Cellular mechanisms of long‐term osmoregulation in magnocellular neurons

Abstract

Osmoregulation is an essential homeostatic process that maintains the osmolality of the extracellular fluid (ECF) close to a physiological setpoint. Vasopressin (VP) plays a key role in osmoregulation and is secreted by the magnocellular neurosecretory cells (MNCs) of the hypothalamus. MNC electrical activity and VP release increase with elevations of ECF osmolality. MNC osmosensitivity depends on a mechanosensitive N‐terminal variant of the transient receptor potential vanilloid type 1 (ΔN‐TRPV1) channel that activates in response to osmotically induced cell shrinkage. ΔN‐TRPV1 mechanosensitivity depends on their association with microtubules in the MNC cytoskeleton and is modulated by a dense layer of submembranous actin in MNC somata. MNCs exposed to sustained increases in osmolality, however, undergo marked somatic hypertrophy, which suggests that other mechanisms may be important to maintain VP release. Recent evidence suggests that the translocation of ΔN‐TRPV1 (and possibly other channels) to the MNC cell surface could contribute to osmotically induced long‐term increases in MNC excitability. Osmotically induced ion channel translocation is dependent on MNC firing, Ca²⁺ influx through L‐type Ca²⁺ channels, the activation of phospholipase C δ1 and protein kinase C, and soluble N‐ethylmaleimide‐sensitive factor attachment protein receptor‐dependent exocytotic fusion. Other recent work has explored osmotically induced changes in the MNC cytoskeleton that may be related to hypertrophy and ion channel translocation. MNCs may also be activated by elevations in extracellular Na⁺ through the activation of the Na⁺‐sensitive Na⁺ channel, Na _X . This review highlights recent advancements in our understanding of long‐term MNC regulation at the cellular level.

1. INTRODUCTION

Osmoregulation is an essential homeostatic mechanism that allows mammals to cope with constant changes in the intake and loss of water and salt, and involves both conscious and unconscious mechanisms. ¹ , ² , ³ Each mammalian species has a different “setpoint” of osmolality, which is typically around 300 mOsmol kg⁻¹ and must be maintained within a narrow range (~3 mOsmol kg⁻¹). ¹ Deviations from this setpoint can have serious consequences. Osmoregulation involves a coordinated response of multiple neuroendocrine systems including the hypothalamus, circumventricular structures, the kidneys, and baro‐ and chemoreceptors in the vasculature, as well as higher cortical structures for thirst and salt appetite. ¹ Low extracellular fluid (ECF) osmolality triggers salt appetite, and high ECF osmolality triggers behavioral thirst. ² The organum vasculosum lamina terminalis (OVLT) and subfornical organ (SFO) are circumventricular organs that possess specialized neurons that respond to changes in ECF osmolality and Na⁺ ([Na]_o) and help coordinate both conscious and unconscious mechanisms of osmoregulation. ² , ⁴ The primary regulators of unconscious osmoregulation are the magnocellular neurosecretory cells (MNCs) of the hypothalamus. ¹ , ⁴ They are primarily located in the supraoptic (SON) and paraventricular (PVN) nuclei and secrete neuropeptide hormones, specifically either vasopressin (VP) or oxytocin (OT). ¹ , ² , ⁴ MNC electrical activity is tightly coupled to VP and OT release from the posterior pituitary (PP). ⁵ , ⁶ , ⁷ , ⁸ , ⁹ VP plays an especially important role in osmoregulation as it acts at the kidneys to cause water reabsorption to increase blood volume and pressure, and acts on the vasculature to cause vasoconstriction. ¹ , ⁴ , ¹⁰ High salt intake increases the excitability of VP‐releasing MNCs and this may contribute to salt‐dependent hypertension in rats. ¹¹ , ¹² , ¹³ OT has been shown to promote natriuresis in rodents and possesses many other roles in the body that have been reviewed extensively elsewhere. ¹⁴ , ¹⁵ , ¹⁶

