You are here: Agriculture > Transgene Plants > Forum 1995 > New pathogenes from GMO?

Virus resistant crops - new viral pathogens from transgenic plants

Table of Contents:

• Summary
• Introduction
• Recombination
• Recombination in CPMP
• Resistance mechanisms
• Heterologous encapsidation

Author: Pia Malnoë

Summary

Tree diffierent events, RNA recombination, heterologous encapsidation and sequence variability have been identified as potential risks when transgenic virus resistant plants are used in an agrosystem. Recombination is considered as the most serious problem because a permanent change is introduced into the viral genome. However, it is likely that recombination between a viral transgenic mRNA and a viral genomic RNA is less frequent than recombination between two different viral genomic RNA molecules co?infecting an untransformed plant. In this case no increased risk factor is introduced into the fields. However, to diminish the risk of recombination, only minimal sequence necessary to induce resistance should be used and one should avoid introducing into the plant the 3' untranslated region of the viral genome. Usually, heterologous encapsidation is considered as a less sever problem because it is limited to a single transfer. In order to avoid heterologous encapsidation a biological containment system is recommended where the amino acids responsible for aphid transmission and virion assembly are either mutated or deleted.

Introduction

In 1986 it was shown for the first time that, by expressing a viral coat protein (CP) gene in a plant cell, it was possible to increase the resistance of the plant against the infection of the homologous virus [1]. Since then it has been possible to obtain a good coat protein mediated protection (CPMT) to at least 50 diffierent viruses belonging to 12 virus groups in 12 different plant species [2]. Commonly, the level of protection is proportional to the degree of relatedness between the infecting virus and the source of the transgene. CPMP is usually effective against different isolates or closely related strains. Lately it has been shown that non structural viral genes, like the viral replicase genes, also induce resistance.

The level of resistance obtained using a transgenic approach is comparable to that conferred by host resistance genes. The possibility to induce virus resistance into a susceptible variety without afflecting the intrinsic properties of the cultivar is of high agronomic value. This year, a transgenic squash resistant to zuchini yellow mosaic virus (ZYMV) has been commercialized in the USA. Several other CPMP resistant plants are expected on the market in the near future. Hence, it is now more and more urgent to consider the eventual risks the use of transgenic plants expressing a viral sequence may cause to our eco? and agrosystem. The principal question one need to ask is "may the presence of such plants create new viruses with altered host andlor vector ranges or may the transmission rate and symptom induction be altered?" These effects could be the results of recombination, heterologous encapsidation or sequence variations. AD these events occur naturally in untransformed plants but the frequency of the events may be different in a transgenic plant.

The genome of more than 70% of the plant viruses consists of one or several RNA molecules which in most cases can serve as a niRNA (+sens RNA). Transmission of the virus from one plantto another requires mechanical inoculation or vectors such as insects, nematodes or fungi. Once the virus has entered the host cell it is decapsidated. The replication and movement of the virus within the plant is depending on the presence of viral and host factors.

RNA viruses have highly variable genomes since the replication of the genome is carried out by a RNA dependent RNA polymerase lacking error reading functions. The error rate ranges within the limits of 10-3 to 10-5 which correspond to one mutation per 10 kilobases [3]. Hence a virus strain consists of a population of closely related RNA genomes. There is a strong selection on these populations depending on the environment they have to adapt to. Most mutations will be deleterious and only the ones given the highest fitness in a specific host and environment will be conserved [4]. Sequence variability and recombination between the viral genomes are considered as the main tools for the evolution of a viral population.

Recombination

Recombination in plant viruses is less well understood than in animal viruses [5]. It seems that recombination occurs in both systems by template switching between either homologous or non-homologous regions [6]. Sequence and structural similarities as well as subcellular concentrations and locations are important factors for the recombination events [7].

Natural recombination between viral genomes has been described very rarely [8, 9]. Lately it has been possible to study recombination in planta during an infection. The data from these experiments indicates that RNA recombination takes place and functional chimeric genomes can be generated through this process. The example of abutflon mosaic virus is interesting. In this case the viral genome contained dispersed point mutations in different essential genes. Mutated genomes were able to complement each other so that the virus population could be propagated at a low frequency. At a later stage a single functional genome was obtained through recombination and the mutated forms were lost [10]. There is several others example where selection has been used to generate new viral particles through both homologous and nonhomologous recombination [11, 12, 13, 14]. Without selection it is almost impossible to detect a recombinant.

Potato with Y virus

We have compared the nucleotide sequences of several different strains of potato virus Y (PVY) [15]. This study shows that the evolution of the strains is relying not only on sequence variability but also on recombination. Recently a new highly virulent strain called PVY?NTN has emmerged in different parts of Europe. This strain causes necroses on tubers resulting in important economical losses for the farmers. The complet sequence of a Hungarian isolate of the PVY?NTN strain has been puplished [16] and reveals a combination of PVY?N and O sequences.

Recombination in CPMP

Recombination between a transgenic viral mRNA and an infecting viral genomic RNA is considered as being the most prominent problem because the changes resulting from a recombination event are permanent. It is important to establish whether the frequency of recombination is different during the infection of a transgenic plant compared to the double infection of an untransformed plant. The following aspects may be considered:

1. Is the overall rate of recombination changed because of an increased opportunity?
2. Is the rate proportional to the concentration of the two parental molecules?
3. Is the site of synthesis important?
4. Can the recombinant compete with the parental virus?

• a better knowledge of resistance mechanisms in the transgenic plants,
• a well defined system to follow specific recombination events and
• a biological assay to be able to determine not only the recombination frequency but also the fitness of a recombined virus, with and without selection.

