Lab on a Chip最新文献

Sensitive and low-cost protein biomarker detection is critical for disease diagnosis. Advanced microfluidic systems can generate miniature reaction compartments for a high-sensitivity assay. However, these platforms often require external instruments, skilled operators, and complex setups. Here, we develop a lab-on-a-capillary (LabCap) platform that integrates photopatterned hydrogel rings within a glass capillary using a reconfigurable stop-flow lithography system. During sample loading and unloading steps, nanoliter-scale aqueous droplets (torodrops) are spontaneously formed around the hydrogel rings, creating isolated reaction compartments without the need for external instruments or an immiscible oil phase. The LabCap platform enables detection of clinically relevant biomarkers, including C-reactive protein (CRP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP). By adjusting the incubation protocol, the detection performance can be tuned to meet different analytical requirements. A 16 min periodic medium exchange protocol yielded calculated limits of detection of 1.3 ng mL^-1 for CRP and 1.8 ng mL^-1 for NT-proBNP, whereas static incubation yielded 0.5 ng mL^-1 for both biomarkers after 1 h and 0.14 ng mL^-1 for CRP at a matched 16 min incubation. In addition, LabCap offers practical advantages, including a low consumable material cost for fabricating one device (<€1), low reagent consumption (<100 μL per assay step), and minimal wash-buffer usage (1 mL). These results demonstrate that LabCap is a simple, cost-effective, and versatile platform for compartmentalized biomarker detection.

Precise microflow control underpins microfluidic technologies in precision medicine, yet conventional approaches rely on anisotropic materials or external fields, limiting fabrication simplicity and scalability. Unlike transformation-based or data-driven approaches, we demonstrate that flow anisotropy can be programmed through the layered organization of locally homogeneous, isotropic media via a series-parallel equivalent fluidic network (SPEFN) realized as single-medium metamaterials (SMMs). SPEFN provides an analytically invertible mapping from network topology to the macroscopic effective-viscosity tensor, from which a connectivity-controlled anisotropy law and an exact non-disturbance condition emerge as direct analytical consequences. We experimentally realize three monolithic SMM devices: a spiral-layered director, an annular cloak, and a fan-shaped concentrator. Particle-tracking velocimetry confirms the predicted responses across all devices, including ≈30° streamline reorientation and a concentrator velocity ratio of 1.37 versus a predicted value of 1.38. The cloak exhibits higher apparent suppression than predicted; tracer retention or adhesion visible in Movie S1 may partly explain this difference. The system further shows robust performance under strongly non-uniform inflow and sub-percent fabrication tolerances.

Label-free brightfield microscopy is widely used to monitor organ-on-chip (OOC) cultures, but manual quality assessment can be subjective, labor-intensive, and difficult to standardize across biological conditions. Using a public dataset of 3072 microscopy images from six cell lines, we systematically evaluated four questions relevant to automated OOC quality assessment: (1) how model architecture and image resolution affect performance, (2) whether predictive information is localized or distributed across the image, (3) whether biological metadata provides complementary information beyond the image, and (4) how well learned quality representations transfer across cell lines. We found that (1) both architecture and input resolution influenced performance, with increasing resolution from 224 × 224 to 768 × 768 improving ConvNeXt-Tiny accuracy from 0.852 to 0.910 and improving performance for five of six evaluated convolutional neural networks (CNNs); (2) multiple instance learning and saliency analyses indicated that quality relevant information was distributed across multiple image regions rather than dominated by isolated local defects; (3) biological metadata provided only modest additional benefit, with the best metadata-conditioned model achieving an accuracy of 0.915 compared with 0.910 for image-only classification; and (4) transfer to previously unseen cell lines varied substantially (ROC-AUC 0.783-0.999), while finetuning models with target-cell-line data substantially recovered performance, increasing pooled accuracy from 0.791 to 0.890. Domain-adversarial training did not consistently improve transfer. Rather than introducing a new deep learning architecture, the novelty of this work lies in systematically evaluating these OOC-specific design and generalization factors within a common framework. These findings highlight image resolution, biological-domain validation, and target-domain adaptation as important considerations for robust automated OOC quality assessment.

