DNA Origami
A virus's genome, folded by hundreds of short staple strands, into any nano-shape you sketch on paper.
DNA origami self-assembles a single ~7,000-base scaffold strand (typically the M13 phage genome) into a precisely defined shape by mixing it with ~200 short ~32-base staple oligonucleotides. Each staple's sequence dictates which two regions of the scaffold it crosses over and pins together. After a thermal anneal from 95 °C to room temperature over a few hours, the molecules find their lowest-energy fold — yielding rectangles, smiley faces, gears, boxes, or arbitrary 3D voxel shapes with ~6 nm feature resolution and ~1-nm placement accuracy for individual atoms.
Paul Rothemund's 2006 method begins with a circular single-stranded DNA scaffold (M13mp18 bacteriophage genome, 7,249 bases). A computer-aided-design tool (caDNAno, scadnano) lays out the desired target shape as a raster of parallel double-helix segments, then auto-generates ~200 staple strands whose sequences are reverse-complements of two short scaffold regions separated along the path. Each staple bridges those regions, forcing a crossover. The scaffold and staples (10× molar excess) are mixed in Mg²⁺-buffered solution and slowly cooled from 95 °C to 20 °C. Watson-Crick base pairing is highly specific, so each staple finds its unique target sites; cooperative folding drives the structure to the designed shape with >90 % yield. 3D variants (Douglas et al. 2009) stack helices into honeycomb or square lattices to build wireframes, polyhedra, and curved shells. Modifications — fluorophores, gold nanoparticles, antibodies, aptamers — can be attached at any individual staple position, giving sub-nanometer placement of functional payloads.
In plain terms
Imagine a really long shoelace mixed with a pile of tiny paper clips. Each paper clip knows exactly which two spots on the shoelace to pinch together, because its shape only fits there. Drop the whole pile into water, warm it up, cool it down — and the shoelace folds itself into a smiley face the size of a virus. DNA origami uses the same trick: a virus's genome is the shoelace; ~200 short DNA strands are the paper clips, and the smiley face can be any shape you want.
Also called: scaffolded DNA nanotechnology, Rothemund origami.
Why it matters
- Programmable nanoscale placement: any of ~200 staple positions can be modified to attach a fluorophore, gold particle, drug, or single antibody — yielding precision unmatched by chemical synthesis.
- Cryo-EM standards: DNA origami nanorulers and brightness standards calibrate single-molecule fluorescence and super-resolution microscopes worldwide.
- Drug-delivery vehicles: barrel and tetrahedron origami carry chemotherapeutics, mRNA, or aptamers and open in response to cell-surface markers — preclinical demonstrations in mice.
- Computing & sensing: tile-based DNA origami arrays implement programmable molecular logic and single-molecule biosensors.
Where it lives
- GATTAquant DNA-PAINT nanorulers (commercial calibration standards)
- Origami-templated patterning for single-photon emitters
- Caltech / TU Munich therapeutic origami nanocarriers
- Bathe lab MIT — automated 3D wireframe origami
Key numbers
- Scaffold length: ~7,249 bases (M13mp18 phage genome)
- Staple count (typical): ~200 strands (32 bases each)
- Feature resolution: ~6 nm (Helix-to-helix spacing)
- Folding yield: >90 % (After thermal anneal in Mg²⁺ buffer)
- First demo: Rothemund, Nature 2006
Surprising facts
- Rothemund's 2006 paper showed that mixing one scaffold strand with hundreds of staples in a tube and heating-then-cooling produces ~100 nm 'smiley faces' or maps of the Western Hemisphere with feature accuracy better than 6 nm — purely by self-assembly, no top-down lithography.
- A single DNA-origami breadboard can position individual gold nanoparticles, fluorophores, or even single proteins to within ~1 nm of a designed target — better than any commercial photolithography.
- Folding takes a tube on a thermocycler, ~$1 of staple oligos, and a few hours; from sketch to nanostructure is faster than ordering a custom PCB.
Open questions
- Stochastic mis-folding pathways: Even with >90 % yield, a small fraction of origami structures lock into kinetic traps — partially folded intermediates that never relax to the design target. The exact pathway by which a scaffold finds the lowest-energy fold among ~10²⁰⁰ possibilities is not fully predicted by current biophysical models; molecular-dynamics simulations track only short fractions of the anneal.
- Mg²⁺ dependence is non-monotonic: Folding yield depends on Mg²⁺ concentration in a non-monotonic way that varies by design — too little and helices repel, too much and unwanted aggregation dominates. The optimum is found empirically per-shape and is not predicted by first-principles electrostatics.
A short history
Ned Seeman proposed branched DNA junctions for nanostructure construction at NYU in the early 1980s, demonstrating the first immobile 4-arm junction in 1983. Two decades of incremental work followed, including DNA tile assemblies. Paul Rothemund's 2006 Nature paper unified the approach with a single long scaffold and many short staples, demonstrating ~100 nm 2D shapes with sub-10 nm precision. Douglas, Dietz, Liedl, Högberg, Shih and others extended origami to 3D in 2009 (multi-layer honeycomb lattice), to wireframes (2015), and to gigadalton-scale assemblies. The field's tooling — caDNAno (Douglas 2009), scadnano (Doty 2020) — makes design routine.
Sources
- Rothemund, P.W.K., Nature (2006) — Folding DNA to create nanoscale shapes and patterns (https://www.nature.com/articles/nature04586)
- Douglas et al., Nature (2009) — Self-assembly of DNA into nanoscale 3D shapes (https://www.nature.com/articles/nature08016)
- Dietz, Douglas & Shih, Science (2009) — Folding DNA into twisted and curved nanoscale shapes (https://www.science.org/doi/10.1126/science.1174251)
- scadnano design suite (https://scadnano.org/)