MicroRNAs (miRNAs) are processed from RNA polymerase II (RNAPII)-specific transcripts of independent genes or from introns of protein-coding genes. In the canonical pathway, primary
precursor (pri-miRNA) processing occurs in two steps, catalysed by two members of the RNase III family of enzymes, Drosha and Dicer, operating in complexes with dsRNA-binding proteins (dsRBPs), for example DGCR8 and transactivation-responsive (TAR) RNA-binding protein (TRBP) in mammals.
In the first nuclear step, the Drosha–DGCR8 complex processes pri-miRNA into an ~70-nucleotide precursor hairpin (pre-miRNA), which is exported to the cytoplasm. Some pre-miRNAs are produced from very short introns (mirtrons) as a result of splicing and debranching, thereby bypassing the Drosha–
DGCR8 step. In either case, cleavage by Dicer, assisted by TRBP, in the cytoplasm yields an
~20-bp miRNA/miRNA* duplex. In mammals, argonaute 2 (AGO2), which has robust RNaseH-like endonuclease activity, can support Dicer processing by cleaving the 3′ arm of some pre-miRNAs,
thus forming an additional processing intermediate called AGO2-cleaved precursor miRNA
(ac-pre-miRNA). Processing of pre-miR-451 also requires cleavage by AGO2, but is
independent of Dicer and the 3′ end is generated by exonucleolytic trimming.