CORDIS - EU research results

RNA interference and plant cell-to-cell silencing signal

Final Activity Report Summary - RNAi MOVEMENT (RNA interference and plant cell-to-cell silencing signal)

The project successfully achieved its three main objectives:

1. to provide in depth characterisation of smd-1, 2 and 3 from the SUC-SUL genetic screen previously set up in the laboratory (see Dunoyer et al. 2005, Nature Genetics 37(12), 1356-1360), as well as of the positional cloning of these mutants.
2. to identify and characterise novel mutants impaired in silencing movement based on the same genetic screen.
3. to determine whether endogenous trans-acting small interfering ribonucleic acid (tasiRNA), in contrast to transgenic plants, was able to move between cells and regulate gene expression at a distance.

The principal results and the most important achievements obtained during the project lifetime are subsequently presented.

In relation to the first project, we started an extensive mapping of the three streptomycin dependent (SMD) mutants that were previously recovered. In RNA interference (RNAi), double-stranded RNA (dsRNA) was processed into short interfering RNA (siRNA) to mediate sequence-specific gene knockdown. The genetics of plant RNAi was not understood, nor were the bases for its spreading between cells. Here, we unravelled the requirements for biogenesis and action of siRNAs directing RNAi in arabidopsis thaliana and demonstrated how alternative routes redundantly mediated this process under extreme dsRNA dosages. We found that SMD1 and SMD2, required for intercellular but not intracellular RNAi, were allelic to RDR2 and NRPD1a respectively, previously implicated in siRNA-directed heterochromatin formation through the action of DCL3 and AGO4. However, neither DCL3 nor AGO4 were required for non-cell autonomous RNAi, uncovering a new pathway for RNAi spreading or detection in recipient cells. Finally, we showed that the genetics of RNAi were distinct from those of antiviral silencing and proposed that this experimental silencing pathway had a direct endogenous plant counterpart.

Because the SUC-SUL screening was far from being saturated, we used the same strategy to identify and characterise novel mutants impaired in silencing movement. Extensive analysis in these mutants showed three additional mutations in DCL4, in which accumulation of 21-nt SUL siRNA was specifically lost. Recovery of dcl4-7, dcl4-8 and dcl4-9 expanded a previously identified series of dcl4 alleles specifically deficient in 21-nt SUL siRNA accumulation. This allowed us to test if the production of 21-nt SUL siRNA involved in virus induced gene silencing (VIGS) and in RNAi could be genetically uncoupled. To perform these experiments we used tobacco rattle virus (TRV) to trigger phytoene desaturase (PDS). The analysis of dcl4-2 to 9 allelic series demonstrated that the production of 21-nt siRNAs did not entail the same genetic requirements during VIGS, RNAi and tasiRNA biogenesis.

The host laboratory had previously demonstrated that, unlike siRNA from transgenes, endogenous miRNA did not move over 10 to 15 cells from the site of their synthesis, as mentioned in Parizzoto et al (2004) 18(18): 2237-2242. TasiRNA were a relative recent class of plant endogenous 21-nt long small RNA that shared characteristics of miRNA activity and siRNA biogenesis. Because of these properties, we aimed to address the following questions:

1. where and when were the tasiRNA precursor transcripts transcribed.
2. whether they were able to regulate gene expression out of the sites in which they were transcribed.

The isolation of a genetrap reporter line for TAS3a, the major locus producing auxin response factor (ARF) that was regulating tasiRNAs in the arabidopsis shoot, was described. Its activity was limited to the adaxial, i.e. upper, side of leaf primordia, thus spatially isolated from ARF-activities, which were located in the abaxial, i.e. lower, side. We demonstrated, through in situ hybridisation and reporter fusions, that the silencing activities of ARF-regulating tasiRNAs were indeed autonomously non-cell manifested to spatially control ARF activities. Endogenous tasiRNAs were thus mediators of a mobile developmental signal and might provide effective gene silencing at a distance beyond the reach of most miRNAs.