T lymphocyte dysfunction represents a pivotal determinant of immunosuppression in sepsis. Our previous studies demonstrated that nuclear fragile X mental retardation-interacting protein 1 (NUFIP1)-mediated ribophagy conferred cytoprotection against apoptosis in CD4⁺ T lymphocytes during sepsis, thereby preserving host immunocompetence. Despite growing evidence linking PANoptosis to the pathogenesis of various diseases, the potential role of ribophagy in modulating CD4⁺ T lymphocytes’ PANoptosis in sepsis remains largely unclear. In the present study, we employed both lipopolysaccharide-stimulated Jurkat T cells and cecal ligation and puncture (CLP)-induced sepsis models to demonstrate marked exacerbation of CD4⁺ T lymphocyte PANoptosis following NUFIP1 knockdown (KD), associated with impaired immune function, as evidenced by diminished cytokine production and T cell proliferation. Tandem mass tagging (TMT) proteomic analysis identified Z-nucleic acid binding protein 1 (ZBP1)-mediated PANoptosome formation and the cyclic GMP-AMP synthase–stimulator of interferon genes (cGAS-STING) pathway as critical nodes in ribophagy-dependent cytoprotection. Mechanistically, sepsis-induced ribosome collision activated the cGAS-STING signaling axis, which in turn recruited NUFIP1 to STING protein complexes. Clinical analysis of septic patients revealed enhanced ribophagy and PANoptosis in peripheral blood CD4⁺ T cells, consistent with the experimental findings. These results suggest that NUFIP1-mediated ribophagy alleviates CD4⁺ T lymphocyte PANoptosis in sepsis via the cGAS-STING pathway, highlighting the therapeutic potential of targeting ribophagy and PANoptosis pathways to mitigate immune paralysis and improve the outcomes following septic insults.

Histidine phosphorylation, the neglected but vital phosphoproteome, is a reversible posttranslational modification catalyzed by histidine kinases and erased by phosphohistidine (pHis) phosphatases (e.g., LHPP). Traditional types of phosphorylation have been implicated with the deadliest adult ocular tumor uveal melanoma (UM), which is lipid metabolism dysfunction related. However, the role of histidine phosphorylation in UM remains unknown. Here, up-regulated histidine phosphorylation is associated with poor UM prognosis. Reversal of histidine phosphorylation by LHPP exerts therapeutic effects. Mechanistically, we identified LHPP as metabolically related protein with mitochondrial targeting sequence. LHPP interacts and reduces excessive histidine phosphorylation of mitochondrial aconitase ACO2 at H73 site, thus restoring ACO2 enzymatic activity and mitochondrial citrate transit in TCA cycle. Reduction of citrate accumulation attenuates overloaded lipid synthesis. Besides, genetic ablation of LHPP in mouse eye exhibits abnormal lipid metabolism. These findings illustrate the antagonistic roles of oncogenic histidine phosphorylation and therapeutic mitochondrial LHPP and provide metabolic insight of pHis modification in ocular diseases.

Background: The hypersensitive state of the central nervous system, commonly known as central sensitization, presumably drives allergic rhinitis (AR) pathogenesis. However, the involvement of central sensitization in AR and its related mechanisms have rarely been studied. Methods: An AR mouse model was induced via ovalbumin treatment. Activation of trigeminal nucleus caudalis (TNC) neurons was assessed by electrophysiological recordings and immunofluorescence staining. The role of TNC neurons in AR was investigated by selectively manipulating them through chemogenetics. The contribution of microglial neuroinflammation to central sensitization and AR was assessed by pharmacologically inhibiting microglial activation. The mechanism of central sensitization in AR was further determined using a microglia–neuron co-culture system. The role of interleukin (IL)-4 in neuroinflammation in AR was explored using in vivo and in vitro methods. Results: The intrinsic neuronal excitability and molecular markers of central sensitization were increased within the TNC of AR mice. Chemogenetic inhibition of TNC neurons ameliorated nasal symptoms, histological changes, autonomic dysfunction, and immune imbalance in AR mice. Moreover, AR mice exhibited an increased number of pro-inflammatory microglia, activation of the NOD-like receptor protein 3 (NLRP3) inflammasome, and enhanced IL-1β production in the TNC. Pharmacological inhibition of microglial activation reduced central sensitization in AR mice and was accompanied by remission of AR. Transcriptomics analysis revealed the inflammatory-prone characteristics of IL-4-treated microglia. IL-4 exposure enhanced lipopolysaccharide-stimulated NLRP3 inflammasome activation in microglia. Intracerebral injection of IL-4 neutralizing antibodies ameliorates neuroinflammation and central sensitization in AR mice and was accompanied by remission of AR. Conclusions: This research uncovers a previously unidentified mechanism wherein microglial neuroinflammation-induced central sensitization in the TNC contributes to AR, providing a new promising approach for the treatment of AR. IL-4 induces noncanonical pro-inflammatory-prone microglia and participates in microglial NLRP3 inflammasome activation in AR.

