Plant Viruses As Gene Transfer Vectors

Plant viruses do not integrate into the genome and therefore have little value as integrative transformation vectors. They also do not pass through the gametes, so their effects are not heritable through meiosis, although they can be transmitted to new stock by grafting. However, they have a number of attractive properties that suggest they could be developed as episomal gene transfer vectors. First, they naturally infect intact plant cells including those of species that have, so far, resisted other transformation methods. Viruses exist that infect all the commercially important crop plants, and in many cases even the naked nucleic acid is infectious if applied to, e.g., the leaf surface. Gene transfer by viral transduction is therefore much easier than Agrobacterium-mediated transformation, much simpler than protoplast transformation, and much cheaper than particle bombardment. Second, because they multiply to high copy numbers in the plant cell, viruses can facilitate very high-level transgene expression. Third, many plant viruses cause systemic infections, i.e., they spread to every cell in the host plant, therefore obviating the need to regenerate transgenic plants from transformed cells. These properties have led to the development of a number of RNA virus-based expression vectors for recombinant protein synthesis, including those based on tobacco mosaic virus (74), potato virus X (75), and cowpea mosaic virus (76).

Plant viruses can also be used as a sensitive assay to detect T-DNA transfer to the plant nucleus. Since viruses replicate to such high copy numbers, viral infection symptoms can be used to confirm the presence of T-DNA in single cells of a given explant. This procedure, termed agroinfection, involves the transfer of a viral genome into the plant cell between T-DNA border repeats and was first used to confirm the genetic transformation of maize (77).


So far, only transformation of the nuclear genome has been considered. It can be desirable, however, to introduce DNA into organelle genomes as this increases the number of transgene copies in the cell and facilitates highlevel transgene expression (78). The first reports of organelle transformation were serendipitous. For example, de Block et al. (79) reported the transformation of tobacco chloroplast DNA using Agrobacterium-mediated transformation of tobacco protoplasts. The rare organellar transformation event was discovered because of the maternal transmission of the transgene, confirmed by Southern blot analysis of chloroplast DNA. The targeted transformation of chloroplasts by particle bombardment was achieved in 1988 by Boynton and colleagues (80) using the unicellular green alga Chlamy-domonas rheinhardtii. The same technique was used by Svab and Maliga (81) to transform tobacco chloroplast DNA. In each case, the particles penetrated the double membrane of the chloroplast so that a biological mechanism to pass through the membrane was unnecessary. More recently, PEG-mediated transformation of protoplasts has been used to introduce DNA into the tobacco chloroplast genome (82) and the two methods have been compared (83). Generally, only one of the many organelles in the cell is transformed, which means that homoplastomic cells must be obtained by intense selection. In most reports, the aad selectable marker gene has been used (Table 2). A recent strategy to track chloroplast transformation involves the fusion of this selectable marker to the gene for green fluorescent protein (84).


A. Overview

While the major goal of plant transformation technology over the last two decades has been to introduce and stably express transgenes in plants, the challenge for the next decade is to refine these techniques and introduce single transgene copies at defined sites without extraneous DNA sequences such as parts of the vector backbone or marker genes. The ultimate aim would be to produce transgenic plants with the transgene integrated at a known site, allowing the experimenter full and predictable control over transgene expression.

B. Clean DNA Transformation

A major disadvantage of all direct DNA transfer transformation methods is the tendency for exogenous DNA to undergo rearrangement and recombination events leading to the integration of multiple fragmented, chimeric, and rearranged transgene copies. Such clusters are prone to internal recombination and may therefore be unstable, as well as promoting transgene silencing. Kohli et al. (47) have shown that such rearrangements may be caused, in part, by recombinogenic elements in the vector backbone of the transformation plasmid. The backbone also contains bacterial markers and other sequences, such as the origin or replication, which are unnecessary for transformation.

