ConspectusNucleic acid nanotechnology has fundamentally transcended the classic paradigm of DNA and RNA as passive carriers of genetic blueprints, which enables the rational design and construction of precise nanostructures with defined shapes, dynamics, and functions. This programmability has revolutionized approaches in biomedicine, facilitating breakthroughs in high-resolution molecular diagnostics, spatially and temporally controlled drug delivery, and the creation of synthetic cellular machinery. However, a central challenge for clinical translation is the inherent immunogenicity of nucleic acid materials. Introducing exogenous DNA or RNA nanostructures risks triggering potent innate immune responses, which can lead to rapid clearance, diminished therapeutic efficacy, inflammation, and toxicity. Rather than pursuing universal immunosuppression, researchers are beginning to rationally exploit defined immunostimulatory pathways, which allows for the strategic incorporation of immune-modulatory cues for vaccine development, immunotherapies, and targeted adjuvant systems.In this Account, we review our efforts to develop framework nucleic acids (FNAs) as a platform with modulable innate immunostimulation for biomedical applications in live cells and in vivo. We briefly summarize structural principles of nucleic acid immune recognition mediated by receptors such as toll-like receptors (TLRs) and cyclic GMP-AMP synthase (cGAS). We highlight that such immune recognition is dictated not merely by the abundance of nucleic acids but by key structural parameters, including size, shape, compactness, and the spatial organization of stimulatory nucleic acid motifs. We illustrate strategies to either enhance or suppress immunostimulation through controlled biodistribution, multivalent ligand display, and dynamic structural reconfiguration. These approaches enable tailored applications such as the development of nanovaccines and cancer immunotherapy, or conversely, anti-inflammatory and antioxidant therapies. Looking forward, we envision FNAs as intelligent tools for precision immunomodulation, bridging nanoscale design with immunological outcomes to advance personalized medicine.

Abstract Deformable condensates offer dynamic interfaces for biomolecule delivery, yet membrane adhesion does not necessarily lead to cellular internalization. The physical transition that determines whether a membrane-bound soft material remains surface-anchored or proceeds through wetting towards productive uptake remains poorly understood, particularly at active living-cell membranes. Here, we engineer sequence-defined DNA condensates through liquid-liquid phase separation (LLPS) and program their interfacial behavior by tuning sticky-end valency and cholesterol organization. These molecular designs precisely regulate condensate fluidity, fusion dynamics and internal organization, generating distinct states of weak contact, persistent anchoring and rapid wetting. Increasing cholesterol-mediated affinity does not enhance uptake. Instead, productive internalization emerges from a balance between membrane adhesion and condensate fluidity and deformability. Native membrane composition further modulates condensate interfacial fate across mammalian cells and plant protoplasts. DNA condensates enrich and deliver CpG ODNs, mRNA (∼2000 nt) and proteins, while cargo loading experiments reveal that preserving condensate architecture is essential for functional delivery. Our findings identify wetting competence as a design parameter for controlling soft material engagement and cellular entry.

Molecular classification of diseases that accurately reflects clinical behavior is fundamental to the realization of precision medicine. Single-cell protein analysis combined with coding technology offers a promising approach for constructing robust molecular classifiers. However, the low abundance of disease-related cells and the technical challenges in parallel profiling of multiple proteins remain major obstacles. Herein, we present a DNA nanostructure-based multicomponent coding strategy that enables multiprotein analysis at the single-cell level by precisely controlling the stoichiometry, orientation, and modularity of the magnetized tags and multicolor fluorescent tags. Compared with conventional linear DNA barcoding methods, our approach allows for the simultaneous magnetic separation of heterogeneous cell populations and multicolor fluorescence-based phenotypic encoding. By integrating single-cell trapping techniques, we demonstrate the accurate molecular subtyping of breast cancer based on fluorescence-encoded phenotypic features. This strategy expands the scope of applications in cell sorting, proteomic profiling, and genomic analysis, thus advancing the frontiers of precision medicine.

