Suppression of beta-1,3-glucanase transgene expression in homozygous plants

Plant pathology and RNAi: a brief history

This article describes the discovery of RNA-activated sequence-specific RNA degradation, a phenomenon now referred to as RNA silencing or RNA interference (RNAi). From 1992 to 1996, a series of articles were published on virus resistant transgenic plants expressing either translatable or nontranslatable versions of the coat protein gene of Tobacco etch virus (TEV). Certain transgenic plant lines were resistant to TEV but not to closely related viruses. In these plants a surprising correlation was observed: Transgenic plant lines with the highest degree of TEV resistance had actively transcribed transgenes but low steady-state levels of transgene RNA. Molecular analysis of these transgenic plants demonstrated the existence of a cellular-based, sequence-specific, posttranscriptional RNA-degradation system that was programmed by the transgene-encoded RNA sequence. This RNA-degradation activity specifically targeted both the transgene RNA and TEV (viral) RNA for degradation and was the first description of RNA-mediated gene silencing.

Molecular mechanisms of RNAi: Implications for development and disease

Research over the past few years has led to dramatic new discoveries on the role of double-stranded RNA (dsRNA) in the cell. RNA duplexes have been shown to orchestrate epigenetic changes, repress translation, and direct mRNA degradation in a sequence-specific manner. These diverse effects of dsRNA on gene expression have been termed RNA interference (RNAi). In addition to playing a role in viral defense and silencing transposons, RNAi also has a critical function in a number of developmental processes in the embryo. In this review, we explore these roles and discuss the molecular mechanisms behind dsRNA-mediated gene silencing. Further, we address the use of RNAi as a tool to study gene function in biology, and as a strategy for treating human disease.

Double-stranded RNA mediates homology-dependant gene silencing of γ-tubulin in the human parasite Entamoeba histolytica

Approaches that eliminate mRNA are a powerful tool for reverse genetics applications in eukaryotic microbes for which gene replacement techniques have not yet been developed. Here, for the first time, we demonstrate that RNA duplexes efficiently inhibit gene expression when introduced into the human parasite Entamoeba histolytica. Chemically synthesized, small interfering RNA (siRNA) were highly specific and efficient in silencing parasite gamma-tubulin mRNA. Use of specific antibodies revealed that microtubules and gamma-tubulin were intra-nuclear in E. histolytica. The RNAi approach to modulation of gamma-tubulin mRNA resulted in loss of the highly organized microtubule array an observation that correlates with a significant reduction of gamma-tubulin as well as of the specific mRNA. Our results suggest that gamma-tubulin is essential for microtubule nucleation in E. histolytica.

Chromatin remodeling and stem cell theory of relativity

The field of stem cell biology is currently being redefined. Stem cell (hematopoietic and non-hematopoietic) differentiation has been considered hierarchical in nature, but recent data suggest that there is no progenitor/stem cell hierarchy, but rather a reversible continuum. The stem cell (hematopoietic and non-hematopoietic) phenotype, the total differentiation capacity (hematopoietic and non-hematopoietic), gene expression as well as other stem cell functional characteristics (homing, receptor and adhesion molecule expression) vary throughout a cell-cycle transit widely. This seems to be dependent on shifting chromatin and gene expression with cell-cycle transit. The published data on DNA methylation, histone acetylation, and also RNAi, the major regulators of gene expression, conjoins very well and provides an explanation for the major issues of stem cell biology. Those features of stem cells mentioned above can be rather difficult to apprehend when a classical hierarchy biology view is applied, but they become clear and easier to understand once they are correlated with the underlining epigenetic changes. We are entering a new era of stem cell biology the era of "chromatinomics." We are one step closer to the practical use of cellular therapy for degenerative diseases.

Revealing the world of RNA interference

The recent discoveries of RNA interference and related RNA silencing pathways have revolutionized our understanding of gene regulation. RNA interference has been used as a research tool to control the expression of specific genes in numerous experimental organisms and has potential as a therapeutic strategy to reduce the expression of problem genes. At the heart of RNA interference lies a remarkable RNA processing mechanism that is now known to underlie many distinct biological phenomena.

The effect of endogenous mRNA levels on co-suppression in tomato

Introduction of truncated polygalacturonase (PG) transgenes into tomato plants caused the production of small interfering RNAs (siRNAs) and co-suppression of both the endogenous and PG transgenes in ripening fruits by post-transcriptional gene silencing. In order to test the possible effect on co-suppression of the endogenous PG mRNA level, we transferred the PG transgenes from a PG-silenced line (wild type background) by crossing to two ripening regulatory mutants with reduced PG: Never-ripe (Nr, approximately 10% endogenous PG mRNA compared to wild type) and ripening-inhibitor (rin, approximately 1% endogenous PG mRNA) and to wild type (as a control). The PG transgenes caused strong co-suppression of the transgenes and the endogenous PG gene in cells with high PG mRNA background (wild type) and silencing appeared to be linked with higher transgene copy number and/or a particular transgene locus. In cells with low endogenous PG mRNA accumulation (Nr), the endogenous PG gene was preferentially suppressed compared to the transgenes, whose expression was not reduced significantly. The expression of the transgenes was also not reduced in the very low PG background (rin), in which endogenous PG was barely detectable. In all the analysed lines with all three PG background levels, siRNAs accumulated in leaves and green fruits, in which the endogenous PG gene is not transcribed. The relatively abundant production of siRNAs in most of the lines was linked with a particular transgene insert. These results suggest that a certain threshold level of endogenous PG mRNA is required for the co-suppression of the truncated PG transgenes and the endogenous PG gene or for extensive silencing of the transgenes.

