Browse Articles
Discover research articles across all indexed journals
Supporting cells orchestrate noise-induced hearing loss via a Gasdermin D-dependent signaling loop with hair cells
Abstract Noise-induced hearing loss (NIHL), a common sensory disorder, is traditionally thought to stem primarily from direct damage to sound-sensing hair cells (HCs). Here, we demonstrate that supporting cells (SCs), neighboring cells not previously implicated in NIHL pathogenesis, orchestrate hearing loss and HC degeneration through Gasdermin D (GSDMD) activation. Mechanistically, noise-induced oxidative stress in HCs triggers activation of epidermal growth factor receptor in SCs, leading to extracellular-regulated kinase phosphorylation and caspase-11-dependent cleavage of GSDMD, thereby establishing an HC-to-SC signaling cascade. Furthermore, GSDMD activation in SCs reciprocally exacerbates oxidative injury in HCs, creating a pathogenic positive feedback loop between the two cell types. Our findings uncover a central role for SCs in noise-induced hearing loss and identify GSDMD-mediated intercellular communication as a potential therapeutic target.
Validation of an in vitro muscle platform to evaluate myogenesis and calcium handling in control and dystrophic human myotubes
Abstract Electrical impedance has emerged as a powerful tool for real-time, label-free, and non-invasive monitoring of cellular processes. Here, we employed an impedance-based assay to characterize the myogenic process of control and dystrophic human myoblasts. First, we conducted a comprehensive analysis of control myoblast differentiation, assessing the effects of initial seeding density and various extracellular matrix coatings. We also evaluated the influence of electrode presence and current application, both of which improved myoblast alignment. Immortalized myoblasts from Duchenne muscular dystrophy patients exhibited marked alterations in early differentiation and maturation, which were readily detected via impedance measurements. We further compared two differentiation protocols using one control and one dystrophic representative cell line. While both protocols supported the formation of mature myotubes, impedance profiles differed depending on the culture medium. Notably, we identified the protocol with superior impedance profile reproducibility over the culture lifespan. Finally, we successfully assessed calcium homeostasis in control and dystrophic myotubes differentiated on 96-well impedance plates. Our findings underscore the potential of impedance-based assays for monitoring myogenesis and identifying disease-associated phenotypes. Moreover, 96-well impedance plates represent a robust tool for high-throughput and high-content functional analysis in muscle disease modeling and therapeutic screening.
NutriSighT: Interpretable Transformer Model for Dynamic Prediction of Underfeeding Enteral Nutrition in Mechanically Ventilated Patients
Evaluation of synergistic and regulatory effects of carbon-reduction and water-saving in the Yangtze River Delta Urban Agglomeration
Peptidoglycan recruitment by a penicillin binding protein
Abstract The cell wall is essential for bacterial survival. Its core component is peptidoglycan (PG), a polymer comprised of disaccharide-peptides (stem peptides) that are cross-linked to one another via transpeptidation by penicillin-binding proteins (PBPs). While much is known about how PBPs are inactivated by β-lactam antibiotics, little is known about how PBPs bind and catalyze the transpeptidation of PG. Here we show how native PG and stem peptides are recruited to PBP5 of E. faecium , a critical ESKAPE pathogen. We discovered that PG binds PBP5 at the periphery of the PBP active site cleft, not the active site, and that the D-Ala leaving group contributes minimally to PBP binding. We show that β-lactam antibiotics and stem peptides can bind PBP5 simultaneously. We also show that only the single central residue of the stem peptide (L-Lys substituted by D-iAsn in E. faecium ) is both necessary and sufficient for peptide recruitment. Finally, we translate our molecular findings by demonstrating that recruitment binding variants are unable to create a PG cell wall in E. faecium . Our studies define the key molecular interactions that govern bacterial cell wall formation and provide opportunities for the development of antibiotics that do not rely on PBP inactivation.