Real-time monitoring of biomolecular dynamics at the subcellular level is essential for understanding organellar functions. DNA nanoprobes represent a powerful tool, yet their rapid degradation upon lysosomal entry results in short imaging windows that preclude long-term dynamic monitoring. To address this challenge, we developed an L-DNA-based framework nucleic acid (L-FNA) nanodevice that leverages mirror-image chirality for intrinsic nuclease resistance. By intergrating an ATP-responsive aptamer module into this L-FNA platform, we constructed the ATP-sensing probe L-FNA-apt, enabling prolonged and stable imaging of intraluminal lysosomal ATP. Compared to conventional right-handed DNA probes, L-FNA exhibited an approximately 7-fold longer intracellular retention half-life (17.6 h vs 2.5 h) and L-FNA-apt maintained a functional integrity window exceeding 24 h. In contrast, D-FNA-apt lost most of its responsiveness by 6 h, revealing a 6 h window in which fluorescence persists without functional activity. Using this platform, we demonstrated that L-FNA-apt remains responsive to energy stress for at least 24 h in living cells. By enabling long-term monitoring of intraluminal metabolites through chiral nuclease resistance, this work establishes a framework to distinguish structural presence from functional integrity, providing a reliable tool for studying organelle energy metabolism and related diseases.

Natural bioactive compounds (NBCs) possess broad therapeutic potential but face major translational barriers, including poor bioavailability and nonspecific biodistribution. Here, we develop a generic NBC delivery platform using computational virtual screening and tetrahedral framework nucleic acids (tFNA). By integrating molecular docking with experimental validation, we demonstrate that 17 of 30 NBCs can be efficiently loaded onto tFNA via noncovalent DNA-groove binding, providing a universal strategy for assessing FNA-based NBCs loading. As a demonstration, a costunolide (COS)-FNA complex was constructed for acute kidney injury (AKI) therapy. This complex leverages the unique properties of tFNA to achieve rapid renal accumulation (5 min) and prolonged retention (12 h). In AKI mice, COS-FNA showed greatly improved bioavailability and achieved significant therapeutic effects at only 1/300 of the conventional COS dose. This study offers a new therapeutic approach for AKI and a flexible platform for broad clinical delivery of NBCs.

The integration of molecular recognition and electronic charge transport within a single-material system is central to the development of bioelectronic interfaces. However, in hybrid bioelectronic systems, these functionalities are often governed by poorly defined structural features, making it difficult to establish clear structure-function relationships. Here, we develop a growth-regulated metallization strategy based on self-assembled DNA nanosheets, enabling the formation of ultrathin, laterally extended metal-nucleic acid hybrid structures. By introducing surface-extending DNA brushes, nanosheet growth is sterically regulated and kinetically controlled, with DNA brush spacing governing nanosheet evolution by modulating the competition between lateral expansion and vertical thickening, thereby defining a kinetically stabilized ultrathin regime. This growth mechanism yields ultrathin, laterally extended amorphous nanosheets across multiple metal surfaces. Within this growth-defined system, structural parameters including brush spacing and metal layer thickness can be systematically tuned to regulate molecular recognition and charge transport. For molecular recognition, probe spacing (i.e., DNA brush spacing) and nanosheet thickness jointly determine hybridization performance by regulating steric hindrance and interfacial accessibility, defining an optimal structural window. For charge transport, nanosheet thickness and compositional matching govern transport behavior: increasing thickness enhances electronic coupling and reduces activation barriers, consistent with hopping-dominated transport, while compositionally matched nanosheet/electrode systems exhibit the most efficient interfacial charge transfer. This combination of enhanced hybridization and efficient charge transport enables high-performance electrochemical biosensing. More broadly, this work establishes a growth-controlled strategy for defining structure-function relationships in metal-nucleic acid hybrid systems.

Electrical signaling in biology is ionic. Safe modulation needs interfaces that couple electrons to ions at soft membranes without injury. We present DNA-FOCUS, a membrane-conformal, DNA-programmed carbon-nanotube biointerface that self-assembles on living cells, weaving DNA-wrapped SWCNTs into an ultraflexible mesh. A subvolt bias is condensed into nanometer hotspots at the membrane, lowering access resistance and engaging endogenous calcium-permeable ion channels. Subvolt stimulation drives sustained Ca 2+ influx across multiple lineages, with voltage-sensitive dye and impedance readouts showing stronger coupling. Responses occur without excess ROS or loss of viability. The focused field also enables gentle electroporation for plasmid delivery, augments mechano-electrical transduction, and modulates neuronal excitability. Transcriptional analysis and functional validation confirm NFAT activation, and in natural killer cells, it boosts perforin and IFN-γ release with higher tumor killing. This programmable, substrate-free interface supports low-voltage electro-modulation for high-throughput analytics and therapeutic augmentation.