Several mechanisms contribute to the osmotic regulation of VP release, such as the astrocyte‐mediated release of taurine to regulate MNC activity, ¹⁷ , ¹⁸ and the phasic firing pattern of VP‐releasing MNCs to promote efficient VP release, ¹⁹ , ²⁰ , ²¹ but this review will focus on mechanisms that are intrinsic to MNCs. MNCs are electrically activated by increases in ECF osmolality. ²² , ²³ , ²⁴ Osmotically induced changes in cell volume are transduced into changes in electrical activity and hormone release. ²³ , ²⁵ MNCs also receive sensory information from the OVLT and SFO regarding hydration status. ² , ²⁵ , ²⁶ MNCs express an N‐terminal variant of the transient receptor potential vanilloid type‐1 (TRPV1) channel called ΔN‐TRPV1. ²⁷ ΔN‐TRPV1 acts as a mechanoreceptor for MNCs and is activated by cell shrinkage caused by increases in ECF osmolality or by negative pressure applied through a patch clamp pipette. ²⁷ , ²⁸ , ²⁹ , ³⁰ MNCs possess a dense and interwoven cytoskeletal scaffold of microtubules (MTs) and a dense subcortical network of actin. ³¹ , ³² ΔN‐TRPV1 mechanosensitivity depends on its connection with the cytoskeleton, especially with MTs. ²⁹ , ³⁰ , ³³ , ³⁴ ΔN‐TRPV1 is tethered to MTs and cell shrinkage (e.g., from high osmolality) is thought to cause ΔN‐TRPV1 activation through mechanical forces exerted on the channel. ²⁹ , ³¹ , ³² , ³⁴ This hypothesis is known as the push activation model. ²⁹ , ³¹ , ³² , ³⁴ The degree of shrinkage‐mediated ΔN‐TRPV1 activation is dependent on the density of MTs, ²⁹ , ³¹ , ³² , ³⁴ and the sensitivity of this activation is regulated by actin and MTs. ³⁵ Actin exists in a dense submembranous layer in MNCs and is increased (i.e., undergoes further polymerization) following hypertonic exposure. ³⁵ , ³⁶ MNC osmosensitivity is enhanced by the actin polymerizing drug jasplakinolide, and actin depolymerization by cytochalasin‐D prevents MNCs from responding to osmotic stimuli, suggesting that actin is required for proper osmotransduction. ³⁵ , ³⁶ Recently, it has been shown that there are gaps or fenestrations in submembranous actin and that ΔN‐TRPV1 and MTs interact in these fenestrations. ³⁷ These sites may serve to enhance the membrane displacement that occurs with small increases in ECF osmolality and may thereby help facilitate osmosensing and osmotransduction. ³⁷

VP and OT are also released from MNC somata and dendrites, which is known as somatodendritic (SD) release. ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² SD release of VP and OT is a slow process that requires sustained increases in intracellular Ca²⁺ (i.e., tens of minutes or longer). ³⁹ , ⁴⁰ , ⁴¹ , ⁴² SD release of VP has been shown to act in autocrine and paracrine fashions to exert negative feedback to MNCs in the SON and PVN. ⁴⁰ , ⁴¹ , ⁴² VP release is maximized by a phasic pattern of firing in MNCs, ⁴³ and the autocrine and paracrine negative feedback may help maintain this firing pattern. ³⁹ , ⁴⁰ VP and OT are located primarily in large dense‐core vesicles (LDCVs) that are trafficked between the MNC somata, dendrites, and axon. ⁴⁴ , ⁴⁵ , ⁴⁶ Sustained osmotic stimuli (e.g., water deprivation or salt loading) increase LDCV trafficking to dendrites. ⁴⁷ , ⁴⁸ It has been hypothesized that the negative feedback of autocrine and paracrine SD VP also helps to “prime” MNCs to release hormone‐packed LDCVs in bursts. ³⁸ , ⁴⁴ , ⁴⁹ The proteins involved in the exocytotic machinery used for axonal and SD LDCV exocytosis are different. ⁴⁶ , ⁵⁰

