Evidence for large complex networks of plant short silencing RNAs.

Background: In plants and animals there are many classes of short RNAs that carry out a wide range of functions within the cell; short silencing RNAs (ssRNAs) of 21-25 nucleotides in length are produced from double-stranded RNA precursors by the protein Dicer and guide nucleases and other proteins to their RNA targets through base pairing interactions. The consequence of this process is degradation of the targeted RNA, suppression of its translation or initiation of secondary ssRNA production. The secondary ssRNAs in turn could then initiate further layers of ssRNA production to form extensive cascades and networks of interacting RNA [1]. Previous empirical analysis in plants established the existence of small secondary ssRNA cascade [2], in which a single instance of this event occurred but it was not known whether there are other more extensive networks of secondary sRNA production.

Methodology/principal Findings: We generated a network by predicting targets of ssRNA populations obtained from high-throughput sequencing experiments. The topology of the network shows it to have power law connectivity distribution, to be dissortative, highly clustered and composed of multiple components. We also identify protein families, PPR and ULP1, that act as hubs within the network. Comparison of the repetition of genomic sub-sequences of ssRNA length between Arabidopsis and E.coli suggest that the network structure is made possible by the underlying repetitiveness in the genome sequence.

Conclusions/significance: Together our results provide good evidence for the existence of a large, robust ssRNA interaction network with distinct regulatory function. Such a network could have a massive effect on the regulation of gene expression via mediation of transcript levels.