Summary
Inspired by nature, modern nanotechnology has enabled the bottom-up construction of molecular machines and nanorobots using two different biomolecular building blocks, DNAs or peptides. As an emerging research field, synergizing these two biomolecular codes into a single nanostructure has provided super-powerful molecular tools into the arsenal of modern nanotechnology. Among them, peptide-DNA conjugates possess both attributes of peptide and DNA and can be arbitrarily predefined in given structural configurations, standing out as unique nanoscale building blocks for de novo design of instrumental nanostructures that otherwise could not be composed by using DNA or peptides only. Herein, the term peptide is used in the broadest sense, including oligopeptide, polypeptide, and protein. In this tutorial review, we survey the main progress made within the past decade in how to use peptide-DNA conjugates as nanoscale bricks to self-assemble hybrid nanostructures for different chemical and biological purposes. A concise perspective is included for existing challenges and potential future research directions. Looking to the horizon, peptide-DNA conjugates may serve as key structural elements in the coming decade to enable the bottom-up construction of advanced molecular machines, even comparable to those cellular organelles evolved by nature.
Introduction
Directed self-assembly of building-block-like molecules is an alluring strategy to form large chemical structures, where the intermolecular interaction patterns between the building-block-like molecules are casted to nano-, micro-, and even up to macroscale levels to drive self-evolved assembling processes. Nature uses double-stranded (ds) DNA/RNA to encode the genetic information of most life forms on Earth via the Watson-Crick hydrogen bonds, namely cytosine paired with guanine (C-G) and thymine/uridine paired with adenine (T/U-A). These two highly specific base-pair interactions have been repurposed to cause a boom in the development of DNA nanotechnology within the past few decades. Since the concept was proposed by Seeman in the 1980s,1,2 a wide range of DNA nanoscale architectures have been assembled to customize different chemical, biological, diagnostic, and/or therapeutic purposes. The DNA nanotechnology leaped into a new era when Rothemund invented the DNA origami technology at the beginning of this century.3 This programmable molecular tool was further empowered by integrating other non-canonical interactions such as Hoogsteen hydrogen bonds, cation chelation, and blunt-end π-π stacking. However, the chemical compositions of DNA nanostructures alone are limited to the functional groups of four nucleotides, namely two purine nucleotides (A and G) and two pyrimidine nucleotides (C and T), which inherently makes them lack diverse chemical functions. Aptamers and DNAzymes are two exceptional modalities, capable of binding to specific targets or catalyzing different types of chemical reactions over a wide range of substrates (DNA, RNA, porphyrin, thymidine dimer, nucleopeptides, amino acids, esters, amides, etc.); however, almost all of them are composed of single-stranded (ss) nucleic acid sequences and thus are not regarded as nucleic acid nanostructures.4,5,6,7,8 Antisense technologies are interfering pathways to regulate gene expression by using ss oligonucleotides (ONs) or ds RNAs with the help of auxiliary enzymes and protein complexes, which also fall out of nucleic acid nanostructures.9,10 The same principle also applies to triplex technology,11 molecular beacons,12 and mRNA vaccines13 where only ssDNA or ssRNA sequences are involved. Toehold-mediated strand displacement has been widely used for signal amplification of various biological cues, but the systems are composed of separate ss ONs.14 Readers are referred to other comprehensive reviews for these applications.4,5,6,7,8,9,10,11,12,13,14,15,16,17,18
When the genetic information is decoded from DNA/RNA, nature uses 20 canonical amino acids to compose other types of biopolymer protein that are responsible for most cellular biological functions. After they are translated in the ribosome, the proteins are transported to the endoplasmic reticulum for correct folding. The primary protein structures fold to form the secondary structures (such as α helices and β sheets) and then to their tertiary and quaternary structures (such as coiled coils and subunit associations) to acquire skeletal mechanics, tensile elastics, catalytic activities, and/or recognition abilities. In the last two decades, highly specific interaction patterns of the secondary/tertiary structures in natural proteins have been leveraged to the surge of peptide nanotechnology for the de novo design of enzyme mimics and the bottom-up construction of artificial protein structures.19 For a thorough overview, we suggest exploring other recently published reviews.20,21,22,23 Despite significant progress that has been made in structural peptide/protein nanotechnology, the intricate sequence-structure relations are not fully understood yet. How to maneuver the protein functions like nature by actuating several amino acid residues and/or short peptide motifs of their primary sequences is not fully known. Given the versatile biological functions offered by 20 amino acid building blocks, the structural peptide/protein nanotechnology is limited by the lack of programmability from one-dimensional (1D) primary sequence to three-dimensional (3D) topological folding.
