The effects of 17α-estradiol treatment on endocrine system revealed by single-nucleus transcriptomic sequencing of hypothalamus
- Department of Obstetrics and Gynecology, National Clinical Research Center for Obstetric and Gynecologic Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, China
- School of Medical Technology，Beijing Institute of Technology, China
- Department of Medical Oncology，Beijing Tsinghua Changgung Hospital, School of Clinical Medicine, Tsinghua University, China
- Department of Cancer Biology and Pharmacology, University of Illinois College of Medicine, United States
- Institute of Reproductive Medicine, Medical School of Nantong University, China
- School of Basic Medical Sciences, Shandong University, China
Figures
Figure 1
Single-nucleus transcriptomic sequencing (snRNA-seq) profiling of the hypothalamus from O, O.T, and Y samples.
(A) UMAP visualization of nuclei colored by 10 cell types from hypothalamus of aged rats (O), 17α-estradiol-treated aged rats (O.T) and young rats (Y). (B) Heatmap showing the classic markers of 10 major cell types in hypothalamus. (C) Cell-type compositions by groups (left panel) or by major cell types with the total cell numbers shown above each column. (D) Circos plot depicting the number of ligand–receptor pairs between Neu and other cell types (color strips) for each group. (E) Dot plot showing significant ligand–receptor interactions between Neurons for each group. Boxes showing the unique ligand–receptor interactions between Neuron.O (black boxes) or between Neuron.O.T (blue boxes). (F) Dot plot of the top six enriched GO biological process terms across three groups of neurons via gene set enrichment analysis (GSEA) analysis. (G) The top 15 changed pathways/gene sets according to the ranks of AUC values in selected pathways related to neuronal synapses and axons from Gene Ontology (GO) biological process, GO molecular function and GO cellular component.
Figure 2 with 3 supplements
Two opposing regulatory signaling networks in neuron metabolism.
(A) Dot plot of the selected pathways representing the prominent changes of overall expression levels across Neuron.O, Neuron.O.T and Neuron.Y in metabolism, signaling, and synaptic activity. (B) Correlation heatmap showing transcription factors (TFs) that correlated with the two opposing regulatory signaling networks in the mixed neurons of O, O.T, and Y. (C) The shared unique markers of each quarter (c1–c4) in six pathways in hypothalamic neurons (O, O.T, and Y). The markers were then collected as c1-up-signature (19 genes) and c4-up-signature (12 genes). (D) The aging-related cell proportions of each quarter are shown by four pathways. (E) The correlation of c1-up-signature and c2-up-signature with the two opposing regulatory signaling networks.
Figure 2—figure supplement 1
Top 20 signaling pathways or gene sets significantly positively or negatively associated with MitoCarta OXPHOS subunits in Neuron.O.
Significant p-values (<0.05) are indicated by a star.
Figure 2—figure supplement 2
The variable response patterns of non-neuron cells to aging and 17α-estradiol treatment in hypothalamus.
Dot plot of overall expression levels of selected pathways from the two opposing signaling networks in nine non-neural cell types.
Figure 2—figure supplement 3
The top enriched pathways of significantly expressed genes in Micro, Astro, and Neuron between O.T and O.
Top 12 enriched GOBP pathways via DAVID Functional Annotation Tools in Micro (O vs Y, O.T vs O), Astro (O vs Y, O.T vs O), and Neuron (O vs Y, O.T vs O) in significantly down-regulated or up-regulated genes, which were calculated via FindMarker function in R package Seurat (test.use=bimod, min.pct=0.1, logfc.Threshold=0.25).
Figure 3 with 1 supplement
Screening of neuron subtypes via supervised clustering, which responded distinctly to aging and 17α-estradiol treatment.
(A) Diagram outlining the features of supervised clustering of neurons in the hypothalamus in comparison with traditional unsupervised clustering. (B) The ranks of cell counts in neuropeptide-secreting neuron subclusters (left panel) and subclusters expressing neuropeptide receptors or hormone receptors (right panel) in sample Y. The cell number (n) in each subset is ≥10. (C, D) The prioritization of the top 20 neuron subclusters across the three types of perturbation (O vs Y, O.T vs Y, and O.T vs O) calculated by the Augur algorithm, in neuropeptide-secreting neurons (C) and neuron subclusters expressing neuropeptide receptors or hormone receptors (D).
Figure 3—figure supplement 1
The similarity of neuropeptide-expressing subclusters or receptor-expressing subclusters in young rat hypothalamus.
(A, B) Heatmaps showing the similarity of neuropeptide-expressing subclusters (A) and receptor-expressing subclusters (B) in the hypothalamus of young rats (left panels). Each subcluster contains no fewer than 10 cells. Venn diagrams (right panels) display the overlap of cell barcodes among neuronal subclusters with higher similarity.
Figure 4
Ranking of neuron subtypes with distinct responses to aging and 17α-estradiol treatment.
(A, B) The top 20 and bottom 20 neuron subtypes based on the mean expression values of five signatures or gene sets, ranked by their values in sample O, in neuropeptide-secreting subtypes (A) and in neuron subtypes expressing neuropeptide receptors or hormone receptors (B).
Figure 5
Responses of Crh neurons to long-term 17α-estradiol treatment.