A major breakthrough has been achieved in the last year in the production of transgenic plants without integrated vector backbone sequences. Fu et al. (85) describe a clean DNA technique involving the use of linear minimal transgene cassettes for particle bombardment. These cassettes lack all vector backbone elements but contain the DNA sequences essential for transgene expression: the promoter, open reading frame, and terminator. Transgenic rice plants regenerated from callus bombarded with such cassettes showed very simple integration patterns, with one or a few hybridizing bands on Southern blots. Comparative transformation experiments using whole plasmid DNA resulted in much more complex integration patterns, with multiple bands of different sizes (Fig. 2). Furthermore, release of a diagnostic fragment from the transgene showed that fewer rearrangement events had occurred in the plants carrying the linear cassettes. The progeny of these plants were examined for transgene expression, and there were no reported instances of transgene silencing. Conversely, whole plasmid transformation by particle bombardment routinely leads to silencing in about 20% of the resulting transgenic plants. This work also showed that linear cassettes promoted the same efficiency of cotransformation as normal plasmids, so the clean DNA strategy is a promising technique for the generation of transgenic plants carrying multiple transgenes.

Agrobacterium- mcdiated transformation was for a long time regarded as a naturally clean system, in which one to three intact T-DNA copies integrated into the genome, and only the DNA between the T-DNA border repeats was transferred. More recent data show, however, that DNA outside the border repeats is cotransferred to the plant genome at a very high frequency (>75% of transformants) (86). In many such cases, the backbone is

Plasmid DNA

Clean fragment



Bombardment of callus

Southern blot

Bombardment of callus

Southern blot

Pinsmid ON A: Complex Intagratlon patterns Some lots of expression stability

Cfun fragment: Simple Integration patterns Stable tnnsgene expression

Pinsmid ON A: Complex Intagratlon patterns Some lots of expression stability

Cfun fragment: Simple Integration patterns Stable tnnsgene expression

Figure 2 The clean DNA transformation system compared to particle bombardment with whole plasmid DNA. The transformation strategy is shown, and two representative Southern blots are compared to demonstrate the simpler integration patterns resulting from transformation with linear minimal cassettes. (Data from Ref. 85.)

linked to the T-DNA, suggesting that the VirD2 protein can skip over one T-DNA border, whereas in other cases the backbone segments are not linked to the T-DNA, suggesting an adventitious integration mechanism (87). Furthermore, T-DNA often integrates as multiple copies, predominantly inverted repeat structures, which are highly correlated with transgene silencing phenomena (88). Multimeric integration also occurs when two different T-DNAs are transferred to the same plant cell by different Agrobacterium strains, suggesting that multimer formation occurs before integration (89). This study showed a preference for linkage involving the right border repeat, although all other possible linkage combinations occurred at a lower frequency (89). Complex T-DNA integration patterns are more prevalent with hypervirulent Agrobacterium strains and/or superbinary vectors.

To address such problems, Chilton and colleagues have pioneered the technique of agrolistic transformation (90). Particle bombardment was used to introduce three plasmids into tobacco and maize cells, one containing a recombinant T-DNA and the others containing the virDl and virD2 genes responsible for T-DNA mobilization. The resulting plants contain single integrated T-DNA sequences at high frequency, indicating that the Vir proteins could function correctly, although T-DNA sequences were also detected as part of the entire plasmid integrated into the plant genome. So, although this technique is useful for the generation of clean integration events, it does not eliminate backbone sequences completely. A possible strategy to limit the production of vector-containing plants is to flank the T-DNA with "killer genes" that confer a lethal phenotype on the transformant (91). Any plants carrying vector backbone sequences would be killed, provided the lethal genes were not disrupted or mutated upon integration and that segments of the vector backbone distal to the lethal genes did not integrate. A more refined technique devised by Gelvin (91) is to produce single-strand T-DNA in vitro using highly purified recombinant VirD2 protein. This development, analogous to the clean DNA strategy because backbone sequences are never introduced into the plant, may allow genuinely clean T-DNA insertions. However, the T-strands must be introduced into protoplasts by electropora-tion; thus, the procedure is conceptually no different from direct DNA transfer.

C. Toward Marker-Free Transgenic Plants

There is some debate in the public domain about the potential dangers of selectable marker genes in transgenic plants. The permanent integration and expression of a selectable marker also prevents the same plant line from being retransformed with the same marker (i.e., if it was necessary to introduce further transgenes into an existing transgenic plant line). This issue has been tackled in a number of ways: (1) the use of genetic systems to eliminate marker genes from transgenic plants, (2) the development of strategies for marker gene segregation in breeding lines, and (3) the development of innocuous markers that do not confer antibiotic or herbicide resistance.

