Evolutionary biology

Small regulatory RNAs pitch in

Article metrics

How did organismal complexity evolve at a cellular level, and how does a genome encode it? The answer might lie in differences, not in the number of genes an organism has, but rather in the regulation of gene expression.

It is commonly believed that complex organisms arose from simple ones. Yet analyses of genomes and of their transcribed genes in various organisms reveal1,2 that, as far as protein-coding genes are concerned, the repertoire of a sea anemone — a rather simple, evolutionarily basal animal — is almost as complex as that of a human. Grimson et al.3 might have the answer to this paradox. They report in this issue (page 1193) that differences in complexity between some evolutionarily ancient, morphologically simple animals and the highly sophisticated primates could be explained, at least in part, by the regulation of gene expression by small RNAs.

One of the three main classes of noncoding regulatory RNA expressed by bilaterians — organisms that have a bilateral symmetry — is the microRNAs (miRNAs). These sequences exist as RNA duplexes (roughly 21–24 nucleotides long), one strand of which usually binds to its target messenger RNA, reducing the protein output from that message. Each miRNA has tens to hundreds of target mRNAs, adding a notable level of complexity to the regulation of gene transcription4,5. In humans, hundreds of miRNAs have been identified, yet the search for related sequences in evolutionarily ancient and morphologically simple non-bilaterian animals (such as sponges, comb jellies and cnidarians) has led to only two to three putative miRNA sequences, found in the sea anemone Nematostella vectensis6,7,8.

To take a closer look at the problem, Grimson et al.3 used deep-sequencing techniques (454 Life Sciences and Illumina sequencing machines) to analyse what was presumably the nearly full complement of small noncoding RNAs in a range of morphologically simple organisms. These included a unicellular choanoflagellate (Monosiga brevicollis) and several representatives of non-bilaterian multicellular phyla: a cnidarian (N. vectensis), a sponge (Amphimedon queenslandica) and a placozoan (Trichoplax adhaerans). The present study3 also demonstrates the greater power of unbiased deep sequencing at near-saturation levels compared with approaches that rely on sequence similarities alone.

In the sponge, the authors find eight different miRNAs. Moreover, they report that the sea anemone has, in fact, not just three but at least 40 different miRNAs. Although, compared with the previous estimates6,7,8, 40 miRNAs is clearly more significant, when compared with the 147 miRNAs identified in the fruitfly Drosophila and the 677 human miRNAs, this is not a high number. A pattern emerges, however: miRNA diversity in these three organisms seems to correlate with an increase in their relative morphological complexity (Fig. 1), even after accounting for some inflation in the number of human miRNAs owing to possible mis-annotation of certain sequences as miRNAs.

Figure 1: Animal miRNAs and morphological complexity.

Grimson et al.3 (data along red lines) reveal the evolutionary origin of animal miRNAs by examining organisms at the base of the animal tree. Combining their data with previous work, three different measures of complexity become apparent: the number of protein-coding genes, total number of neurons and number of miRNAs. There is relatively little correlation between morphological complexity and the number and diversity of protein-coding genes. However, miRNA number correlates well with the organism's total number of neurons. Indeed, a large proportion of vertebrate miRNAs are expressed in the nervous system. These data also show the dynamic nature of the miRNA complement in each lineage, particularly visible in rapidly evolving species (Oikopleura and fruitfly).

All but one of the sea anemone miRNAs seem to lack any resemblance to bilaterian miRNAs, with the one exception being almost identical in sequence to the bilaterian miRNA known as miR-100. The similarity of miR-100 sequences among sea anemone, fly and vertebrate pushes the origin of animal miRNAs back to at least the last common ancestor of sea anemones and vertebrates, some 600 million years ago. The eight sponge miRNAs do not match those of either the sea anemone or bilaterians. But, as the miRNA biosynthesis machinery is clearly fully conserved in sponges3, it seems plausible that the origin of animal miRNA dates back to the last ancestors of all animals — the Urmetazoa.

