Peptide–oligonucleotide conjugates as nanoscale building blocks for assembly of an artificial three-helix protein mimic

Designed peptide-based structures have been shown to yield artificial proteins¹ and even in a few cases nanoscale objects^2,3,4. Simultaneously, oligonucleotides (ONs) have been intensively used for nanotechnology^5,6, including so-called DNA origami⁷, to create structurally advanced objects such as DNA-based nanoboxes⁸. In contrast, formation of nanoscale objects and artificial proteins from peptide-ON conjugates (POCs) as building blocks with combined involvement of two separate self-assembly principles have, to our knowledge, not yet been realized. Protein de novo design involves rational design of peptide or protein molecules to fold into a target protein or protein-like structure, rather than the use or re-design of a naturally occurring sequence. Protein de novo design offers a test of our understanding of the factors controlling protein structure, folding and stability. The approach also offers the prospect of access to tailor-made proteins^{9,10,11,12,13,14,15,16,17}. One way to overcome the complexity of protein folding is the concept of template-assembled synthetic proteins (TASPs)^18,19. Several groups have explored this approach with a diverse set of templates^{20,21,22,23,24,25}, and we have reported the first low-resolution structure of a TASP in the form of a carboprotein, using small-angle X-ray scattering (SAXS)²⁶.

Well-defined DNA secondary structures such as double helices²⁷, triple helices²⁸, multi-way junctions²⁹ and quadruplexes³⁰ are potential scaffolds in designs of TASPs, requiring POCs as building blocks. Different methods have previously been applied to couple peptides or proteins with ONs³¹, including azide-alkyne cycloaddition reactions due to their remarkable compatibility with diverse functional groups and their high second-order rate constant under mild conditions^32,33. However, to the best of our knowledge, only three examples have been reported on peptide and protein assembly driven by the formation of DNA secondary structures^34,35,36. In one report it was demonstrated that quadruplex formation could orient two very short peptides to form an adjacent two-loop protein-like surface³⁴, in another that three- and four-way DNA junctions positioned one recombinant eADF4(C16) element at each terminal³⁵, and in a more recent report that two model substrates (a maltose-binding protein and an antibody fragment against urokinase plasminogen activator receptor) positioned on DNA junctions imitated the geometry of an antibody³⁶. Previously, POCs have thus not been explored in de novo protein design using two orthogonal self-assembly principles.

For the present proof-of-concept study we relied on a sequence derived from CoilV_aL_d^37,38 which was N-terminally extended with an azidohexanoyl-Tyr linker. The α-helical coiled coil is a ubiquitous protein motif that exists in 5–10% of all protein sequences³⁹. One of the main characteristics of a coiled coil is the simplicity of its sequence, as it consists of a motif that repeats itself every seven residues, (abcdefg)_n. In coiled coils, two or more helices wrap around each other in a left- or right-handed helical twist conformation. These helices can adopt different topologies, as they can assemble in parallel or antiparallel orientation. The solution structure analysis of CoilV_aL_d also revealed a cooperative monomer–dimer–trimer equilibrium, with the dimer state being an intermediate³⁷. Later the crystal structure of CoilV_aL_d revealed a parallel triple-helical structure³⁸.

Herein, we report an efficient and high-yielding preparation of POCs by copper-free alkyne-azide cycloaddition reactions, and we show that ON triple helix-formation can be used to organize peptide strands leading to the formation of a highly stable three-helix bundle protein mimic that dimerizes at higher concentrations. Locked nucleic acid (LNA) was central to our design⁴⁰.

猜你喜欢