Planting the seeds of a new paradigm

RNA-mediated gene silencing has emerged in recent years as an important mechanism for regulating gene expression. Some of the key discoveries have been made in plants

The presence of a chromatin boundary appears to shield a transgene in tobacco from RNA silencing

We present isogenic transgenic tobacco lines that carry at a given chromosomal position a beta-glucuronidase (GUS) reporter gene either with or without the presence of the matrix-associated region known as the chicken lysozyme A element. Plants were generated with the Cre-lox site-specific recombination system using heterospecific lox sites. Analysis of GUS gene expression in plant populations demonstrates that the presence of the A element can shield against RNA silencing of the GUS gene. Protection was observed in two of three independent tobacco transformants. Plants carrying an A element 5' of the GUS gene always had stable GUS activity, but upon removal of this A element, the GUS gene became silenced over time in two lines, notably when homozygous.

The genetics of RNA silencing

Although initially recognized as a handy tool to reduce gene expression, RNA silencing, triggered by double-stranded RNA molecules, is now recognized as a mechanism for cellular protection and cleansing: It defends the genome against molecular parasites such as viruses and transposons, while removing abundant but aberrant nonfunctional messenger RNAs. The underlying mechanisms in distinct gene silencing phenomena in different genetic systems, such as cosuppression in plants and RNAi in animals, are very similar. There are common RNA intermediates, and similar genes are required in RNA silencing pathways in protozoa, plants, fungi, and animals, thus indicating an ancient pathway. This chapter gives an overview of both biochemical and genetic approaches leading to the current understanding of the molecular mechanism of RNA silencing and its probable biological function.

Epigenetic interactions between Arabidopsis transgenes: characterization in light of transgene integration sites

The stochastic variability of expression that is a characteristic of eukaryotic nuclear transgenes is often attributed to epigenetic mechanisms that are triggered by repetitive transgene locus structures and influenced by chromosomal position effects. In order to address the contribution of chromosomal position effects in the context of a fully sequenced genome, a novel set of transgene loci was established in the compact genome of Arabidopsis thaliana. Transgenes expressing GFP-tagged or GUS-tagged fusion proteins of Arabidopsis COP1 collectively displayed three types of gene silencing, which are distinguished by their developmental timing, gene dosage dependence, (post)transcriptional control, and extent of endogene co-suppression. Subsequently, the heritability of epistatic interactions between allelic and non-allelic transgene loci was investigated in light of both intrinsic transgene features, in particular T-DNA copy number per locus, and chromosomal insertion sites. The notion that chromosomal flanking sequences underlie the ability of transgenes to function as masters or targets of epigenetically heritable trans-silencing interactions was generally not favored by our data. Moreover, among single T-DNA loci at different chromosomal locations the great majority showed homozygosity-dependent posttranscriptional silencing. However, spontaneous silencing (in cis) may be promoted by a pericentromeric location. Instead, intrinsic transgene features correlated with all major aspects of silencing behavior tested.

Gene silencing

Gene silencing has evolved in a broad range of organisms probably as defense mechanisms against invasive nucleic acids. Two major strategies are utilized. Transcriptional gene silencing (TGS) acts to prevent RNA synthesis and posttranscriptional gene silencing (PTGS) acts to degrade existing RNA. Although the final effects are similar, the mechanisms of TGS and PTGS are species specific. In most eukaryotes, gene silencing is associated with de novo DNA methylation. However, Caenorhabditis elegans shows an efficient PTGS-like mechanism but lacks a DNA methylation system. Additionally, key enzymes involved in plant and nematode PTGS, the cellular RNA-directed RNA polymerases, appear to be missing in Drosophila melanogaster. In this review, we discuss common features of TGS and PTGS that have been identified across species but for TGS we will concentrate only on methylation-mediated gene inactivation. This effort is complicated by the vague borders between gene silencing and normal gene regulation. Mechanisms that are involved in gene silencing are also used to regulate controlled expression of endogenous genes. To outline the general aspects, gene silencing will be defined as narrowly as possible. The intention behind this review is to stimulate discussion and we seek to facilitate this by introducing speculative concepts that could lead to some reappraisal of the literature.

Epigenetic history of an Arabidopsis trans-silencer locus and a test for relay of trans-silencing activity

BACKGROUND

Meiotically heritable epimutations affecting transgene expression are not well understood, even and in particular in the plant model species, Arabidopsis thaliana. The Arabidopsis trans-silencer locus, C73, which encodes a fusion protein between the repressor of photomorphogenesis, COP1, and green fluorescent protein (GFP-COP1), heritably modifies the expression pattern and cop1-like cosuppression phenotypes of multiple GFP-COP1 target loci by transcriptional gene silencing.

RESULTS

Here we describe three additional features of trans-silencing by the C73 locus. First, the silencing phenotype of C73 and of similar complex loci was acquired epigenetically over the course of no more than two plant generations via a stage resembling posttranscriptional silencing. Second, imprints imposed by the C73 locus were maintained heritably for at least five generations in the absence of the silencer with only sporadic spontaneous reversion. Third, the pairing of two other GFP-COP1 transgene loci, L91 and E82, showed an increased tendency for epigenetic modification when L91 carried an epigenetic imprint from C73, but not when E82 bore the imprint.

CONCLUSIONS

The latter data suggest a transfer of trans-silencing activity from one transgene locus, C73, to another, namely L91. These results extend our operational understanding of interactions among transgenes in Arabidopsis.

RNA-mediated RNA degradation in transgene- and virus-induced gene silencing

In the 'RNA world' hypothesis it is postulated that RNA was the first genetic molecule. Recent discoveries in gene silencing research on plants, fungi and animals show that RNA indeed plays a key role not only in controlling invading nucleic acids, like viruses and transposable elements, but also in regulating the expression of transgenes and endogenous genes. Double-stranded RNAs were identified to be the triggering structures for the induction of a specific and highly efficient RNA silencing system, in which enzyme complexes, like Dicer and RISC, facilitate as 'molecular machines' the processing of dsRNA into characteristic small RNA species. RNA silencing can be transmitted rapidly from silenced to non-silenced cells by short and long distance signaling. There is evidence that at least one component of the signal is a specific, degradation-resistant RNA.