Amorphous atomically thin metals offer an exceptionally high density of catalytically active sites, yet their practical electrocatalytic performance is often constrained by kinetic limitations arising from inefficient charge transport and restricted mass exchange within disordered ultrathin domains. This mismatch between local surface reactivity and macroscopic reaction kinetics represents a fundamental barrier to fully exploiting amorphous metallic catalysts. Here, we report a hierarchically integrated amorphous PtCu architecture in which atomically thin PtCu nanosheets and interconnected PtCu nanotubes are spatially interwoven to establish distinct yet strongly coupled reaction and transport domains. Within this architecture, the amorphous nanosheets function as highly active catalytic interfaces, while the contiguous nanotube network provides a continuous transport backbone that enables rapid electron percolation and efficient reactant diffusion. This deliberate functional partitioning effectively decouples surface reactivity from transport constraints, thereby alleviating the intrinsic kinetic bottlenecks of amorphous two-dimensional metals. As a consequence, the PtCu nanotube-nanosheet hybrid delivers markedly enhanced hydrogen evolution kinetics and Pt utilization efficiency, surpassing most reported Pt-based electrocatalysts. More broadly, this work demonstrates that transport-enabled structural integration can fundamentally reshape the electrocatalytic behavior of amorphous metals, establishing a transferable structural design strategy for converting atomic-scale disorder into macroscopic catalytic efficiency.

Membrane curvature orchestrates essential processes such as vesicle fission, fusion, and cell migration, yet its real-time quantification has remained elusive. Here, we introduce the Molecular Tensiometer, a fluorescence-lifetime-based framework for noninvasive, quantitative mapping of membrane curvature using the tension-sensitive probe, Flipper-Tension Reporter (Flipper-TR). A monodisperse liposome library prepared by DNA brick-assisted sorting established a power-law dependence on the probe lifetime of the curvature radius, refined through all-atom molecular dynamics and quantum-chemical analyses. This calibration enabled direct translation of lifetime into curvature with nanoscale precision in model systems, demonstrating that Flipper-TR served as the sensing element of the Molecular Tensiometer. While preserving global morphology, the Molecular Tensiometer resolved nanoscale fluctuations and localized curvature hotspots that are invisible to intensity imaging, providing a high-fidelity platform for probing dynamic membrane mechanics in model lipid membranes.

Identifying novel gene fusions is critical for cancer diagnosis and drug development. While a few advanced methods have shown the capability to detect gene fusions involving unknown partners, comprehensive detection of gene fusions, especially of those with low copy numbers, remains a challenge. Indeed, most current panel-based sequencing methods fall short in reliability and cost efficiency. Here we present a method for detecting potentially novel gene fusions using anchored random reverse primers (ARRP) during PCR-based library construction, allowing the simultaneous capture of mutations and RNA splicing variants. Furthermore, the combination with blocker displacement amplification technology enables a median of 22-fold allele enrichment for gene fusions, achieving a limit of detection ~10-fold lower than that of current technologies and resulting in an 8-fold cost reduction. Using ARRP-seq, we identify numerous novel fusions in 98 clinical tissue samples, showcasing its diagnostic potential in prostate cancer and capacity for personalized diagnostics in cervical cancer. A library preparation method using anchored random reverse primer sequencing for quantitative analysis of novel gene fusions from low-input samples.

The brain is the most complex organ in the human body. For over a century, the classical Golgi staining method has been a cornerstone in neuroanatomy, but its low efficiency and uneven staining in large samples have limited its utility for systematic neural network analysis. This long-standing challenge persisted without a satisfactory solution—until we embarked on a daring exploration inspired by cross-disciplinary curiosity.