Soluble N‐ethylmaleimide‐sensitive factor attachment protein receptors (SNAREs) are a superfamily of proteins that govern most exocytosis and endocytosis in neurons. SNARE assembly requires the interaction of 3–4 specific proteins (VAMP‐2, SNAP‐25, synaptotagmin, syntaxin) located on the plasma membrane and the intracellular vesicle. ⁵¹ Each SNARE protein has multiple isoforms that are tailored for specific types of exocytosis and endocytosis and can be highly specialized for type of cargo (e.g., hormone transport vs. ion channel transport) and speed (e.g., fast axonal transport vs. slow hormone release). ⁵¹ SNAP‐25 and VAMP‐2 are highly expressed in MNC axons and are not expressed in the somata, yet dissociation of the complete SNARE complex prevents both axonal and SD VP and OT release. ⁴⁶ , ⁵⁰ The above studies suggest that different SNARE machinery govern axonal and SD hormone release, which may explain why SD LDCV release is much slower than axonal LDCV release. The exocytotic proteins involved in SD release remain unidentified.

Prolonged increases in osmolality (i.e., hours to days) have been shown to cause dramatic structural and functional changes in MNCs, including retraction of glial processes from around MNC somata and axon terminals, an increase in the density of subcortical and cytoplasmic actin and MT networks, increases in the expression of many proteins, and an increased density of a variety of channels and receptors on the MNC cell surface. ³² , ³⁶ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ It has been proposed that ion channels may be translocated to the plasma membrane following chronic hyperosmotic exposure. ⁵⁷ , ⁵⁸ Sustained osmotic stimuli cause a marked hypertrophy of MNC somata, ⁵² , ⁵⁹ but the functional significance of this hypertrophy remains unclear. Osmotically induced hypertrophy appears to depend at least partly on intrinsic properties of the MNCs since it is evoked in acutely isolated MNCs following tens of minutes of hyperosmotic exposure. ⁶⁰ The expansion of the MNC soma during the hypertrophic response could alter both the physical relationship between ΔN‐TRPV1 and the cytoskeleton as well as the mechanical forces experienced by the channels, which indicates that the push activation model may be insufficient to explain long‐term osmosensitivity. It is unclear whether the increase in soma size is due to insertion of reserves of plasma membrane, ⁶¹ or to the fusion of intracellular vesicles with the plasma membrane, ⁴⁷ , ⁶² or some combination of both. This review aims to review the current literature available on the mechanisms of long‐term osmoregulation, to identify gaps in our understanding of how MNCs cope with chronic osmotic stimuli, and to identify key questions that need to be answered regarding long‐term MNC regulation.

2. STRUCTURAL AND FUNCTIONAL ADAPTATIONS TO SUSTAINED HIGH OSMOLALITY

2.1. Structural adaptations and MNC hypertrophy

Osmotically evoked hypertrophy of MNCs in the SON and PVN has been documented since the 1970s, but the mechanisms surrounding how and why this hypertrophy occurs remain unclear. In one study, 10 days of salt loading rats revealed 170% increases in SON size compared to euhydrated rats, ⁵² while in another, 7 days of hypo‐osmotic challenge led to a 40% reduction in MNC size. ⁶³ , ⁶⁴ MNCs exposed to sustained increases in osmolality in vitro (i.e., administration of hypertonic solution to isolated MNCs) or in vivo (i.e., salt loading or water deprivation of whole animals) maintain a high degree of excitability for as long as the ECF osmolality is elevated. ¹ , ¹² , ⁶⁵ It is unclear whether ΔN‐TRPV1 activity in these conditions depends on mechanical activation, which would imply a cytoskeletal reorganization that somehow maintains the force on the channels despite the change in cell size.