Sequence-, tissue-, and delivery-specific targeting of RNA during post-transcriptional gene silencing

Transgenic Nicotiana benthamiana plants expressing an untranslatable version of the coat protein (CP) gene from the Tamarillo mosaic virus (TaMV) were either resistant to TaMV infection or recovered from infection. These phenotypes were the result of a post-transcriptional gene silencing (PTGS) mechanism that targeted TaMV-CP sequences for degradation. The TaMV-CP sequences were degraded when present in the wild-type TaMV potyvirus, in transgene mRNA, or in chimeric viral vectors based on White clover mosaic virus. The more efficiently targeted region was mapped to a 134-nt segment. Differences were observed in the efficiency of targeting during cell-to-cell and long-distance movement of the chimeric viruses. However, the TaMV-CP sequences do not appear to be targeted for degradation when delivered by biolistics.

RNA interference

A conserved biological response to double-stranded RNA, known variously as RNA interference (RNAi) or post-transcriptional gene silencing, mediates resistance to both endogenous parasitic and exogenous pathogenic nucleic acids, and regulates the expression of protein-coding genes. RNAi has been cultivated as a means to manipulate gene expression experimentally and to probe gene function on a whole-genome scale.

Intrinsic direct repeats generate consistent post-transcriptional gene silencing in tobacco

It is well documented that transgenes with inverted repeats can efficiently trigger post-transcriptional gene silencing (PTGS), presumably via a double stranded RNA induced by complementary sequences in their transcripts. We show here that transgenes with intrinsic direct repeats can also induce PTGS at a very high frequency (80-100%). A transgene with three or four repeats induced PTGS in almost 100% of the primary transformants, regardless of whether a strong (enhanced 35S promoter) or a relatively weak (chlorophyll a/b binding protein promoter) promoter was used. The PTGS induced by three or four repeats is consistently inherited in subsequent generations, and can inactivate homologous genes in trans. Based on the high frequency and consistent heritability, we propose that the intrinsic direct repeat within a transgene may act as a primary determinant of PTGS referred to as direct repeat-induced PTGS (driPTGS). Silencing occurred in all five genes, in this and two previous reports, suggesting that driPTGS might be a universal gene silencing mechanism both in dicotyledonous tobacco plants and monocotyledonous rice cells. In addition, driPTGS may help dissect the gene silencing mechanism and generate silenced phenotypes useful for research and plant biotechnology products.

The rest is silence

Over the past several years, RNAi and its related phenomena have emerged not only as a powerful experimental tool but also as a new mode of gene regulation. Through a combination of genetic and biochemical approaches we have learned much about the mechanisms underlying dsRNA responses. However, many of the most intriguing aspects of dsRNA-induced gene silencing have yet to be illuminated. What has become abundantly clear is that the complex and highly conserved biology underlying RNA interference is critical both for genome maintenance and for the development of complex organisms. However, it seems probable that we have only begun to reveal the diversity of biological roles played by RNAi-related processes.

Tissue-specific patterns of a maize Myb transcription factor are epigenetically regulated

The maize p1 gene encodes a Myb-homologous regulator of red pigment biosynthesis. To investigate the tissue-specific regulation of the p1 gene, maize plants were transformed with constructs combining promoter and cDNA sequences of two alleles which differ in pigmentation patterns: P1-wr (white pericarp/red cob) and P1-rr (red pericarp/red cob). Surprisingly, all promoter/cDNA combinations produced transgenic plants with red pericarp and red cob (RR pattern), indicating that the P1-wr promoter and encoded protein can function in pericarp. Some of the RR patterned transgenic plants produced progeny plants with white pericarp and red cob (WR pattern), and this switch in tissue-specificity correlated with increased transgene methylation. A similar inverse correlation between pericarp pigmentation and DNA methylation was observed for certain natural p1 alleles, which have a gene structure characteristic of standard P1-wr alleles, but which confer red pericarp pigmentation and are consistently less methylated than standard P1-wr alleles. Although we cannot rule out the possible existence of tissue-specific regulatory elements within the p1 non-coding sequences or flanking regions, the data from transgenic and natural alleles suggest that the tissue-specific pigmentation pattern characteristic of the P1-wr phenotype is epigenetically controlled.

Enabling technologies for manipulating multiple genes on complex pathways

Many complex biochemical pathways in plants have now been manipulated genetically, usually by suppression or over-expression of single genes. Further exploitation of the potential for plant genetic manipulation, both as a research tool and as a vehicle for plant biotechnology, will require the co-ordinate manipulation of multiple genes on a pathway. This goal is currently very difficult to achieve. A number of approaches have been taken to combine or 'pyramid' transgenes in one plant and have met with varying degrees of success. These approaches include sexual crossing, re-transformation, co-transformation and the use of linked transgenes. Novel, alternative 'enabling' technologies are also being developed that aim to use single transgenes to manipulate the expression of multiple genes. A chimeric transgene with linked partial gene sequences placed under the control of a single promoter can be used to co-ordinately suppress numerous plant endogenous genes. Constructs modelled on viral polyproteins can be used to simultaneously introduce multiple protein-coding genes into plant cells. In the course of our work on the lignin biosynthetic pathway, we have tested both conventional and novel methods for achieving co-ordinate suppression or over-expression of up to three plant lignin genes. In this article we review the literature concerning the manipulation of multiple genes in plants. We also report on our own experiences and results using different methods to perform directed manipulation of lignin biosynthesis in tobacco.

