Molecular machines that assemble polymers in a programmed sequence are ubiquitous in nature and are necessary for life. Nanotechnology can mimic such assembly – synthetic machines made of self-assembling DNA molecules can be used to link natural and artificial building blocks in defined sequences by covalent bonds.

Now, a team of researchers led by Andrew Turberfield of the Department of Physics at the University of Oxford and Rachel O’Reilly at the Department of Chemistry at Warwick University is reporting on a new DNA nanomachine that can be used to execute a molecular programme that produces peptides or olefin oligomers with a defined sequence. To be more precise, alternate steps in a so-called DNA hybridization chain reaction trigger “acyl transfer” or “Wittig” reactions that link specified building blocks to a growing covalently bonded oligomer by peptide or carbon–carbon double bonds, respectively. Both these chemistries involve transfer reactions in which a covalent bond formed at the active end of the growing oligomer occurs at the same time as the building block is cleaved from its DNA adapter.

Making biomimetic peptides or completely unnatural oligomers

This process is very similar to one that occurs in nature, in which amino-acid residues are transferred from aminoacyl transfer-RNAs to a growing polypeptide chain in the ribosome. Both chemistries can accommodate building blocks with a wide range of side chains, allowing researchers to synthesize a variety of sequence-controlled products, from biomimetic peptides to completely unnatural oligomers.

“Initially, we attach individual chemical reactants to DNA ‘hairpins’ that don’t react with one another since they only interact at a low rate,” explains team member Richard Muscat of Oxford University. “When we add an initiator strand, however, the DNA hairpins start to bind to each other, and as they do so, bring the reactants closer together, causing them to react.”

A chain of DNA hairpins and a finished chemical product

As more DNA hairpins bind, more reactants are added, triggering additional reactions. “In the end, we are left with a chain of DNA hairpins and a finished chemical product.”

The order in which the hairpins assemble is determined by sequences known as toeholds that control the interactions between the different hairpins, he adds. “By changing the sequencers in the toeholds, the order in which the hairpins react and therefore the order in which the chemical reactants are brought together can be controlled.”

And that is not all: the DNA attached to the finished molecules contains a record of the individual reaction stages. “Through processes of amplification and DNA sequencing, it is possible for us to actually read this record back,” Muscat tells “It might therefore be possible to use the molecular assembler to template a whole range of products, creating an entire library of products. The DNA tag could even help identify a specific product in this library – for example one that binds to a target protein that would otherwise be difficult to identify directly because of its low concentration. Such a technique might come in useful for screening for molecules that bind targets or for producing molecules in response to specific biological signals.”

The research is detailed in Nature Chemistry doi:10.1038/nchem.2495.