Global mapping of miRNA-target interactions in cattle (Bos taurus)

With regulatory roles in development, cell proliferation and disease, micro-RNA (miRNA) biology is of great importance and a potential key to novel RNA-based therapeutic regimens. Biochemically based sequencing approaches have provided robust means of uncovering miRNA binding landscapes on transcriptomes of various species. However, a current limitation to the therapeutic potential of miRNA biology in cattle is the lack of validated miRNAs targets. Here, we use cross-linking immunoprecipitation (CLIP) of the Argonaute (AGO) proteins and unambiguous miRNA-target identification through RNA chimeras to define a regulatory map of miRNA interactions in the cow (Bos taurus). The resulting interactome is the deepest reported to date for any species, demonstrating that comprehensive maps can be empirically obtained. We observe that bovine miRNA targeting principles are consistent with those observed in other mammals. Motif and structural analyses define expanded pairing rules with most interactions combining seed-based pairing with distinct, miRNA-specific patterns of auxiliary pairing. Further, miRNA-target chimeras had predictive value in evaluating true regulatory sites of the miR-17 family. Finally, we define miRNA-specific targeting for >5000 mRNAs and determine gene ontologies (GO) for these targets. This confirmed repression of genes important for embryonic development and cell cycle progress by the let-7 family, and repression of those involved in cell cycle arrest by the miR-17 family, but it also suggested a number of unappreciated miRNA functions. Our results provide a significant resource for transcriptomic understanding of bovine miRNA regulation, and demonstrate the power of experimental methods for establishing comprehensive interaction maps. Overall design: miRNA-target chimeras were identified in MDBK cells from standard AGO-CLIP or CLEAR-CLIP experiments as indicated. Experiments were included where cells had been exposed to infection with bovine viral diarrhea virus or tinyLNA-17 treatment. These are indicated. Processed reads were aligned to the host genome (BosTau7). Alignment statistics of CLIP reads are given in the processed data file "Table S1". A complete table of significant peaks from standard AGO-CLIP reads of the 39 data sets is given in "Table S2". This table additionally contains information regarding the associated CLIP read cluster (clustering of all CLIP reads) and potential overlapping miRNA-target chimeras, as well as genomic annotation. A complete list of individual miRNA-target chimeras is given in "Table S3". In the "RNA-seq: tinyLNA-17 vs. Mock" subseries, mRNA-seq was performed on two replicates each of MDBK cells treated with the miR-17 family inhibitor, tinyLNA-17. Differential gene expression analysis was performed and the associated data is given as processed data file "Table S5".