Grimson et al. could not convincingly detect miRNAs in either the placozoan (Trichoplax is a very simple organism, more like a multicellular amoeba than an animal) or the choanoflagellate, putatively the closest unicellular sister group of multicellular animals. Both of these species also seem to lack Pasha3, a protein crucial for the biosynthesis of miRNAs. Unlike other proteins of the miRNA biosynthesis machinery, Pasha's only known function is in miRNA processing. Loss of the gene encoding Pasha should therefore directly and only affect miRNA generation. Consequently, as the latest, but not uncontested, molecular phylogeny9 places Placozoa between the sponges and Cnidaria, the absence of miRNAs in Placozoa is probably due to secondary loss, whereby a single gene loss may have led to the complete loss of miRNAs in these organisms. Moreover, assuming a role for miRNAs in supporting higher morphological complexity, their loss, together with other gene losses9, may have led to the secondary morphological simplification of the placozoan body plan.

Besides miRNAs, bilaterians possess another type of (slightly longer) small RNA called piRNA. This comes in two forms, both typically expressed in the germ line — the cells that produce gametes. Grimson et al. find high numbers of both piRNA subclasses in sponges and sea anemones. Strictly speaking, members of these two phyla do not have a classical germ line. Instead, multipotent stem cells, which can differentiate into diverse cell types — both somatic and germline — ensure lifelong production of gametes. Future work will undoubtedly aim to elucidate the function of piRNAs during stem-cell differentiation and gamete formation in such basal metazoans.

The finding that two basal metazoans possess relatively high numbers of miRNAs and piRNAs that have almost no related sequences in Bilateria suggests that cnidarian and sponge miRNAs have largely diversified independently of bilaterian miRNAs. But is this a rule or an exception in the evolution of these organisms? One view is that, once fixed in the population, miRNAs are seldom lost. Consequently, there would be a constant increase in diversity, correlating with an increase in morphological complexity7. Indeed, Grimson and colleagues' data confirm a correlation between miRNA diversity and morphological complexity as measured, for instance, by the total number of neurons in an organism (Fig. 1). But work on rapidly evolving species, such as several species of Drosophila and the planktonic urochordate Oikopleura, revealed10,11 both extensive gain and loss of many miRNAs over relatively short evolutionary timescales. This indicates the dynamic evolutionary nature of many miRNAs, and explains the diversity of the miRNA complement in different lineages.

What could simple organisms such as the sea anemone and sponges use their 'respectable' number of miRNAs for? With no functional data available, one can only speculate. Many cnidarians have a rather complex, multi-stage life cycle. So, as in worms, flies and mammals, miRNAs could be involved in the timing of these different developmental stages. Another striking feature of cnidarians is their enormous capacity to regenerate. This property relies on the continuous differentiation of stem cells, possibly requiring miRNAs as part of a well-buffered molecular patterning system to ensure homeostasis. As for the role of piRNAs, in Bilateria they are usually involved in silencing mobile genetic elements called transposons. Intriguingly, compared with other animal genomes, that of the sea anemone also contains a remarkable diversity of transposable elements. Control of genome integrity by piRNAs during gamete formation might therefore be crucial for the survival of these basal metazoans over evolutionary timescales.


  1. 1

    Putnam, N. H. et al. Science 317, 86–94 (2007).

  2. 2

    Technau, U. et al. Trends Genet. 21, 633–639 (2005).

  3. 3

    Grimson, A. et al. Nature 455, 1193–1197 (2008).

  4. 4

    Bartel, D. P. Cell 116, 281–297 (2004).

  5. 5

    Ambros, V. Nature 431, 350–355 (2004).

  6. 6

    Pasquinelli, A. E. et al. Nature 408, 86–89 (2000).

  7. 7

    Sempere, L. F., Cole, C. N., McPeek, M. A. & Peterson, K. J. J. Exp. Zool. B Mol. Dev. Evol. 306, 575–588 (2006).

  8. 8

    Prochnik, S. E., Rokhsar, D. S. & Aboobaker, A. A. Dev. Genes Evol. 217, 73–77 (2007).

  9. 9

    Srivastava, M. et al. Nature 454, 955–960 (2008).

  10. 10

    Fu, X., Adamski, M. & Thompson, E. M. Mol. Biol. Evol. 25, 1067–1080 (2008).

  11. 11

    Lu, J. et al. Nature Genet. 40, 351–355 (2008).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.