Because of their outstanding weak light detection ability, multiplication-type photodetectors have great application prospects in fields such as environmental detection, biological science, and night vision imaging. Traditional photomultiplication-type organic photodetectors (PM-OPDs) usually achieve a large external quantum efficiency (EQE) through the mechanism of interfacial trap-assisted charge tunneling injection but inevitably produce a large dark current. Here, we demonstrate a novel dual-function photon-regeneration multiplication-type organic photodetector (PRM-OPD) with photocurrent gain by absorbing additional photons generated by an integrated organic light-emitting diode unit. The optimized PRM-OPD exhibits a maximum EQE of 2,484% and maintains a low dark current density (J_d) of only 10⁻⁸ A/cm². More importantly, the resulting PRM-OPD can not only exhibit efficient detection performance but also directly detect the shape of the light spot, showing a dual-function working mode. Furthermore, by combining optical resonant microcavities, the response spectrum is successfully extended to the near-infrared region and a narrowband PRM-OPD with excellent performance is obtained. This designed device structure provides a new idea for the development of high-performance PM-OPDs.

Nonviral intracellular delivery based on pulsed-electric-field-induced electroporation is one of the most effective and widely used platforms in basic biological and biomedical research. However, the conventional bulk electroporation technique has exhibited limited performance in improving delivery efficiency with a single type of pulse, especially for in vivo small interfering RNA (siRNA) delivery. Pulse modulation has been confirmed effective in facilitating intracellular delivery. Nonetheless, pore evolution and regulation during and after electric exposure plays an essential role in the effective intracellular delivery of molecules with variable sizes. Here, we propose a progressive electroporation (PEP) strategy on the basis of multiple-pulse combination, which decouples the perforation process and delivery process compared to conventional bulk electroporation, efficiently improving delivery efficiency with regulation of the perforated pores. We demonstrated an important correlation between delivery efficiency enhancement and delayed pore resealing by quantitative investigations. The performance of this disruption-and-field-enhancement method also showed delivery advantages over conventional chemical systems. Moreover, we validated the improvement for siRNA knockdown efficacy in vivo. Overall, PEP helps provide a unique insight into improving intracellular delivery, by regulating pore dynamics rather than just inducing perforation. This strategic advancement of PEP may pave the way for the development of advanced wearable delivery systems with reduced energy consumption.

Traditional solid metal catalysts in organic reaction have been hindered by drawbacks, including limited surface catalytic region, restricted reactant diffusion, catalyst deactivation, low atom utilization efficiency, and harsh reaction condition. To overcome these drawbacks, we introduce liquid metal into organic reaction. An efficient mechano-assisted reaction system was proposed integrating dynamic metal catalyst and liquid metal reaction medium. With a low-energy ball-milling process (at a speed of 270 rad·min⁻¹), the dynamic metal atoms in liquid state could achieve maximum atomic utilization and fast reaction kinetics compared to solid metal. Furthermore, liquid metal also act as reaction media. It not only offers a homogeneous electron-rich environment for reactants but also continuously disperses active metals to construct a multimetal catalytic system in one step. These would revolutionize the understanding of metal catalysts and reaction media, providing a novel platform for exploration of high-throughput catalysis and discovery of novel chemical reaction in future.

Highly pathogenic avian influenza H5N1 virus has spread to over 1,080 dairy farms across 18 states in the United States, resulting in 41 human infections and posing serious risks to both animal and public health. To address these risks, a hemagglutinin-based mRNA–lipid nanoparticle vaccine was developed, and its safety, immunogenicity, and protective efficacy in high-yielding lactating dairy cows were evaluated. The vaccine was well tolerated, had no adverse effects on health or milk production, and induced strong antibody responses. Two weeks after the second immunization, all the immunized cattle were fully protected against a high-dose H5N1 virus challenge. Notably, two-thirds of the cattle were still completely protected even at the 19th week after the first vaccination, when their serum antibody levels were very low. These data demonstrate that the mRNA vaccine confers robust, lasting protection against H5N1 virus in lactating dairy cows, providing a foundation for clinical trials.

