others appearing within a few weeks of each other Grierson D: A transgene with repeated DNA causes high [15••,17••,18••] provide indirect but compelling evidence that RMD is a gen- frequency, post- transcriptional suppression of ACC-oxidase gene eral defense mechanism against plant viruses and that viruses have evolved expression in tomato. Plant J 1998, 15:737-746. counter-defense strategies. Provides supporting evidence for the possible role of double stranded RNA as an initiator of gene silencing. 17. Beclin C, Berthome R, Palauqui JC, Tepfer M, Vaucheret H: Infection •• 33. Waterhouse PM, Graham HW, Wang MB: Virus resistance and of tobacco or Arabidopsis plants by CMV counteracts systemic •• post-transcriptional silencing of non-viral (trans)genes. Virology gene silencing in plants can be induced by simultaneous expression of sense and antisense RNA. 1998, in press. Proc Natl Acad Sci USA 95 This paper and the others appearing within a few weeks of each other 1998, :13959-13964. [15••,16••,18••] provide indirect but compelling evidence that RMD is a gen- Provides supporting evidence for the possible role of double stranded RNA eral defense mechanism against plant viruses and that viruses have evolved in gene silencing. The presented data also explain why viral antisense trans- counter-defense strategies. genes do not confer virus resistance. It is probably because the gene silenc- ing requires the presence of both sense and antisense RNAs. 18. Anandalakshmi R, Pruss GJ, Ge X, Marathe R, Smith TH, Vance VB: A 34. Metzlaff M, O’Dell M, Cluster PD, Flavell RB: RNA-mediated RNA •• viral suppressor of gene silencing in plants. Proc Natl Acad Sci • degradation and chalcone synthase A silencing in petunia. Cell USA 1998, 95:13079-13084. 1997, 88:845-854. This paper and the others appearing within a few weeks of each other Provides supporting evidence for the possible role of double stranded RNA [15••,16••,17••] provide indirect but compelling evidence that RMD is a gen- in gene silencing. eral defense mechanism against plant viruses and that viruses have evolved counter-defense strategies. 35. Tabara H, Grishok A, Mello CC: RNA in C. elegans: soaking in the genome sequence. Science 1998, 282:430-431. 19. Carrington JC, Whitham SA: Viral invasion and host defense: • strategies and counter-strategies. Curr Opin Plant Biol 1998, 36. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC: 1:336-341. • Potent and specific genetic interference by double-stranded RNA A concise and informative review that puts RMD in the context of the vari- in . Nature 1998, 391:806-811. ous virus resistance mechanisms deployed in plants and the corresponding There is a remarkable similarity between this genetic interference in nema- variety of counter-defense strategies developed by plant viruses. todes and gene silencing in plants. See also [35]. 20. Palauqui J-C, Elmayan T, Pollien J-M, Vaucheret H: Systemic acquired 37. Dehio C, Schell J: Identification of plant genetic loci involved in •• silencing: transgene-specific post-transcriptional silencing is post-transcriptional mechanism for meiotically reversible transmitted by grafting from silenced stocks to non-silenced transgene silencing. Proc Natl Acad Sci USA 1994, 91:5538-5542. scions. EMBO J 1997, 16:4738-4745. With [21••] the first report that gene silencing is associated with systemic 38. Elmayan T, Balzergue S, Beon F, Bourdon V, Daubremet J, Guenet Y, • signalling. The experiments involved graftings of silenced plants to non Mourrain P, Palauqui JC, Vernhettes S, Vialle T et al.: Arabidopsis silenced plants. The systemic signal has not yet been characterised but is mutants impaired in cosuppression. Plant Cell 1998, 10:1747-1757. likely to be a nucleic acid. PTGS and RMD will probably not be fully understood until mutations have been isolated and characterised in genes required for silencing. This report 21. Voinnet O, Baulcombe DC: Systemic signalling in gene silencing. provides the first step to that end. It also describes plant