Circulating tumor cells (CTCs) hold great promise as liquid biopsy biomarkers for cancer diagnosis and treatment monitoring. However, their efficient enrichment and accurate identification from blood remain challenging due to their extreme rarity and the complex background of blood cells. Herein, we present an integrated strategy combining an inertial microfluidic chip with a deep learning algorithm for label-free isolation and intelligent identification of CTCs. The microfluidic chip incorporates periodic contraction-expansion arrays within a spiral channel, harnessing synergistic inertial lift force, Dean drag force, and vortex-induced lift force to achieve size-based, label-free, and rapid separation of MCF-7 cells. As a result, the microfluidic system achieved a CTC recovery rate of 89.2% and a white blood cell depletion rate of 86.9%. For CTC identification, the YOLOv8 deep learning model was applied to brightfield images of sorted cells, achieving 94.6% for both precision and recall without the need for cell labeling or specialized equipment. This integrated platform has significant potential for precision cancer diagnosis and therapeutic monitoring.
Microfluidic technologies, with advantages in low sample consumption, controllable fluid manipulation, structural integration, and system miniaturization, provide a promising route for adapting flow cytometry to point-of-care testing (POCT). Nevertheless, the practical adaptation of microfluidic flow cytometry to POCT scenarios still faces multiple challenges, such as bulky supporting components, relatively high system cost, complex workflows, and difficulties in multi-module integration. This review focuses on key technologies that can promote the development of microfluidic flow cytometry toward POCT applications. We introduce the overall architecture and basic working principles of microfluidic flow cytometry and then summarize recent advances in four major functional modules: fluid driving, cell focusing, signal detection, and data analysis. Specifically, this review discusses the transition from external pump-based control to low-external-dependency driving strategies, from sheath-flow focusing to sheathless single-cell focusing, and from conventional detection schemes to miniaturized detection approaches, together with the development of mobile-device-based readout and cloud-based analysis. Finally, we discuss key system-integration challenges in POCT-oriented microfluidic flow cytometry and highlight the need to develop cartridge-based, automated, intelligent, and user-friendly single-cell detection platforms tailored to specific diagnostic scenarios.
Biomacromolecules—including proteins, nucleic acids, and gene-editing tools—have demonstrated immense potential for therapeutic applications. However, their intracellular bioavailability is severely constrained by the cell membrane, which acts as a natural barrier to macromolecular entry. Conventional endocytic uptake pathways often lead to cargo entrapment within endosomes or lysosomes, followed by enzymatic degradation and consequently inefficient endosomal escape. To overcome this limitation, non-endocytic transmembrane delivery strategies have emerged as a compelling alternative. This review provides a systematic overview of recent advances in this field, with a particular emphasis on direct cytosolic delivery mechanisms, including physical membrane poration, cell-penetrating peptides (CPPs), membrane fusion, phase-separated carriers, virus-like particles (VLPs), and biological machines. By circumventing endosomal sequestration, these approaches enable the direct delivery of functional biomacromolecules into the cytosol, thereby offering innovative solutions for both therapeutic delivery and cellular engineering.
Diabetes-induced hyperglycemia leads to oxidative stress, inflammation, and impaired wound healing, particularly in gangrene. In this study, silver nanoparticles (AgNPs) were synthesized using flower extracts of Ixora coccinea and Rhododendron arboreum (IR-AgNPs) through a microwave-assisted green synthesis method. The nanoparticles were characterized using UV–spectroscopy, FTIR, XRD, SEM-EDX, and zeta size analysis. IR-AgNPs exhibited strong antioxidant activity surpassing that of ascorbic acid and significant anti-inflammatory effects exceeding diclofenac sodium. Broad-spectrum antimicrobial activity was observed against E. coli, S. aureus, P. aeruginosa, and Clostridium perfringens, key pathogens associated with gangrene. Cytotoxicity studies revealed minimal toxicity at concentrations ≤ 20 µg/mL. Under hyperglycemic conditions (25 mM glucose), 3T3-L1 fibroblasts showed decreased viability, elevated ROS, reduced glucose uptake, and downregulation of insulin signaling genes. Treatment with IR-AgNPs (25–100 µg/mL) restored cell viability, reduced ROS, enhanced glucose uptake, and promoted wound closure. Gene expression analysis confirmed upregulation of IRS-1, PI3K, AKT, and GLUT-4, indicating restored insulin sensitivity. These findings highlight the potential of IR-AgNPs as a multifunctional therapeutic approach for managing hyperglycemia-induced cellular dysfunction and promoting tissue repair in diabetic gangrene.
