Skip to main content

Argonaute-associated factors required for translational repression in plant RNA silencing

Final Report Summary - ASTRIR (Argonaute-associated factors required for translational repression in plant RNA silencing)

In Arabidopsis thaliana, microRNAs (miRNAs) are loaded in ARGONAUTE 1 (AGO1) as part of RNA induced silencing complexes (RISCs) to regulate target mRNAs via base pairing. AGO1 possesses an intrinsic endonuclease activity responsible for the ‘slicing’ of mRNA targets, an activity that is abrogated when bulges or mismatches face nucleotides 10-11 of the miRNAs. Such mismatches promote in turn alternative forms of target repression such as the translational inhibition and/or decay of mRNAs. Intriguingly, the fact that in Arabidopsis most of the miRNAs regulates their target mRNAs via perfect or near-perfect complementarity has contributed to the widespread belief that plant miRNAs, unlike their animal counterparts, exert their effect mostly through target mRNA slicing. Yet, a previous work notably carried out in our lab suggests that plant miRNAs can concurrently slice and transnationally inhibit a given pool of mRNAs, and this raises the fundamental question of how slicing is avoided during translational inhibition.

One possibility is that translational repressor proteins associate to AGO1 in order to change further the fate of the mRNA targets. Following this idea, 4 factors were identified as potential AGO1-associated factors. It concerns the two RRM proteins RSP40/41, the helicase of the translational initiation complex eIF4a1, and the polyA binding protein PABP2. Here, we show that eIF4a1 genetically interacts with AGO1. This protein does not affect the miRNAs biogenesis, nor the stability of the main silencing factors, but change the mode of action of AGO1. Indeed, in an eif4a1 mutant background, the slicing activity of AGO1 is greatly improved, favouring thereby the cleavage of the target mRNAs. Consistent with this result, this property also gives to the mutant plants a stronger resistance upon infection by the Tobacco rattle virus (TRV).
In order to decipher more carefully which components of the plant RNA silencing interfere with the protein synthesis, we have also developed in parallel two methods to isolate the entire translational machinery, the immunoprecipitation of polysomes and the classical ribosomal profiling onto sucrose gradients. These two complementary approaches allowed us identifying different silencing factors directly associated with both monosomes and polysomes. If eIF4a1 does not cosediment with polysomes, we can however show that in addition to AGO1, AGO4, AGO5, AGO9 and AGO10 are present on polysomes. Interestingly, we also observed that only a part of the miRNAs can be co-purified together with the translational machinery, strongly suggesting that only a specific pool of miRNAs is responsible for translational repression in plant. At last, and surprisingly, deep sequencing analysis show the presence of a large number of 24 nucleotides long siRNAs in polysomal fractions. If for the moment the role of these siRNAs is unclear, at least these results are consistent with the presence of AGO4, AGO5 or AGO9, which are mainly loaded with 24 mers siRNAs.

The work described above suggests that eIF4a1 is important to modulate the slicing activity of AGO1. As multiple pools of AGO1 co-exist in Arabidopsis, we propose that eIF4a1 is required in one of these pools to programme RISC in a non-slicing mode of action, and force an alternative type of gene regulation such as translational repression. Likewise, our polysomes profiling experiments show that only a particular pool of miRNAs associate with the translational machinery. It is therefore tempting to speculate that the biochemical properties of small RNAs (sizes, mismatches with targets, GC content, etc…) might also be important to initiate the translational repression. Following this hypothesis, it would be worth to seriously consider the 24 mers siRNAs, and therefore others AGOs such as AGO4 or AGO9, as potential effectors of the translational repression mechanism.

Overall, our results aimed, through highly innovative approaches, at improving our understanding of the mode of action of small RNAs, which are at the core of gene regulation in most, if not all, higher organisms. Largely because of its substantial added value in terms of scientific excellence, the outcome of our work gives new insights into the antiviral defence but also the small RNA dependent gene regulation, in general. This will be beneficial to attract and encourage companies as well as other academic groups to develop new tools and/or therapeutic solutions for the future. In addition, these new possibilities will also increase the attractiveness of Europe to worldwide researchers in the longer run.