Osmotically induced hypertrophy in acutely isolated MNCs has been used to probe its mechanisms. ⁶⁰ Both the initiation and the maintenance of osmotically induced hypertrophy were shown to be dependent on AP firing, Ca²⁺ influx through L‐type Ca²⁺ channels, the activation of phospholipase C (PLC) and protein kinase C (PKC), activation of ΔN‐TRPV1 channels, and SNARE‐mediated exocytosis (Figure 1A–E). ⁶⁰ Blockade of any of these processes (e.g., with TTX, nifedipine, U‐73122, bisindolylmaleimide‐I, SB366791, and TAT‐NSF700, respectively) prevented the onset of hypertrophy when applied prior to the onset of the hypertonic stimulus (Figure 1B,D,E). Application of TTX, SB366791, or nifedipine triggered a reversal of hypertrophy when applied after hypertrophy had occurred (Figure 1C), ⁶⁰ indicating that action potential firing and Ca²⁺ influx are required to maintain MNC hypertrophy. The recovery from hypertrophy was blocked by the nonselective dynamin inhibitor dynasore (St. Louis, MO, USA; Figure 1F). ⁶⁰ Dynasore has been shown to inhibit the endocytic machinery in ways independent of dynamin, ⁶⁶ and therefore some caution is warranted, but these data suggest that the endocytic processes that underlie recovery from hypertrophy may be dynamin dependent.

FIGURE 1.

Sustained increases in osmolality cause somatic hypertrophy in isolated MNCs. (A) Hypertonic saline causes hypertrophy of acutely isolated cells over tens of minutes. (B) Hypertrophy is prevented by inhibiting APs (with TTX), TRPV1 (SB366791), L‐type Ca²⁺ channels (nifedipine), and by intracellular Ca²⁺ chelation (BAPTA‐AM). (C) Hypertrophy is reversed by application of TTX, SB366791, and nifedipine. (D) Hypertrophy is prevented by inhibiting PLC (U‐73122) or PKC (bisindolylmaleimide‐I). (E) Hypertrophy is prevented by inhibiting SNARE‐mediated exocytosis (TAT‐NSF700). TAT‐NSF700scr is an inactive analogue of the inhibitor and has no effect on hypertrophy. (F) The recovery from hypertrophy is prevented by dynasore. For (B–F), the hypertonic stimulus was 325 mosmol kg⁻¹. Adapted from Reference 60. CSA, cross‐sectional area; MNCs, magnocellular neurosecretory cells; PLC, phospholipase C; PKC, protein kinase C; SNARE, soluble N‐ethylmaleimide‐sensitive factor attachment protein receptor.

2.2. Ion channel translocation and functional adaptations of MNCs to chronic osmotic stress

Exocytosis is an essential cellular process. ⁵¹ , ⁶⁷ , ⁶⁸ The fusion of intracellular membranous structures with the plasma membrane can be secretory (i.e., it can mediate the release of neurotransmitters or hormones into the ECF), non‐secretory (i.e., it can mediate the translocation of proteins from internal storage sites to the plasma membrane), or a combination of the two. ⁶⁹ , ⁷⁰ The translocation of ion channels to the plasma membranes of excitable cells like neurons has been identified as a mechanism to cause long‐term increases in cell excitability. ⁷¹ , ⁷² , ⁷³ A classic example is long‐term potentiation (LTP), in which repeated stimulations cause the translocation of specific ion channels (AMPA and NMDA) to the post‐synaptic membrane to cause a long‐lasting enhancement of synaptic strength. ⁷³