Colorectal cancer (CRC) remains a leading cause of cancer-related mortality worldwide, driven by tumor heterogeneity, late-stage detection, and variable therapeutic response. The integration of machine learning (ML) with multi-omics and probiotic research offers a transformative pathway toward precision CRC management. ML enables high-dimensional analysis of genomic, proteomic, metabolomic, and microbiome data, facilitating biomarker discovery, early disease detection, and individualized therapeutic design. Supervised and deep learning models have identified key microbial taxa, including Fusobacterium, Parvimonas, and Peptostreptococcus, as potential diagnostic biomarkers, while explainable AI enhances interpretability and clinical trust. Multi-omics integration bridges microbial and host metabolic interactions, revealing mechanisms linking dysbiosis with tumorigenesis. In parallel, ML-driven frameworks accelerate probiotic discovery, safety profiling, and the design of next-generation strains with anti-tumor and immunomodulatory properties. Reinforcement and adaptive learning models further enable personalized probiotic interventions by simulating host–microbiome dynamics. Despite challenges related to data heterogeneity, algorithmic bias, and regulatory validation, the convergence of ML, synthetic biology, and explainable AI heralds a new era of precision oncology. This integrative paradigm has the potential to transform probiotics from empirical supplements into intelligent, personalized therapeutics. Future directions emphasize the need for multi-cohort validation, robust model development to minimize overfitting, and the integration of ethical and regulatory frameworks to ensure reliable and clinically translatable ML applications in CRC management.
Direct mass spectrometry has become an important analytical approach in traditional Chinese medicine (TCM) because it enables rapid, direct, and minimally destructive characterization with little or no chromatographic pretreatment. In current TCM research, Direct mass spectrometry has shown clear advantages in rapid fingerprinting, authenticity assessment, process monitoring, and spatial chemical analysis, particularly in applications that require high throughput and in situ detection. However, its broader use remains constrained by several unresolved issues, including matrix effects, signal instability, limited quantitative robustness, poor interlaboratory reproducibility, and the lack of standardized spectral annotation and dedicated databases. These limitations mean that, at present, Direct mass spectrometry is more established for rapid qualitative assessment than for absolute quantification, cross-study comparison, and mechanism-related analysis. This review summarizes the major direct mass spectrometry platforms, their analytical principles, and their applications to different TCM analytical tasks and constituent classes, drawing mainly on representative studies from the past decade and including earlier foundational and early TCM-related reports where needed. It further highlights a central challenge in the field, namely the gap between rapid signal acquisition and reliable analytical interpretation, which continues to limit the extension of direct mass spectrometry from rapid screening to more robust practical application. Future progress will depend on methodological standardization, improved quantitative correction, stronger database support, and multimodal validation. Portable platforms and AI-assisted data analysis may further support field deployment and evidence interpretation when integrated into standardized and chemically interpretable workflows, thereby enabling direct mass spectrometry to play a more substantial role in TCM quality evaluation, process studies, and mechanism-related research.
