Adaptation or dysfunction? Magnocellular vasopressin neurones’ responses to chronic salt‐loading

Since the pioneering work of Verney (1947) proposed the existence of central sensory elements termed osmoreceptors (Verney, 1947), the neural mechanisms underlying osmotic detection and integration have been extensively investigated. These are mediated by specialised neurones in the brainstem and in nuclei adjacent to the anterior third ventricle, which transduce osmotic disturbances into electrical activity (Bourque, 2008). Magnocellular neurosecretory cells (MNCs) in the paraventricular and supraoptic (SON) nuclei in hypothalamus release oxytocin (OXT) and vasopressin (AVP). Their axons project to the neurohypophysis, where hormone secretion is triggered by action potentials generated at their soma. MNCs releasing AVP (MNCsVP) are excited by changes in sodium concentration and plasma osmolality; even a minor (1%–2%) increase is sufficient to initiate AVP release (Verney, 1947). When osmolality exceeds the osmotic threshold, AVP secretion increases proportionally, enhancing renal water reabsorption through V2 receptor activation in distal nephron segments. Behavioural responses, including augmented thirst and reduced sodium consumption, are simultaneously engaged. OXT can also promote natriuresis in rats, though not in humans (Bourque, 2008). AVP secretion is modulated by osmoreceptors in circumventricular organs (CVOs), such as the organum vasculosum of the laminae terminalis (OVLT) and the subfornical organ, which lack a blood–brain barrier and project to MNCsVP. Under hyperosmotic conditions water efflux and membrane deformation depolarise CVO neurones, elevating MNCsVP firing (Bourque, 2008; Levi et al., 2021). Accordingly several studies report salt-loading (SL)-induced hyperexcitability in these neurones (Choe et al., 2015; Levi et al., 2021), promoting sustained AVP secretion. A previous study showed FOS expression of MNCsVP in the SON in two SL models (Ang II–salt and via 2% NaCl intake) and demonstrated an increased neuronal firing following hyperosmotic stimulation using angled-slice patch-clamp preparations that preserve OVLT–SON connectivity (Levi et al., 2021). This Journal Club article highlights the recent study by Lozic et al. (2025), which challenges this classical view by showing, through in vivo electrophysiology, that SON MNCsVP exhibit distinct adaptive changes in response to chronic osmotic stimulation, presumably as a mechanism to conserve neurohypophysial hormone stores. An important aspect of the study is that very few investigations have used in vivo electrophysiological recordings to characterise the electrical activity of SON MCNs following SL. Sprague–Dawley rats were allocated to three groups: euhydrated controls; SL rats receiving 2% NaCl instead of water for 7 days; and rehydrated rats, exposed to the same SL protocol followed by 2 days of water access. In anaesthetised animals extracellular recordings were performed at the pituitary stalk to identify SON neurones antidromically activated by neurohypophysial stimulation. Cells were classified as OXT, continuously firing AVP, or phasic AVP neurones, the latter defined by abrupt alternations between high-frequency firing and quiescence. Acute osmotic stimulation via intravenous 2 M NaCl increased MNCsVP firing linearly by ∼0.2 spikes s−1 per 1 mOsm kg−1. After 7 days of SL plasma osmolality increased by ∼39 mOsm kg−1; thus, had acute sensitivity been preserved, a ∼7.8 spikes s−1 increase would be expected. Instead the authors observed that although basal firing was higher in SL rats, particularly in phasic AVP neurones, which efficiently promote AVP release, the chronic response amounted to only 0.9 spikes s−1, far below the predicted value. MNCs from euhydrated animals fired at 5.1 ± 2.5 spikes s−1; SL rats at 6.0 ± 3.0 spikes s−1; and rehydrated rats at 5.3 ± 3.0 spikes s−1. These findings contrast with previous evidence from Choe et al. (2015), who reported a 3.9 spikes s−1 increase in MNCsVP after 7 days of SL (Choe et al., 2015), a greater but still submaximal response representing approximately 50% of the predicted value. Collectively these results suggest attenuated osmosensitivity, rather than a complete functional habituation. Some interpretations are possible: overactivation of the system may trigger a feedback mechanism leading to a reduction in MNCsVP responsiveness; alternatively, in addition to its intrinsic osmosensitivity, the SON receives osmosensitive afferent inputs from the CVOs, the primary detectors of the sodium increase resulting from SL, which may have influenced the MNCsVP firing