The translocation of ion channels has been observed in MNCs in response to sustained increases in osmolality, ⁵⁷ , ⁵⁸ which led to the hypothesis that the processes underlying osmotically evoked hypertrophy could mediate the osmotically evoked translocation of ion channels in MNCs. A recent study showed that ΔN‐TRPV1 undergoes osmotically induced translocation to the MNC plasma membrane following sustained increases in osmolality with a time course similar to that observed with osmotically induced hypertrophy. ⁶² The mechanisms that govern this translocation mirror those underlying osmotically induced somatic hypertrophy: AP firing, Ca²⁺ influx through L‐type Ca²⁺ channels, PLC and PKC activation, and SNARE‐dependent exocytosis are all required for osmotically induced ΔN‐TRPV1 translocation (Figure 2). ⁶² Application of dynasore also prevented ΔN‐TRPV1 internalization and recovery from hypertrophy (Figure 2A‐iv). ⁶⁰ , ⁶² Similar results were observed for MNCs isolated from water deprived rats (Figure 3) except that the degree of hypertrophy and the increase of ΔN‐TRPV1 immunofluorescence were greater. These data demonstrated that ion channel translocation and hypertrophy are parallel mechanisms that may be essential parts of a larger set of structural and functional adaptations that MNCs undergo in response to sustained high osmolality. Various Ca²⁺ and Na⁺ channels have been shown to play roles in long‐term osmoregulation ⁷⁴ , ⁷⁵ and it is possible that the translocation of these channels could contribute to MNC osmosensitivity.

FIGURE 2.

Sustained increases in osmolality cause reversible ΔN‐TRPV1 translocation to the plasma membrane. (A) ICC images of MNCs isolated from normally hydrated rats using an antibody directed against ΔN‐TRPV1 (green) in various treatments. Membrane staining for ΔN‐TRPV1 is increased by hypertonic treatment (A‐ii) and this increase is reversed by a return to isosmotic solution (A‐iii). Recovery is blocked by treatment with dynasore (A‐iv). The increase in ΔN‐TRPV1 staining is prevented by a PLC inhibitor (U‐73122, 2A‐v), by a PKC inhibitor (bisindoylmaleimide‐I, A‐vi) and an inhibitor of exocytotic fusion (TAT‐NSF700, A‐vii). (B) Bar scatter plot representing the plasma membrane TRPV1 immunofluorescence of the images depicted in (A). (C) Bar scatter plot representing the cross‐sectional area of the images depicted in (A). For all bar scatter plots, ***p < .001. Adapted from reference 62. ICC, immunocytochemistry; PKC, protein kinase C; PLC, phospholipase C; ΔN‐TRPV1, N‐terminal variant of the transient receptor potential vanilloid type 1.

FIGURE 3.

Twenty‐four hours of water deprivation causes significant reversible ΔN‐TRPV1 translocation to the plasma membrane. (A) ICC images of MNCs isolated from 24‐h WD rats using an antibody directed against ΔN‐TRPV1 (green) in various treatments. MNCs isolated from WD rats show high levels membrane staining for ΔN‐TRPV1 channels (A‐i) and this is decreased by the return to isotonic solution (A‐ii), by treatment with TTX (A‐iii) by treatment with nifedipine (A‐iv), or by treatment with the PLC inhibitor U‐73122 (A‐v). (B) Bar scatter plot representing the cross‐sectional area of the images depicted in (A). (C) Bar scatter plot representing the plasma membrane TRPV1 immunofluorescence of the images depicted in (A). For all bar scatter plots, ***p < .001. Adapted from reference 62. ICC, immunocytochemistry; MNCs, magnocellular neurosecretory cells; PLC, phospholipase C; WD, water‐deprived; ΔN‐TRPV1, N‐terminal variant of the transient receptor potential vanilloid type 1.

2.3. PLCδ1 in osmoregulation

The dependence of both hypertrophy and ΔN‐TRPV1 translocation on PLC activation suggests that this enzyme may play a key role in osmosensitivity and osmoregulation. PLC inhibition prevents full activation of osmosensitive non‐selective cation current in isolated MNCs. ⁷⁶ The observation that PLC is activated in isolated MNCs in an activity‐ and Ca²⁺ influx‐dependent manner led to the hypothesis that a Ca²⁺‐dependent isoform of PLC might underlie these effects. ⁷⁶ Subsequent work demonstrated that a highly Ca²⁺‐sensitive isoform of PLC called PLCδ1 plays a key role in MNC osmosensitivity (Figure 4). ⁷⁶ , ⁷⁷ , ⁷⁸ , ⁷⁹ Mice that lack PLCδ1 (PLCδ1 KO mice) display dysfunctional systemic osmoregulation; PLCδ1 KO mice that are deprived of water for 24 h exhibit significantly elevated serum osmolality compared with water‐deprived control mice (Figure 4B). ⁷⁸ MNCs isolated from PLCδ1 KO mice do not display the increased submembranous actin density in response to either sustained increases in osmolality or activation by angiotensin II (a known activator of actin polymerization) ⁸⁰ observed in control mice (Figure 4C). ⁷⁸ They also do not display osmotically induced ΔN‐TRPV1 translocation or somatic hypertrophy (Figure 4D,E). ⁶² These data suggest that PLCδ1 plays an essential role in MNC osmosensitivity and osmoregulation.