Antibodies safeguard our health through their precise and potent binding to specific antigens, demonstrating promising therapeutic efficacy in the treatment of numerous diseases, including COVID-19. Recent advancements in biomedical language models have shown the great potential to interpret complex biological structures and functions. However, existing antibody-specific models have a notable limitation that they lack explicit consideration for antibody structural information, despite the fact that both 1-dimensional sequence and 3-dimensional structure carry unique and complementary insights into antibody behavior and functionality. This paper proposes the Sequence-Structure multi-level pre-trained Antibody Language Model (S²ALM), combining holistic sequential and structural information in one unified, generic antibody foundation model. We construct a hierarchical pre-training paradigm incorporated with 2 customized multi-level training objectives to facilitate the modeling of comprehensive antibody representations. S²ALM’s representation space uncovers inherent functional binding mechanisms, biological evolution properties, and structural interaction patterns. Pre-trained over 75 million sequences and 11.7 million structures, S²ALM can be adopted for diverse downstream tasks: accurately predicting antigen–antibody binding affinities, precisely distinguishing B cell maturation stages, identifying antibody crucial binding positions, and specifically designing novel coronavirus-binding antibodies. Remarkably, S²ALM outperforms well-established and renowned baselines and sets new state-of-the-art performance across extensive antibody-specific understanding and generation tasks. S²ALM’s ability to model comprehensive and generalized representations further positions its potential to advance real-world therapeutic antibody development, potentially addressing unmet academic, industrial, and clinical needs.

Solvation free energies play a fundamental role in various fields of chemistry and biology. Accurately determining the solvation Gibbs free energy (\mathrm{\Delta }{G}_{\text{solv}}) of a molecule in a given solvent requires a deep understanding of the intrinsic relationships between solute and solvent molecules. While deep learning methods have been developed for \mathrm{\Delta }{G}_{\text{solv}} prediction, few explicitly model intermolecular interactions between solute and solvent molecules. The molecular modeling graph neural network more closely aligns with real-world chemical processes by explicitly capturing atomic-level interactions, such as hydrogen bonding. It achieves this by initially establishing indiscriminate connections between intermolecular atoms, which are subsequently refined using an attention-based aggregation mechanism tailored to specific solute–solvent pairs. However, its sharply increasing computational complexity limits its scalability and broader applicability. Here, we introduce an improved framework, molecular merged hypergraph neural network (MMHNN), which leverages a predefined subgraph set and replaces subgraphs with supernodes to construct a hypergraph representation. This design effectively mitigates model complexity while preserving key molecular interactions. Furthermore, to handle noninteractive or repulsive atomic interactions, MMHNN incorporates an interpretation mechanism for nodes and edges within the merged graph, leveraging the graph information bottleneck theory to enhance model explainability. Extensive experimental validation demonstrates the efficiency of MMHNN and its improved interpretability in capturing solute–solvent interactions.

Hydrogen sensing is of increasing importance in conjunction with the development and expanded utilization of hydrogen as an energy carrier or chemical reactant. This study focuses nanoscale hydrogen sensors that incorporate nanohybrid structural innovations to fabricate various systems, thereby enhancing detection efficiency and accuracy. Concurrently, advancements in optical hydrogen sensing and next-generation hybrid functional mechanisms have provided greater precision and universality with the aid of artificial intelligence of things (AIoT). For instance, optical hydrogen sensors offer high sensitivity and accurate gas detection with strong immunity to electromagnetic interference. Beyond optics, emerging models of next-generation composite multifunctional detection mechanisms provide operational advantages in hydrogen sensing, such as self-powering and long-range capabilities. In addition, the continual advancements in machine learning methods provide a feasible solution for data processing in hydrogen sensing applications through their integration with AIoT. This paper not only highlights the application of machine learning to enhance hydrogen sensor detection but also underscores its potential to improve the accuracy of future detection systems. In summary, these advances in nanohybrid structures, optical sensing, hybrid functional mechanisms, and machine learning integration represent strides in improving the performance, reliability, and versatility of hydrogen sensors, offering promising solutions for diverse hydrogen-related applications.

