Intronic microRNA: discovery and biological implications.

Identification of microRNAs (miRNAs) is essential to studying their physiological functions. Due to the difficulties in discovering truly expressed miRNAs from genomic random hairpin secondary structure sequences, it is beneficial to predict them from expressed sequences--expressed sequence tags (ESTs) and intronic sequences. We used a modified scanning pipeline using criteria based on the features of known pre-miRNAs and phylogenetic conservation for predicting intronic miRNAs. Upon examination, we found that 25% of known human miRNAs belong to intronic regions of known protein-coding genes. About 50% of these intronic miRNAs reside in introns whose length is longer than 5,000 bps. It is likely that these intronic miRNAs can have their own independently regulated transcription units, which can be regulated by RNA polymerase II (Pol II) or RNA polymerase III (Pol III). It was recently demonstrated that RNA Pol III could transcribe human miRNAs through associated repetitive elements. Since various repetitive elements are often found to be present in the intronic regions, the distribution of intronic miRNAs and their possible transcription regulation are presented. Although the intronic miRNAs and their host genes could be regulated independently, it is possible that the intronic miRNA can still down-regulate its own host protein-coding gene by targeting the untranslated region (UTR) of the host gene. Another biological implication is that intronic miRNAs could play an important role as negative feedback regulators. We propose hypothetical models of such feedback regulation on host protein-coding genes by selecting the transcription factors as miRNA targets or by protein-protein interactions between intronic miRNA host gene product and miRNA target gene products.