FIGURE 4.

PLCδ1 Is essential for osmoregulation. (A) Electrophysiological recordings showing that inhibiting PLC attenuates osmotically evoked increases in ΔN‐TRPV1 currents. (B) Bar scatter plot showing that the serum osmolality following 24 h of water deprivation are much higher in PLCδ1 KO mice than in control mice. (C) ICC images depicting actin (red) across various treatments. Actin density is increased in response to hypertonic solution or angiotensin II in MNCs isolated from control mice but not in PLCδ1 KO mice. (D) ICC images of acutely isolated MNCs from hydrated control (top panel) and PLCδ1 KO (bottom panel) mice using an antibody directed against ΔN‐TRPV1. MNCs from PLCδ1 KO mice failed to exhibit osmotically induced ΔN‐TRPV1 translocation or hypertrophy. (E) Bar scatter plots depicting the statistical relationships of the images in (D). For all bar scatter plots, **p < .01, ***p < .001. Adapted from references 62, 76, 78, and. ICC, immunocytochemistry; MNCs, magnocellular neurosecretory cells; PLCδ1, phospholipase C δ1; ΔN‐TRPV1, N‐terminal variant of the transient receptor potential vanilloid type 1.

3. MULTIPLE MECHANISMS FOR MNC REGULATION

Multiple ions and other osmolytes contribute to ECF osmolality. Na⁺ is, however, the most abundant extracellular cation and in most physiological conditions, [Na⁺]_o varies in parallel with ECF osmolality. ⁸¹ , ⁸² , ⁸³ High [Na⁺]_o (hypernatremia) typically occurs with increases in osmolality, ⁸¹ , ⁸² , ⁸³ but it is possible to develop hypernatremia without significant change in osmolality (e.g., in a condition known as hypervolemic hypernatremia). ⁸⁴ , ⁸⁵ , ⁸⁶ In hypervolemic hypernatremia, which occurs most commonly following acute kidney injury, there is an increase in both ECF volume and [Na⁺]_o, but ECF osmolality is not significantly affected. ⁸⁴ , ⁸⁵ , ⁸⁶ This example suggests the need for mechanisms to respond to elevated Na⁺ that are independent of the mechanisms for responding to increased osmolality. Cells in the SFO and OVLT respond to changes in ECF Na⁺ concentrations via the non‐voltage gated, Na⁺‐sensitive Na⁺ channel Na _X , and Na _X ‐mediated changes in OVLT and SFO activity can direct salt‐intake behaviour. ⁸⁷ , ⁸⁸ , ⁸⁹ Na _X is also expressed in MNCs, ⁹⁰ and recent data suggest that Na _X plays a role in MNC osmosensitivity. ⁹⁰ , ⁹¹ The Bourque laboratory recently demonstrated that hypernatremia can increase MNC activity in the absence of an increase in osmolality (i.e., a similar physiological condition to hypervolemic hypernatremia). ⁹¹ They observed that knocking down the expression of Na _X in MNCs blocks increases in MNC activity caused by high Na⁺ but does not affect increases in MNC activity caused by high osmolality, suggesting that there are distinct mechanisms for Na⁺ sensing and osmosensitivity. ⁹¹ They also showed that ΔN‐TRPV1 inhibition does not prevent increases in MNC activity from occurring in the presence of high Na⁺⁹¹, suggesting that ΔN‐TRPV1 does not play a role in Na⁺ sensing under isotonic conditions.

