Antiviral RNA silencing viral counter defense in plants

The research described in this thesis centres around the mechanism of RNA silencing in relation to virus-host interaction, an area of increasing importance. It shows how this recently disclosed mechanism can be used to produce virus-resistant plants. Based on the activity of the RNA silencing machinery, which is an important line of defense of plants against viruses, plants can be pre-programmed to produce so-called small-interfering RNAs (siRNAs) that specifically target the genome or transcripts of incoming viruses. This principle works like an antivirus software package on a computer, in which signatures recognizing specific viruses are included. Any incoming file potential virus) is scanned for the recognition sequence much like an siRNA does on an incoming viral RNA.

In order to produce multiple virus resistance the work presented in this thesis combined the very high silencing activity of double stranded RNA (dsRNA) with a fusion strategy involving small, 150 basepairs long segments of the nucleocapsid (N) gene of four different tomato-infecting tospoviruses. The virus-derived sequences were arranged in an inverted repeat leading to the production of dsRNA. Transgenic plants stably expressing these dsRNAs were shown to process them into siRNAs thereby pre-programming the plants to fight incoming tospoviruses. Indeed, these transgenic plants showed multiple virus resistance at a high frequency. The high frequency is important for the production of transgenic plants that are difficult to transform.

As RNA silencing is an important antiviral defense mechanism of plants, viruses have to overcome this hurdle and they do this by encoding proleins capable of blocking this pathway at certain steps. At the onset of this research project a number of such ("RNA si!encing suppressor") proteins had been characterized for positive strand RNA viruses and DNA viruses. In Chapter 3 it is shown that also negative strand RNA viruses, i.e. Tomato spotted wilt virus (TSWV) and Rice hoja blanca virus (RHBV), encode suppressors of RNA silencing QSISs and NS3, respectively). Interestingly, while the silencing suppressors are encoded on analogous genomic positions, they show different silencing suppression activities.

Having identified the first negative strand RNA viral silencing suppressors the next step was to investigate whether also mammalian negative strand RNA viruses could also encode such proteins. This was a logical next step to take for two reasons. 1. TSWV is a member of the Bunyaviridae, a virus family of which most other members infect mammals, and 2. viruses of this family are propagated in their insect vectors while it has been shown that silencing suppressors can be active in insects. To test this possibility the well characterized NSl protein of the Influenza A virus was chosen to be tested, indeed, it could be demonstrated that NSl is capable of suppressing RNA silencing in plants, possibly by binding siRNAs. These findings were strengthened by the observation that NSl enhances the virulence of Potato virus X when it was incorporated into its genome. The chimeric PVXMSl virus induced more pronounced disease symptoms, the NSI still being capable of suppressing RNA silencing when expressed from that viral vector. Whether NSl also has RNA silencing suppressor activity in mammalian cell systems though, still awaits to be confirmed. An intriguing fact is that NSl is an inhibitor of the interferon pathway, a mammalian antiviral mechanism in which - again - dsRNA plays a key role. This may indicate that in mammalian cells the interferon response and RNA silencing are somehow interlinked to act in concert. It is also possible that NSl binds long dsRNA to hide it from the interferon response and that the resulting silencing suppression in plants is only a side effect caused by that activity.

A further aim of this thesis was to better understand the interactions between viral silencing suppressors and the host. The suppressors of RNA silencing used were the previously mentioned NSs, NS3, NSl and additionally HC-Pro of Cowpea aphid borne mosaic virus and 2b of Cucumber mosaic virus. For the analysis, cDNA-amplified fragment length polymorphism (AFLP) technology was used. This technique allows a very broad analysis of transcriptional changes in plant tissues upon certain treatments. This combined with the A. tumefaciens mediated transient gene expression allowed monitoring of transcription profile changes caused by the expression of viral silencing suppressors. Out of approximately 25,000 mRNAs tested 362 were found to be differentially expressed due to the activity of silencing suppressors. Interestingly a large majority of genes (80%) was down-regulated. The fact that quite a number of genes were differentially expressed was to be expected since RNA silencing also plays a role in regulating certain mRNA levels via the activity of so-called micro-RNAs (miRNAs). Any disturbances in this processes will produce complex transcriptional changes for instance by directly inhibiting the activity of miRNAs (as it has been shown for HC-Pro for instance).

To conclude it can be stated that the interactions between antiviral RNA silencing and the countermeasures viruses have evolved to frustrate such process are on one hand a very important topic in virology and on the other hand a strong starting point for breakthroughs in other fields of research such as functional genomics and development. In an application environment, RNA silencing has allowed us to develop efficient and broad virus resistance in plants.