Rongrong Zhu

Spinal cord injury (SCI) remains a severe condition with an extremely high disability rate. The challenges of SCI repair include its complex pathological mechanisms and the difficulties of neural regeneration in the central nervous system. In the past few decades, researchers have attempted to completely elucidate the pathological mechanism of SCI and identify effective strategies to promote axon regeneration and neural circuit remodeling, but the results have not been ideal. Recently, new pathological mechanisms of SCI, especially the interactions between immune and neural cell responses, have been revealed by single-cell sequencing and spatial transcriptome analysis. With the development of bioactive materials and stem cells, more attention has been focused on forming intermediate neural networks to promote neural regeneration and neural circuit reconstruction than on promoting axonal regeneration in the corticospinal tract. Furthermore, technologies to control physical parameters such as electricity, magnetism and ultrasound have been constantly innovated and applied in neural cell fate regulation. Among these advanced novel strategies and technologies, stem cell therapy, biomaterial transplantation, and electromagnetic stimulation have entered into the stage of clinical trials, and some of them have already been applied in clinical treatment. In this review, we outline the overall epidemiology and pathophysiology of SCI, expound on the latest research progress related to neural regeneration and circuit reconstruction in detail, and propose future directions for SCI repair and clinical applications.

Spinal cord injury (SCI) involves diverse injury responses in different cell types in a temporally and spatially specific manner. Here, using single-cell transcriptomic analyses combined with classic anatomical, behavioral, electrophysiological analyses, we report, with single-cell resolution, temporal molecular and cellular changes in crush-injured adult mouse spinal cord. Data revealed pathological changes of 12 different major cell types, three of which infiltrated into the spinal cord at distinct times post-injury. We discovered novel microglia and astrocyte subtypes in the uninjured spinal cord, and their dynamic conversions into additional stage-specific subtypes/states. Most dynamic changes occur at 3-days post-injury and by day-14 the second wave of microglial activation emerged, accompanied with changes in various cell types including neurons, indicative of the second round of attacks. By day-38, major cell types are still substantially deviated from uninjured states, demonstrating prolonged alterations. This study provides a comprehensive mapping of cellular/molecular pathological changes along the temporal axis after SCI, which may facilitate the development of novel therapeutic strategies, including those targeting microglia.

The differentiation of neural stem cells (NSCs) into neurons is proposed to be critical in devising potential cell-based therapeutic strategies for central nervous system (CNS) diseases, however, the determination and prediction of differentiation is complex and not yet clearly established, especially at the early stage. We hypothesize that deep learning could extract minutiae from large-scale datasets, and present a deep neural network model for predictable reliable identification of NSCs fate. Remarkably, using only bright field images without artificial labelling, our model is surprisingly effective at identifying the differentiated cell types, even as early as 1 day of culture. Moreover, our approach showcases superior precision and robustness in designed independent test scenarios involving various inducers, including neurotrophins, hormones, small molecule compounds and even nanoparticles, suggesting excellent generalizability and applicability. We anticipate that our accurate and robust deep learning-based platform for NSCs differentiation identification will accelerate the progress of NSCs applications.