Post-transcriptional gene silencing in plants

Post-transcriptional gene silencing (PTGS) in plants is an RNA-degradation mechanism that shows similarities to RNA interference (RNAi) in animals. Indeed, both involve double-stranded RNA (dsRNA), spread within the organism from a localised initiating area, correlate with the accumulation of small interfering RNA (siRNA) and require putative RNA-dependent RNA polymerases, RNA helicases and proteins of unknown functions containing PAZ and Piwi domains. However, some differences are evident. First, PTGS in plants requires at least two genes – SGS3 (which encodes a protein of unknown function containing a coil-coiled domain) and MET1 (which encodes a DNA-methyltransferase) – that are absent in C. elegans and thus are not required for RNAi. Second, all Arabidopsis mutants that exhibit impaired PTGS are hypersusceptible to infection by the cucumovirus CMV, indicating that PTGS participates in a mechanism for plant resistance to viruses. Interestingly, many viruses have developed strategies to counteract PTGS and successfully infect plants – for example, by potentiating endogenous suppressors of PTGS. Whether viruses can counteract RNAi in animals and whether endogenous suppressors of RNAi exist in animals is still unknown.

Transgene silencing of invertedly repeated transgenes is released upon deletion of one of the transgenes involved

To analyse experimentally the correlation between transgene silencing and the presence of an inverted repeat in transgenic Arabidopsis thaliana plants, expression of the beta-glucuronidase (gus) gene was studied when present as a convergently transcribed inverted repeat or as a single copy in otherwise isogenic lines. In transformants containing two invertedly repeated gus genes separated by a 732 bp palindromic sequence, gus expression was low, as exemplified by the expression levels in the parental line KH15. The parental KH15 locus could induce efficiently in trans silencing of gus copies at allelic and non-allelic positions. In transformants containing two invertedly repeated gus genes separated by a 826 bp non-repetitive spacer region, gus expression was high or intermediate, especially in hemizygous state and at late developmental stages, as demonstrated in detail for line KHsb67. Removal of one of the gus copies by Cre recombinase resulted in all cases in constitutively high gus expression in hemizygous as well as in homozygous state. The derived deletion lines could no longer induce in trans silencing of homologous gus copies. The results show that convergent transcription of transgenes in an inverted repeat is an important parameter to trigger their silencing and that co-transformation of two T-DNAs with identical transgenes can be used to obtain inverted repeats and targeted co-suppression of the homologous endogenes. Moreover, the data suggest that the spacer region in between the inverted genes plays a role in the efficiency of initiating and maintaining silencing.

Gene silencing as an adaptive defence against viruses

Gene silencing was perceived initially as an unpredictable and inconvenient side effect of introducing transgenes into plants. It now seems that it is the consequence of accidentally triggering the plant's adaptive defence mechanism against viruses and transposable elements. This recently discovered mechanism, although mechanistically different, has a number of parallels with the immune system of mammals.

Tobacco ribosomal DNA spacer element stimulates amplification and expression of heterologous genes

Here we show that the cis-acting genetic element aps (amplification-promoting sequence), isolated from the nontranscribed spacer region of tobacco ribosomal DNA (rDNA), increases the level of expression of recombinant proteins. Transgenic tobacco plants, transformed with expression cassettes containing the herbicide-resistant acetolactate synthase (hr-ALS) gene or the green fluorescent protein (GFP) gene fused to the aps sequence, had greater levels of corresponding messenger RNAs (mRNAs) and proteins compared to transformants lacking aps. Analysis of transgenic plants showed that aps increased the copy number and transcription of the adjacent heterologous genes and, in the case of hr-ALS, enhanced the herbicide resistance phenotype. Both the increased transgene copy number and enhanced expression were stably inherited. These data provide the first evidence that the aps sequence can be used for gene amplification in transgenic plants and possibly other multicellular organisms.

Functional anatomy of a dsRNA trigger: differential requirement for the two trigger strands in RNA interference

In RNA-mediated interference (RNAi), externally provided mixtures of sense and antisense RNA trigger concerted degradation of homologous cellular RNAs. We show that RNAi requires duplex formation between the two trigger strands, that the duplex must include a region of identity between trigger and target RNAs, and that duplexes as short as 26 bp can trigger RNAi. Consistent with in vitro observations, a fraction of input dsRNA is converted in vivo to short segments of approximately 25 nt. Interference assays with modified dsRNAs indicate precise chemical requirements for both bases and backbone of the RNA trigger. Strikingly, certain modifications are well tolerated on the sense, but not the antisense, strand, indicating that the two trigger strands have distinct roles in the interference process.

Hierarchical patterns of transgene expression indicate involvement of developmental mechanisms in the regulation of the maize P1-rr promoter

The maize P1-rr gene encodes a Myb-homologous transcription factor that regulates the synthesis of red flavonoid pigments. Maize plants transformed with segments of the P1-rr promoter driving a GUS reporter gene exhibit significant variation in transgene expression, both between independent transformation events and among sibling plants derived from a single event. Interestingly, variability in spatial expression is not random; rather, transgene activity occurs predominantly in five patterns that fit a hierarchy: expression is most common in kernel pericarp, with sequential addition of expression in cob glumes, husk, silk, and tassel. The hierarchical expression pattern of P-rr::GUS transgenes suggests a possible model for developmental regulation of the P1-rr gene. Our results demonstrate that variability in transgene expression, a common occurrence in transgenic plant studies, can be informative if adequately analyzed to uncover underlying patterns of gene expression.

Plants as bioreactors for protein production: avoiding the problem of transgene silencing

Plants are particularly attractive as large-scale production systems for proteins intended for therapeutical or industrial applications: they can be grown easily and inexpensively in large quantities that can be harvested and processed with the available agronomic infrastructures. The effective use of plants as bioreactors depends on the possibility of obtaining high protein accumulation levels that are stable during the life cycle of the transgenic plant and in subsequent generations. Silencing of the introduced transgenes has frequently been observed in plants, constituting a major commercial risk and hampering the general economic exploitation of plants as protein factories. Until now, the most efficient strategy to avoid transgene silencing involves careful design of the transgene construct and thorough analysis of transformants at the molecular level. Here, we focus on different aspects of the generation of transgenic plants intended for protein production and on their influence on the stability of heterologous gene expression.

