Tanycyte-Independent Control of Hypothalamic Leptin Signaling

Abstract

Leptin is secreted by adipocytes to regulate appetite and body weight. Recent studies have reported that tanycytes actively transport circulating leptin across the brain barrier into the hypothalamus, and are required for normal levels of hypothalamic leptin signaling. However, direct evidence for leptin receptor (LepR) expression is lacking, and the effect of tanycyte-specific deletion of LepR has not been investigated. In this study, we analyze the expression and function of the tanycytic LepR in mice. Using single-molecule fluorescent in situ hybridization (smfISH), RT-qPCR, single-cell RNA sequencing (scRNA-Seq), and selective deletion of the LepR in tanycytes, we are unable to detect expression of LepR in the tanycytes. Tanycyte-specific deletion of LepR likewise did not affect leptin-induced pSTAT3 expression in hypothalamic neurons, regardless of whether leptin was delivered by intraperitoneal or intracerebroventricular injection. Finally, we use activity-regulated scRNA-Seq (act-Seq) to comprehensively profile leptin-induced changes in gene expression in all cell types in mediobasal hypothalamus. Clear evidence for leptin signaling is only seen in endothelial cells and subsets of neurons, although virtually all cell types show leptin-induced changes in gene expression. We thus conclude that LepR expression in tanycytes is either absent or undetectably low, that tanycytes do not directly regulate hypothalamic leptin signaling through a LepR-dependent mechanism, and that leptin regulates gene expression in diverse hypothalamic cell types through both direct and indirect mechanisms.

Keywords: hypothalamus; leptin; metabolism and obesity; radial glia; single cell RNA sequencing; tanycyte.

FIGURE 1

LepR mRNA expression is not detected in tanycytes of adult mice fed ad libitum. (A) Representative images for smfISH analysis using Rax (red) and LepR (green) probes showing α (left) and β tanycytes (right). (B) RT-qPCR analysis for LepR transcript in Rax-expressing tanycytes that were isolated from Rax-CreER^T2/lsl-Sun1-GFP mice. Crym is a tanycyte marker. Gapdh expression, used as a loading control, showed similar abundance in each sample. (C) Differential expression analysis between GFP-positive and GFP-negative population sorted from Rax-EGFP mice displaying examples of highly enriched in each fraction and LepR. (D) tSNE plots using previously reported data from 20,921 cells isolated from the ArcN-ME. Rax and Crym were strongly expressed in the tanycyte cluster, and LepR in the Syt1-positive neuronal cluster, particularly in Npy-positive neurons. Scale bar: 20 um (A).

FIGURE 2

Leptin receptor mRNA expression is not detected in tanycytes of neonatal or fasted adult mice. (A) SmfISH analysis using Rax (red) and LepR (green) probes in ad lib and fasted animals. (a–f) are the higher magnification images of the boxed area in (A). (B) SmfISH analysis using Rax (red) and LepR (green) probes in P1 and P4 mice. Rax-expressing red fluorescence labeled region represents the tanycytic layer. (C) Violin plots of LepR and known tanycytic markers Col23a1 and Slc16a2 in tanycyte cluster comparing between ad lib and fasted conditions. (D) Violin plots of LepR and known neuronal marker genes in Npy-positive cluster. The upper panel shows β1 tanycytes and the lower panel shows β2 tanycytes. Scale bar: 50 um (A), 20 um (B), 20 um (a–f).

FIGURE 3

Tanycyte-specific genetic ablation of LepR. (A) Schematic diagram showing mouse LepR isoforms and the knockout strategy used in this study. (B) PCR analysis showing the 740 bp DNA fragment amplified from the deleted allele. (C) SmfISH analysis using Rax (red) and LepR (green) probes in LepR^lox/+;Rax-CreER^T2 (HET) and LepR^lox/lox;Rax-CreER^T2 (KO). (a–c) and (d–f) are the higher magnification images of the boxed area in (A,B), respectively. Scale bar: 50 um (A,B), 20 um (a–f).

FIGURE 4

Leptin-induced STAT3 phosphorylation in the hypothalamus of control and LepR cKO mice. (A) Representative images of pSTAT3 immunohistochemistry 5 min (upper) and 45 min (lower) after i.p. leptin injection. (B) Representative images of pSTAT3 immunohistochemistry 30 min after i.c.v. aCSF (upper) and leptin (lower) injection. (C) Quantification of the pSTAT3 positive cells in the images shown in (A), n = 3–4. (D) Quantification of the pSTAT3 positive cells in the images shown in (C), n = 3–4. (E) Percentage of cases showing the tanycytic pSTAT3 in each indicated condition. DMH, dorsomedial nucleus; VMH, ventromedial nucleus; ArcN, arcuate nucleus; ME, median eminence. Scale bar: 200 um (A,C).

FIGURE 5

Single-cell RNA sequencing analysis identifies differentially expressed genes in mediobasal hypothalamus following i.c.v. leptin infusion. (A) Schematic diagram describing the pipeline of generating act-Seq data (left) and 2D tSNE plot with annotated clusters (right). (B) Correlation plot showing log-normalized gene-expressions between aCSF (x-axis)- and leptin (y-axis)-infused samples in tanycytes (left) and ependymocytes (right). (C) Violin plots showing changes in gene expressions between aCSF- and leptin-infused samples in tanycytes (top left), neurons (top right), endothelial cells (bottom left), and ependymocytes (bottom right). (D) Bar graphs showing fold changes between two groups (positive fold change indicates an increased expression in leptin-infused group) in all 7 clusters.