MicroRNA evolution in animals: origins of miRNA biogenesis machinery.

MicroRNAs (miRNAs) are central components of post-transcriptional gene regulation in animals, acting through Argonaute-containing effector complexes to modulate the stability and translation of target mRNAs. In bilaterian animals, including humans, large miRNA repertoires are deeply integrated into gene regulatory networks that control development, cell differentiation, and physiological homeostasis. The pervasive regulatory role of miRNAs in animals is well established through mechanistic and functional studies spanning multiple model systems^1,2.

Given their broad functional impact, a long-standing question in evolutionary biology has been how and when miRNAs (and the molecular machinery required for their biogenesis) first emerged. Early perspectives were strongly influenced by bilaterian model organisms, in which miRNA complements are large and evolutionarily conserved. This led to the hypothesis that miRNA evolution was closely tied to the emergence of multicellularity and increasing regulatory complexity. More recent comparative work, combining small RNA sequencing with comparative genomics across early-branching animals and unicellular holozoans, has provided the experimental basis to test these ideas directly³.

The first decisive experimental evidence for an early origin of animal miRNAs came from small RNA sequencing studies in non-bilaterian animals. Sponges, among the earliest-branching metazoan lineages, were shown to express bona fide miRNAs with defining characteristics of canonical animal miRNAs, including precise processing from hairpin precursors and reproducible expression patterns.

These findings demonstrated that miRNA-based regulation predates the emergence of bilaterians and is not restricted to animals with complex tissues or organs³. This conclusion was reinforced by genomic analyses of early-branching animals. For example, the genome of the coral reef sponge Amphimedon queenslandica revealed a surprisingly rich repertoire of regulatory genes, supporting the idea that regulatory complexity can precede morphological complexity during evolution⁴.

Importantly, while early branching animals express miRNAs, their miRNA complements are modest in size compared with those of bilaterians. This pattern suggests that miRNAs originated early but underwent substantial expansion and diversification later in animal evolution, particularly along the bilaterian stem.

Central to understanding miRNA evolution is the Microprocessor complex, which defines canonical animal miRNA biogenesis. In bilaterians, the Microprocessor consists of the RNase III enzyme Drosha and its RNA-binding partner DGCR8 (also known as Pasha). This nuclear complex cleaves primary miRNA transcripts (pri-miRNAs) to release precursor hairpins (pre-miRNAs), thereby establishing the precise ends of mature miRNAs. The resulting pre-miRNAs are subsequently exported from the nucleus and further processed by Dicer in the cytoplasm, before being loaded into Argonaute proteins to form the RNA-induced silencing complex (RISC). This sequence of processing steps defines canonical animal miRNA biogenesis and distinguishes miRNAs from other small RNA pathways, such as siRNAs and piRNAs, which follow a different route⁵.

For many years, the Microprocessor was assumed to be a metazoan innovation. This view was overturned by comparative genomic and transcriptomic analyses of unicellular relatives of animals. A pivotal study demonstrated that ichthyosporeans (single-celled relatives of animals) encode clear homologs of Drosha and Pasha and express bona fide miRNAs detectable by small RNA sequencing. This combination of genomic and experimental evidence strongly supports a pre-metazoan origin of the canonical animal miRNA machinery, indicating that miRNAs did not arise strictly as a consequence of multicellularity⁶.

The evolutionary history of miRNAs is not, however, a linear story of retention and expansion. Genomic and transcriptomic analyses of the ctenophore Mnemiopsis leidyi (a comb jelly) failed to identify canonical miRNAs or core Microprocessor components such as Drosha and DGCR8. This absence suggests either secondary loss of the miRNA pathway in this lineage or the evolution of a fundamentally different regulatory architecture⁷. This finding is particularly notable because it contrasts with the presence of miRNAs in other early-branching animals such as sponges and cnidarians (corals, anemones and jellyfishes).

For example, small RNA sequencing in cnidarian species such as Nematostella vectensis (the starlet sea anemone) revealed dozens of expressed miRNAs, confirming the presence of miRNA-based regulation outside Bilateria. However, only a limited subset of these miRNAs is shared with bilaterians, indicating extensive lineage-specific miRNA innovation and turnover^3,8.

These evolutionary insights have been enabled by advances in next-generation sequencing. Small RNA sequencing allows direct detection of miRNAs, while genome sequencing permit the anchoring of small RNA reads to hairpin-forming loci, distinguishing true miRNAs from degradation products or other small RNAs. At the same time, next-generation sequencing introduces challenges in non-model organisms, including incomplete genomes, contamination, and permissive annotation criteria, which can inflate false positives.

To address these issues, curated databases have become essential. miRBase has long served as a repository for published miRNA sequences, but heterogeneity in annotation standards has complicated deep evolutionary comparisons. In response, MirGeneDB was developed as a curated database applying uniform criteria to miRNA gene annotation across metazoans, with particular value for early-branching and non-canonical species⁹.

Taken together, experimental and comparative evidence supports a model in which canonical miRNAs and their processing machinery originated before the emergence of animals and subsequently expanded and diversified in bilaterians. At the same time, lineage-specific losses and divergent evolutionary trajectories, exemplified by ctenophores, caution against simplistic narratives. Continued progress will depend on improved genomic resources for non-model organisms, careful application of next-generation sequencing technologies, and consistent annotation frameworks capable of supporting robust evolutionary inference across the full diversity of animal life.