The first genetic system for marker elimination was reported by Golds-borough et al. (92) and involved the incorporation of transgenes within a maize Ac or Ds transposon, cloned within the T-DNA sequence. Two marker genes were used, and plants could be recovered that lacked one or other of the markers. In a subsequent system, the Agrobacterium isopentenyl transferase (ipt) gene was cloned in an Ac element within the T-DNA borders, while a second reporter gene carried in the T-DNA was outside the transposon (93). The ipt gene affects cytokinin metabolism so transformed plants showed a shooty phenotype. Ac elements often jump from site to site in the genome, but in about 15% of cases there is excision without subsequent reintegration. In these cases, the marker was removed and normal plant growth resumed as long as further copies of the T-DNA were not present. The excision events left the remainder of the T-DNA intact and did not affect expression of the linked reporter gene. This system is advantageous for a number of reasons: (1) the marker is innocuous, (2) the marker can be eliminated, and (3) there is automatic selection for single-copy insertions of the T-DNA.

In sexual populations, one method that could produce marker-free transgenic plants is to use cotransformation to introduce the nonselected gene and selectable marker at different loci and screen progeny in the R1 generation for segregants lacking the selectable marker gene. In direct DNA transfer cotransformation, there are occasional dispersive integration events resulting in transgene segregation and aberrant segregation ratios, although cointegration at a single locus predominates. The majority of T-DNA co-transformation events also result in cointegration, so that only a low frequency of marker-free plants is generated (13). The use of superbinary vector systems with two T-DNA plasmids may facilitate marker segregation, and segregation has been shown in 50% of the progeny of rapeseed and tobacco plants produced by Agrobacterium-mediated cotransformation with two binary plasmids (94).

The preceding strategy cannot be used for vegetatively propagated plants. An alternative strategy utilizes the bacteriophage site-specific recom-binase system Cre-loxP. Tobacco plants were transformed with two marker genes, each flanked by loxP recombination target sites (95). The plants were then retransformed with the ere gene, and expression of Cre recombinase resulted in recombination between the loxP sites and excision of the original markers. One problem with this system is that transformation with the cre gene must itself involve an additional selectable marker. This requirement was circumvented by the transient expression of Cre, resulting in the production of a few marker-free plants.

D. Toward Precise Integration and Control of Expression

Another way in which Cre-loxP, and other site-specific recombination systems, can be exploited is to target transgenes to precise loci in the plant genome. Even in the absence of multiple and rearranged transgene copies, there is still much variation in transgene expression levels, and this has been attributed to so-called position effects reflecting the nature of the local molecular environment at the site of integration (e.g., the presence of regulatory elements, repetitive DNA, chromatin structure, and DNA methylation). In animal cells, a number of loci have been identified that are favorable to transgene expression because transgenes integrated there tend to be expressed faithfully and strongly. The controlled integration of DNA at such sites can be facilitated by flanking the transgene with loxP sites and placing a loxP site at the desired target site. A similar approach has been attempted in plants (96), but it appears that transgenes integrated at sites on several different chromosomes still display variable levels of expression within lines of identical transformants, suggesting that other factors are responsible for the variation (91). This agrees with data showing variable levels of transgene expression and different modes of transgene silencing involving different methylation patterns in a line of rice plants carrying the same single-copy, three-gene transgenic locus (97).

A different strategy to avoid position effects is to isolate the transgene from the surrounding molecular environment by flanking it with matrix attachment regions (MARs) that attach it to the nuclear matrix and isolate the transgene into an independent chromatin domain (98). The variability in transgene expression was reduced when reporter genes such as gusA were flanked by MARs, although as with the Cre-loxP system, there was still some variability that could be attributed to other factors (98).

Was this article helpful?

0 0
Body Detox Made Easy

Body Detox Made Easy

What exactly is a detox routine? Basically a detox routine is an all-natural method of cleansing yourbr body by giving it the time and conditions it needs to rebuild and heal from the damages of daily life and the foods you eat and other substances you intake. There are many different types of known detox routines.

Get My Free Ebook

Post a comment