Characterization of the dynamics of expression of the AGO1 gene and miR168 during the formation of Brassica neo-allotetraploids
Résumé
In our laboratory, we use the Brassica napus model (AACC, 2n=38) to add to our knowledge of the evolutionary success of allopolyploidy that has marked the evolution of eukaryotes and notably of plants. To characterize the molecular mechanisms at the origins of a new allopolyploidisation event, we analyze resynthesized Brassica allotetraploids (coll. A.-M. Chèvre, INRA Rennes) that represent a highly valuable plant material to follow the genomic changes occurring during the formation of the neo-polyploids in comparison with the diploid progenitors B. rapa (AA, 2n=20) and B. oleracea (CC, 2n=18). In the present study, we completed previous results that have demonstrated the mobilization of small non coding RNAs immediately after allopolyploidization, supporting the model of a PTGS-to-TGS shift involved in the control of reactivated components of the genome like transposable elements (Martinez Palacios et al. in prep.). We have thus characterized the dynamics of expression of the AGO1 gene and the microRNA miR168 that represent the central molecular actors of the PTGS pathway, in order to validate our model. First, we performed the bioanalysis of all the copies of the AGO1 gene present in Brassica genomes to identify and characterize any duplicate retention of the AGO1 gene copies during the evolution of the Brassica genome. We then characterized the dynamics of expression of the AGO1 gene and its regulatory partner miR168 along the allopolyploidisation event (represented by the F1 hybrids, the S0, S1, S3 and S5 allotetraploids in comparison with the diploid progenitors), by developing quantitative RT-PCR analysis. Our main conclusion is that the sRNA-based regulation of reactivated TEs, EVEs and other heterochromatic sequences depicted during the formation of the Brassica neo-allotetraploids, and the one that has been well characterized during viral attacks share similar mechanisms that are in accordance with the characterization of the expression dynamics of AGO1 and miR168 we have made during the present study.