Paramutation in maize

Paramutation is a heritable change in gene expression induced by allele interactions. This review summarizes key experiments on three maize loci, which undergo paramutation. Similarities and differences between the phenomenology at the three loci are described. In spite of many differences with respect to the stability of the reduced expression states at each locus or whether paramutation correlates with DNA methylation and repeated sequences within the loci, recent experiments are consistent with a common mechanism underlying paramutation at all three loci. Most strikingly, trans-acting mutants have been isolated that prevent paramutation at all three loci and lead to the activation of silenced Mutator transposable elements. Models consistent with the hypothesis that paramutation involves heritable changes in chromatin structure are presented. Several potential roles for paramutation are discussed. These include localizing recombination to low-copy sequences within the genome, establishing and maintaining chromatin domain boundaries, and providing a mechanism for plants to transmit an environmentally influenced expression state to progeny.

RNA degradation and models for post-transcriptional gene-silencing

Post-transcriptional gene silencing (PTGS) is a form of stable but potentially reversible epigenetic modification, which frequently occurs in transgenic plants. The interaction in trans of genes with similar transcribed sequences results in sequence-specific degradation of RNAs derived from the genes involved. Highly expressed single-copy loci, transcribed inverted repeats, and poorly transcribed complex loci can act as sources of signals that trigger PTGS. In some cases, mobile, sequence-specific silencing signals can move from cell to cell or even over long distances in the plant. Several current models hold that silencing signals are 'aberrant' RNAs (aRNA), which differ in some way from normal mRNAs. The most likely candidates are small antisense RNAs (asRNA) and double-stranded RNAs (dsRNA). Direct evidence that these or other aRNAs found in silent tissues can induce PTGS is still lacking. Most current models assume that silencing signals interact with target RNAs in a sequence-specific fashion. This results in degradation, usually in the cytoplasm, by exonucleolytic as well as endonucleolytic pathways, which are not necessarily PTGS-specific. Biochemical-switch models hold that the silent state is maintained by a positive auto-regulatory loop. One possibility is that concentrations of hypothetical silencing signals above a critical threshold trigger their own production by self-replication, by degradation of target RNAs, or by a combination of both mechanisms. These models can account for the stability, reversibility and multiplicity of silent states; the strong influence of transcription rate of target genes on the incidence and stability of silencing, and the amplification and systemic propagation of motile silencing signals.

Plants as bioreactors for protein production: avoiding the problem of transgene silencing

Plants are particularly attractive as large-scale production systems for proteins intended for therapeutical or industrial applications: they can be grown easily and inexpensively in large quantities that can be harvested and processed with the available agronomic infrastructures. The effective use of plants as bioreactors depends on the possibility of obtaining high protein accumulation levels that are stable during the life cycle of the transgenic plant and in subsequent generations. Silencing of the introduced transgenes has frequently been observed in plants, constituting a major commercial risk and hampering the general economic exploitation of plants as protein factories. Until now, the most efficient strategy to avoid transgene silencing involves careful design of the transgene construct and thorough analysis of transformants at the molecular level. Here, we focus on different aspects of the generation of transgenic plants intended for protein production and on their influence on the stability of heterologous gene expression.

(TRANS)GENE SILENCING IN PLANTS: How Many Mechanisms?

Epigenetic silencing of transgenes and endogenous genes can occur at the transcriptional level (TGS) or at the posttranscriptional level (PTGS). Because they can be induced by transgenes and viruses, TGS and PTGS probably reflect alternative (although not exclusive) responses to two important stress factors that the plant's genome has to face: the stable integration of additional DNA into chromosomes and the extrachromosomal replication of a viral genome. TGS, which results from the impairment of transcription initiation through methylation and/or chromatin condensation, could derive from the mechanisms by which transposed copies of mobile elements and T-DNA insertions are tamed. PTGS, which results from the degradation of mRNA when aberrant sense, antisense, or double-stranded forms of RNA are produced, could derive from the process of recovery by which cells eliminate pathogens (RNA viruses) or their undesirable products (RNA encoded by DNA viruses). Mechanisms involving DNA-DNA, DNA-RNA, or RNA-RNA interactions are discussed to explain the various pathways for triggering (trans)gene silencing in plants.

Virus recovery is induced in Brome mosaic virus p2 transgenic plants showing synchronous complementation and RNA-2-specific silencing

Nicotiana benthamiana plants expressing Brome mosaic virus (BMV) p2 protein complemented replication of RNAs1 + 3 but, surprisingly, supported little or no replication of RNA-2. Despite this, the p2 transgenic plants were able to support systemic migration of RNAs-1 and -3. Kinetic analyses showed identical degradation rates for RNAs-2 and -3, greatly detracting from the concept of an induction of an RNA-2-specific degradation system. Deletion analysis identified a 200-nucleotide sequence that may contribute to silencing in a context-specific manner. When R1 progeny of a severely silencing p2 transgenic line were tested for virus resistance, three different classes of reactions were observed. In class 1 and class 3 plants, the virus moved systemically and showed various extents of RNA-2 silencing. However, in class 2 plants, there was a stochastic onset of post-transcriptional silencing in the systemic leaves that was reminiscent of virus recovery. Plants showing recovery tended to have a greater number of transgene loci than did those exhibiting component-specific silencing. The induction of silencing did not appear to be dependent solely on the combined steady state levels of the transgene and viral RNA. Some plants transformed with a p2 frameshift construct showed a complete silencing phenotype, but none showed RNA-2-specific silencing. While the relationship between the two types of silencing remains unclear, we speculate that our observations reflect early events in the induction of virus recovery.

