Hypothalamic tanycytes: potential roles in the control of feeding and energy balance

Abstract

Tanycytes, glial-like cells that line the third ventricle, are emerging as components of the hypothalamic networks that control body weight and energy balance. They contact the cerebrospinal fluid (CSF) and send processes that come into close contact with neurons in the arcuate and ventromedial hypothalamic nuclei. Tanycytes are glucosensitive and are able to respond to transmitters associated with arousal and the drive to feed. At least some tanycytes are stem cells and, in the median eminence, may be stimulated by diet to generate new neurons. The quest is on to understand how tanycytes detect and respond to changes in energy balance and how they communicate with the rest of the nervous system to effect their functions.

Figure 1

Hypothalamic tanycytes and their relationship to hypothalamic nuclei. (a) Immunostaining against vimentin (green) in the Djungarian hamster reveals the ependymal cells and tanycytes in the layer around the third ventricle (3V). Ependymal cells (ec) occur dorsally, whereas the tanycytes (tc) appear more ventrally – note their processes extending into the parenchyma, especially into the dorsomedial hypothalamus (DMH), ventromedial hypothalamus (VMH), and the arcuate nucleus (Arc) and median eminence (ME). Scale bar, 100 μm. (b) Detailed view of tanycytes genetically labeled in mouse with green fluorescent protein (GFP) and counterstained against glial fibrillary acidic protein (GFAP), a typical marker of glial cells . Note the single process that terminates in a specialization. Scale bar, 20 μm. (c) Schematic diagram illustrating the hypothalamic nuclei involved in energy balance and the relationship of tanycytes to these nuclei. Abbreviations: ACh, acetylcholine; His, histamine; LH, lateral hypothalamus; PVN, paraventricular nucleus. Adapted, with permission, from (b).

Figure 2

Hypotheses for glucose sensing in tanycytes. (a) The pancreatic β cell paradigm. Glucose is taken up into the cell (1), to be converted via glucokinase (GK) (2) to glucose-6-phosphate (G-6-P) which enters into the Krebs cycle. The net increase in ATP production (3) causes closure of the K-ATP channel (Kir6.2) (4) via a specific SUR subunit. This causes depolarization (V_m), opening of Ca²⁺ channels (5), Ca²⁺ influx, and increased exocytosis (6). Most of these components are present in tanycytes, leading to the suggestion that this may be the mechanism by which tanycytes sense glucose. However, there are no reports of any voltage-gated channels in tanycytes. Furthermore, the ability of tanycytes to respond to non-metabolizable analogs of glucose has called this model into question. (b) Alternative models for tanycyte glucosensing. In the upper model (bi), glucose is taken up via a Na⁺-linked glucose transporter (1). The resulting Na⁺ accumulation and depolarization may reverse the Na⁺–Ca²⁺ exchanger (2), causing an increase in intracellular Ca²⁺. This increase may lead, via an unknown mechanism (3), to the release of ATP (4), which activates P2Y1 receptors resulting in a further increase in intracellular Ca²⁺ through mobilization of intracellular stores. The lower model (bii) is similar except that glucose and its analogs act via one or more G-protein-coupled receptors (1) to mobilize intracellular Ca²⁺ directly (2), which again leads to release of ATP (3) and activation of P2Y1 receptors. Modified, with permission, from .

Figure 3

Glucose-sensing in tanycytes. (a) Fura-2 ratio images of acute hypothalamic slice from rats showing the increase in intracellular Ca²⁺ (denoted in red) evoked by a puff from a glucose-containing patch pipette (position indicated). This evoked a long-lasting Ca²⁺ wave (around 20 s in duration) in the tanycyte layer (delineated by dashed lines in top left image; note that the response started in the tanycyte layer and spread into the slice) . (b) Recordings with an ATP biosensor of glucose-evoked ATP release from tanycytes. The inset shows recording arrangement and corresponding positions of the glucose pipette. A larger signal was observed when glucose was applied close to the biosensor (1) compared to when it was applied further away (2,3) . 3v, third ventricle. Reproduced, with permission, from (a,b).

Figure I

Gene expression in tanycytes changes with photoperiod. (a) Cartoon illustrating tanycytes under long and short photoperiod (LD and SD, respectively). The gene products regulating thyroid hormone signaling (i.e., Dio2 in hamster and rat [31,35–37], TTR and OATP1C1 in rat [31,42]), retinoic acid signaling (RALDH1, CRBP1, and STRA6 in rat [42]), as well as GPR50 (hamster) and NMUR2 (rat) , are upregulated under LD compared to SD. (b,c) Examples of changes in mRNA levels for Dio2 and OATP1C1 in the F344 rat under LD (upper panels) and SD (lower panels). Representative autoradiographs showing, in coronal section, mRNA expression in the ependymal layer (arrows, containing the tanycytes) around the third ventricle (3V). Scale bars represent 1 mm. Adapted from .

Figure I

Proposed model for tanycyte proliferation. By analogy to radial glia, tanycytes could either generate new neurons directly (1), or via intermediate progenitor cells that could differentiate into neurons and/or astrocytes (2), or migrate as progenitors and divide further before differentiating (3). Inset: dividing vimentin-positive tanycytes revealed by BrdU staining . These new neurons and progenitor cells could use the tanycyte processes as a means to migrate into the relevant hypothalamic nuclei. On this basis, tanycyte stem cells would form distinct populations to subserve the nuclei into which their processes project. Tanycytes are also capable of producing and releasing both retinoic acid (RA) and ATP , which are proposed to act as inhibitors or activators of proliferation, respectively. The gray cells represent other tanycytes and are shaded merely for clarity of presentation. Inset adapted, with permission, from .