Protease profiling in fecal samples: a novel non-invasive diagnostic tool for gastrointestinal disorders
Abstract Fecal protease profiling represents a promising frontier in the non-invasive diagnosis of gastrointestinal disorders, particularly inflammatory bowel diseases (IBD), such as Crohn’s disease and ulcerative colitis, and irritable bowel syndrome (IBS). These conditions share overlapping symptoms but differ significantly in etiology and pathology, making accurate differentiation essential for appropriate management. This pilot study investigated protease activity in stool samples using a custom panel of fluorogenic peptide substrates across varying pH conditions to uncover disease-specific enzymatic signatures. Samples from IBD patients revealed broad protease activation involving both serine and cysteine classes, while the small IBS cohort showed a tendency toward a pattern enriched in furin-like serine proteases pattern dominated by furin-like serine proteases, especially at alkaline pH. Notably, one substrate, Ac-RSVL-AMC, showed higher activity in UC than in CD at acidic pH and moderate discriminatory ability in this pilot cohort. Inhibition assays confirmed the enzymatic contribution of furin-like proteases, and follow-up analysis in remission-phase IBD patients indicated persistent dysregulation, suggesting potential biomarker utility beyond active inflammation. The observed substrate-specific activity profiles highlight the importance of sequence context in proteolytic cleavage and underscore the complexity of protease involvement in gastrointestinal pathology. These findings support fecal protease profiling as a promising, rapid and low-cost approach that could contribute to distinguishing IBD from IBS and differentiating IBD subtypes, providing a foundation for future minimally invasive diagnostic strategies that require validation in larger cohorts.
GSDME-dependent pyroptosis drives abdominal aortic aneurysm via promoting vascular senescence
Mechanistic analysis of flow functions in tertiary enriched gas injection following secondary lean gas injection
Enzymatic synthesis of key RNA therapeutic building blocks using simple phosphate donors
Abstract The rapid emergence of RNA therapeutics has highlighted the need for more efficient, scalable and sustainable methods for their manufacture. Biocatalytic approaches hold particular promise, but rely on a secure, sustainable and low-cost supply of nucleoside triphosphate (NTP) building blocks, including those containing chemical modifications. Here we report the development of a biocatalytic approach and engineered enzymes to convert widely available nucleosides into NTPs featuring pharmaceutically relevant modifications using inexpensive phosphate donors. Importantly our strategy obviates the need for ATP as a phosphate donor that complicates NTP isolation using existing methods. To showcase the utility of our approach, we employ an engineered acid phosphatase, polyphosphate kinase and acetate kinase to produce 2′- O -methoxyethyl-ATP (2′-MOE-ATP) and 2′-fluoro-ATP, key building blocks of commercial therapeutics. Finally, we show that crude NTPs from our process can be used directly in enzymatic oligonucleotide synthesis, obviating the need for costly NTP isolation or purification steps.
High-efficiency removal of methyl orange from wastewaters using polyimide/chitosan-MoS2-UiO-66 nanofiber adsorbents
Abstract Conventional adsorbents such as chitosan and layered MoS 2 show promising functional groups or surface activity but suffer from intrinsic drawbacks such as limited surface area, and nanosheet aggregation, which restrict their efficiency in rapid dye removal. To overcome these limitations, we designed polyimide/chitosan-MoS 2 -UiO-66 (PI-CS-MoS 2 -UiO-66) nanofibrous adsorbents that integrate structural stability, abundant functional groups, and high porosity within a single hybrid system. The nanofibers were prepared by electrospinning PI as a robust support, followed by dip-coating with CS-MoS 2 -UiO-66 and crosslinking. XRD, BET, FT-IR, TGA, and SEM confirmed the successful incorporation of the composite. Adsorption tests on methyl orange (MO) showed strong dependence on pH, dosage, initial concentration, and temperature. The maximum adsorption capacity reached 195.61 $$\:\frac{\text{m}\text{g}}{\text{g}}$$ with excellent correlation to the Langmuir isotherm (R 2 = 0.9841). Thermodynamic parameters (ΔG < 0, ΔH < 0) indicated a spontaneous and exothermic process. Moreover, the adsorbent maintained good performance over three adsorption-desorption cycles, demonstrating reusability. These results demonstrate that combining PI, CS, UiO-66, and MoS 2 offers a synergistic platform that addresses the shortcomings of the individual components and enables high-efficiency dye removal from wastewater.
