Redox-amplified pyrolytic glassy carbon sensor for ultrasensitive simultaneous detection of nitrate and nitrite.

Microfluidic reactors for the synthesis of inorganic and hybrid nanoparticles for drug delivery.

The Kirsten rat sarcoma viral oncogene homolog (KRAS) is one of the most frequently mutated oncogenes and is associated with poor prognosis. Long considered an undruggable target, KRAS has recently become actionable with the development of direct inhibitors, particularly against the G12C mutation. Our group previously reported promising efficacy and safety results for glecirasib (JAB-21822), a novel KRASG12C inhibitor, in phase I/II trials involving solid tumors harboring KRASG12C mutations (ClinicalTrials.gov NCT05009329, NCT05194995). Nevertheless, primary (5.61%) and acquired (9.64%) resistance were observed. This study analyzed 18 patients with advanced solid tumors harboring the KRASG12C mutation from the JAB-21822 cohort. Longitudinal blood samples (N = 45) were collected at baseline, during partial response or stable disease, and at disease progression. Circulating tumor cells (CTCs) were isolated via a microfluidics platform (CTC100, Cellomics) and categorized into epithelial (E-CTCs), mesenchymal (M-CTCs), and epithelial/mesenchymal mixed (E/M-CTCs) subtypes. At progression, the proportion of E-CTCs showed a decreased trend, while that of E/M-CTCs increased significantly (p = 0.03). In long-term responders, the inflection points of declining M-CTC levels correlated with clinical progression and were consistent with radiographic outcomes. Baseline CTC counts > 1 were associated with shorter progression-free survival (PFS; p = 0.046). E-CTC ≤ 1 correlated with longer PFS (p = 0.025), and M-CTC ≤ 1 with longer overall survival (OS; p = 0.033). E/M-CTC ≤ 1 showed a trend toward improved OS (p = 0.086). In addition, patients with > 1 CTC who received local radiotherapy for progressive lesions after glecirasib targeted therapy had significantly prolonged PFS and OS compared to those who did not (p < 0.05).

Engineering ECM-mimetic scaffolds from biological macromolecules for physiologically relevant tumoroid models.

Non-enzymatic oxidative stress biomarkers such as thiobarbituric acid reactive substances (TBARS), protein carbonyls (PC), and reduced glutathione (GSH) are among the most widely applied endpoints in aquatic biomonitoring and ecotoxicology, with relevance for assessing pollution across coastal, estuarine, and marine ecosystems. Despite decades of use, however, their translation into robust environmental monitoring tools remains constrained by analytical and interpretative bottlenecks, including poor interlaboratory comparability, lack of harmonized sampling and storage procedures, matrix interferences, inconsistent normalization strategies, and insufficient quality assurance frameworks. This review critically examines the use of non-enzymatic oxidative stress biomarkers in aquatic systems, shifting the focus from biomarker relevance toward methodological and technological barriers that currently limit operational deployment. We synthesize evidence on methodological variability, analytical artefacts, and confounding factors affecting biomarker responses across biological matrices and environmental contexts. Our analysis highlights that TBARS, PC, and GSH are highly sensitive indicators of oxidative damage associated with exposure to diverse contaminant classes, including metals, organic pollutants, and complex contaminant mixtures. However, substantial heterogeneity in analytical approaches and experimental designs across studies continues to hinder interstudy comparability and the establishment of ecologically meaningful baseline values. Finally, we propose a technology-oriented roadmap for improving biomarker standardization and field applicability, including minimum reporting requirements, validation priorities, and scalable analytical strategies such as miniaturized assays, microfluidic platforms, biosensing technologies, and smartphone-based detection systems. Overall, the integration of these approaches may facilitate the transition of non-enzymatic oxidative biomarkers from research endpoints to standardized and operational tools for environmental monitoring and decision-making in aquatic ecosystems.

Microfluidic sensing technologies for exosome-based isolation, detection and therapy in panvascular diseases.