Label-free particle analysis across micro- and nanoscale particle systems requires multiple specialized instruments, making it difficult to compare measurements across scales within a single workflow. Here we show the use of NanoSpacer/MicroSpacer, a glass-based confinement platform, for particle detection, relative particle counting, and diffusion-based sizing at the single particle level. The platform confines liquid samples between spacer-defined glass surfaces, restricting the vertical observation volume while allowing particles to move freely within the imaging plane. This improves particle visibility under darkfield illumination and supports more reliable particle tracking for diffusion-based sizing in solutions. Diffusion coefficients are obtained from mean-squared displacement (MSD) analysis using open-source software and converted to hydrodynamic size via the Stokes-Einstein relationship. The platform was evaluated across four model systems spanning micro- to nanoscale: (i) binary GFP/mKate bacterial mixtures were quantified at the single-cell level, with GFP ratios of 59%, 18%, and 89% in agreement with correlative flow cytometry (65%, 23%, 86%) and colony-forming unit counts (54%, 17%, 86%); (ii) bacteriophages Polaria and Mystique were detected by darkfield and apparent size estimates returned modal hydrodynamic diameters of approximately 250 nm and 190 nm, consistent with electron microscopy; (iii) putative nanobubbles were produced by chemical and mechanical methods, with modal diameters of approximately 170 nm and 600 nm; (iv) polystyrene nanoplastic particles (100, 200, and 1000 nm) were imaged in freshwater and seawater, where salt-induced aggregation produced clusters spanning orders of magnitude in size. Together, these results establish NanoSpacer/MicroSpacer as a cost-efficient, microscopy-compatible workflow for particle metrology across micro- and nanoscale systems relevant to marine science and microbiology.

Immunosuppression induced by metabolic stress within the solid tumor microenvironment (TME) critically limits the efficacy of cancer immunotherapies. Here, we develop a tumor microenvironment-on-a-chip (TMoC) platform integrated with single-cell microfluidic analysis to precisely model both hypoxic and lipid-enriched conditions commonly found in solid tumors. Utilizing this system, we elucidate the distinct effects of these two metabolic stressors on natural killer (NK) cell function. Hypoxia broadly suppresses NK cell cytotoxicity and downregulates multiple functional molecules. In contrast, lipid accumulation exerts a selective inhibitory effect, reducing granzyme B and IL-10 expression while relatively preserving IFN-γ mRNA expression. Single-cell analysis reveals stress-induced functional heterogeneity and secretory synergy among NK cells. Notably, we identified a relatively high-IFN-γ-secreting subpopulation retained under lipid-rich conditions despite an overall reduction in IFN-γ secretion at the single-cell level. Furthermore, single-cell pairing assays dissociate global killing efficacy from killing kinetics, revealing that hypoxia and lipid accumulation impair NK cells through distinct kinetic modes. Using the TMoC platform to evaluate combination therapies targeting both hypoxia and lipid metabolism, we show that dual intervention synergistically restores NK cell cytolytic function. The platform demonstrates superior screening sensitivity compared to conventional in vitro models. Collectively, this work highlights the potential of integrating engineered microphysiological systems with single-cell analysis to decipher complex TME interactions and guide the development of immunotherapeutic strategies.

The integration of real-time sensors in microphysiological systems (MPS) faces steep challenges due to the mass transport limitations inherent to microfluidic environments, complex microdevice fabrication methods, and maintenance of long-term performance at high throughput. Reactive oxygen species (ROS) represent attractive yet elusive sensing targets for MPS due to their integral role in cellular metabolism, oxidative homeostasis, and response to disease and injury. Nevertheless, without real-time measurement, their short-lived nature limits practical investigations of ROS as biomarkers in biomedical research. In this work, we have integrated a 96-plex, real-time ROS measurement system into PREDICT96, a high-throughput MPS platform, as a first demonstration of a highly multiplexed electrochemical sensor capability. The ROS measurement system comprises a miniaturized amperometric sensor array capable of rapidly measuring H₂O₂, a prototypical ROS, at micromolar concentrations (1-10 μM) with high sensitivity and selectivity across 96 microphysiological devices. Integrated microfluidic pumping enables active mixing during measurement and supports the detection of targets with low concentrations or short lifetimes. This novel capability will enable future work investigating ROS as biomarkers for disease modeling and medical countermeasure evaluation, targeting viral respiratory infections of lung-immune tissue models.

