The deep evolution of metazoan microRNAs Benjamin M. Wheeler, a Alysha M. Heimberg, b Vanessa N. Moy, b Erik A. Sperling, c Thomas W. Holstein, d Steffen Heber, e and Kevin J. Peterson b,Ã a Department of Computer Science, North Carolina State University, Raleigh, NC 27695, USA b Department of Biological Sciences, Dartmouth College, Hanover, NH 03755, USA c Department of Geology and Geophysics, Yale University, P.O. Box 208109, New Haven, CT 06520, USA d Molekulare Evolution und Genomik, Heidelberger Institut fu ¨r Zoologie, Neuenheimer Feld 230, 69120 Heidelberg, Germany e Bioinformatics Research Center, North Carolina State University, Raleigh, NC 27695, USA Ã Author for correspondence (email: kevin.j.peterson@dartmouth.edu) SUMMARY microRNAs (miRNAs) are approximately 22-nucleotide noncoding RNA regulatory genes that are key players in cellular differentiation and homeostasis. They might also play important roles in shaping metazoan macro- evolution. Previous studies have shown that miRNAs are continuously being added to metazoan genomes through time, and, once integrated into gene regulatory networks, show only rare mutations within the primary sequence of the mature gene product and are only rarely secondarily lost. However, because the conclusions from these studies were largely based on phylogenetic conservation of miRNAs between model systems like Drosophila and the taxon of interest, it was unclear if these trends would describe most miRNAs in most metazoan taxa. Here, we describe the shared complement of miRNAs among 18 animal species using a combination of 454 sequencing of small RNA libraries with genomic searches. We show that the evolutionary trends elucidated from the model systems are generally true for all miRNA families and metazoan taxa explored: the continuous addition of miRNA families with only rare substitutions to the mature sequence, and only rare instances of secondary loss. Despite this conservation, we document evolutionary stable shifts to the determination of position 1 of the mature sequence, a phenomenon we call seed shifting, as well as the ability to post-transcriptionally edit the 5 0 end of the mature read, changing the identity of the seed sequence and possibly the repertoire of downstream targets. Finally, we describe a novel type of miRNA in demosponges that, although shows a different pre-miRNA structure, still shows remarkable conservation of the mature sequence in the two sponge species analyzed. We propose that miRNAs might be excellent phylogenetic markers, and suggest that the advent of morphological complexity might have its roots in miRNA innovation. INTRODUCTION microRNAs (miRNAs) are a recently discovered group of small RNA regulatory genes that have captured the attention of investigators interested in the control of cellular differen- tiation (Ambros 2004; Zhao and Srivastava 2007; Hobert 2008; Makeyev and Maniatis 2008; van Rooij et al. 2008; Yi et al. 2008), its misregulation (Lu et al. 2005; Meltzer 2005; Calin and Croce 2006; Esquela-Kerscher and Slack 2006; He et al. 2007; Sevignani et al. 2007; Barbarotto et al. 2008; Medina and Slack 2008; Yang et al. 2008), and its evolution (Sempere et al. 2006; Heimberg et al. 2008; Pierce et al. 2008). With the demonstration that morphologically complex ani- mals like vertebrates and fruit flies possess a protein-coding genetic tool kit that is largely conserved across Metazoa (Peterson and Sperling 2007; Putnam et al. 2007; Ryan et al. 2007; Simionato et al. 2007; Yamada et al. 2007; Larroux et al. 2008), investigators began to look at other components of gene regulatory networks for molecules that might show a different evolutionary pattern. Initial surveys of miRNAs across the animal kingdom demonstrated a very compelling feature of miRNA evolution when compared with the evolution of the protein-coding repertoire: miRNA families are continuously being added to bilaterian lineages through evolutionary time such that vertebrates, for example, were characterized by the possession of miRNA families not found in arthropods, and vice versa (Hertel et al. 2006; Sempere et al. 2006; Prochnik et al. 2007; Heimberg et al. 2008). Fur- ther, these authors showed that miRNAs, once integrated into the genomic regulatory circuitry, are only rarely secondarily lost, and the mature miRNA sequences are under intense negative selection. Because of these features, evolutionary biologists have attempted to reconstruct the phylogenetic history of the miRNAs found in vertebrates, insects, and nematodes, and noted the following. First, known miRNAs neither could be found nor could be detected in sponges (Sempere et al. 2006; Prochnik et al. 2007). Second, bilaterians had a greatly EVOLUTION & DEVELOPMENT 11:1, 50–68 (2009) DOI: 10.1111/j.1525-142X.2008.00302.x & 2009 The Author(s) Journal compilation & 2009 Wiley Periodicals, Inc. 50