(A) The top 20 and bottom 20 neuropeptide-secreting neuron subtypes, ranked by their mean expression values of five signatures or gene sets in sample O. (B) Expression profiles of selected pathways from two opposing signaling networks in Crh, Kiss1, and Prlh neurons. (C) Downregulated and upregulated differentially expressed genes (DEGs) associated with mitochondria or the adherens junction pathway in Crh neurons, comparing O.T vs O. (D) Top 25 transcription factor (TF) activities in Crh and Gnrh1 neurons. (E) Serum levels of Crh, cortisol, and aldosterone in Y, O, and O.T groups as measured by enzyme immunoassay; two-tailed unpaired t-tests were performed, with p-values indicated.
Figure 6 with 2 supplements
The response of oxytocin (Oxt) neurons to 17α-estradiol and the causal effects of Oxt on other endocrine factors.
(A) The relative cell proportions of peptide-expressing subclusters (upper panel) and receptor-expressing subclusters (lower panel) across Y, O, and O.T (sorted in descending order of proportions in Y). Only subclusters with a cell count of n≥10 in sample Y were included for calculation. (B) Dot plots showing the expression profiles of the selected pathways from the two opposing signaling pathways in four types of food uptake-related neurons, which decreased or increased among the top 10 ranks in (A) or (B). Blue arrows: c1-up-signature and c4-up-signature. (C) Volcanic plots showing the differentially expressed genes (DEGs) between Neuron.O.T and Neuron.O in the pathway synaptic membrane. (D) Enzyme immunoassay of the plasma levels of Oxt in three groups. (E) Top 25 transcription factor (TF) activities in neuron Oxt. (F) Significant causal effects (p<0.05, inverse-variance weighting IVW) between exposure OXT (id: prot-a-2159) and 204 endocrine-related outcomes, which were not significant in reverse Mendelian randomization (MR) analysis. Significant heterogeneity (Q_pval <0.05). Significant horizontal pleiotropy (pval <0.05).
Figure 6—figure supplement 1
The expression profiles of selected pathways from the two opposing signaling networks in 6 cardiovascular system-related neurons.
Among the 26 selected pathways, eight were metabolism-related pathways and usually elevated during aging (orange) and 18 were either negatively correlated with MitoCarta OXPHOS subunits or pathways related to synapse activity (dark green).
Figure 6—figure supplement 2
Bidirectional two-sample MR analysis of causal effects between 203 endocrine-related factors and Oxt (id: prot-a-2159).
(A) Significant causal effects (p<0.05, IVW) related to exposure Oxt (id: prot-a-2159) and 204 endocrine-related outcomes in both bidirectional Mendelian randomization (MR) analysis. (B) Significant causal effects (p<0.05, inverse-variance weighting IVW) related to 204 endocrine-related exposures and outcome Oxt (id: prot-a-2159) in both bidirectional MR analysis.
Figure 7 with 2 supplements
The response of hypothalamic-pituitary-gonadal (HPG) axis in males to 17α-estradiol and the causal effects of gonadotropin-releasing hormone (Gnrh) on other endocrine factors.
(A) The expression profiles of pathways from the two opposing signaling networks in Gnrh1-, Esr2-, Esr1-, or Ar-positive neurons. (B) Enzyme immunoassay of the serum levels of Gnrh, total testosterone (T), and estrogen (E) in Y, O, and O.T samples. Two-tailed unpaired t-test was performed. (C) Inflammation of seminiferous tubules in testes of O and O.T. Left two panels: representative HE staining of testis inflammation in O and the normal seminiferous tubules of O.T. Right panel: the mean testis inflammation index of O and O.T. Bar, 50 μm (D) The top 25 TF activities in Gnrh1 neurons in three groups. (E) The activities of 14 pathways in Gnrh1-, Esr2-, Esr1-, or Ar-positive neurons. (F) Significant causal effects (inverse-variance weighting IVW, p<0.05) between exposure GNRH1 (id: prot-a-1233) and 204 endocrine-related outcomes, which were not significant in reverse MR analysis. (G) Items with significant causal effects (IVW, p<0.05) in both directions of MR analysis between GNRH1 (id: prot-a-1233) and 204 endocrine-related outcomes.
Figure 7—figure supplement 1
The top 20 and bottom 20 neuron subtypes based on the mean expression values of c4-up-signature, ranked by the values in sample O.T, in neuropeptide-, or hormone-secreting subtypes (upper panel) and in neuron subtypes expressing neuropeptide receptors or hormone receptors (lower panel).
Figure 7—figure supplement 2
Two-sample MR analysis of causal effects of 204 endocrine-related exposures on outcome GNRH1.
(A) Significant causal effects (p<0.05, inverse-variance weighting IVW), which were not significant in reverse MR analysis between 204 endocrine-related exposures and outcome GNRH1 (id: prot-a-1233). (B) Significant causal effects (p<0.05, IVW) in both directions of MR analysis.
Additional files
-
Supplementary file 1
auROC Analysis of 97 Synapse Activity-Related Pathways.
- https://cdn.elifesciences.org/articles/100346/elife-100346-supp1-v1.xlsx
-
Supplementary file 2
List of Neuropeptides, Receptors, and Signatures.
- https://cdn.elifesciences.org/articles/100346/elife-100346-supp2-v1.xlsx
-
Supplementary file 3
Protein Quantitative Trait Locus (pQTL) Data of 204 Human Endocrine-Related GWAS Summary Datasets.
- https://cdn.elifesciences.org/articles/100346/elife-100346-supp3-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/100346/elife-100346-mdarchecklist1-v1.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
The effects of 17α-estradiol treatment on endocrine system revealed by single-nucleus transcriptomic sequencing of hypothalamus
eLife 13:RP100346.
https://doi.org/10.7554/eLife.100346.4