Sequences throughout the basic beta-1,3-glucanase mRNA coding region are targets for homology dependent post-transcriptional gene silencing

In the transgenic tobacco line T17, plants homozygous for the gn1 transgene display developmentally regulated post-transcriptional silencing of basic beta-1,3-glucanase genes. Previously, it has been shown that silencing involves a markedly increased turnover of silencing-target glucanase mRNAs. Using a two-component viral reporter system facilitated a comparison, in a quantitat- ive manner, of the relative silencing efficiencies of various sequences derived from the gn1 transgene. The results show that target sites for the silencing mechanism are present throughout the coding region of the gn1 mRNA. Similar-sized coding region sequences along the entire gn1 mRNA display a similar susceptibility to the silencing mechanism. The susceptibility to silencing increases as the coding region elements increase in size. Relative to internal sequences, the 5' and 3' terminal regions of the gn1 mRNA are inefficient targets for the silencing machinery. Importantly, sequences of the gn1 transgene that are not part of the mature gn1 mRNA are not recognized by the silencing machinery when expressed in chimeric viral RNAs. These results show that the glucanase silencing mechanism in T17 plants is primarily directed against gn1 mRNA-internal sequences and that terminal sequences of the gn1 mRNA are relatively unaffected by the silencing mechanism.

Posttranscriptional gene silencing of gn1 in tobacco triggers accumulation of truncated gn1-derived RNA species

Posttranscriptional silencing of basic beta-1,3-glucanase genes in the tobacco line T17 is manifested by reduced transcript levels of the gn1 transgene and homologous, endogenous basic beta-1,3-glucanase genes. An RNA ligation-mediated rapid amplification of cDNA ends (RLM-RACE) technique was used to compare the 3' termini of gn1 RNAs present in expressing (hemizygous and young homozygous) and silenced (mature homozygous) T17 plants. Full-length, polyadenylated gn1 transcripts primarily accumulated in expressing plants, whereas in silenced T17 plants, mainly 3'-truncated, nonpolyadenylated gn1 RNAs were detected. The relative abundance of these 3'-truncated gn1 RNA species gradually increased during the establishment of silencing in homozygous T17 plants. Similar 3'-truncated, nonpolyadenylated gn1 RNA products were observed in an independent case of beta-1,3-glucanase posttranscriptional gene silencing. This suggests that these 3'-truncated gn1 RNAs are a general feature of tobacco plants showing posttranscriptional silencing of the gn1 transgene.

RNA-triggered gene silencing

Double-stranded RNA (dsRNA) has recently been shown to trigger sequence-specific gene silencing in a wide variety of organisms, including nematodes, plants, trypanosomes, fruit flies and planaria; meanwhile an as yet uncharacterized RNA trigger has been shown to induce DNA methylation in several different plant systems. In addition to providing a surprisingly effective set of tools to interfere selectively with gene function, these observations are spurring new inquiries to understand RNA-triggered genetic-control mechanisms and their biological roles.

Expression and sequence requirements for nitrite reductase co-suppression

We have previously reported that the introduction of a full-length tobacco nitrite reductase Nii1 cDNA under the control of the 35S promoter triggers co-suppression of endogenous Nii genes in 25% of tobacco transformants. Here we show that introduction of chimeric Nii1-uidA, uidA-Nii1 and Nii1-uidA-Nii1 transgenes carrying 186 bp of the 5' end and/or 241 bp of the 3' end of the Nii1 cDNA do not trigger co-suppression of endogenous Nii genes. In addition, we show that when introduced by crossing or transformation into co-suppressed transgenic tobacco lines carrying full-length Nii1 transgenes, these chimeric transgenes are not silenced. These results therefore suggest that the 5' and 3' ends of the Nii1 cDNA are not sufficient to trigger co-suppression and are not targets for homology-dependent RNA degradation. Surprisingly, co-suppression was released in a double transformant obtained by introduction of one of these constructs into the co-suppressed transgenic tobacco line 461-2.1 homozygous for a full-length Nii1 transgene, and in one plant regenerated from untransformed leaf discs (plant 461-2.1*). The reappearance of co-suppression at very low frequency (less than 10(-3)) in the F2 progeny of plant 461-2.1* and the apparent absence of structural modification of the transgene locus suggest a metastable epigenetic modification. The steady-state level of Nii mRNAs in the plant 461-2-.1* was higher than in wild-type plants but lower than in hemizygous plants 461-2.1 which never trigger silencing. These results therefore confirm that transcription of the transgene above a particular threshold is required to trigger co-suppression.

CUT1, an Arabidopsis gene required for cuticular wax biosynthesis and pollen fertility, encodes a very-long-chain fatty acid condensing enzyme

Land plants secrete a layer of wax onto their aerial surfaces that is essential for survival in a terrestrial environment. This wax is composed of long-chain, aliphatic hydrocarbons derived from very-long-chain fatty acids (VLCFAs). Using the Arabidopsis expressed sequence tag database, we have identified a gene, designated CUT1, that encodes a VLCFA condensing enzyme required for cuticular wax production. Sense suppression of CUT1 in transgenic Arabidopsis plants results in waxless (eceriferum) stems and siliques as well as conditional male sterility. Scanning electron microscopy revealed that this was a severe waxless phenotype, because stems of CUT1-suppressed plants were completely devoid of wax crystals. Furthermore, chemical analyses of waxless plants demonstrated that the stem wax load was reduced to 6 to 7% of wild-type levels. This value is lower than that reported for any of the known eceriferum mutants. The severe waxless phenotype resulted from the downregulation of both the decarbonylation and acyl reduction wax biosynthetic pathways. This result indicates that CUT1 is involved in the production of VLCFA precursors used for the synthesis of all stem wax components in Arabidopsis. In CUT1-suppressed plants, the C24 chain-length wax components predominate, suggesting that CUT1 is required for elongation of C24 VLCFAs. The unique wax composition of CUT1-suppressed plants together with the fact that the location of CUT1 on the genetic map did not coincide with any of the known ECERIFERUM loci suggest that we have identified a novel gene involved in wax biosynthesis. CUT1 is currently the only known gene with a clearly established function in wax production.

