A molecule that stores information, folds itself into working shapes, and reshapes itself when a signal arrives would be a gift to any engineer of the very small. Biology has been hiding one in plain sight. RNA, long treated as the passive courier that carries instructions between DNA and the protein factories of the cell, is being drafted into construction duty: genetic switches controlled by pills, logic circuits made of genes, origami structures folded from a single strand, and self-assembling scaffolds that form inside living cells. A survey of the field published in Nature’s Nanoscience and Nanotechnology supplement argues that RNA’s versatility makes it programmable matter, a material whose information and function share one four-letter code.
RNA’s appeal as a construction material comes from a dual nature. Alone among the cell’s working molecules, it both carries genetic information and performs the catalytic and binding functions normally assigned to proteins. Its code is minimal: four nucleotides, adenine, uracil, cytosine, and guanine, which pair with each other in predictable ways, allowing a strand to fold back on itself into loops, stems, and complex three-dimensional shapes with binding sites on their surfaces. Unlike DNA nanostructures, which are famously stable but rigid, RNA can remodel itself when a specific molecule arrives or the environment changes. This dynamism is what makes it programmable matter: a machine that can be switched, gated, and reshaped on command.
The simplest machines are riboswitches, messenger RNA structures that act as molecular toggles, turning protein production on or off when they bind a target. Naturally occurring riboswitches respond to metabolites inside cells, which are awkward handles for a therapist or an engineer, so researchers have been screening RNA sequences for switches that respond to drugs already approved for human use. In 2025, a team led by Jorg Hartig at the University of Konstanz demonstrated a riboswitch built on a bacterial aptamer that recognizes oxypurinol, the active metabolite of the gout drug allopurinol. In mammalian cells, the switch boosted expression of a target gene by up to 154-fold at a two-millimolar dose of oxypurinol, and by 65-fold at one millimolar. The researchers also solved the crystal structure showing how the drug-like molecule nestles into the RNA. Hartig’s lab says newer switches can amplify expression as much as 1,000-fold.
The therapeutic vision is a gene therapy with a dimmer switch. A patient with a metabolic deficiency might receive a corrected gene delivered by a viral vector or lipid nanoparticle, with a riboswitch embedded in its message. Swallowing a pill would turn the gene on for the day; skipping the pill would let it rest, avoiding the toxic buildup of a metabolite from a gene that never sleeps. The approach has been tested in mouse models, and the group is working toward clinical applications. More elaborate designs chain switches together: Jongmin Kim’s team at the Pohang University of Science and Technology has built RNA logic gates that activate a gene only when the right combination of signals is present, combining drug-responsive riboswitches with toehold switches that detect specific RNA molecules. The circuits are being developed to steer engineered therapeutic bacteria that could limit tissue damage in inflammatory bowel disease.
Folding, not switching, is the frontier. RNA origami applies the base-pairing rules of DNA origami to a more pliable material. Hao Yan’s group at Arizona State University has used RNA origami as the basis of a cancer vaccine: free RNA outside cells is a danger signal, associated in the immune system’s mind with viral infection, and structures built from it can be tuned to wake up dormant immune cells around a tumor. Coupled with immune-stimulating molecules and injected into mouse models of cancer, the constructs produced strong tumor-specific immune responses and improved survival in breast cancer and melanoma models, and the group plans to move toward clinical testing. Inside cells, the challenge is different: RNA folds almost as soon as it is transcribed, so separately encoded strands rarely get a chance to meet. Most designs therefore rely on a single strand engineered to adopt its shape as it is made, with stem-loop modules that pair into kissing-loop assemblies. In recent work, Fei Zhang’s group at Rutgers University showed that such strands self-assemble inside the nuclei of human cells into zigzag scaffolds, rings, and fishnet-like meshes, structures that could display protein-binding domains or reach out to chromosomes and regulate genes at their source.
A parallel line of work builds synthetic condensates, fluid droplets that mimic the phase-separated compartments cells use to concentrate enzymes and coordinate reactions. Because they are fluid rather than walled, these droplets can merge, split, and redistribute their contents, pooling the ingredients for a particular reaction on demand. Proteins are notoriously hard to engineer into such structures because their folding is difficult to control; RNA’s predictable base pairing makes it an attractive alternative for building them.
The field’s limits are acknowledged in the same breath as its promise. Researchers still do not fully understand how RNA behaves inside living cells, where crowding, RNA-binding proteins, and degradation machinery all intervene between a design and its function. Translating folded structures into working technologies will require closing that gap. But the direction of travel is clear: the molecule that carried the cell’s messages is being drafted into the cell’s machine shop. A strand of RNA that folds itself into a switch, a logic gate, or a scaffold is not just a biological curiosity. It is a material in which information and function share a single code, exactly what a builder of nanoscale machines would order.
References
Michael Eisenstein, The dark horse of biology: how RNA is becoming a nanotool maker’s dream. Nature 655, S2-S5 (2026). DOI: 10.1038/d41586-026-02180-6.
Vera Hedwig et al., Engineering oxypurinol-responsive riboswitches based on bacterial xanthine aptamers for gene expression control in mammalian cell culture. Nucleic Acids Research 53, gkae1189 (2025). DOI: 10.1093/nar/gkae1189.