The establishment of monoclonal, stably transduced cell lines is a critical step in functional genomics and drug discovery. However, conventional methods are often time-consuming, labor-intensive, and prone to compromising cell viability. Here, we present a microfluidic single-cell sorting system based on laser-induced jetting (LIJet) that significantly improves the efficiency and quality of stable cell line generation. This system integrates a light-responsive substrate with metal coating and a PDMS microfluidic chip featuring an array of microwells, enabling single-cell capture, identification, and non-contact precision release. A 532nm nanosecond pulsed laser is used to generate localized microjets, which accurately eject target cells from the microwells. In addition to achieving a 100% sorting success rate and maintaining over 95.3% post-sorting cell viability, the system supports long-term on-chip culture and viral transduction with full real-time monitoring. We demonstrated the platform's functionality by performing on-chip ZsGreen lentiviral transduction of human lung adenocarcinoma PC9 cells, followed by fluorescence-based single-cell selection, ultimately establishing monoclonal cell lines with stable transgene expression. This platform offers notable advantages in low-damage manipulation, dynamic monitoring, and functional perturbation, providing a robust and efficient solution for the construction of stably transduced cell lines, gene function screening, and phenotypic analysis across a variety of biomedical applications.

Sensitivity-improved nanozyme LFIA with optimized fluid control on centrifugal microfluidic platform.

Minimalist digital microfluidics for high-performance multiplexed protein POCT: A wash-free, sample-to-answer, and trace-sample paradigm.

Diabetic peripheral neuropathy (DPN) is a common, incurable complication of diabetes that causes sensory loss, pain, and motor problems. Conventional treatments like blood glucose management, pain relief, and neuroprotective drugs have limited success and do not prevent disease progression. Advances in neurobiology, regenerative medicine, and bioengineering have led to novel therapies that target underlying mechanisms and promote regeneration. Monitoring and evaluating the onset and progression of DPN are essential for effective clinical management. Given rapid advances in understanding DPN and developing new treatments, a comprehensive review that covers clinical progress, molecular pathology techniques, and emerging bioengineering strategies is both timely and essential. This review addresses: (1) DPN pathophysiology; (2) drug therapies from clinical trials since 2020; (3) animal models used in DPN research; (4) progress and challenges in biomaterial-based drug delivery systems; (5) developments and limitations of microfluidic platforms for DPN modeling; and (6) bioengineered devices used for DPN diagnosis and monitoring. Integrating clinical insights, molecular techniques, and bioengineering innovations seeks to create a forward-looking framework for next-generation DPN treatment and management.

N-Nitrosodimethylamine (NDMA) is a genotoxic nitrosamine that is commonly found in water due to its ready formation from commonly available precursor compounds. Routine monitoring of NDMA in water is challenging due to the need for complex sample preparation and instrumentation. Here, we propose the first example of an NDMA assay on a paper-based microfluidic platform. Moreover, it is the first study to integrate the required photochemical nitrosation reaction directly on-chip, enabling a complete photochemical-colorimetric workflow. The device is constructed with hollow PET capillary channels, double-sided adhesive, and glass fiber/paper pads with immobilized reagents, enabling sample analysis by simply dipping the device into water. On-chip detection of NDMA with sodium 1-naphthol-4-sulfonate is followed by Fe2+- and Co2+-based complexation, producing complementary green and red chromogenic signals. Its user-friendly dip-and-fold operation facilitates on-site, rapid, and high-frequency monitoring. The Fe2+- and Co2+-based channels provide linear detection ranges of 0-100 ppm and 0-30 ppm, with limits of detection of 4 ppm and 1.5 ppm, respectively. Interference studies, proper masking strategies, and dual-channel readout ensured high selectivity, while stability tests demonstrated excellent device stability over four weeks. Spike recovery experiments in tap, river, and lake water showed consistent recovery performance with RSD < 10%. This sustainable platform costs less than $0.2 per device, is compatible with smartphone-based readout, and offers a practical alternative to conventional analytical techniques. The findings presented here open possibilities to expand on-chip photochemical reactions in paper-based and other microfluidic systems for the detection of a broader range of contaminants.