A rational design principle is to coherently combine these two powerful nanoworld codes, DNA and proteins/peptides, into a single macromolecule to synergize the high structural programmability of DNA nanotechnology and versatile chemical functions offered by peptides/protein sequences. This strategy is surmised to unveil a new chemical dimension that the single use of each modality is unable to access. Gratifyingly, although the research field is only slowly emerging in recent years, significant progress has been made to illuminate the power of this unique molecular tool for various chemical and nanotechnological applications. In this tutorial review, we introduce the seminal works and milestone advances made within the past decade to provide a retrospective for this new research area. We set the following criteria to highlight the vital role of cross-domain synergy between DNA and peptides/proteins in the hybrid nanostructures: (1) DNA component strands and peptides/proteins are covalently conjugated, (2) both the DNA domain and the peptide/protein domain can be self-assembled in an orthogonal manner, (3) each domain has more than one component strand, (4) the self-assembly processes are highly cooperative via intermolecular interactions, and (5) cross-domain interactions are allowed for higher-order structures. A few non-covalent conjugation examples are also included to demonstrate a broad sense of DNA-peptide/protein conjugates for nanotechnological uses. However, due to the space limit, we apologize that we are unable to cover all the progress made in this emerging research field. For an in-depth overview of DNA-peptide/protein conjugates for other applications such as surface display, templated functional assembly, enzyme cascades, protein/peptide encapsulation, and targeted delivery, we refer the readers to other excellent reviews/articles published recently.11,24,25,26,27,28
In the following sections, we briefly introduce the commonly used chemical methods for bioconjugation of DNA to peptides/proteins. Afterward, four different subsections are included in order of increasing hierarchical complexity to demonstrate that peptide-DNA conjugates can be exploited as building-block-like parts to assemble various hybrid nanostructures: (1) using peptide-ON conjugates to assemble hybrid nanostructures, (2) using peptide-DNA nanostructure conjugates to assemble hybrid nanostructures, (3) using protein-ON conjugates to assemble hybrid nanostructures, and (4) using protein-DNA nanostructure conjugates to assemble hybrid nanostructures. Finally, concise conclusions are given for the current challenges and our perspectives.
Chemical methods to compose peptide-DNA conjugates
Various chemical methods have been developed to connect peptides to DNAs either covalently or non-covalently, catering for different chemical and biological purposes. The covalent ligation produces a stable DNA-peptide hybrid macromolecule and, in most cases, is realized via aminoacylations, orthogonal reactions, Michael additions, and thiol oxidations (Figures 1A–1G). The non-covalent conjugation, however, takes advantage of the high binding affinity between two cognate entities (one attached to DNAs and the other fused to peptides) to confine the DNA and peptide modalities in proximity via protein-protein, protein-ligand, and ligand-ligand interactions (Figure 1H). Given the main topic as the recent progress of peptide-DNA conjugates in chemical and nanotechnological applications, in this section, we first elaborate on several user-friendly covalent methods that have been widely used for quantitative DNA-peptide ligations. In these cases, small linkages are formed between peptides and DNAs and thus are considered not to significantly perturb the DNA and peptide functions. Then we briefly introduce the non-covalent conjugation method. Other means that generate bulky linkers are out of the scope in this section, and we recommend that interested readers explore recent reviews on the topic.24,26,27…
