Single-cell Transcriptomic Profiling of the Hypothalamic Median Eminence during Aging

Aging is a slow and progressive natural process that compromises the normal functions of cells, tissues, organs and systems. The aging of the hypothalamic median eminence (ME), a structural gate linking neural and endocrine systems, may impair hormone release, energy homeostasis and central sensing of circulating molecules, leading to systemic and reproductive aging. However, the molecular and cellular features of ME aging remain largely unknown. Here we describe the transcriptional landscape of young and middle-aged mouse ME at single-cell resolution, revealing the common and cell-type-specific transcriptional changes with age. The transcriptional changes in cell-intrinsic programs, cell-cell crosstalk and cell-extrinsic factors highlight five molecular features of ME aging and also implicate several potentially druggable targets at cellular, signaling and molecular levels. Importantly, our results suggest that vascular and leptomeningeal cells may lead the asynchronized aging process among diverse cell types and drive local inflammation and cellular senescence via a unique secretome. Together, our study uncovers how intrinsic and extrinsic features of each cell type in the hypothalamic ME are changed by the aging process, which will facilitate our understanding of brain aging and provide clues for efficient anti-aging intervention at the middle-aged stage. Single cell RNA-seq of median eminence in wild type mice at 3-month old and 12-month old.衰老是一种缓慢且渐进的自然过程，会损害细胞、组织、器官及系统的正常生理功能。作为连接神经与内分泌系统的结构门户，下丘脑正中隆起（hypothalamic median eminence, ME）的衰老可能损伤激素释放、能量稳态以及循环分子的中枢感知能力，进而引发全身衰老与生殖衰老。然而，目前学界对ME衰老的分子与细胞特征仍知之甚少。本研究以单细胞分辨率刻画了年轻与中年小鼠ME的转录组全景，揭示了随衰老发生的共性转录变化与细胞类型特异性转录变化。细胞内在程序、细胞间串扰以及细胞外在因子的转录变化，凸显了ME衰老的五大分子特征，并在细胞、信号通路与分子层面指出了多个潜在可靶向干预的靶点。值得注意的是，本研究结果表明，血管细胞与软脑膜细胞可能主导了不同细胞类型间的异步衰老进程，并通过独特的分泌组驱动局部炎症与细胞衰老。综上，本研究阐明了下丘脑正中隆起内各类细胞的内在与外在特征随衰老发生的改变机制，这将有助于加深我们对脑衰老的理解，并为中年阶段高效的抗衰老干预提供新思路。本数据集包含野生型小鼠3月龄与12月龄正中隆起的单细胞RNA测序（single cell RNA-seq）数据。