Submicrometre Superstructure Co-assembled from Protein and DNA

Functional supramolecular complexes co-assembled with multiple proteins and nucleic acids are ubiquitous in nature, such as ribosome and viruses. It is fundamentally important to understand and utilize these heterogeneous structures. Here, we reported a protein-DNA submicrometre superstructure fabricated from the self-assembly of protein CsgA and DNA nanostructures. By inspecting the physiological conditions of protein-DNA co-assembly, we found that the originally soluble CsgA could polymerize into insoluble fibers in a particular buffer (30 mmol/L Tris-HCl, 450 mmol/L NaCl, pH = 7.2), and such fibers benefited the storage of tetrahedron DNA nanostructure (TDN). Concentration and fibrillation time were optimized for the aggregation-free conversion of CsgA into amyloid fibers. Specifically, the optimum conversion of monomeric CsgA into micrometer-scale fibers could be realized at a concentration of 5 mu mol/L. Meanwhile, atomic force microscopy (AFM) suggested that CsgA assembled into mature fibers in 5 days and formed larger aggregates after 7 days. The height of mature fibers and aggregates was about 3.9 and 6.1 nm, respectively. Afterwards, TDN was modified with NTA molecule and conjugated to CsgA through the chelation of Ni2+, His-tag, and NTA. A submicrometre complex CsgA fiber-dTDN was further generated by hybridizing two copies of TDN in dimeric structure (dTDN) through beta-sheet interactions and Watson-Crick hybridization. This approach could fabricate a series of dTDN structures precisely without inducing the random aggregation of molecules. The yield of up to 44% was higher than that obtained from the direct connection of DNA modules via DNA technology. In summary, our findings demonstrated that CsgA fibers could act as a sort of novel scaffold for the assembly of protein-templated DNA nanostructure. Particularly, this model provides a deep insight into the generation of functional superstructures through self-assembly of protein and DNA-based building blocks.