Characterization of the GTPγS release function of a G protein-coupled receptor
Unraveling the spatiotemporal clustering of malaria incidence and modeling a decade of epidemiological data in Jimma Zone, Oromia, Ethiopia
Reconstructing epigenomic dynamics through a single-cell multi-epigenome data integration framework
Multi-sensor observer-based residual learning with Auto-Permutation Feature Importance for fault diagnosis of multistage centrifugal pumps under variable pressures
Structural snapshots of Pseudomonas aeruginosa LptB2FG and LptB2FGC reveal insights into lipopolysaccharide recognition and transport
Abstract Gram-negative bacteria are intrinsically resistant to many antibiotics because of densely packed lipopolysaccharides (LPS) in the outer leaflet of their outer membrane (OM), which acts as a highly effective barrier towards the spontaneous permeation of toxic molecules, including antibiotics. LPS are extracted from the inner membrane by the ABC transporter LptB 2 FGC and translocated across the periplasm via a protein bridge to the OM. While structural studies have elucidated aspects of Lpt function in enterobacteria, little is known about how this system operates in divergent species such as Pseudomonas aeruginosa , a major human pathogen. Here, we report five cryo-electron microscopy structures of P. aeruginosa LptB 2 FG and LptB 2 FGC, revealing a rigid body movement in the periplasmic β-jellyroll domains necessary for LPS to shuttle through the periplasmic space. Notably, these structures exhibit a significantly smaller LPS binding cavity compared to previously determined models, suggesting the ligand-unbound states of the transporter. Mass spectrometry and molecular dynamics simulations indicate that the phosphate groups of LPS are the key determinants for binding and that the transporter can also accommodate cardiolipin. Together, these findings reveal previously unappreciated structural diversity in the Lpt system and provide mechanistic insight into how pathogenic Gram-negative bacteria tailor LPS recognition and transport. This understanding offers new avenues for the development of novel inhibitors targeting membrane biogenesis.
Evaluating the impact of housing modifications on milk infrared spectra as indicators of dairy cow welfare status
Accelerated discovery of crystalline materials with record ultralow lattice thermal conductivity via a universal descriptor
LBNet: an optimized lightweight CNN for mammographic breast cancer classification with XAI-based interpretability
Abstract Breast cancer represents a major worldwide health burden, marked by high incidence and mortality rates across diverse socioeconomic populations. While deep learning has enabled advances in automated mammographic analysis, existing models often suffer from high computational complexity. They also face limited generalizability and a lack of interpretability. To overcome these challenges, this research introduces LBNet, a lightweight and interpretable convolutional neural network (CNN) built for accurate and efficient breast cancer detection, particularly in resource-constrained settings. With only 2.4 million trainable parameters, LBNet consists of five convolutional layers, leveraging ReLU activation, batch normalization, and max-pooling to optimize feature extraction while maintaining computational efficiency. LBNet was trained on the RSNA dataset using the Adam optimizer and five-fold cross-validation. It achieved 97.28% accuracy. For cancer cases, precision was 99% and recall was 96%. For non-cancer cases, precision was 96% and recall was 99%. In comparison, baseline models such as VGG19, SE-ResNet152, and ResNet152V2 yielded lower accuracies of 87.54%, 87.50%, and 85.24%, respectively, while transfer learning approaches peaked at 87.37% accuracy. LBNet’s generalizability was validated in external datasets, achieving 99.54% accuracy on CBIS-DDSM and 98.50% on MIAS. To enhance clinical trust, this work integrated SHAP (SHapley Additive exPlanations) and Grad-CAM (Gradient-weighted Class Activation Mapping). These methods effectively highlighted diagnostically relevant regions in mammograms. This improved prediction transparency. LBNet demonstrates strong potential as an accurate, efficient, and interpretable solution for breast cancer screening, and future studies could explore its extension to multi-view mammography and real-time clinical deployment.