Stochastic and nonstochastic post-transcriptional silencing of chitinase and beta-1,3-glucanase genes involves increased RNA turnover-possible role for ribosome-independent RNA degradation

Stochastic and nonstochastic post-transcriptional gene silencing (PTGS) in Nicotiana sylvestris plants carrying tobacco class I chitinase (CHN) and beta-1,3-glucanase transgenes differs in incidence, stability, and pattern of expression. Measurements with inhibitors of RNA synthesis (cordycepin, actinomycin D, and alpha-amanitin) showed that both forms of PTGS are associated with increased sequence-specific degradation of transcripts, suggesting that increased RNA turnover may be a general feature of PTGS. The protein synthesis inhibitors cycloheximide and verrucarin A did not inhibit degradation of CHN RNA targeted for PTGS, confirming that PTGS-related RNA degradation does not depend on ongoing protein synthesis. Because verrucarin A, unlike cycloheximide, dissociates mRNA from ribosomes, our results also suggest that ribosome-associated RNA degradation pathways may not be involved in CHN PTGS.

Isolation of an RNA-directed RNA polymerase-specific cDNA clone from tomato

A 3600-bp RNA-directed RNA polymerase (RdRP)-specific cDNA comprising an open reading frame (ORF) of 1114 amino acids was isolated from tomato. The putative protein encoded by this ORF does not share homology with any characterized proteins. Antibodies that were raised against synthetic peptides whose sequences have been deduced from the ORF were shown to specifically detect the 127-kD tomato RdRP protein. The immunoresponse to the antibodies correlated with the enzymatic activity profile of the RdRP after chromatography on Q-, poly(A)-, and poly(U)-Sepharose, hydroxyapatite, and Sephadex G-200 columns. DNA gel blot analysis revealed a single copy of the RdRP gene in tomato. RdRP homologs from petunia, Arabidopsis, tobacco, and wheat were identified by using polymerase chain reaction. A sequence comparison indicated that sequences homologous to RdRP are also present in the yeast Schizosaccharomyces pombe and in the nematode Caenorhabditis elegans. The previously described induction of RdRP activity upon viroid infection is shown to be correlated with an increased steady state level of the corresponding mRNA. The possible involvement of this heretofore functionally elusive plant RNA polymerase in homology-dependent gene silencing is discussed.

Isolation of an RNA-directed RNA polymerase-specific cDNA clone from tomato

A 3600-bp RNA-directed RNA polymerase (RdRP)-specific cDNA comprising an open reading frame (ORF) of 1114 amino acids was isolated from tomato. The putative protein encoded by this ORF does not share homology with any characterized proteins. Antibodies that were raised against synthetic peptides whose sequences have been deduced from the ORF were shown to specifically detect the 127-kD tomato RdRP protein. The immunoresponse to the antibodies correlated with the enzymatic activity profile of the RdRP after chromatography on Q-, poly(A)-, and poly(U)-Sepharose, hydroxyapatite, and Sephadex G-200 columns. DNA gel blot analysis revealed a single copy of the RdRP gene in tomato. RdRP homologs from petunia, Arabidopsis, tobacco, and wheat were identified by using polymerase chain reaction. A sequence comparison indicated that sequences homologous to RdRP are also present in the yeast Schizosaccharomyces pombe and in the nematode Caenorhabditis elegans. The previously described induction of RdRP activity upon viroid infection is shown to be correlated with an increased steady state level of the corresponding mRNA. The possible involvement of this heretofore functionally elusive plant RNA polymerase in homology-dependent gene silencing is discussed.

Silencing of beta-1,3-glucanase genes in tobacco correlates with an increased abundance of RNA degradation intermediates

Post-transcriptional gene silencing of beta-1,3 glucanase genes in the transgenic tobacco line T17 is characterised by an increased turnover and, as a consequence, reduced levels of gn1 transgene and endogenous beta-1,3 glucanase mRNAs. Here, additional gn1 RNAs, both larger and smaller than the full-length messenger, are shown to accumulate in silenced plants of the transgenic tobacco line T17. The longer-than-full-length gn1 RNAs are the result of cryptic processing of the gn1 messenger. The small gn1 RNAs in silenced plants correspond to distal and proximal parts of the mature gn1 messenger. The proximal RNA products are intact at their 5' extremity, but terminate at different positions at the 3'-end. The distal RNA products contain a poly(A) tail and are truncated to various positions at the 5'-end. These observations indicate that degradation of the mature gn1 transcript does not start at the 5'- or 3'-end, but rather are consistent with degradation of the gn1 transcript starting with an endonucleolytic cleavage followed by internal exonuclease digestion. Importantly, the truncated products are more abundant in silenced plants than in expressing plants. This suggests, together with the previously reported silencing-related increased gn1 mRNA turnover and the similar rates of gn1 transcription in silenced and expressing T17 plants, that the predominant decay route for the gn1 transcripts differs between silenced and expressing conditions.