Resistance mechanisms

The mechanisms involved in coat protein mediated virus resistance can be grouped into two main classes. In one class, the resistance depends on the synthesis of viral coat protein in the plant. The efficiency of protection is generally correlated with the abundance of the said protein. An example of this type of resistance is illustrated by TMV [17]. The authors have shown that an efficient resistance to TMV is obtained if the CP accumulates in the tissue which was initially infected. The virion disassembly seems to be affected by the presence of the transgenic CP and the decapsidation of the infecting virion might be inhibited [18]. The transgenic CP may also interfere with the long distance transport of the virus particles.

In the second class, described as RNA-mediated resistance, the situation is reversed. The degree of resistance is negatively correlated with the amount of transgenic mRNA and coat protein present in the cell. This is the case for the viruses belonging to the potyvirus group. RNA-mediated virus resistance cannot be explained easily using our present knowledge, but it shares common features with a phenomenon observed in transgenic plants referred to as co-suppression or silencing of gene expression.

The concentration of the transgenic viral mRNA in the cell is different depending on the type of resistance. In the case of TMV resistance, the concentration of the CP mRNA is relatively high but still at least ten times inferior to the concentration of viral genomic RNA during an infection. In a transgenic plant resistant to PVY, there is very little transgenic mRNA present in the cytoplasm. However, if for any reason the resistance is lost, the amount of CP mRNA may increase in the cell.

The replication of most plant RNA viruses takes place in the cytoplasm and often in specialized compartments within the cell. These compartments are formed by viral proteins as illustrated by the formation of the cytoplasmic inclusion bodies during a PVY infection. Such kind of compartmentalization could prevent the cytoplasmic mRNA from entering into direct contact with the genomic viral RNA and at the same time prevent recombination.

RNA recombination is supposed to take place by template switching during RNA replication [6, 22]. There is a possibility that the replicase complex of an infecting virus is able to recognize the 3' untranslated region in a CP RNA molecule and use it as a template for RNA synthesis. If the template switching occurs from the mRNA to the viral genomic RNA strand, a full length hybrid RNA could be created. Such a situation can be avoided by deleting the Yuntranslated region from the CP construct introduced into the plant genorne. If template switching occurs from the genon-fic RNA to the viral mRNA a non infectious molecule will be obtained. To obtained a full length infectious recombined RNA molecule a second recombination event is required.

This data indicates that it is likely that recombination between a viral transgenic mRNA and a viral genomic RNA is less frequent than recombination between two different viral genomic RNA molecules infecting an untransformed plant. However, proving it experimentally will require a very well defined system with an infectious cDNA copy of the viral genome.

Heterologous encapsidation

When two viruses infect the same plant, different kinds of interactions can take place [23]. For example,

• virus A may suppress the expression of virus B. this is referred to as cross protection.
• Virus A may enhance the expression of virus B or
• virus A and B may act synergetically.
• In some cases the genome of virus A may be encapsidated by the coat protein of virus B.

Heterologous encapsidation has been studied under field conditions for PLRV [27] and two potyviruses, ZYMV and PRSV, [28]. The efficiency of heterologous encapsidation is variable, depending on the viruses involved. Phenotypic mixing occurs mainly between closely related strains. Transcapsidation occurs between less related strain and even between different viruses although not necessarily in both directions. This has been well documented for barley yellow darf virus (BYDV) [29].

The risks associated with heterologous encapsidation in CP resistant plants have been discussed in detail by Tepfer [30] and Palukaitis [31]. In infected transgenic plants the situation is not quite the same as in plants with double infection. In the latter case the CP of virus A and B as well as their genornic RNA will be present in the cell and there will be a preference for a homologous encapsidation. In the CPW plant the transgenic coat protein will be present but not the corresponding genomic RNA, which means that there will be an excess of free transgenic CP. Hence, it is possible that the frequency of transcapsidation is higher in an infected transgenic plant than in a plant infected by two viruses.

During our study using a transgenic potato line resistant to PVY-N, we were able to show that a phenotypic mixing took place when the plants were infected with a related PVY-O strain [32]. Lecoq and coworkers [33] have also been able to show that a non aphid transmissible strain of the potyvirus ZYMV could be transmitted by aphids after having infected a transgenic plant expressing the coat protein of PLRV. A third example is the transencapsidation of CMV by the coat protein of AIMV [34]. The two last examples demonstrate a significant rate of heterologous encapsidation between two unrelated viruses in transgenic plants.

Consequently, there is no doubts that heterologous encapsidation occurs in untransformed and transformed plants but the main question is "what are the potential risks of heterologous encapsidation in CP resistant plants". The general opinion does not consider heterologous encapsidation as a problem because the phenomenon is limited to a single transfer [2]. The transcapsidated virus becomes defective with regards to the new host and should not be able to propagate without a helper virus. However, the fact that transcapsidation in transgenic plants may contribute to the introduction of a new virus into a new ecological niche can not be excluded [33].

For safety reasons it is recommend

1. not to express a coat protein in a plant that is not its natural host.
2. to create a biological containment system. For example, in the potyvirus group it has been shown that an amino acid triplet (DAG) in the N-terminal part of the CP determines aphid transnmission. Deletion of the sequence encoding these amino acids in the CP gene will create a transgenic coat protein that cannot be transmitted to another plant. Jagadish and collaborators [35] have been able to determine the amino acids in the core region of the coat protein of a potyvirus that are essential for assembly of the viral particle. Deletions or mutations of these amino acids in the transgenic CP would be another way of hindering transcapsidation.