The spinal cord is a crucial component of the central nervous system that facilitates sensory processing and motor performance. Despite its importance, the spatiotemporal codes underlying human spinal cord development have remained elusive. In this study, we have introduced an image-based single-cell transcription factor (TF) expression decoding spatial transcriptome method (TF-seqFISH) to investigate the spatial expression and regulation of TFs during human spinal cord development. By combining spatial transcriptomic data from TF-seqFISH and single-cell RNA-sequencing data, we uncovered the spatial distribution of neural progenitor cells characterized by combinatorial TFs along the dorsoventral axis, as well as the molecular and spatial features governing neuronal generation, migration, and differentiation along the mediolateral axis. Notably, we observed a sandwich-like organization of excitatory and inhibitory interneurons transiently appearing in the dorsal horns of the developing human spinal cord. In addition, we integrated data from 10× Visium to identify early and late waves of neurogenesis in the dorsal horn, revealing the formation of laminas in the dorsal horns. Our study also illuminated the spatial differences and molecular cues underlying motor neuron (MN) diversification, and the enrichment of Amyotrophic Lateral Sclerosis (ALS) risk genes in MNs and microglia. Interestingly, we detected disease-associated microglia (DAM)-like microglia groups in the developing human spinal cord, which are predicted to be vulnerable to ALS and engaged in the TYROBP causal network and response to unfolded proteins. These findings provide spatiotemporal transcriptomic resources on the developing human spinal cord and potential strategies for spinal cord injury repair and ALS treatment.

The Transparent Embedding Solvent System, an innovative imaging method, integrates tissue clearing and sectionreconstruction imaging, ensuring consistent sub-micron resolution. This technique is adaptable and compatible with various imaging platforms and specimens. Recently, there has been a profound interest in gaining deeper insights into large biological samples to unravel structural and functional intricacies. 3D large-volume optical imaging, in particular, has received considerable attention due to its high resolution and potential to register molecular and cellular identities via various fluorescent labeling methodologies, such as transgenic mouse models, viral vector expression, and immunolabeling. Large-volume optical imaging of mammalian tissues has historically been difficult due to heterogeneous tissue properties and light scattering. Imaging an intact nervous system is even more challenging due to the long-range projections and complex connections. In recent years, a number of tissue processing and imaging techniques have enabled large-volume optical imaging, which has begun to reveal the 3D cellular profiles and neural circuits in mammals. There are two main types of large-volume imaging workflows: tissue clearing and block-face imaging. Tissue-clearing techniques have evolved significantly in recent years and been extensively reviewed. [1] [2] [3] It removes lipids from the tissues with different detergents or solvents to render the tissue optically transparent. Together with the advancement of lightsheet microscopy, it has been applied to image whole organs, such as brains from rodents and nonhuman primates, as well as some entire rodents. [4] [5] [6] These methods enable detailed examination of cellular and sub-cellular structures in their native 3D environments. Even though tissue clearing can be time-consuming, imaging acquisition is generally quick and can be completed in minutes to a few hours with contemporary lightsheet microscopy. However, the lightsheet also tends to compromise axial resolution. Moreover, a common concern with tissue clearing is diminished optical resolution in the deeper parts of the samples due to inadequate clearing and improper refractive index (RI) matching. Lastly, samples could be distorted during clearing and mounting steps, making quantitative comparisons between samples or registration to established atlases challenging. Block-face imaging, on the other hand, uses various mechanical and chemical methods to achieve thin sections for continuous imaging and reconstruction. Several automated block-face imaging methods have been developed, such as the micro-optical sectioning tomography series. 7, 8 They feature simple tissue preparation, rigid sample properties, and high imaging resolution, making quantitation and registration easier. However, they are less compatible with immunolabeling and can be time-consuming due to the need for precise slicing.

Dear Editor

Zygotic genome activation (ZGA) marks the beginning of the embryonic program for a totipotent embryo, which gives rise to the inner cell mass (ICM) where pluripotent epiblast arises, and extraembryonic trophectoderm. However, how ZGA is connected to the first lineage segregation in mammalian embryos remains elusive. Here, we investigated the role of nuclear receptor (NR) transcription factors (TFs), whose motifs are highly enriched and accessible from the 2-cell (2C) to 8-cell (8C) stages in mouse embryos. We found that NR5A2, an NR TF strongly induced upon ZGA, was required for this connection. Upon Nr5a2 knockdown or knockout, embryos developed beyond 2C normally with the zygotic genome largely activated. However, 4–8C-specific gene activation was substantially impaired and Nr5a2-deficient embryos subsequently arrested at the morula stage. Genome-wide chromatin binding analysis showed that NR5A2-bound cis-regulatory elements in both 2C and 8C embryos are strongly enriched for B1 elements where its binding motif is embedded. NR5A2 was not required for the global opening of its binding sites in 2C embryos but was essential to the opening of its 8C-specific binding sites. These 8C-specific, but not 2C-specific, binding sites are enriched near genes involved in blastocyst and stem cell regulation, and are often bound by master pluripotency TFs in blastocysts and embryonic stem cells (ESCs). Importantly, NR5A2 regulated key pluripotency genes Nanog and Pou5f1/Oct4, and primitive endoderm regulatory genes including Gata6 among many early ICM genes, as well as key trophectoderm regulatory genes including Tead4 and Gata3 at the 8C stage. By contrast, master pluripotency TFs NANOG, SOX2, and OCT4 targeted both early and late ICM genes in mouse ESCs. Taken together, these data identify NR5A2 as a key regulator in totipotent embryos that bridges ZGA to the first lineage segregation during mouse early development.

Combination therapy with PD-1 blockade and IL-2 substantially improves anti-tumor efficacy comparing to monotherapy. The underlying mechanisms responsible for the synergistic effects of the combination therapy remain enigmatic. Here we show that PD-1 ligation results in BATF-dependent transcriptional induction of the membrane-associated E3 ubiquitin ligase MARCH5, which mediates K27-linked polyubiquitination and lysosomal degradation of the common cytokine receptor γ chain (γc). PD-1 ligation also activates SHP2, which dephosphorylates γcY357, leading to impairment of γc family cytokine-triggered signaling. Conversely, PD-1 blockade restores γc level and activity, thereby sensitizing CD8+ T cells to IL-2. We also identified Pitavastatin Calcium as an inhibitor of MARCH5, which combined with PD-1 blockade and IL-2 significantly improves the efficacy of anti-tumor immunotherapy in mice. Our findings uncover the mechanisms by which PD-1 signaling antagonizes γc family cytokine-triggered immune activation and demonstrate that the underlying mechanisms can be exploited for increased efficacy of combination immunotherapy of cancer.