Wearable molecularly imprinted polymer sweat testosterone sensor for noninvasive auxiliary early-stage polycystic ovary syndrome at rest.

A ROS-based droplet microfluidic platform for rapid antimicrobial susceptibility testing and single-cell heterogeneity analysis

For the precise molecular detection of bloodstream infection, sample processing is critical, as it involves purifying pathogenic DNA to circumvent the inhibitory effect of complex blood components. To integrate the cumbersome workflow encompassing the entire process to realize automated operation, passive methods such as gravity-driven microfluidics eliminate additional driving and controlling modules to simplify device setup and operations. Nevertheless, gravity-driven microfluidics encounters challenges in handling complex flow control and necessitates a relatively large volume of fluid to initiate the flow. In this study, to harness the simplicity of gravity for manipulating the cumbersome processing and achieving the automated detection of bacterial pathogens from whole blood, a novel rotation-programmed microfluidic platform was developed. Through simple procedural rotation of the chip, DNA extraction, LAMP reaction, and detection were integrated into the gravity-driven microfluidic platform. Complex handling procedures, such as sample processing including adsorption, washing, and elution, can be readily automated by antibiotic-modified magnetic nanoparticles on the chip. The sample-to-answer process was completed within 1.5 h, and the limit of detection for S. aureus from whole blood was 102 CFU mL-1. Machine learning-based image processing enabled rapid and accurate result readout through colorimetric detection. Beyond the diagnosis of bloodstream infections, this platform can be potentially extended to other complex and multistep processing applications, such as protein assay, by adjusting the design and programming.

Diatoms are environmentally responsive photosynthetic microorganisms whose growth dynamics and biosilicification processes are tightly regulated by external physicochemical conditions. However, conventional bulk cultivation and existing microfluidic platforms often fail to provide stable three-dimensional confinement together with dynamic environmental control, limiting long-term quantitative analysis at single-cell resolution. Here, we present a permeable hydrogel microreactor system integrated with microfluidic perfusion and a neural network-based image analysis workflow for on-chip investigation of diatom growth dynamics. Monodisperse alginate/carboxymethyl chitosan hydrogel microspheres were engineered to stably confine individual Cyclotella cryptica cells while permitting efficient molecular exchange. The microreactors were immobilized within a perfused microfluidic device, enabling long-term cultivation and real-time imaging under dynamically regulated conditions. Coupled with this neural network-based approach for segmentation and contour extraction, we quantitatively reconstructed single-cell growth trajectories and division events, achieving a specific growth rate of 1.874 d-1 under perfusion, which represents a 5.5-fold increase over batch controls. Furthermore, dynamic copper exposure enabled concentration-dependent stress profiling, yielding EC50 values of 8.53 μM (growth inhibition) and 7.25 μM (proliferation inhibition) at single-cell resolution. This platform offers a versatile analytical framework for resolving cellular heterogeneity and environmental responses in photosynthetic microorganisms under precisely controlled microenvironments.

