Microarray Platforms Based on 3D Printing.

This paper introduces an innovative method for the fabrication and infusion of microwell arrays based on digital light processing (DLP) 3D printing. A low-cost DLP 3D printer is employed to fabricate microstructures rapidly with a broad dynamic range while maintaining high precision and fidelity. We constructed microwell arrays with varying diameters, from 200 to 2000 µm and multiple aspect ratios, in addition to microchannels with widths ranging from 45 to 1000 µm, proving the potential and flexibility of this fabrication method. The superimposition of parallel microchannels onto the microwell array, facilitated by positive or negative pressure, enabled the transfer of liquid to the microwells. Upon removal of the microchannel chip, a dispensed microdroplet array was obtained. This array can be modulated by adjusting the volume of the microwells and the inflow fluid. The filled microwell array allows chip-to-chip dispensing to the microreactor array through binding and centrifugation, facilitating multistep and multireagent assays. The 3D printing approach also enables the fabrication of intricate cavity designs, such as micropyramid arrays, which can be integrated with parallel microchannels to generate spheroid flowcells. This device demonstrated the ability to generate spheroids and manipulate their environment. We have successfully utilized precise modulation of spheroids size and performed parallel drug dose-response assays to evaluate its effectiveness. Furthermore, we managed to execute dynamic drug combinations based on a compact spheroids array, utilizing two orthogonal parallel microchannels. Our findings suggest that both the combination and temporal sequence of drug administration have a significant impact on therapeutic outcomes.

Large-Scale Integration of Single-Cell RNA-Seq Data Reveals Astrocyte Diversity and Transcriptomic Modules across Six Central Nervous System Disorders.

The dysfunction of astrocytes in response to environmental factors contributes to many neurological diseases by impacting neuroinflammation responses, glutamate and ion homeostasis, and cholesterol and sphingolipid metabolism, which calls for comprehensive and high-resolution analysis. However, single-cell transcriptome analyses of astrocytes have been hampered by the sparseness of human brain specimens. Here, we demonstrate how large-scale integration of multi-omics data, including single-cell and spatial transcriptomic and proteomic data, overcomes these limitations. We created a single-cell transcriptomic dataset of human brains by integration, consensus annotation, and analyzing 302 publicly available single-cell RNA-sequencing (scRNA-seq) datasets, highlighting the power to resolve previously unidentifiable astrocyte subpopulations. The resulting dataset includes nearly one million cells that span a wide variety of diseases, including Alzheimer's disease (AD), Parkinson's disease (PD), Huntington's disease (HD), multiple sclerosis (MS), epilepsy (Epi), and chronic traumatic encephalopathy (CTE). We profiled the astrocytes at three levels, subtype compositions, regulatory modules, and cell-cell communications, and comprehensively depicted the heterogeneity of pathological astrocytes. We constructed seven transcriptomic modules that are involved in the onset and progress of disease development, such as the M2 ECM and M4 stress modules. We validated that the M2 ECM module could furnish potential markers for AD early diagnosis at both the transcriptome and protein levels. In order to accomplish a high-resolution, local identification of astrocyte subtypes, we also carried out a spatial transcriptome analysis of mouse brains using the integrated dataset as a reference. We found that astrocyte subtypes are regionally heterogeneous. We identified dynamic cell-cell interactions in different disorders and found that astrocytes participate in key signaling pathways, such as NRG3-ERBB4, in epilepsy. Our work supports the utility of large-scale integration of single-cell transcriptomic data, which offers new insights into underlying multiple CNS disease mechanisms where astrocytes are involved.

Design and High-Resolution Analysis of an Efficient Periodic Split-and-Recombination Microfluidic Mixer.

We developed a highly efficient passive mixing device based on a split-and-recombine (SAR) configuration. This micromixer was constructed by simply bonding two identical microfluidic periodical open-trench patterns face to face. The structure parameters of periodical units were optimized through numerical simulation to facilitate the mixing efficiency. Despite the simplicity in design and fabrication, it provided rapid mixing performance in both experiment and simulation conditions. To better illustrate the mixing mechanism, we developed a novel scheme to achieve high-resolution confocal imaging of serial channel cross-sections to accurately characterize the mixing details and performance after each SAR cycle. Using fluorescent IgG as an indicator, nearly complete mixing was achieved using only four SAR cycles in an aqueous solution within a device's length of less than 10 mm for fluids with a Péclet number up to 8.7 × 104. Trajectory analysis revealed that each SAR cycle transforms the input fluids using three synergetic effects: rotation, combination, and stretching to increase the interfaces exponentially. Furthermore, we identified that the pressure gradients in the parallel plane of the curved channel induced vertical convection, which is believed to be the driving force underlying these effects to accelerate the mixing process.