The basic flow of genetic information is “DNA → RNA → protein,” yet regulation in the RNA phase is crucial for development and disease. In 1993, while studying developmental-timing mutants of the nematode C. elegans, Ambros and Ruvkun found that the gene lin-4 produces a 22-nucleotide non-coding RNA. They showed that this short RNA pairs partially with the 3′ untranslated region of lin-14 mRNA and blocks its translation, switching the worm from larval to adult programs. Because of its tiny size, the molecule was named “microRNA,” and hundreds to thousands of similar RNAs were later discovered in humans and many other animals. MicroRNAs are processed by the enzymes Drosha and Dicer and loaded onto Argonaute proteins to form a complex that silences target mRNAs. This mechanism acts like a fine-tuner for processes such as cell differentiation, tumor suppression and neural circuit formation. Thanks to this knowledge, measuring microRNAs in blood is being tested for early cancer diagnosis, and drugs that block harmful microRNAs are under development. The discovery overturned the old idea that only proteins control genes and rewrote biology textbooks.

Gene regulation extends beyond transcription, involving post-transcriptional mechanisms that control mRNA stability and translational efficiency. Ambros and Ruvkun focused on the heterochronic gene circuit lin-4/lin-14 in C. elegans and reported that the lin-4 locus encodes 22- and 61-nt small RNAs. They identified several eight-base complementary elements in the lin-14 3′UTR; disrupting this pairing recapitulated the developmental-timing phenotype, proving functional interaction. Subsequent work by Ruvkun uncovered let-7, a microRNA conserved up to mammals, establishing that microRNA-mediated regulation is evolutionarily ancient. Today about 2,600 mature miRNAs are annotated in the human genome, and they are thought to hierarchically regulate a large fraction of protein-coding genes. miRNAs are transcribed by RNA polymerase II as primary transcripts (pri-miRNAs) that are processed in the nucleus by the microprocessor complex (Drosha–DGCR8) into precursor miRNAs (pre-miRNAs). The pre-miRNA is exported to the cytoplasm by Exportin-5, trimmed by Dicer, and one strand is loaded onto an Argonaute-containing RISC. RISC binds target mRNAs via partial complementarity centered on the seed region (nucleotides 2–8) and recruits deadenylase complexes such as CCR4–NOT to shorten poly(A) tails. Consequently the mRNA is degraded or its translation is repressed, with the degree of regulation tuned by pairing outside the seed and by site context. Evolutionary analyses show that bursts of new miRNA genes coincide with the emergence of multicellularity and organ diversification, suggesting that miRNA expansion enabled greater regulatory complexity. Clinically, miRNA profiling is being adopted as a biomarker for cancers and cardiovascular diseases, and nucleic-acid therapeutics such as antagomirs are undergoing clinical trials. Thus, microRNAs have produced widespread impact, bridging fundamental biology with translational research.

lin-4-dependent repression produces a graded reduction in LIN-14 protein via cooperative action of multiple miRNA-binding sites in the 3′UTR, as quantified by Ruvkun’s in-vivo reporter assays. Subsequent development of target-prediction algorithms incorporating seed 7-mer complementarity and thermodynamic stability, exemplified by the let-7 family, enabled systematic mapping of miRNA-regulated networks. RISC repression relies primarily on CCR4-NOT mediated deadenylation followed by DCP1/2 decapping; ribosome profiling indicates that translational inhibition can precede or coincide with deadenylation. Experimental evolution systems have demonstrated that miRNA gene duplication and seed divergence increase network redundancy and phenotypic robustness, whereas deletion of miRNA clusters provokes epistatic expression noise. High-resolution CLIP methods (HITS-CLIP, PAR-CLIP, CLEAR-CLIP) identify Argonaute binding sites in vivo at single-nucleotide accuracy and reveal non-canonical interactions in exon–intron junctions and upstream ORFs. Structural analyses of Drosha/DGCR8 show that the upstream carboxylate side chain binds a proline-rich motif, recognizing the double- to single-stranded junction in pri-miRNA. Cryo-EM structures establish that the distance between Dicer’s PAZ and RNase III domains defines the 22-nt step size of pre-miRNA cleavage. Disease-related hotspot mutations in the RNase IIIb domain of DICER1 create 5p/3p processing biases and contribute to altered mRNA signatures in specific tumors. Loss or gain of 3′UTR target sites through SNPs or RNA editing correlates with GWAS signals, and miRNA-based post-GWAS analyses aid functional variant annotation. Recent small-molecule screens have identified compounds that selectively inhibit the Lin28–let-7 axis and show therapeutic benefit in diabetes and nerve-regeneration models. Clinical-stage LNA-modified antagomirs such as MRG-110 and RG-012 are combined with GalNAc conjugation or exosome loading to optimize delivery to organs with high target-miRNA expression. These advances accelerate system-level understanding of miRNA networks and their translation into precision medicine.