Infected wounds feature an acidic microenvironment, providing an endogenous stimulus for targeted therapy. Taking advantage of the acid-triggered decomposition of zeolitic imidazolate framework-8 (ZIF-8), we constructed a pH-responsive theranostic composite (denoted as SH@ZIF-8/AgNPs, with shikonin encapsulated within ZIF-8 and silver nanoparticles (AgNPs) decorated on its surface) by encapsulating the bioactive compound shikonin (SH) within ZIF-8 and subsequently growing AgNPs in situ on its surface. The composite exhibited a high SH loading capacity (44.2%) and encapsulation efficiency (73.7%). Upon exposure to acidic infectious conditions, the ZIF-8 framework degraded, leading to on-demand release of SH and Ag+ ions, which together exhibited synergistic antibacterial effects against wound pathogens. Notably, the released components not only displayed enhanced antibacterial activity but also demonstrated potent concentration-dependent radical scavenging ability, mitigating oxidative stress. This work presents a dual-functional platform integrating antimicrobial intervention with antioxidant protection, offering a promising non-antibiotic strategy for smart, pH-responsive wound dressings in infected wound management.
The misuse of the broad-spectrum fungicide imazalil (IMZ) poses significant challenges to food safety and environmental safety, which requires rapid and accurate methods for detecting IMZ in complex food matrix and environmental samples. Herein, a novel anti-IMZ nanobodies (Nbs) with excellent stability and high expression yield was from a high-efficiency capacity of over 1014 pfu/mL alpaca phage display nanobody library. Furthermore, the molecular recognition mechanism of anti-IMZ Nbs was simulated, and it was found that the causes affecting antibody sensitivity were mainly attributed to the hydrophobicity scales of active binding pocket, with synergistic enhancement of hydrogen bonding forces, and then proposed a directed evolution strategy for optimizing the anti-hydrophobic small molecule antibodies. Subsequently, a nanobody-based nanogold immunochromatography assay (Nb-GICA) was developed for IMZ detection in fruits and environmental water samples. The Nb-GICA achieved rapid detection (8 min per test), with a broad linear range (8.87–380.57 ng/mL) and a limit of detection (LOD) of 6.05–16.56 ng/mL. Additionally, the average spiked recovery rates in samples ranged from 82.6% to 117.5%, showing good consistency with GC-MS/MS results. The results demonstrated that the developed nanobody was a promising recognition element for the development of immunological methods to monitor IMZ residues in food and environmental samples.
DNA methylation and active demethylation are key epigenetic mechanisms that dynamically regulate gene expression in multicellular organisms. Intermediates of active DNA demethylation—5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC)—are essential for genomic stability, embryonic development, and tissue-specific gene regulation. Their dysregulation is closely associated with diseases like cancer, making accurate detection vital for epigenetic research, early diagnosis, and treatment monitoring. However, detecting these intermediates is challenging due to their extremely low abundance, structural similarity to cytosine, and incompatibility with PCR-based amplification. This review summarizes recent advances in signal amplification methods for sensitive detection of these products, grouped into three categories: small molecule-based, nucleic acid amplification-based, and catalytically active or multi-molecule systems. These approaches overcome technical limitations through targeted chemical derivation, specific enrichment, and enzyme-assisted amplification, enabling precise quantification to epigenetic bases. Such progress has deepened our understanding of epigenetic networks. Future efforts should focus on improving sensitivity for ultra-rare species, simplifying protocols, and enhancing compatibility with clinical samples.
Hydrogels demonstrate significant value in the biomedical field, particularly in wound management and tissue engineering, owing to their unique three-dimensional network structure, excellent biocompatibility, and functional design flexibility. This study successfully developed a multifunctional hydrogel (QPHP) constructed from natural polymers of chitosan (CS) and hyaluronic acid (HA). CS was functionalized via quaternization and 3-nitro-4-carboxyphenylboronic acid to obtain QCSP, while HA was modified by grafting 4-fluorophenylboronic acid to prepare HAP. The hydrogel network is formed through dynamic boronate ester bonds and electrostatic interactions between QCSP and HAP. Experiments indicate that stable, self-supporting gels form when 2 % (w/v) QCSP is mixed with HAP at a concentration not less than 4 %. Performance characterization results show that the optimized formulation (2 % QCSP/5 % HAP) exhibited favorable swelling ratio and water retention capacity. Scanning electron microscopy images further confirm that the hydrogel possesses a uniform, porous three-dimensional network structure. Additionally, the QPHP hydrogel demonstrates good stretchability, rapid self-healing capability, and strong adhesion to various biological substrates. In summary, the QPHP hydrogel integrates adjustable physical properties, self-healing characteristics, and reliable bioadhesion, showing promising potential for application in advanced wound dressings.
