Zhuxin Dong

Nanopore sequencing for full-length proteins is daunting due to inaccuracy of fingerprinting amino acids with ionic current and lack of theoretical models. Here, we demonstrate that nanopores drilled in 5 nm-thick membranes with constriction diameter <1 nm, i.e., subnanopores, can site-specifically read the primary structure of amyloid-β (Aβ) isoforms after 50-fold coverage. Molecular dynamics simulations reveal that charged side chains near the constriction swing and modulate the volume exclusion of neighboring residues, resulting in a sequence context-dependent signal altered between translocation orientations. Orientation of individual label-free molecules is inferred by clustering, which yields high-fidelity empirical consensus that strongly correlates with computational trajectories (PCC = 0.78 ± 0.03). A homemade self-supervised learning method categorizes Aβ42 and Aβ40 at the single-molecule level (AUC = 0.88 ± 0.02) and predicts the mole fraction in agreement with actual ratios. Solid-state subnanopores offer a straightforward strategy for proteoform single-molecule classification and sequence extraction in mixtures.

Oral tumors are not shaped by biochemical signals alone; they also evolve within mechanically active and thermally fluctuating microenvironments that critically regulate malignant progression, immune remodeling, and therapeutic resistance. Here, we develop a dynamic bioorthogonal organoid platform that couples a 3D‐printed photothermal elastic scaffold with an interpenetrating hydrogel network derived from decellularized oral mucosa extracellular matrix, enabling the reconstruction of in vivo‐like structural support, matrix‐derived biochemical cues, spatial cellular compartmentalization, and dynamic physical stimulation. Through click chemistry‐mediated cell anchorage and hydrogel encapsulation, oral cancer cells and macrophages are spatially organized into a compartmentalized cancer–immune niche. Under cyclic mechanical stretch and near‐infrared‐triggered periodic photothermal stimulation, dynamic culture markedly enhances cancer cell proliferation and glycolytic reprogramming while concomitantly polarizing macrophages toward a pro‐tumoral M2 phenotype. In both monoculture and coculture settings, biochemical–mechano–thermal stimulation substantially increases cancer cell chemoresistance. Mechanistically, these biochemical‐physical cues activate stress‐adaptive and survival programs while enriching cancer stem‐like populations. In a xenograft mouse model exposed to in vivo biochemical, mechanical and thermal stimulation, dynamic conditioning markedly accelerates tumor progression, enhances metastatic potential, and confers pronounced chemoresistance. Together, this study establishes a biomimetic multicellular tumor platform for uncovering how dynamic biochemical‐physical cues orchestrate cancer–immune crosstalk and therapeutic failure.

Both enzyme‐free DNA circuits and polymerase‐mediated amplification represent powerful and versatile platforms for nucleic acid signal transduction. However, DNA polymerases’ intrinsic polymerizing/exonucleolytic activities cause nonspecific extension/degradation of hairpins, restricting the combination of these two reactions to stepwise operations with limited efficiency. Herein, we report a guanine‐locked DNA circuit resisting polymerase‐mediated side reactions, enabling homogeneous coupling of catalytic hairpin assembly and dCTP‐free rolling circle amplification. Modified onto magnetic nanoparticles for optomagnetic biosensing, this platform unifies target recognition, amplification, and real‐time transduction in a one‐pot isothermal format, achieving a detection limit of 9 aM with a wide dynamic range in 1 h for Botrytis cinerea DNA. The biosensor discriminates Botrytis species and correlates strongly with qPCR ( r = −0.986) for infected rose petal samples. This work resolves the incompatibility between hairpin cascades and DNA polymerases, providing a generalizable strategy for crosstalk‐free homogeneous nucleic acid circuits.

Enthalpy-driven catalytic hairpin assembly (CHA) suffers from an inherent trade-off, wherein enhanced reaction kinetics is inevitably accompanied by elevated nonspecific background. Catalytic hairpin assembly-and-cyclization (CHAC) integrates CHA with enzymatic ligation, allowing single-target-triggered multiple circularization events and thus bypassing the stoichiometric limitation of conventional padlock probe ligation. Nevertheless, how the introduction of a ligase reaction remodels the enthalpy-driven DNA circuit remains unclear. Herein, we systematically characterize kinetic profiles and background leakage behaviors of CHAC, confirming that T4 DNA ligase enhances overall product accumulation, with a more pronounced kinetic advantage at moderate to high target concentrations. Notably, the spontaneous upregulation of nonspecific leakage, typically induced alongside kinetic improvement, is effectively restricted in CHAC. Leveraging this enhanced kinetic performance with restricted background leakage, we further combined CHAC with nicking-enhanced rolling circle amplification (NickRCA) to construct an optomagnetic biosensor targeting the mecA gene in methicillin-resistant Staphylococcus aureus. CHAC-derived circular templates guide NickRCA to generate abundant short single-stranded amplicons, which cross-link detection probe-modified magnetic nanoparticles and elicit quantitative optomagnetic responses. This study characterizes ligase-dependent kinetic and leakage modulation of the present CHAC system, with the underlying mechanism proposed to involve both covalent cyclization and noncovalent duplex binding effects, which may be generalizable to other DNA strand displacement circuits.