Recent studies by Stocker and colleagues have shown that MNCs isolated from mice that lack TRPV1 (TRPV1 KO mice) are able to osmoregulate. ⁹² They showed that 7 days of salt loading or 24 h of water deprivation activate equal numbers of MNCs in the SONs of TRPV1 KO mice and control mice. ⁹² They also showed that infusion of 2 M NaCl solution elicited an increase in MNC activity and VP release in TRPV1 KO mice comparable to those observed in control mice. ⁹² Their results might be explained by a Na⁺‐dependent activation of MNCs that occurs independently of TRPV1 channels.

4. UNANSWERED QUESTIONS

The observation that the absence of PLCδ1 prevents mice from effectively osmoregulating in response to water deprivation ⁷⁸ suggests that PLCδ1 is essential for MNC osmosensitivity. This could involve the lack of direct PLCδ1‐mediated regulation of ΔN‐TRPV1 channels, the inability of these MNCs to hypertrophy or undergo ΔN‐TRPV1 translocation to the plasma membrane, or from a defect in the mechanisms that underlie osmotically evoked cytoskeletal reorganization. Further research will be required to determine the importance of these mechanisms. Water deprivation, however, causes both hyperosmolality and hypernatremia and the loss of osmoregulation in PLCδ1 KO mice suggests that PLCδ1 is essential for the response to both. A plausible hypothesis is that cytoskeletal reorganization, MNC hypertrophy, and ion channel translocation all contribute to the response to sustained hyperosmolality or hypernatremia and that their absence suppresses normal osmoregulation. The fact that osmotically evoked hypertrophy and ion channel translocation are activated in an action potential‐ and Ca²⁺ influx‐dependent fashion suggests that hypernatremia could activate these processes in the absence of changes in osmolality through activation of Na _X , which would be expected to increase MNC firing. ⁹¹ These processes could therefore occur in TRPV1 KO mice exposed to sustained water deprivation or salt loading, ⁹² and could contribute to their ability to osmoregulate in the absence of the ΔN‐TRPV1‐mediated mechanisms. ΔN‐TRPV1 translocation cannot occur in these MNCs, but it is possible that the translocation of other types of excitatory ion channels (such as Na_X and voltage‐gated Na⁺ and Ca²⁺ channels) could contribute to sustained osmosensitivity in MNCs. It will be important to determine if this is the case to better understand long‐term MNC regulation.

Several other key questions about these processes remain unanswered. The slow time course of the exocytotic fusion that underlies ΔN‐TRPV1 translocation suggests that it is not directly Ca²⁺‐dependent and that its Ca²⁺ dependence may be indirect. This is consistent with the hypothesis that it is mediated by the activation of PLCδ1 and that this activation somehow triggers SNARE‐mediated exocytosis in the MNC soma. Most known examples of exocytotic release (such as neurotransmission and hormone release) are triggered by an increase in Ca²⁺ leading to the activation of one of several different Ca²⁺‐dependent isoforms of synaptotagmin. There are, however, isoforms of synaptotagmin that are activated in a Ca²⁺‐independent fashion, and it is possible that one of those isoforms is responsible for osmotically evoked ion channel translocation in MNCs. The identification of the isoform of synaptotagmin involved in this process will not only clarify mechanisms of sustained osmosensitivity in MNCs but might also give insight into mechanisms of long‐term neuromodulation in other neurons.