Cancer remains one of the most challenging diseases worldwide due to limitations in conventional diagnostic and therapeutic approaches, including non-specific targeting, multidrug resistance, and systemic toxicity. The integration of nanotechnology with radiopharmaceutical science has given rise to radio conjugated nanoparticles (RNPs)—multifunctional platforms that combine molecular imaging and targeted radiotherapy within a single system. These nanoscale constructs, engineered with diagnostic and therapeutic radionuclides, offer enhanced tumor localization, real-time monitoring, and improved therapeutic outcomes while minimizing off-target effects. This review provides a comprehensive overview of recent advances in the design, radiolabelling strategies, and surface modifications of RNPs aimed at achieving personalized cancer management. It discusses the selection of radionuclides, nanoparticle carriers, and targeting ligands, as well as analytical evaluation methods such as radiochemical purity, stability, and pharmacokinetics. Furthermore, the clinical translation challenges—including large-scale synthesis, regulatory frameworks, and long-term safety concerns—are critically examined. The review also explores multimodal imaging approaches, combination therapies, and the emerging role of artificial intelligence in enhancing diagnosis and treatment planning. This work consolidates current knowledge on RNPs, highlighting their transformative potential in precision oncology and future directions toward enhanced clinical translation and patient well-being.
The convergence of consumer electronics and digital health has spurred the rise of earables—advanced in-ear wearable devices that exploit their unique anatomical position to monitor a wide range of physiological, biomechanical, and environmental signals. By enabling continuous, unobtrusive, and personalized health tracking, earables are shifting healthcare from clinic-based to decentralized models. This review outlines the technological foundations of earables, highlighting core sensing modalities such as photoplethysmography, electroencephalography, inertial measurement, and acoustic sensing, which collectively capture parameters including heart rate, oxygen saturation, temperature, and brain activity. We discuss their expanding applications in cardiometabolic monitoring, neuropsychiatry, vestibular assessment, respiratory health, and hearing augmentation. Key challenges—spanning data reliability, user adherence, privacy, clinical validation, and regulatory approval—are addressed alongside advances in machine learning and artificial intelligence (AI) that enable real-time data interpretation and personalized feedback. Looking forward, the development of novel biosensors, closed-loop interventions, and strategies for clinical integration positions earables as a pivotal platform for next-generation digital health systems.
Ischemic stroke is an acute neurological emergency caused by cerebral blood flow obstruction, with current diagnosis heavily dependent on time-consuming neuroimaging techniques that often delay critical intervention. Extracellular vesicles (EVs) have emerged as promising biomarker carriers for brain disorders due to their ability to cross the blood-brain barrier, yet their clinical translation has been hindered by complex isolation and detection requirements. Here, we develop a novel wash-free analytical platform leveraging liposome-EV fusion mediated by membrane lipid fluidity and integrated with Förster resonance energy transfer (FRET) technology. This innovative approach enables direct, rapid quantification of both EV concentration and EV-encapsulated miRNA in native biofluids without requiring prior EV separation, overcoming key limitations of conventional methods. When applied to clinical plasma samples, our method demonstrated high sensitivity with detection limits of 4.272 × 1011 particle/mL for EVs and 0.779 nM for miRNA. While no significant difference in total EV concentration was observed between ischemic stroke patients and hypertensive controls, EV-derived miRNA-21 levels were markedly elevated in patient samples, showing exceptional diagnostic performance (AUC = 0.915). This fusion-mediated FRET platform represents a significant advancement in EV-based diagnostics, offering a rapid, sensitive tool with substantial potential for point-of-care stroke diagnosis and timely intervention.