The organizational complexity of biominerals has long fascinated scientists seeking to understand biological programming and implement new developments in biomimetic materials chemistry. Nonclassical crystallization pathways have been observed and analyzed in typical crystalline biominerals, involving the controlled attachment and reconfiguration of nanoparticles and clusters on organic templates. However, the understanding of templated amorphous silica mineralization remains limited, hindering the rational design of complex silica-based materials. Here, we present a systematic study on the stabilization of self-capping cationic silica cluster (CSC) and their assembly dynamics using DNA nanostructures as programmable attachment templates. By tuning the composition and structure of CSC, we demonstrate high-fidelity silicification at single-cluster resolution, revealing a process of adaptive templating involving cooperative adjustments of both the DNA framework and cluster morphology. Our results provide a unified model of silicification by cluster attachment and pave the way towards the molecular tuning of pre- and post-nucleation stages of sol-gel reactions. Overall, our findings provide new insights for the design of silica-based materials with controlled organization and functionality, bridging the gap between biomineralization principles and the rational design of biomimetic material.

Translating in situ dynamic changes of key signaling molecules into actionable clinical readouts remains a formidable challenge for noninvasive diagnostics. Here, focusing on reactive oxygen species (ROS) as pivotal signaling mediators, we developed defect-programmed DNA origami ROS sensors (DOSs) for portable urinalysis of localized oxidative stress. Using triangular DNA origami (DO) nanostructures as two-dimensional synthetic soft crystals, we programmed the number of discontinuity defects between adjacent staple strands and established a positive correlation between defect number and ROS-triggered degradation kinetics. To transform this programmable degradation into a diagnostic function, we then engineered DOSs via orthogonal assembly of targeting and signaling modules onto DO. In a murine model of acute liver injury (ALI), DOSs selectively accumulated in the liver and underwent ROS-triggered fragmentation into renal-clearable debris, converting hepatic ROS levels into quantifiable urinary signals. Notably, this transformation efficiency depended positively on defect number in DOSs, enabling portable urinalysis that detected ALI onset at least 4 h earlier than conventional alanine aminotransferase (ALT) testing, with a maximum area under the curve of 0.94. This defect-engineering strategy establishes a generalizable platform for early, noninvasive diagnosis of ROS-related diseases.

Single-molecule detection (SMD) holds considerable promise in biomedical research. Although atomic force microscopy (AFM) provides an important technique with nanoscale resolution for SMD, its broader application is limited by labeling challenges and slow data processing. Here, we present a machine learning (ML)-powered strategy combining AFM and DNA nanotags for SMD and cancer diagnosis. Nickases are applied to create specific single-strand breaks in target DNA, allowing insertion of exogenous DNA to attach shape-distinct nanotags for AFM imaging. A YOLOv5l algorithm is adopted to automatically recognize target objects in AFM images, which can classify 370 structures in 1.21 seconds with 98% accuracy. The proof of concept of this strategy is confirmed by identifying nickase-edited sites on both linear and circular DNA. Its practical applicability is demonstrated by detecting KRAS Gly12Arg (G12R) and p53 Arg175His (R175H) mutations in samples from patients with pancreatic and colorectal cancer, with accuracy rivaling Sanger sequencing and quantitative polymerase chain reaction, opening avenues for SMD.

Electrochemiluminescence (ECL) biosensors, which convert electrochemically generated excited states into optical signals, have attracted considerable interest for the ultrasensitive detection of disease-related biomarkers. Compared with conventional immunoassays and chemiluminescent or fluorescent methods, ECL offers low background signals, a wide dynamic range, and precise temporal and spatial control of light emission via potential modulation, making it particularly suitable for point-of-care testing in cancer, cardiovascular and cerebrovascular diseases, neurodegenerative disorders, and infectious diseases, where biomarkers often exist at ultra-low levels in complex biological matrices. However, further improving analytical sensitivity and detection reliability remains a key challenge, driving the rapid development of diverse signal amplification strategies. These approaches include nanomaterial-assisted signal enhancement, enzyme- and nanozyme-catalyzed generation or consumption of co-reactants, nucleic acid-based amplification schemes, and intrinsic amplification enabled by the rational design of co-reactants and co-reactant accelerators. This review systematically categorizes these amplification strategies and highlights representative applications across major disease classes. Finally, current limitations and future perspectives are discussed to promote the clinical translation of signal-amplified ECL biosensors for early diagnosis and precision medicine.

