Artificial DNA letters beyond A, T, G, C break a fundamental pairing rule to produce nanostructures with new shapes, far greater durability, and an unexpected ability to self-sort.
(Nanowerk Spotlight) Among the defining features of the double helix, as Watson and Crick described it in 1953, is size complementarity: large purines always pair with small pyrimidines across the helix. A with T, G with C. This pairing rule gives DNA its uniform diameter and regular geometry, and when researchers began engineering synthetic nanostructures from DNA, the same rule carried over. Every origami box, tile lattice, and walking molecular robot assembled to date has followed it. Size complementarity has been a central constraint of DNA nanotechnology.
But the four-letter alphabet that serves genetics so well creates practical problems for engineering. Just as DNA strands occasionally mispair in living cells, they do the same in synthetic assemblies, and these errors multiply as structures grow larger and require more strands. The largest constructions take days to form, and fewer than 1 in 100 copies assembles correctly. Even successfully assembled structures face a second challenge: natural DNA degrades quickly inside living cells, where enzymes called nucleases break it apart, severely limiting biomedical applications.
These barriers, however, are not fundamental. The hydrogen-bonding logic of base pairing leaves room for far more combinations than nature explored. By reshuffling the pattern of hydrogen bond donors and acceptors on the bases, chemists can create artificial nucleotides that pair according to the same rules as natural DNA but with different partners. Up to 12 such building blocks forming six distinct pairs are possible, all fitting within the standard double-helical geometry.
This synthetic platform, known as AEGIS (anthropogenic evolvable genetic information system), shares DNA’s sugar-phosphate backbone, so it behaves like DNA in solution and can be handled with familiar laboratory tools. AEGIS already underpins billion-dollar diagnostic products and has enabled laboratory evolution of new receptors and catalysts.
But in all of these applications, AEGIS molecules function the way DNA does in biology: they store and transmit information through their sequences. DNA nanotechnology asks something different of the molecule. It treats DNA not as a message but as a physical building material, exploiting its base-pairing rules to fold and tile strands into engineered shapes. Whether AEGIS could also work as a structural material remained an open question.
The reseaerchers assembled DNA nanostructures from synthetic base pairs that pair purines with purines to create “fat” double helices wider than normal, and pyrimidines with pyrimidines to create “skinny” helices narrower than normal. Both types self-assembled into well-defined architectures, proved far more durable than conventional structures, and spontaneously sorted themselves by type when mixed.
Design of DNA nanostructures with AEGIS base pairs. (A) Comparison of canonical Watson-Crick base pairs (A-T and G-C) and fat (B-P and D-X) and skinny (S-Z and T-K) AEGIS base pairs. (B) Schematics of DNA double helices formed by canonical Watson-Crick base pairs and AEGIS base pairs. (C to F) A modified version of “double crossover (DX)” DNA tile design was used to investigate self-assembly of fat DNA and skinny DNA nanostructures. The DNA tiles are connected via sticky-end (colored coded) adhesion. (C) Schematics showing the fat DNA tile design and assembly. (D) SYBR Gold–stained native polyacrylamide gel showing the assembly of fat DNA strands. (E) Schematics showing the skinny DNA tile design and assembly. (F) SYBR Gold–stained native polyacrylamide gel showing the assembly of skinny DNA strands. The skinny DNA bases with nitro group (Z and K) quenched fluorescence, resulting in a negative signal. (Image: Reproduced from DOI:10.1126/sciadv.aeb6713, CC BY) (click on image to enlarge)
To test the concept, the team chose a well-established design motif: the double-crossover tile, a small DNA unit that links to neighbors through complementary sticky ends to form extended periodic structures. They built fat tiles from two purine-purine base pairs and skinny tiles from two pyrimidine-pyrimidine pairs, annealed them, and examined the products. The two tile types produced markedly different architectures from the same basic design. Fat tiles curved and closed into micrometer-long nanotubes averaging about 13.7 nm in diameter. Skinny tiles assembled into flat two-dimensional lattices.
The reason for this divergence lay in the geometry of the tile-to-tile junctions. The angle between three connected fat tiles measured about 37°, tight enough to drive the curvature that rolls a lattice into a tube. Skinny tiles opened to roughly 54°, most likely because pyrimidine-pyrimidine stacking at the junctions is weaker, producing a flatter conformation that resists curling. Standard DNA tiles formed nanotubes with nearly the same 36° angle as fat tiles but with a smaller average diameter of 9.8 nm, consistent with the thinner profile of conventional double helices.
The same design rules produced three distinct structural outcomes depending solely on the type of base pair used. Beyond geometry, a central question was whether the new base pairs would also make nanostructures more durable.
When the researchers heated preassembled nanotubes, fat DNA nanotubes remained largely intact up to 50 °C and some survived even at 60 °C. Regular DNA nanotubes began falling apart at 45 °C and disappeared above 50 °C. Resistance to nuclease digestion proved equally decisive. After treatment with DNase I, standard DNA nanotubes mostly vanished within 30 minutes, while fat DNA nanotubes and skinny DNA lattices remained largely intact after 3 hours. Even individual AEGIS strands, not yet assembled into structures, degraded far more slowly than their natural DNA equivalents.
The researchers then asked whether fat and skinny strands could coexist within a single structure. Because a purine-only fat strand can pair with a complementary pyrimidine-only skinny strand through conventional purine-pyrimidine pairing, hybrid tiles are feasible in principle. Five of six hybrid tile designs successfully formed nanotubes, though quality diminished as more skinny strands replaced fat ones, likely because of slight size mismatches between the different duplex geometries.
A final experiment tested whether fat and skinny strands would segregate when given the choice. The team mixed two different hybrid nanotubes and reannealed them. Rather than producing scrambled products, the fat and skinny strands sorted themselves, reassembling into separate fat nanotubes and skinny two-dimensional lattices. Microscopy confirmed that the resulting structures matched the dimensions and angles of their pure counterparts.
This self-segregation points toward programmable shape transformations, where structures could reorganize through strand displacement or temperature changes. Moving from four nucleotide letters to eight also increases the information density of each strand, from two bits per position to three. With more letters available, each sequence becomes more distinctive, reducing the chance that strands find wrong partners during assembly.
Combined with enhanced thermal and enzymatic stability, these properties open design territory that conventional DNA cannot reach, particularly in cellular and medical settings where natural DNA structures fall apart before they can function. This initial exploration tested only a subset of what AEGIS can offer. The system can accommodate up to 12 building blocks forming six orthogonal pairs, all fitting within the same helical geometry that governs natural DNA. That combinatorial space, together with the demonstrated compatibility with established design rules, suggests that synthetic genetic alphabets could substantially extend the reach of DNA nanotechnology.
For authors and communications departmentsclick to open
Lay summary
Prefilled posts
Nanowerk Newsletter
Get our Nanotechnology Spotlight updates to your inbox!
Thank you!
You have successfully joined our subscriber list.
Become a Spotlight guest author! Join our large and growing group of guest contributors. Have you just published a scientific paper or have other exciting developments to share with the nanotechnology community? Here is how to publish on nanowerk.com.