It is also important to understand other aspects of the mechanisms underlying activated ion channel translocation in MNCs. The identity of the vesicles for trafficking channels to and from the plasma membrane, for example, is not yet known. It is also unknown whether specific motor proteins are involved in the movement of these vesicles and how those proteins associate with the cytoskeleton. The cytoskeleton itself undergoes a remarkable reorganization during hypertrophy and during the rapid recovery from hypertrophy when MNCs are returned to isosmotic solutions. It will be fascinating to learn what happens to the cytoskeleton and its association with ion channels and the exocytotic machinery during those periods of flux. The observation that the dynamin inhibitor dynasore blocks both recovery from hypertrophy, ⁶⁰ and the internalization of ΔN‐TRPV1 channels, ⁶² suggests that these processes could be dynamin‐dependent. This suggests that these processes may be similar to Fast Endophilin‐Mediated Endocytosis (FEME), which is a dynamin‐dependent, clathrin‐independent endocytic process that is rapidly activated by receptor‐ligand interactions between endophilin, actin, dynamin, the SNARE complex, and the MT motor protein dynein. ⁹³ , ⁹⁴ FEME occurs on a timescale ranging from a few seconds to a few minutes ⁹³ , ⁹⁴ and involves interactions between actin, endophilin A1 (a protein involved in endocytosis), and synaptotagmin‐11 (a non‐Ca²⁺‐dependent isoform of synaptotagmin). ⁹⁵ Interestingly, synaptotagmin‐11 is involved in slow, non‐secretory processes, and both synaptotagmin‐11 and endophilin A1 are involved in LTP maintenance in the basal forebrain. ⁹⁶ , ⁹⁷ , ⁹⁸ Endophilin A1 has also been shown to be involved in hormone‐rich LDCV trafficking in adrenal chromaffin cells (another type of neurosecretory cell). ⁹⁹ Synaptotagmin‐11 and endophilin A1 are both expressed in MNCs. ¹⁰⁰

5. CONCLUDING REMARKS

Sustained increases in ECF osmolality trigger MNCs to undergo a dramatic and reversible hypertrophy that is associated with the translocation of ΔN‐TRPV1 channels to the cell surface. ⁶² These processes appear to be part of the osmotically evoked functional reorganization that enables MNCs to maintain high levels of VP release during prolonged osmotic stress. ³³ , ⁶² Many of the processes leading to hypertrophy and ion channel translocation have been identified, but many mechanisms remain unknown. Some of these elements, and some of the gaps in our knowledge, are summarized in Figure 5. The structural and functional relationship between ΔN‐TRPV1 channels and the tubulin and actin molecules in the cytoskeleton is critical for osmosensing, and it will be important for us to understand how that relationship changes during hypertrophy and the rapid recovery from hypertrophy that has been observed. ⁶⁰ , ⁶² Identification of the isoforms of the SNARE proteins involved in ion channel translocation may yield critical clues to the function of this system and perhaps to other examples of long‐term neuromodulation. We also have much to learn about the process of trafficking ion channels in MNCs—the identity of the internal trafficking structures, how they are associated with the cytoskeleton, whether other ion channels are translocated along with ΔN‐TRPV1, and whether Na _X ‐mediated MNC excitation is sufficient to activate hypertrophy and ion channel translocation in the absence of changes in osmolality. Answering these questions will further our understanding of the cellular mechanisms of this essential homeostatic mechanism and may also yield tools that will help us to understand the physiological significance of these processes by, for example, allowing us to selectively prevent ion translocation while not affecting other aspects of osmosensitivity. Much work remains to be done.

FIGURE 5.

Understanding the structural and functional adaptations of MNCs with sustained osmotic challenge requires further investigation. This schematic diagram highlights various structural and functional adaptations of MNCs to sustained increases in osmolality and their recovery from these increases that have been highlighted in this review (left column of text). The right column of text summarizes unanswered questions proposed in this review that require further investigation to elucidate. This figure was made using BioRender.com. MNCs, magnocellular neurosecretory cells.

AUTHOR CONTRIBUTIONS

Kirk D. Haan: Writing – original draft; writing – review and editing; visualization; conceptualization. Thomas E. Fisher: Supervision; resources; funding acquisition; project administration; writing – review and editing; visualization; conceptualization.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Haan KD, Fisher TE. Cellular mechanisms of long‐term osmoregulation in magnocellular neurons. J Neuroendocrinol. 2025;37(12):e70090. doi: 10.1111/jne.70090

DATA AVAILABILITY STATEMENT

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.