Piezoelectric drop-on-demand inkjet printing enables controlled microdroplet deposition in microfluidics, biomanufacturing, and precision manufacturing. However, conventional pressure chamber piezoelectric printheads are susceptible to viscous dissipation and nozzle clogging, exhibit limited adaptability to high-viscosity, low-surface-tension liquids and particle/cell suspensions, and struggle to suppress satellite droplets effectively. These limitations hinder their practical deployment in high-precision and high-stability inkjet applications. Here, a dynamic tapered nozzle generates droplets through local self-pumping induced by geometric deformation rather than pressure-wave propagation in a conventional pressure chamber. Structural optimization enabled droplet velocities of 5.42-12.67 m s^-1 and stable ejection at viscosities up to 40.3 mPa s, surface tensions of 24.44-72.75 mN m^-1, and Z values of 1.01-38.11. Mechanistic analysis identified high-frequency residual nozzle oscillations as the primary cause of ligament stretching and secondary breakup. Based on this mechanism, a multiple waveform modulation strategy is devised to dynamically regulate the nozzle displacement response. This strategy effectively suppresses satellite droplets during high-speed ejection while maintaining printing throughput and achieves a 32% reduction in droplet diameter. The satellite-droplet suppression effect is further validated in a 0.5 wt% SiO₂ nanoparticle suspension and a 10 wt% PVP K30 solution. The platform enables high-viability printing of MC3T3 cells and carrier-free intracellular delivery of 70 kDa fluorescein isothiocyanate (FITC)-dextran, establishing a feasible pathway for high-speed and stable ejection across broad fluid property ranges with low clogging risk.

The precise manipulation of micro-scale targets is a foundational technology driving advancements in biomedical engineering, clinical diagnostics, and micro-manufacturing. Here, we present a multifunctional acoustic valve microrobotic system capable of the precise manipulation of cells and microorganisms. The focused surface acoustic wave transducers are utilized to form valve-like potential wells, thereby enabling stable entrapment and transportation of micro targets. Based on this ability, it can further realize direct sorting and enrichment, by sweeping the acoustic field through the targeted regions. Experimental results demonstrate that this platform can achieve high-speed manipulation exceeding 1000 μm s^-1, and can achieve sub-10 μm navigation accuracy. Furthermore, it supports highly efficient sorting and enrichment, delivering a target recovery rate of over 90% and an enrichment ratio of 5 to 10 fold in a single cycle. We verified the system's biocompatibility through thermal stability experiments and cell viability assays, and successfully applied the acoustically enriched microalgae to the photocatalytic degradation of tetracycline. Collectively, this versatile and highly adaptable acoustic platform provides a powerful tool for complex micro-manipulation, and has the potential to be used in cutting-edge applications such as cell analysis and biological assembly.

The liver and kidneys are essential for drug metabolism and clearance and are highly sensitive to toxicity. Most in vitro models assess these organs in isolation, limiting the study of inter-organ interactions. Here, an induced pluripotent stem cell (iPSC)-derived kidney-liver organ-on-a-chip model was developed that co-cultured hepatocyte-like cell (HLC) organobodies and proximal tubular-like cells (PTL) under microfluidic flow. In parallel, a PTL-HepaRG spheroid co-culture model was established to investigate bioactivation of the prodrug ifosfamide. Co-culture induced distinct transcriptional responses in PTL and HLC, as revealed by gene set enrichment analysis. In PTL, co-culture with HepaRG enriched signaling and cell polarity pathways, while co-culture with HLC enhanced metabolic and transport-related programs. In HLC, co-culture with PTL affected intermediary metabolism, lipid processing, redox regulation, and plasma protein synthesis. The PTL-HepaRG model demonstrated bioactivation of ifosfamide into metabolite chloroacetaldehyde, highlighting the utility of these systems for studying human-relevant kidney-liver interactions and drug metabolism in vitro. However, the concentration of chloroacetaldehyde was too low to cause adverse effects in the renal tissue in the current setup.

The blood-brain barrier (BBB) protects the central nervous system by restricting entry of harmful blood-borne factors, but this selectivity actively limits delivery of therapeutics to the brain. Because BBB function is shaped by dynamic interactions within the neurovascular unit (NVU), particularly between brain endothelial cells, pericytes, and astrocytes, there is a need for human-relevant models that capture both barrier properties and neurovascular crosstalk in a controlled setting. Here, we present a novel human induced pluripotent stem cell (iPSC)-based NVU-on-chip that couples a perfusable, vessel-like BBB compartment to an adjacent open-top chamber housing a human brain organoid (hBO) via a membrane-free, hydrogel-based interface. This design enables co-culture of BBB and parenchymal components in close proximity while allowing direct cell-cell interactions at the barrier-tissue boundary without an artificial porous membrane. Using this platform, we demonstrate stable on-chip co-culture and show that hBO-derived astrocytes can sprout, migrate toward the BBB compartment and establish direct contact with endothelial cells, recapitulating the NVU structure. This system offers a human-specific framework for modeling NVU biology in health and disease and for evaluating BBB penetration together with downstream effects of candidate therapeutics on hBO.