The global diabetes epidemic necessitates precise, affordable, and continuous glucose monitoring technologies. Graphene and its derivatives, with their high conductivity, tunable chemistry, and mechanical flexibility, have enabled major advances in glucose biosensing. Unlike prior reviews, this article emphasizes the translation of graphene-based sensors from laboratory studies to clinically relevant, wearable, and non-invasive systems. We highlight recent progress in laser-induced graphene (LIG) electrodes for low-cost, flexible platforms; hybrid nanostructures with transition metals and oxides for enzyme-free sensing with improved stability; and field-effect transistor (FET) and plasmonic devices achieving attomolar–femtomolar detection in sweat, saliva, and tears. Comparative analysis with commercial continuous glucose monitors (CGMs) underscores graphene’s potential to overcome challenges of enzyme degradation, biofouling, and cost. Fabrication strategies such as printing, microfluidics, and smartphone integration further advance prospects for real-time point-of-care use. This review provides a critical evaluation of enzymatic and non-enzymatic graphene-enabled glucose sensors across electrochemical, optical, and transistor modalities. It discusses benchmarks of sensitivity, selectivity, and stability, and identifies gaps in reproducibility, standardization, and regulatory approval. By focusing on clinical translation, wearable integration, and commercialization pathways, this work outlines how graphene-based biosensors could transform diabetes management beyond conventional enzymatic strips and current CGMs.
This computational study aims to explore the inhibition property of two plant-derived AMPs (antimicrobial peptides), defensin VvK1 and snakin-1, against SHV-1 β-lactamase from Klebsiella pneumoniae by molecular docking, MDS (molecular dynamics simulation) and MM-PB/GBSA (Molecular Mechanics-Poisson-Boltzmann/Generalized Born Surface Area) analysis. The AMPs (defensin VvK1 and snakin-1) were docked to SHV-1 β-lactamase for intermolecular (protein-peptide) interaction analysis and binding affinity prediction. SHV-1-defensin-VvK1 and SHV-1-snakin-1 complexes were subjected to MM-PB/GBSA free binding energy calculations after 10 ns MDS. To compare the efficacy of stronger AMP, defensin-VvK1, with reference inhibitor, cymal-6, 50 ns MDS was accomplished. Safety profiles of the AMPs were predicted by allergenicity and toxicity testing. Docking study revealed lower binding free energy for top-ranked SHV-1-snakin-1 complex than SHV-1-defensin-VvK1. SHV-1-defensin-VvK1 was energetically stronger than SHV-1-snakin-1 at 10 ns MDS run. After 50 ns MDS run, defensin-VvK1 (MM-PBSA: −70.73 ± 0.57 kcal/mol; MM-GBSA: −62.53 ± 0.43 kcal/mol) showed stronger efficacy than cymal-6 (MM-PBSA: −21.46 ± 0.50 kcal/mol; MM-GBSA: −26.79 ± 0.34 kcal/mol) against SHV-1. Both the AMPs demonstrated acceptable physicochemical properties, and were found non-allergenic and non-toxic. Thus, current in silico study validated the anti-Klebsiella pneumoniae activity of naturally occurring AMPs (snakin-1 and defensin-VvK1), wherein defensin-VvK1 showed higher activity against SHV-1 β-lactamase. Therefore, the AMPs currently being studied could be used as alternative biotherapeutic agents to combat life-threatening infections caused by antibiotic resistant bacterial pathogens. Overall, the findings can be translated into the process of designing and developing peptide-based therapeutics by targeting bacterial β-lactamases.