ABSTRACT Long‐chain nucleic acids (LCNAs) are increasingly important as diagnostic biomarkers, yet their quantitative detection is often hindered by compact conformations that limit probe accessibility. Here, we report a DNA nanoruler‐assisted linearization strategy that mechanically regulates LCNA conformation by employing a rigid DNA tetrahedral framework (DTF) dimer as a nanoruler. This approach anchors the target termini at a predefined nanoscale spacing (∼10–15 nm), imposing defined spatial constraints that stretch and pre‐open secondary structures, leading to linearization and enhanced site accessibility. In homogeneous assays, this strategy delivers robust signal enhancement across LCNAs with diverse structures. Integrated into an electrochemical platform, it achieves LODs of 1 copy/µL for the E and N genes and 10 copies/µL for ORF1ab, with linear ranges of 10–10 5 copies/µL for E gene and 10 2 –10 5 copies/µL for both N gene and ORF1ab from SARS‐CoV‐2. Furthermore, when incorporated into a fluorescence chip, this method enables high‐throughput analysis of circulating cfDNA (ALU115) for prostate cancer discrimination with an AUC of 90.35%. Collectively, we establish a DNA nanoruler‐based conformational regulation strategy as a structure‐tolerant approach for high‐performance LCNA detection in both electrochemical and fluorescent modes.

DNA cryptography has been explored for data encryption because of the high storage capacity of DNA molecules, the considerable parallelism of DNA computing, and the diversity of DNA nanostructures. Nevertheless, integrating multiple encryption protocols in a single DNA origami communication workflow remains challenging. Here, we develop a DNA multilayer encryption device that integrates multiple cryptographic algorithms for secure communication. The encoding system exploits the addressability of rectangular DNA nanostructures to create spatial patterns into nano-Morse code. In a codebook-based symmetric encryption framework, ciphertext messages encoded by nano-Morse code patterns are transmitted through shared key. Furthermore, the transformation between the rectangular and tubular DNA nanostructures allows the steganography and conformation-gated verification with a key space of 2576. We demonstrate multilayer secure communication in block-based message normalization mode by transmitting the message "JUNE6 INVASION NORMANDY," achieving confidentiality, integrity, and authenticity. This work advances DNA nanotechnology from a structural scaffold to a programmable information-processing platform with implications for intelligent molecular systems.

DNA has emerged as a promising medium for the post-silicon era of information storage due to its ultrahigh density and longevity. However, current systems are bifurcated, with solid-state systems providing robust cold archival but lacking accessibility, while fluidic molecular computing systems offer dynamic processing but suffer from low density and instability. This mutual exclusivity has hindered the development of hierarchical memory, a standard in modern computing, within molecular storage systems. Here, we bridge this gap by engineering a reconfigurable DNA memory architecture driven by programmable liquid-liquid phase separation (LLPS). Our system leverages sequence-based encoding to achieve an ultrahigh storage density of 7×10^10 GB/g, approaching the theoretical limits of DNA accessibility. In its fluidic hot state, DNA droplets enable rapid data loading (~83.8% in 5 min) and function as an in-memory editing platform supporting versatile, addressable bit-level operations including selective erasure (~65.1%) and high-efficiency rewriting and replacement (>99%) via programmable strand displacement. Importantly, to resolve the stability trade-off, we engineered a programmable phase transition whereby the triggered assembly of a rigid tetrahedral DNA framework (TDF) armor transforms liquid condensates into robust armored droplets. This cold state confers exceptional resistance to enzymatic and physical degradation, projecting multi-millennial data stability. By enabling reversible transitions between an editable, high-density computing mode and a stabilized archival mode, this work establishes the architectural foundation for scalable molecular information storage capable of hierarchical data management. ### Competing Interest Statement The authors have declared no competing interest. National Natural Science Foundation of China, 32571596

Directional and long-distance signal transmission is crucial for high-performance molecular computation and biomimetic systems. DNA origami offers precise surface addressability for programmable DNA signal propagation but is limited by the area of single structures. Here, we develop a bridged connection strategy to enable scalable and seamless assembly of double-layer square DNA origami nanostructures (dsDONs) to realize cross-origami signal transmission over a long distance. We demonstrate that the seamless assembly of DNA origami addresses spatial gaps and structural bending in conventional 2D assemblies, making them suitable for cross-origami signal transmission. We introduce an encoded bridging rule to break rotational symmetry, enabling the scaling up of DNA origami templates. Experimentally, we demonstrate directional single-molecule DNA signal transmission over distances up to 500 nm and the implementation of DNA logic circuits across two origamis. This strategy provides the basis for developing large-scale molecular communication networks and computing circuits.