lines that would be •• Nature 1997, 389:553. useful for gene over-expression. With [20••] the first report that gene silencing is associated with systemic signalling. The experiments involved localised introduction of green fluores- 39. Angell SM, Baulcombe DC: Consistent gene silencing in cent protein (GFP) DNA into a GFP transgenic plant and demonstrated that transgenic plants expressing a replicating potato virus X RNA. local events could initiate systemic gene silencing. The systemic signal has EMBO J 1997, 16:3675-3684. not yet been characterised but is likely to be a nucleic acid. 40. Atkinson RG, Bieleski LRF, Gleave AP, Jannsen BJ, Morris BAM: Post- 22. Voinnet O, Vain P, Angell S, Baulcombe DC: Systemic spread of transcriptional silencing of chalcone synthase in petunia using a • sequence-specific transgene RNA degradation is initiated by geminivirus-based episomal vector. Plant J 1998, 15:593-604. localised introduction of ectopic promoterless DNA. Cell 1998, 41. Kaido M, Mori M, Mise K, Okuno T, Furusawa I: Inhibition of brome 95:177-187. mosaic virus (BMV) amplification in protoplasts from transgenic Further characterises the mechanisms of systemic silencing but does not tobacco plants expressing replicable BMV RNAs. J Gen Virol demonstrate the chemical identity of the signal. 1995, 76:2827-2833. 23. Palauqui JC, Vaucheret H: Transgenes are dispensable for the RNA 42. Suzuki M, Masuta C, Takanami Y, Kuwata S: Resistance against • degradation step of cosuppression. Proc Natl Acad Sci USA 1998, cucumber mosaic virus in plants expressing the viral replicon. 95:9675-9680. FEBS Lett 1996, 379:26-30. Further characterises the mechanisms of systemic silencing but does not demonstrate the chemical identity of the signal. 43. Yamaya J, Yoshioka M, Meshi T, Okada Y, Ohno T: Expression of tobacco mosaic virus RNA in transgenic plants. Mol Gen Genet 24. English JJ, Baulcombe DC: The influence of small changes in 1988, 211:520-525. transgene transcription on homology-dependent virus resistance and gene silencing. Plant J 1997, 12:1311-1318. 44. AzpirozLeehan R, Feldmann KA: T-DNA insertion mutagenesis in Arabidopsis: Going back and forth. Trends Genet 1997, 13:152-156. 25. Sijen T, Wellink J, Hiriart J-B, van Kammen A: RNA-mediated virus resistance: role of repeated transgenes and delineation of 45. Briggs SP: Plant genomics: more than food for thought. Proc Natl targeted regions. Plant Cell 1996, 8:2277-2294. Acad Sci USA 1998, 95:1986-1988. 26. English JJ, Mueller E, Baulcombe DC: Suppression of virus 46. Martienssen RA: Functional genomics: probing plant gene function accumulation in transgenic plants exhibiting silencing of nuclear and expression with transposons. Proc Natl Acad Sci USA 1998, genes. Plant Cell 1996, 8:179-188. 95:2021-2026. 27. de Carvalho F, Gheysen G, Kushnir S, Van Montagu M, Inzé D, 47. Quick WP, Schurr U, Scheibe R, Schulze ED, Rodermel SR, Bogorad Castresana C: Suppression of b-1,3-glucanase transgene L, Stitt M: Decreased ribulose-1,5-bisphosphate carboxylase- expression in homozygous plants. EMBO J 1992, 11:2595-2602. oxygenase. In Transgenic Tobacco Transformed With Antisense Rbcs 1. Impact On Photosynthesis In Ambient Growth-Conditions Planta 28. Wassenegger M, Heimes S, Riedel L, Sänger HL: RNA-directed de 1991, 183:542-554. novo methylation of genomic sequences in plants. Cell 1994, 76:567-576. 48. Hudson GS, Evans JR, Voncaemmerer S, Arvidsson YBC, Andrews TJ: Reduction of ribulose-1,5-bisphosphate carboxylase 29. Matzke MA, Matzke AJM: Epigenetic silencing of plant transgenes oxygenase content by antisense RNA reduces photosynthesis. In as a consequence of diverse cellular defence responses. Cell Mol Transgenic Tobacco Plants Plant Physiol 1992, 98:294-302. Life Sci 1998, 54:94-103. 49. Dean C, Pichersky E, Dunsmuir P: Structure, evolution, and 30. Lister RM, Murant AF: Seed transmission of nematode-borne regulation of Rbcs genes. In Higher-Plants Annu Rev Pl Physiol viruses. Ann Appl Biol 1967, 59:49-62. Plant Mol Biol 1989, 40:415-439.