Position-dependent methylation and transcriptional silencing of transgenes in inverted T-DNA repeats: implications for posttranscriptional silencing of homologous host genes in plants

Posttranscriptional silencing of chalcone synthase (Chs) genes in petunia transformants occurs by introducing T-DNAs that contain a promoter-driven or promoterless Chs transgene. With the constructs we used, silencing occurs only by T-DNA loci which are composed of two or more T-DNA copies that are arranged as inverted repeats (IRs). Since we are interested in the mechanism by which these IR loci induce silencing, we have analyzed different IR loci and nonsilencing single-copy (S) T-DNA loci with respect to the expression and methylation of the transgenes residing in these loci. We show that in an IR locus, the transgenes located proximal to the IR center are much more highly methylated than are the distal genes. A strong silencing locus composed of three inverted T-DNAs bearing promoterless Chs transgenes was methylated across the entire locus. The host Chs genes in untransformed plants were moderately methylated, and no change in methylation was detected when the genes were silenced. Run-on transcription assays showed that promoter-driven transgenes located proximal to the center of a particular IR are transcriptionally more repressed than are the distal genes of the same IR locus. Transcription of the promoterless Chs transgenes could not be detected. In the primary transformant, some of the IR loci were detected together with an unlinked S locus. We observed that the methylation and expression characteristics of the transgenes of these S loci were comparable to those of the partner IR loci, suggesting that there has been cross talk between the two types of loci. Despite the similar features, S loci are unable to induce silencing, indicating that the palindromic arrangement of the Chs transgenes in the IR loci is critical for silencing. Since transcriptionally silenced transgenes in IRs can trigger posttranscriptional silencing of the host genes, our data are most consistent with a model of silencing in which the transgenes physically interact with the homologous host gene(s). The interaction may alter epigenetic features other than methylation, thereby impairing the regular production of mRNA.

Irregular patterns of transgene silencing in allohexaploid oat

An irregular pattern of transgene silencing was revealed in expression and inheritance studies conducted over multiple generations following transgene introduction by microprojectile bombardment of allohexaploid cultivated oat (Avena sativa L.). Expression of two transgenes, bar and uidA, delivered on the same plasmid was investigated in 23 transgenic oat lines. Twenty-one transgenic lines, each derived from an independently selected transformed tissue culture, showed expression of both bar and uidA while two lines expressed only bar. The relationship of the transgenic phenotypes to the presence of the transgenes in the study was determined using (1) phenotypic scoring combined with Southern blot analyses of progeny, (2) coexpression of the two transgenic phenotypes since the two transgenes always cosegregated, and (3) reactivation of a transgenic phenotype in self-pollinated progenies of transgenic plants that did not exhibit a transgenic phenotype. Transgene silencing was observed in 19 of the 23 transgenic lines and resulted in distorted segregation of transgenic phenotypes in 10 lines. Silencing and inheritance distortions were irregular and unpredictable. They were often reversible in a subsequent generation of self-pollinated progeny and abnormally segregating progenies were as likely to trace back to parents that exhibited normal segregation in a previous generation as to parents showing segregation distortions. Possible causes of the irregular patterns of transgene silencing are discussed.

A transgene with repeated DNA causes high frequency, post-transcriptional suppression of ACC-oxidase gene expression in tomato

Gene silencing with sense genes is an important method for down-regulating the expression of endogenous plant genes, but the frequency of silencing is unpredictable. Fifteen per cent of tomato plants transformed with a 35S-ACC-oxidase ( ACO 1) sense gene had reduced ACC-oxidase activity. However, 96% of plants transformed with an ACC-oxidase sense gene, containing two additional upstream inverted copies of its 5' untranslated region, exhibited reduced ACC-oxidase activity compared to wild-type plants. In the three plants chosen for analysis, there were substantially reduced amounts of both endogenous and transgenic ACO RNA, indicating that this was an example of co-suppression. Ribonuclease protection assays using probes spanning intron-exon borders showed that the reduced accumulation of endogenous ACO mRNA occurred post-transcriptionally since the abundance of unprocessed transcripts was not affected. The ACO1 transgene with the repeated 5'UTR also strongly inhibited the accumulation of RNA from the related ACO 2 gene in flowers, although there is little homology between the 5'UTRs of ACO 1 and ACO 2. These results indicate that although repeated DNA in a transgene greatly enhances the probability of gene silencing of an endogenous gene, it also involves generation of a trans -acting silencing signal produced, at least partly, from sequences external to the repeat.

Transgenes are dispensable for the RNA degradation step of cosuppression

Cosuppression results in the degradation of RNA from host genes and homologous transgenes after transcription in the nucleus. By using grafting experiments, we have shown previously that a systemic signal mediates the propagation of cosuppression of Nia host genes and 35S-Nia2 transgenes from silenced 35S-Nia2 transgenic stocks to nonsilenced 35S-Nia2 transgenic scions but not to wild-type scions. Here, we examined the requirements for triggering and maintenance of cosuppression in various types of scions. Grafting-induced silencing occurred in 35S-Nia2 transgenic lines over-accumulating Nia mRNA whether they are able to spontaneously trigger cosuppression or not and in 35S-Nia2 transgene-free plants over-accumulating host Nia mRNA caused by metabolic derepression. When grafting-induced silenced scions were removed from the silenced stocks and regrafted onto wild-type plants, silencing was not maintained in the 35S-Nia2 transgene-free plants and in the 35S-Nia2 transgenic lines that are not able to trigger cosuppression spontaneously. Conversely, silencing was maintained in the 35S-Nia2 transgenic lines that are able to trigger cosuppression spontaneously. Our results indicate that the presence of a 35S-Nia2 transgene is dispensable for the RNA degradation step of posttranscriptional silencing when host Nia mRNA over-accumulate above the level of wild-type plants. They also suggest that grafting-induced RNA degradation does not result in the production of the systemic silencing signal required for spontaneous triggering and maintenance.

RNA-mediated virus resistance in transgenic plants

In recent years the concept of pathogen-derived resistance (PDR) has been successfully exploited for conferring resistance against viruses in many crop plants. Starting with coat protein-mediated resistance, the range has been broadened to the use of other viral genes as a source of PDR. However, in the course of the efforts, often no clear correlation could be made between expression levels of the transgenes and observed virus resistance levels. Several reports mentioned high resistance levels using genes incapable of producing protein, but in these cases, even plants accumulating high amounts of transgene RNA were not most resistant. To accommodate these unexplained observations, a resistance mechanism involving specific breakdown of viral RNAs has been proposed. Recent progress towards understanding the RNA-mediated resistance mechanism and similarities with the co-suppression phenomenon will be discussed.