Metabolism-regulating microspheres have evolved from conventional drug carriers into active platforms capable of spatiotemporally reprogramming pathological metabolic networks. Chronic diseases are increasingly understood to be driven by metabolic dysregulation, highlighting the need for therapeutic strategies that enable localized and precise metabolic intervention. This review systematically outlines the core design principles of these microspheres, emphasizing the synergistic integration of engineered chemical properties, such as ionic signaling, metabolite delivery, and pathway modulator release, with tailored physical characteristics, including stiffness, porosity, and size of the microspheres. Together, these features construct “metabolic instruction systems” that correct dysregulated pathways at the tissue level. Their versatile applications include orthopedic diseases, such as osteoporosis, osteoarthritis, and bone defects; ophthalmic conditions, including glaucoma and diabetic retinopathy; and gynecological disorders, such as premature ovarian insufficiency, ovarian cancer, and endometriosis. These systems target key metabolic abnormalities, such as glycolytic dysregulation, mitochondrial dysfunction, and oxidative stress, which are recognized as central drivers of disease pathogenesis across multiple organ systems. Despite considerable progress, clinical translation remains limited by tissue-specific delivery barriers, interindividual metabolic heterogeneity, and long-term safety concerns within dynamic metabolic networks. Emerging strategies, such as personalized formulations, artificial-intelligence-driven designs, and organ-on-a-chip validation platforms, are being developed to address these challenges. With ongoing interdisciplinary innovation, metabolism-regulating microspheres hold great promise as precise therapeutic modalities for a spectrum of chronic diseases rooted in metabolic imbalance, offering targeted and sustained metabolic correction.

Cytoskeletal remodeling, particularly actin dynamics, is a central driver of tumor metastasis. However, actin-targeting agents have faced major translational barriers due to poor specificity and the absence of defined druggable sites. Here, we report a bladder tumor-targeting polyarginine peptide, R11, as a precision modulator of actin dynamics capable of disrupting the cytoskeletal architecture of bladder cancer (BCa) to suppress its lung metastasis potently and persistently. R11 directly interacts with actin, weakening the actin–plectin–vimentin/integrin β4 axis and initiating a cascade of cytoskeletal disorganization that ultimately impairs cellular motility and metastatic potential. Remarkably, nanoscale multivalent assemblies of R11 amplify these effects through enhanced multivalent binding to actin. This study unveils a new strategy for cytoskeleton-targeted intervention through peptide-based precision materials, highlighting R11 assemblies as a promising therapeutic platform for the treatment of metastatic BCa and potentially other cytoskeleton-dependent malignancies.

Drug-resistant bacterial infections threaten global health and economies. Stimulus-responsive antibacterial materials are promising for combating resistance and improving treatment precision. Among them, electroresponsive strategies stand out for being practical, economical, tunable, and safe. This perspective examines various strategies, including electroluminodynamic therapy, electrodynamic therapy, and electrocontrolled drug release, which enable precise antibacterial action with minimal side effects and demonstrate strong application potential. Finally, we outline current challenges and future directions to guide further research and practical applications.

