Abstract
How newly generated microRNA (miRNA) genes are integrated into gene regulatory networks during evolution is fundamental in understanding the molecular and evolutionary bases of robustness and plasticity in gene regulation. A recent model proposed that after the birth of a miRNA, the miRNA is generally integrated into the network by decreasing the number of target genes during evolution. However, this decreasing model remains to be carefully examined by considering in vivo conditions. In this study, we therefore compared the number of target genes among miRNAs with different ages, combining experiments with bioinformatics predictions. First, we focused on three Drosophila miRNAs with different ages. As a result, we found that an older miRNA has a greater number of target genes than a younger miRNA, suggesting the increasing number of targets for each miRNA during evolution (increasing model). To further confirm our results, we also predicted all target genes for all miRNAs in D. melanogaster, considering co-expression of miRNAs and mRNAs in vivo. The results obtained also do not support the decreasing model but are reasonably consistent with the increasing model of miRNA-target pairs. Furthermore, our large-scale analyses of currently available experimental data of miRNA-target pairs also showed a weak but the same trend in humans. These results indicate that the current decreasing model of miRNA-target pairs should be reconsidered and the increasing model may be more appropriate to explain the evolutionary transitions of miRNA-target pairs in many organisms.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1621-1633 |
| Number of pages | 13 |
| Journal | Genome Biology and Evolution |
| Volume | 8 |
| Issue number | 5 |
| DOIs | |
| State | Published - Apr 27 2016 |
Bibliographical note
KAUST Repository Item: Exported on 2020-10-01Acknowledgements: We thank Shu Kondo for advice regarding the Drosophila
crossing experiments and Miu Kubota for help with the experiments. We are also grateful to Junichi Imoto, Sonoko
Kinjo, Norikazu Kitamura, Kaoru Matsumoto, Masatoshi Nei,
Masa-aki Yoshida, and Ikuko Yuyama for their comments on
earlier versions of the manuscript. We also thank the associate
editor and the two reviewers for their constructive comments
on our work. Computations were partially performed on the
NIG supercomputer at National Institute of Genetics. This
work was supported by grants from the National Institute of
Genetics, the Center for the Promotion of Integrated Sciences
(CPIS), and JSPS KAKENHI Grant Number 25711023 to M.N.