Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Gathering a bouquet of miRNA targets

Abstract

Pulling down microRNA-induced silencing complexes (miRISCs) allows researchers to collect microRNAs and their mRNA targets in vivo.

Main

Tiny and elusive, microRNAs (miRNAs) went unnoticed for years. Only a serendipitous mutation in a well-characterized model organism brought them to light. But because genetic tools are slow to identify miRNAs and the mRNAs they target, many researchers use bioinformatics to predict miRNA pathways. The success rate of these in silico searches, however, has not been satisfactory. Now, with his colleagues at University of Colorado at Boulder, Min Han reports an efficient and robust biochemical screen for miRNAs and their targeted mRNAs, allowing unprecedented in vivo characterization of these elusive species.

RNA-induced silencing complexes (RISCs) form around miRNA precursors to process the miRNA and disrupt the expression of targeted mRNAs. Members of the GW182 protein family are essential components of the miRISC, but mutations in ain-1, a Caenorhabditis elegans gene encoding a GW182 protein, had mild miRNA-deficient phenotypes.

Suspecting redundancy, Han and colleagues sought and identified a homolog of ain-1, which they called ain-2. They found that worms with mutations in both genes had an miRNA pathway block. AIN-1 and AIN-2, however, were not needed for miRNA production, but instead for miRNA function. The researchers proposed that AIN-1– and AIN-2–associated miRISCs acted at an 'effector stage' of the pathway and might contain mature miRNAs and their targeted mRNAs.

In coimmunoprecipitation assays for either AIN-1 or AIN-2, Han and colleagues characterized other components of these effector miRISCs, hoping to capture active complexes from living tissue. They first sequenced the collected small RNAs and found an impressive 106 of the 130 previously known miRNAs between the AIN-1– and AIN-2–derived samples. Additionally, the small RNA samples had 9 new putative miRNAs that had been missed in earlier screens, perhaps because of low expression and sequence conservation. This catalog of miRNAs suggested that this approach was effective and comprehensive.

Then, co-first authors Liang Zhang and Lei Ding isolated mRNA from these samples. In preliminary experiments with quantitative reverse transcription–PCR, they found that “many known target mRNAs appeared to be stably associated with these complexes,” says Zhang. Encouraged, the researchers ran the AIN-1– and AIN-2–containing miRISCs on microarrays. They detected almost 90% of the 12 previously established targeted mRNAs in the samples, reporting a sensitivity not yet seen in computational predictions of miRNA targets or from pulldowns with other components of miRISCs.

In total, the researchers identified almost 3,600 transcripts associated with AIN-1 or AIN-2. Han and colleagues could follow mRNAs that are known to be developmentally regulated, detecting these transcripts only when they are regulated by miRNAs, demonstrating these data reflected the contents of active miRISCs. The investigators also found that 28% of AIN-1– or AIN-2–associated transcripts had been predicted to be miRNA targets by bioinformatic analysis, a significantly higher proportion than the 13% predicted of the entire transcriptome. This observation suggests that these AIN-1– or AIN-2–derived samples were enriched for transcripts that had already been predicated to be miRNA targets based on characteristics other than association with effector miRISCs.

Han hopes that this list of putative miRNA targets will become a valuable training set for algorithms, improving their predictive power. He says that “if we have enough data from many complexes under different physiological conditions, we may be able to generate a miRNA-miRNA target interaction map through correlation analysis,” matching miRNAs to targeted mRNA and deducing miRNA networks.

Han says that the biochemical screen is a “very effective way to identify miRNA targets,” and that this approach could be used in other systems as well. GW182 proteins have divergent sequences, but their conserved function suggests that the protein family may be a foothold into the effector miRISC, giving researchers a new vantage point on the world of miRNAs.

References

  1. 1

    Zhang, L. et al. Systematic identification of C. elegans miRISC proteins, miRNAs, and mRNA targets by their interactions with GW182 proteins AIN-1 and AIN-2. Mol. Cell 28, 598–613 (2007).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

Download references

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Stevens, K. Gathering a bouquet of miRNA targets. Nat Methods 5, 123 (2008). https://doi.org/10.1038/nmeth0208-122b

Download citation

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing