Spatial Transcriptomics of Supraoptic Nucleus and Paraventricular Nucleus of the Hypothalamus

Abstract

Magnocellular neurosecretory cells (MNCs) in the supraoptic nucleus (SON) and paraventricular nucleus (PVN) of the hypothalamus release oxytocin (OXY) and arginine vasopressin (AVP) into the peripheral circulation, playing a key role in the regulation of fluid and electrolyte homeostasis. Several groups have investigated transcriptomics in these regions; however, the data lack spatial context important for better understanding physiological functions of genes. Our study leverages spatially-resolved transcriptomics to visualize the relationship between cells, their gene expression profiles, and their relative locations within a tissue sample. Here, we use a novel spatial transcriptomics technology, Visium Spatial Gene Expression, manufactured by 10x Genomics to obtain spatially-resolved gene expression data for the SON and PVN of an adult male Sprague-Dawley rat. Briefly, a fresh frozen brain was sectioned at 10μm thickness to collect four sections (~4x4mm) containing SON and PVN regions. The tissue was then mounted and fixed on the capture areas of a Visium slide containing probes that bind mRNA. Next, hematoxylin and eosin (H&E) staining and high-resolution imaging of the slide was performed prior to tissue permeabilization. The permeabilization releases mRNA from cells that then binds to capture probes on the slide beneath the tissue. cDNA is then synthesized from captured mRNA from which sequencing libraries are prepared. Sequence data is then visualized, using 10x Genomics’ Space Ranger and Loupe Browser applications, to overlay spatial gene expression data with our H&E stained images. Foremost, our results show that gene cluster analysis successfully differentiated myelinated fiber tracts from nuclei and identified several distinct neuronal populations. Gene clusters overlaying fiber tracts contain glia-specific markers (e.g., Mbp and Gfap) and those overlaying SON and PVN regions contain neuron-specific markers (e.g., Syn1 and Nrgn). These data also have the resolution to resolve OXY- and AVP-specific clusters in the SON via targeted gene analysis, with anatomical distribution consistent with immunohistochemistry results for the regions. Additionally, our results show clustering of SON with magnocellular PVN, while parvocellular PVN was a different cluster. Differential expression analysis of these two clusters revealed relevant genes for the distinct regions and neuronal subpopulations in the top 20 globally distinguishing genes (e.g., Avp and Oxt for SON and magnocellular PVN cluster and Trh for parvocellular PVN cluster). This spatially-resolved transcriptomic data allows us to obtain whole transcriptomic data without sacrificing morphological context (e.g., SON and PVN regions). Future spatial transcriptomic studies can be used to investigate changes in MNC gene expression in terms of sex differences and disease models that can potentially inform cellular mechanisms involved in pathophysiology.