Polycystic ovary syndrome (PCOS) is a multifactorial endocrine and metabolic disorder characterized by hormonal imbalances, metabolic dysfunction, and chronic inflammation. To elucidate the molecular mechanisms underlying PCOS pathogenesis, this study employed an integrated multi-omics approach, combining serum metabolomic and proteomic analyses of clinically diagnosed PCOS patients and healthy controls. The findings revealed three key synergistic pathways contributing to disease progression: (1) dysregulation of the sex hormone - uric acid axis and altered polyunsaturated fatty acid (PUFA) metabolism; (2) oxidative stress-induced NAD+ depletion, driven by xanthine oxidase (XO) overactivity and aberrant MAPK/HSP/GSTP1 signaling; and (3) disturbances in lipid metabolism that trigger activation of the arachidonic acid - cyclooxygenase (COX)/prostaglandin inflammatory cascade. These interconnected molecular networks establish a pathological cycle characterized by hyperandrogenism, oxidative stress, and sustained inflammation, ultimately resulting in anovulation and morphological alterations of the ovaries. This study provides novel mechanistic insights into the molecular crosstalk governing PCOS and identifies potential therapeutic targets, including the XO/NAD+ redox axis and the arachidonic acid - COX inflammatory pathway. The systems biology framework presented here offers a foundation for developing personalized intervention strategies tailored to the heterogeneous phenotypes of PCOS.
The increasing incidence of gout necessitates monitoring of uric acid (UA) levels for early risk assessment and clinical management. Although nanomaterials with oxidoreductase-mimicking properties demonstrate great promise in natural enzyme-free UA detection, accurate quantification in authentic samples remains a challenge due to common interference from coexisting reductants and matrix complexity. To overcome this limitation, here we integrate nanozyme catalysis-based smartphone sensing with machine learning to develop regression models that can effectively eliminate matrix interference, thereby achieving the accurate quantification of UA in real urine. A new peroxidase-mimetic nanozyme (Na2V6O16) was prepared to catalyze the H2O2-mediated oxidation of 3,3’,5,5’-tetramethylbenzidine (TMB) to blue oxTMB. When UA existed, it could suppress the chromogenic process due to its reducibility, leading to a smartphone-based assay exhibiting linear response to UA (0.05–0.4 mM) with a detection limit of 0.045 mM. Crucially, to eliminate the interference of redox species (ascorbic acid, L-cysteine, glutathione, etc.) coexisting in samples, machine learning algorithms optimized through parameter tuning were employed to correct background signals from actual specimens, significantly enhancing detection accuracy and reliability. Our work achieves the precise quantitative prediction of UA in complex urine matrices, providing a sensitive, reliable, and non-invasive approach for gout risk stratification and management.
Over the past few decades, advancements in microfabrication and biomaterials have driven the innovation of microfluidic technologies, significantly advancing the development of lab-on-a-chip (LOC) systems. As transformative tools in precision medicine and high-throughput biomolecular analysis, LOC platforms can process and analyze numerous samples rapidly and cost-effectively, all within compact individual devices. This review summarizes the latest developments in the fabrication and detection technologies of LOC devices, along with their applications in precision medicine and biomolecular analysis. Furthermore, we anticipate a future where LOC technologies will redefine the landscape of medical diagnostics and biomolecular research, ultimately transforming biomedical research and clinical practice.
MicroRNA-155 (miR-155) is a critical biomarker implicated in various pathological processes, including cancer progression, immune response, and cardiovascular diseases. Consequently, accurate detection of miR-155 at low concentrations is essential for early diagnosis and effective treatment monitoring. To address this need, this study introduces a novel application of reduced graphene oxide/silver sulfide-gold (rGO/Ag₂S-Au) nanocomposites as signal tracers for the sensitive detection of miR-155. First, Ag₂S-Au nanoparticles were synthesized via in-situ reduction of HAuCl₄ on Ag₂S surfaces and then incorporated into rGO to form the nanocomposites. Subsequently, the composites were hybridized with signal probes to fabricate signal labels. In the detection mechanism, miR-155 triggers local catalytic hairpin assembly (L-CHA) to amplify signals. Finally, the products were immobilized on capture probe-modified electrodes, and electrochemical responses were measured by differential pulse voltammetry (DPV). As a result, the biosensor exhibits a linear detection range from 1 fM to 1 μM, with a detection limit as low as 0.17 fM. Overall, this work advances electrochemical biosensor technology and provides valuable insights for high-performance biosensor design in rapid clinical diagnostics.
