Expanding RNA physiology: microRNAs in a unicellular organism: Figure 1.

Abstract

Until recently, the world viewed through the eyes of molecular biologists seemed simple with a well-known slogan hailing a central dogma: “DNA makes RNA makes protein.” This has been interpreted to mean that genetic output is entirely or almost entirely transacted by proteins (Mattick 2004). However, within a few years of the discovery of RNA interference (RNAi) (Fire et al. 1998), an explosion of data has lifted the veil off many previous puzzling findings. Thus, our view of the field was irrevocably changed with a new slogan—that went alongside the classical central dogma—“RNA makes small RNA makes no protein” (Fig. 1). Sequence information in long RNA transcripts is converted into small, ∼20to 30-nucleotide (nt)-long, non-protein-coding RNAs such as microRNAs (miRNAs) and small interfering RNAs (siRNAs). These small RNAs trigger various forms of sequence-specific gene silencing, including RNA cleavage and degradation, cleavage-independent mRNA decay, translational repression, methylation of protein-coding DNA, and heterochromatin formation in a variety of eukaryotic organisms ranging from fission yeast, plants, fungi, nematode, and fly to humans. This process is now collectively referred to as RNA silencing (Tomari and Zamore 2005; Zaratiegui et al. 2007). In addition to classical small non-proteincoding RNAs—including tRNAs, small nuclear RNAs (snRNAs), and small nucleolar RNAs (snoRNAs)—recent studies have shown that our cells express a surprisingly large number of new regulatory small RNAs that seem to be particularly abundant in roles that direct the binding of protein complexes to specific nucleic acid sequences (Huttenhofer and Schattner 2006). These include miRNAs, several groups of natural or endogenous siRNAs, and Piwi-interacting RNAs (piRNAs) (Kim 2006). The ongoing efforts to uncover the full biological scope of RNA silencing have already modified our notions about how gene expression is controlled. It is becoming clear that RNA silencing comprises an important tier of gene expression in eukaryotes, enabling the integration and networking of complex suites of gene activity, thereby elaborating multicellular complexity (Bartel 2004; Kloosterman and Plasterk 2006). The first small RNA trigger discovered in eukaryotes was the lin-4 miRNA, which was found by mapping a developmental timing (heterochronic) mutant Caenorhabditis elegans locus (Lee et al. 1993). The discovery of the phylogenetic conservation of let-7 miRNA then really opened up the field (Pasquinelli et al. 2000). Thousands of miRNAs have now been identified in various multicellular organisms (miRBase; Griffiths-Jones et al. 2006), and an astonishing number of miRNA genes have been predicted to exist in our genome (Miranda et al. 2006). Transcripts of genes encoding self-complementary RNA that forms imperfect hairpins, termed primary microRNA (pri-miRNA), are cleaved by the Drosha– Pasha/DGCR8 complex in the nucleus to generate premiRNA. The pre-miRNA is exported to the cytoplasm and further processed by Dicer into an ∼21to 23-nt miRNA duplex. Then, one strand over the other is often preferentially loaded into a RNA-induced silencing complex (RISC), which contains at its core a member of the Argonaute family of proteins (Parker and Barford 2006). Once incorporated into the silencing complex, mature miRNAs interact with target mRNAs at specific sites to induce cleavage of the message, cleavage-independent mRNA decay, or inhibit translation (Tomari and Zamore 2005; Pillai et al. 2007). Cleavage of mRNA targets is catalyzed by Argonaute proteins (Liu et al. 2004; Parker and Barford 2006). In both animals and plants, miRNAs play important roles in diverse developmental processes, including developmental timing, cell death, cell proliferation, asymmetric cell fate decision, organogenesis, and patterning of the nervous system. However, the biological functions of most miRNAs remain unknown (Alvarez-Garcia and Miska 2005; Jones-Rhoades et al. 2006). Recent studies have shown that miRNAs target 30% or more of all animal protein-coding genes, most of which are regulated through base pairing between specific sites in the 3 untranslated region of the messenger RNAs (mRNAs) and a small region termed the “seed” (nucleotides 1–8, from the 5 end) near the 5 end of the miRNA (Brennecke et al. 2005; Lewis et al. 2005; Xie et al. 2005). These rather lax criteria can result in numerous targets being predicted for a given miRNA. Many genes have several target sites for either one miRNA or a few different miRNAs. Additionally, many miRNAs are expressed in a cell typeor tissue-specific manner, and miRNAs and their targets have, to a large extent, mutually exclusive expression patterns (“para-expression”) (Farh et al. 2005; Stark et al. 2005). Therefore, the general function of miRNAs may be that miRNAs diversify cell types or organ types by fine-tuning the proteome, and then retain Corresponding author. E-MAIL siomi@genome.tokushima-u.ac.jp; FAX 81-88-6339451. Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1559707.