Programmed cell death (PCD) pathways such as apoptosis and pyroptosis play a pivotal role in cancer therapy, yet achieving precise and selective activation of these pathways using nanomaterials remains a major challenge. Here, we report the synthesis of two quasi‐structurally isomeric 13‐nuclear copper nanoclusters that share an identical Cu ₁₃ core but differ in peripheral ligands, i.e. Cu ₁₃ ‐OH and Cu ₁₃ ‐F . These ligand‐engineered nanoclusters elicit different cellular localization and reactive oxygen species (ROS) generation profiles, and thereby selectively activate distinct PCD mechanisms for efficient and low‐toxicity tumor therapy. Specifically, Cu ₁₃ ‐OH anchors onto cell membranes and forms nano‐spike protrusions that disrupt cytoskeletal organization and suppress PI3K‐AKT signalling, leading to caspase‐dependent apoptosis with a concentration‐threshold response. In contrast, Cu ₁₃ ‐F shows enhanced cellular uptake/internalization and stronger ROS‐generation capability, which may be associated with fluorination‐modulated copper redox behavior, thereby activating Caspase‐4/Gsdmd‐mediated pyroptosis. Both in vitro and in vivo studies conclude potent tumor‐selective inhibition, particularly that Cu ₁₃ ‐F exhibits exceptional efficacy at low‐doses due to its enhanced membrane permeation and robust ROS generation. These findings confirm ligand engineering as a versatile strategy for precise spatial modulation of PCD pathways, offering a new paradigm for feasible design of tumor‐selective nanomedicines through subtle atomic‐level modifications.

Capturing the dynamic variations of nanoscale viscosity within biological fluids is pivotal for deciphering the mechanisms of critical physiological and pathological processes, notably blood coagulation. However, this endeavor remains challenging due to the limitations of conventional techniques. Standard assays relying on bulk optical or rheological measurements lack the spatial resolution required for nanoscale viscosity analysis, failing to distinguish cascade processes in complex coagulation reactions. We present an optomagnetic nanorelaxometry platform that probes nanoscale viscosity in real time through analysis of magnetic nanoparticles’ Brownian relaxation dynamics. Requiring a sample volume of 50 μL, the platform demonstrated quantitative accuracy in glycerol solutions and successfully tracked dynamic viscosity changes during collagen gelation. Most notably, during platelet-rich plasma coagulation, our platform revealed a nonmonotonic viscosity profile inaccessible to conventional assays, suggesting a novel methodological perspective on coagulation mechanisms. Evaluation with clinical anticoagulants and a thrombolytic agent further confirmed its practical utility, underscoring the platform’s potential for pharmacological screening and point-of-care coagulation monitoring with nanoscale resolution.

Rolling circle amplification (RCA) is a powerful isothermal nucleic acid amplification technique prized for its robustness and simplicity. However, conventional RCA-based detection is fundamentally limited by the stoichiometry of padlock probe ligation, wherein each target molecule ideally yields only one circular template. Existing strategies to improve ligation efficiency often sacrifice key benefits of RCA or focus on specificity rather than catalytic turnover. Herein, we developed catalytic hairpin assembly-and-cyclization (CHAC), a homogeneous cascade that integrates catalytic hairpin assembly with enzymatic ligation, enabling each target to initiate multiple circularizations and breaking the 1:1 stoichiometric limit. As the target serves only to initiate catalytic assembly without acting as the substrate for ligation or the primer for RCA, CHAC overcomes constraints on target identity and topology. Applied to detection of the mpox (monkeypox) virus E9L gene using a single-tube format (incorporating detection probe-modified magnetic nanoparticles) with real-time optomagnetic sensing, CHAC-RCA achieved detection limits of 0.3–2 fM (depending on amplification duration) within a total assay time of ~2–2.5 h, representing at least a 100-fold improvement over conventional ligation-RCA. The assay showed high specificity, robustness, and clinical concordance with quantitative PCR, establishing CHAC-RCA as a versatile and efficient platform for ultrasensitive nucleic acid detection.

Nanopores are cutting-edge interdisciplinary tools that can analyze biomolecules at the single-molecule level for many applications, e.g. DNA sequencing. Efforts are underway to extend nanopores to proteomics, including the development of machine learning algorithms for protein sequencing and identification. However, single-molecule data are intrinsically noisy and hard to process. Moreover, the development and performance of machine learning for nanopore is jeopardized by data scarcity. Self-supervised learning is an emerging method that may yield advantages in nanopore scenarios. We propose and experimentally validate Nanopore analysis using Self-Supervised Learning (NanoSSL), a generative self-supervised learning framework based on attention mechanisms for the identification of protein signals from nanopores. Leveraging a two-step approach consisting of self-supervised pre-training and supervised fine-tuning, NanoSSL learns useful feature representations from empirical data to facilitate downstream classification tasks. Inspired by the concept of fragmentation in conventional protein sequencing technologies, during pretraining each translocation event is split into multiple non-overlapping fragments of equal size, some of which are randomly masked and reconstructed using a masked autoencoder. Learning the feature representations of the reconstructed nanopore events facilitates molecular identification in fine-tuning. In this study, we retested a publicly available nanopore multiplexed protein sensing dataset for model iteration, and subsequently measured Alzheimer's disease biomarker Aβ1-42 using homemade solid-state nanopores. Empirical results indicated NanoSSL achieved an unprecedented performance across four metrics: accuracy, precision, recall, and F1 score, when classifying two mutated Aβ1-42, E22G and G37R. The self-supervised learning and attention mechanism were verified as the source of performance gains. The main program is available at https://doi.org/10.5281/zenodo.17172822.