Competition between mammalian RNAi-related gene silencing pathways is well documented. It is therefore important to identify all classes of small RNAs to determine their relationship with RNAi and how they affect each other functionally. Here, we identify two types of 59-phosphate, 39-hydroxylated human tRNA-derived small RNAs (tsRNAs). tsRNAs differ from microRNAs in being essentially restricted to the cytoplasm and in associating with Argonaute proteins, but not MOV10. The first type belongs to a previously predicted Dicer-dependent class of small RNAs that we find can modestly down-regulate target genes in trans. The 59 end of type II tsRNA was generated by RNaseZ cleavage downstream from a tRNA gene, while the 39 end resulted from transcription termination by RNA polymerase III. Consistent with their preferential association with the nonslicing Argonautes 3 and 4, canonical gene silencing activity was not observed for type II tsRNAs. The addition, however, of an oligonucleotide that was sense to the reporter gene, but antisense to an overexpressed version of the type II tsRNA, triggered robust, >80% gene silencing. This correlated with the redirection of the thus reconstituted fully duplexed double-stranded RNA into Argonaute 2, whereas Argonautes 3 and 4 were skewed toward less structured small RNAs, particularly single-strand RNAs. We observed that the modulation of tsRNA levels had minor effects on the abundance of microRNAs, but more pronounced changes in the silencing activities of both microRNAs and siRNAs. These findings support that tsRNAs are involved in the global control of small RNA silencing through differential Argonaute association, suggesting that small RNA-mediated gene regulation may be even more finely regulated than previously realized.

SUMMARY Short-hairpin RNA (shRNA)-induced RNAi is used for biological discovery and therapeutics. Dicer, whose normal role is to liberate endogenous miRNAs from their precursors, processes shRNAs into different biologically active siRNAs, affecting their efficacy and potential for off-targeting. We found that in cells, Dicer induced imprecise cleavage events around the expected sites based on the previously described 5′/3′-counting rules. These promiscuous non-canonical cleavages were abrogated when the cleavage site was positioned 2 nt from a bulge or loop. Interestingly, we observed that the ~1/3 of mammalian endogenous pre-miRNAs that contained such structures were more precisely processed by Dicer. Implementing a new “loop-counting rule”, we designed potent anti-HCV shRNAs with substantially reduced off-target effects. Our results suggest that Dicer recognizes the loop/bulge structure in addition to the ends of shRNAs/pre-miRNAs for accurate processing. This has important implications for both miRNA processing and future design of shRNAs for RNAi-based genetic screens and therapies.

The Sleeping Beauty (SB) transposon is an emerging tool for transgenesis, gene discovery, and therapeutic gene delivery in mammals. Here we studied 1,336 SB insertions in primary and cultured mammalian cells in order to better understand its target site preferences. We report that, although widely distributed, SB integration recurrently targets certain genomic regions and shows a small but significant bias toward genes and their upstream regulatory sequences. Compared to those of most integrating viruses, however, the regional preferences associated with SB-mediated integration were much less pronounced and were not significantly influenced by transcriptional activity. Insertions were also distinctly nonrandom with respect to intergenic sequences, including a strong bias toward microsatellite repeats, which are predominantly enriched in noncoding DNA. Although we detected a consensus sequence consistent with a twofold dyad symmetry at the target site, the most widely used sites did not match this consensus. In conjunction with an observed SB integration preference for bent DNA, these results suggest that physical properties may be the major determining factor in SB target site selection. These findings provide basic insights into the transposition process and reveal important distinctions between transposon-and virus-based integrating vectors.Approximately half of the mammalian genome is derived from ancient transposable elements. Although the two general types of transposable elements, (DNA) transposons and retrotransposons, are often regarded as "selfish DNA parasites" or "junk DNA," their frequent movement in and out of host cell chromosomes has played a significant role in genome diversification and evolution. Members of the Tc1/mariner family of DNA transposons are extremely widespread in nature (44) and can function independently of species-specific host factors (29,56). Although the vast majority of elements present in vertebrate genomes are nonfunctional (14, 32), an active Tc1-like element called Sleeping Beauty (SB) was recently reconstructed from ancient transposon fossils found within fish genomes (20).SB elements transpose by a cut-and-paste mechanism that requires the sequence-specific binding of the SB transposase to the transposon ends (25). This transposition process involves the precise excision and reintegration of the transposon from one DNA site to another site, which invariably contains a TA dinucleotide that is duplicated upon insertion. The transfer of DNA strands at the insertion site is mediated by the transposase catalytic core domain, which contains a conserved DDE motif shared by a large group of recombinase proteins, including the V(D)J recombinase and retrovirus integrases (44). SB is capable of efficient transposition in a variety of cell types (24), including human, mouse, and fish cells, and is an emerging tool for genetic research on vertebrates, with potential applications for transgenesis (22), functional genomics (1, 5-9, 11, 12, 16, 18, 19, 21, 33), and human gene therapy...