Droplet digital enzyme-linked immunosorbent assay (ddELISA) enables ultrasensitive protein quantification; however, its performance is often limited by low single-bead encapsulation efficiency due to Poisson statistics and significant background interference from empty droplets. We present an integrated microfluidic system that overcomes both limitations through synergistic magneto-inertial manipulation (SMIM). By coupling an external magnetic field with laminar inertial focusing, the system applies coordinated hydrodynamic lift and magnetophoretic forces on immunomagnetic beads, thereby enabling three interdependent functions within a continuous workflow: ordered single-file bead focusing, high-throughput single-bead encapsulation, and active sorting of bead-containing droplets. Under optimized conditions (30 μL min-1; 300 mT), magnetic beads are focused into a stable single-file train with a lateral deviation of only 0.95 ± 1.21 μm. This deterministic ordering prior to droplet generation fundamentally overcomes the Poisson limit, achieving single-bead encapsulation efficiency of 81.31%─a 2.21-fold improvement over stochastic loading─and a total encapsulation efficiency of 91.11%. The integrated magnetic sorting module generates localized high-gradient fields, capturing bead-containing droplets with 93.97 ± 1.18% efficiency while eliminating >91% of empty droplets. The platform's quantitative capability was validated using a dual-color fluorescent bead assay (R2 = 0.9977; 0.50 - 1.75 × 107 beads mL-1) and an E. coli O157 ddELISA (R2 = 0.9999; LOD = 18.23 cells mL-1). This work establishes a fully integrated microfluidic platform that addresses the long-standing challenges of encapsulation inefficiency and empty-droplet interference in ddELISA by transforming bead encapsulation from a stochastic to a deterministic process, thereby providing a broadly applicable framework for high-throughput, single-particle-resolved digital bioanalysis.

An all-in-one microfluidic system via data-driven design for on-site genotyping of genetically modified foods.

Immune-driven stromal inflammation in pancreatic cancer within a microfluidic platform

Heavy metal ion pollution has emerged as a global issue. These contaminants are not only present in water sources but are also commonly detected in air, soil, food, and consumer products, posing serious risks to ecosystems and human health. Even at very low concentrations, heavy metal ions can exhibit substantial toxicity. Traditional methods for the detection of heavy metal ions typically require complex laboratory equipment and specialized technicians, making them inadequate for rapid on-site monitoring. Microfluidic technology, as an innovative platform capable of precisely controlling and manipulating minute volumes of fluid, has demonstrated enormous potential in analytical chemistry, biomedicine, and environmental monitoring. In the rapidly developing field of microfluidics, paper-based microfluidic platforms have become prominent due to their low cost, straightforward fabrication, and eco-friendly nature, offering powerful tools for the detection of heavy metal ions in diverse samples. This survey consolidates the major advances reported from 2015 to 2025 in utilizing paper-based microfluidic systems for identifying heavy metal ion pollutants in diverse sample types, including air, explosive residues, water sources, herbal supplements, skin-whitening cosmetics, environmental aerosols, urine, soil, gunshot residues, cucumber plants, and food. The review analyzes in detail the principles and applications of detection strategies based on colorimetric methods, fluorescent methods, electrochemical methods, dual-detection systems, and other methods, as well as the role of nanomaterials and selective recognition elements in improving detection sensitivity and specificity. These portable, low-cost, and easy-to-operate detection systems provide viable solutions for environmental and public health monitoring, particularly suitable for resource-limited regions and scenarios requiring rapid detection.

Bottom-up synthetic biology has achieved remarkable success in designing molecular architectures at the nanoscale. However, constructing functional, cell-sized artificial systems at the mesoscale remains a significant challenge due to the stochastic nature of pure molecular self-assembly. Natural cells overcome this thermodynamic limitation by employing active, energy-consuming mechanisms to maintain size, identity, and structural hierarchy. In this Perspective, I argue that, to construct consistent and robust molecular systems that can withstand practical use, we must seek a synergy between intrinsic self-assembly and deterministic engineering. Microfluidics should be viewed not merely as a replacement for biological regulation, but as a platform that provides extrinsic physical boundary conditions to guide and amplify the potential of molecular self-assembly. By categorizing microfluidic platforms into four hierarchical levels─from molecular assembly to multicellular bodies─I illustrate how engineering can recapitulate or replace complex biological regulatory systems. Finally, I discuss the necessary paradigm shift from bespoke laboratory craftsmanship to standardized biofoundries, outlining the applications and manufacturing breakthroughs required to democratize artificial cell technology for industrial and biomedical applications.