Two-dimensional transition metal dichalcogenides (TMDs) are promising candidates for next-generation electronics, but their future application is hindered by the inherently slow growth kinetics of conventional vapor deposition methods, particularly for the synthesis of large-area single-crystal films. Here, we demonstrate a source-confined chemical vapor deposition strategy that enables the fast synthesis of centimeter-scale MoS₂ single-crystal films within just 10 min. An optimized sandwich-structured Mo source was employed to ensure a concentration-balanced metal supply under sodium chloride catalysis, followed by sulfurization to form MoS₂. The films nucleate uniformly and directionally on the miscut C/A sapphire substrate positioned 2 cm upstream of the Mo source, achieving high crystal quality with a low sulfur vacancy density of 8.49 × 10¹² cm⁻². Additionally, these films support the development of high-performance enhancement-mode MoS₂ field-effect transistors, exhibiting excellent transport performances, including a high on–off ratio of 10⁸, an average positive threshold voltage of 1.71 ± 0.32 V, an average mobility of 34.28 ± 0.46 cm² V⁻¹ s⁻¹, and an average subthreshold swing of 155.8 ± 33.7 mV dec⁻¹. Furthermore, high-performance rail-to-rail inverter gates and logic circuits with low power consumption (<0.3 nW) were successfully demonstrated, underscoring the potential of these MoS₂ films for integrated circuit applications. This work offers a scalable and reliable approach for the fast growth of large-scale TMDs single-crystal films, accelerating their future applications in next-generation electronics.

In liver ischemia–reperfusion injury (LIRI), macrophage clearance of apoptotic cells via efferocytosis is crucial to prevent excessive inflammation and tissue damage. Here, we investigate the role of nucleotide-binding oligomerization domain-like receptor protein 3/cysteine-aspartate protease-1 (NLRP3/Caspase-1) signaling in modulating macrophage efferocytosis during LIRI. We observed robust activation of the NLRP3/Caspase-1 pathway during the early phase of LIRI. Genetic ablation of Nlrp3 or Caspase-1 substantially reduced LIRI severity. Notably, myeloid-specific Nlrp3 knockout mice exhibited less severe LIRI compared to hepatocyte-specific Nlrp3 knockouts, whereas macrophage-specific overexpression of Caspase-1 exacerbated tissue injury. Mechanistically, NLRP3/Caspase-1 activation enhances a disintegrin and metalloprotease protein-17 (ADAM17)-mediated cleavage of Mer proto-oncogene tyrosine kinase (MerTK), leading to impaired efferocytosis. Pharmacological inhibition of ADAM17 restored macrophage efferocytic capacity and alleviated LIRI. Clinically, elevated serum levels of soluble MerTK (s-Mer) correlated with hepatic injury severity and Caspase-1 activation in patients after partial hepatectomy or liver transplantation. Our findings suggest a potential therapeutic strategy for LIRI prevention and treatment.

Self-driving vehicles rely closely on interactions with humans, vehicles, and the surrounding environment. However, the interactive analysis of self-driving is impacted by multiple perception sources, heterogeneous data, and complex environments in actual scenes. Due to the above issues, we are often unclear about the behavior of self-driving vehicles and do not understand their decisions, and it is also difficult to achieve synergy with our human intentions. We introduce the significance of research in the field of self-driving interactive cognition, detailing its components and underlying infrastructure. Furthermore, we demonstrate how self-driving interactive cognition, inspired by the Wiener model, embodies intelligence in complex environments with the purpose of stressing the importance of interactive cognition in complex environments and scientifically evaluating the analysis of machine interactive cognition. Then, a multidimensional analysis model of self-driving interactive cognition is established based on perceptual information acquisition, multichannel and cross-modal data registration, attention mechanism, visual recognition and understanding, and embodied dynamic control. Supported by the above, we build a multiview spatiotemporal graph convolutional network (MV-STGCN) model for action recognition to realize vehicle-to-human body language interactive cognition. Most importantly, we innovatively propose a nonlinear-CRITIC–technique for order preference by similarity to an ideal solution (TOPSIS)-based method to analyze the interactive cognition analyses of different action recognition algorithms efficiently, such as MV-STGCN. Future self-driving vehicles are bound to demonstrate multichannel and cross-modal intelligence perception and human–vehicle-friendly interaction, and we are committed to better realizing humanoid driving analysis and the embodied intelligence of self-driving vehicles. “Self-driving + interactive cognition” could make future vehicles become interactive wheeled robots that can be trusted and better serve human society.