Leydig cells’ (LCs’) senescence is an important reason for the decline of testicular function in elderly men. Cellular communication network factor 5 (CCN5) regulates lipid metabolism and cellular fibrosis through multiple mechanisms. However, its role in LCs’ aging and the underlying molecular mechanisms remain unclear. This study aimed to elucidate the effects and molecular mechanisms by which CCN5 drives aging phenotypes in LCs and to evaluate the potential of targeting CCN5 as a therapeutic strategy for testicular aging. CCN5 expression was located in LCs and elevated in aged testis. Overexpression of CCN5 led to LCs’ aging and testis dysfunction. Extracellularly, CCN5 activated β-catenin and SMAD2/3 phosphorylation, promoting the expression of fibrosis-related genes. Intracellularly, CCN5 did not affect de novo cholesterol synthesis-related genes but changed the balance of cholesterol transporters. CCN5 bound to and reduced ring finger protein 213 (RNF213) protein levels. RNF213 knockdown activated forkhead box O, p16, and p21, resulting in SA-β-gal activation, reduced cell proliferation, and lipid droplet loss. In aged mice, CCN5 knockdown improved testicular atrophy, restored lipid droplet content and testosterone synthesis, and enhanced physical endurance and sexual behavior. In summary, CCN5 drives LCs’ aging and testicular dysfunction maybe via promoting fibrosis and lipid droplet loss. Targeting CCN5 offers a promising strategy to treat testicular aging and associated reproductive endocrine disorders.

Terahertz waves, positioned between infrared and microwave frequencies, have important potential in various fields, but their potential for in vivo biomedical applications has mostly remained untapped. In the present study, we focused on testing the potential of noninvasive high-frequency terahertz stimulation (HFTS) as a treatment for anxiety in mice. Mice were subjected to acute restraint stress to induce anxiety and then clustered into anxiety-susceptible and anxiety-resilient groups using the K-means algorithm. We developed an anxiety phenotype prediction classifier utilizing the naïve Bayes algorithm to accurately categorize mice. Noninvasive HFTS was targeted at the anterior cingulate cortex across the skulls of anxiety-susceptible mice, resulting in a marked anxiolytic effect. The underlying mechanism of HFTS’s anxiolytic effect was subsequently elucidated through in vivo and in vitro electrophysiological and morphological methods, revealing that HFTS decreases the excitability of pyramidal neurons in the anterior cingulate cortex by enhancing voltage-gated K⁺ channel and leak K⁺ channel conductance. The study not only expands the potential applications of HFTS, particularly its noninvasive use, in the regulation of anxiety-like disorders but also introduces innovative methodologies and insights that have the potential to lay the groundwork for future research in the field of physical biomedicine.

Breaking through cell membrane barriers is a crucial step for intracellular drug delivery in antitumor chemotherapy. Hereby, a magnetic nanorobot, capable of exerting mechanical agitation on cellular membrane to promote intracellular drug delivery, was developed. The main body of the nanorobots was composed of nano-scaled gold nanospikes that were deposited with Ni and Ti nanolayers for magnetic activation and biocompatibility, responsively. The nanorobots can be precisely navigated to target cancer cells under external magnetic field control. By virtue of the sharp nanospike structures, the magnetically powered rotation behavior of the nanorobots can impose mechanical agitation on the living cell membrane and thus improve the membrane permeability, leading to promoted transmembrane cargo delivery. Coarse-grained molecular dynamics simulation revealed that the mechanism of mechanical intervention regulated permeability of the bilayer lipid membrane, allowing for enhanced transmembrane diffusion of small cargo molecules. An in vitro study demonstrated that these nanorobots can markedly enhance the efficiency of drug entry into tumor cells, thus improving the effectiveness of tumor therapy under magnetic activation in vivo. This work paves a new way for overcoming cell membrane barriers for intracellular drug delivery by using a magnetic nanorobotic system, which is expected to promote further application of magnetically controlled nanorobot technology in the field of precision medicine.

Due to the low concentration of amyloid-beta (Aβ) in plasma and the high content of interfering factors, the conventional detection method for the quantification of Aβ still faces the problem of insufficient limit of detection (LOD). In this work, we propose a new light-triggered graphene–black phosphorus heterostructure (G-BP) field-effect transistor (FET) biosensing platform that achieves a ​​marked​​ reduction in the LOD. The LOD for Alzheimer’s disease (AD) biomarker Aβ₄₂ detection using the G-BP FET is as low as 235.1 zM (2.351 × 10⁻¹⁹ M), which is the lowest value reported to date and is approximately 2 to 3 orders of magnitude lower than other reported biosensing platforms. The G-BP FET platform provides precise, real-time guidance for non-invasive early diagnosis, disease monitoring, and personalized treatment plans for AD. Moreover, this method has good scalability and potential applications in other areas, including early detection of cancer and other major chronic diseases.