pattern observed. The authors also highlight the need to reconsider experimental designs for studying osmosensitive neurones in vitro. If these neurones operate chronically in a hypertonic environment, their transfer to the artificial conditions commonly used for both euhydrated and SL animals may effectively be hyposmotic, thereby inducing electrical imbalances and responses that do not faithfully reflect the physiological state in vivo. This consideration underscores the importance of experimental standardisation and the integration of in vivo approaches. To examine electrophysiological properties the authors analysed interspike interval parameters, including spike patterns reflecting hyperpolarising and depolarising afterpotentials (DAP). Notably, despite the preservation or even enhancement of the DAP in MNCsVP neurones, their responsiveness to an acute osmotic stimulus is reduced, which is functionally paradoxical and points to the involvement of additional adaptive mechanisms. This dissociation suggests that intrinsic excitability and stimulus–response gain may be independently regulated during chronic osmotic stress, potentially at stages upstream of spike generation, such as osmotic sensing or synaptic integration, which are not directly addressed in the present study. The study also found no reversal of GABAergic signalling in SL animals, whereas Choe et al. (2015) reported a switch from inhibition to excitation in SL hypothalamic explants. The divergence suggests that GABAergic effects differ between in vitro and in vivo environments, likely due to differing intracellular chloride dynamics (Choe et al., 2015). The available data are still inconsistent in the literature, requiring further studies to broaden the understanding of the mechanisms underlying the reversal of GABA effects in response to SL. Lozic et al. additionally reviewed previously published data concerning osmolarity and circulating OXT/AVP levels in SL models. Most published studies show pituitary AVP content generally reaches its lowest levels around day 5 of SL, and plasma AVP elevated but does not rise progressively. In their study SL rats exhibited higher plasma sodium, markedly lower pituitary AVP content and elevated plasma AVP. Interestingly although the reduced osmosensitivity of MNCsVP proposed by the authors, plasma AVP levels are still double, whereas pituitary AVP content is markedly reduced after SL. Combined with electrophysiological data, this suggests that even a modest (∼1 spike s−1) increase in MNCsVP firing may sustain circulating AVP without fully depleting pituitary stores, despite an ∼80% reduction in pituitary content. This discrepancy challenges the established linear relationship between firing rate and plasma AVP, supporting an adaptive, rather than pathological, response to chronic osmotic stress. The study presents important methodological limitations. The first concerns the use of urethane as an anaesthetic, which is known to elevate plasma osmolality and thereby stimulate AVP release. The authors note that under urethane urine production is minimal, which may indeed compromise the renal clearance of AVP; however given that the study challenges prevailing views, measuring plasma osmolality during the experiments would have helped exclude anaesthetic effects and strengthen the conclusions. Another important observation is that the conclusions rely solely on extracellular electrophysiological recordings, limiting the ability to determine whether a true conversion from continuous to phasic firing occurs or whether this reflects a sampling bias inherent to random single-unit recordings, as extracellular approaches do not allow longitudinal tracking of firing transitions within the same neurone. The classical Guytonian model, in which increased sodium intake raises plasma tonicity to drive thirst and renal water retention, has been increasingly re-evaluated. The brain plays a multifaceted role in fluid and electrolyte homeostasis, and the relationship between AVP release and osmolarity is more complex than previously assumed. The main findings of Lozic et al. show that the basal firing rate of both MNCsVP was elevated in SL rats, particularly in phasic cells, though less than predicted. Responses to osmotic challenge were reduced rather than enhanced, and no GABAergic stimulatory effect was observed. These results suggest homeostatic plasticity that maintains stable AVP production during chronic SL and highlight the need to investigate compensatory mechanisms supporting both short-term accommodation and long-term adaptation, via receptor regulation or altered neurotransmitter release. Overall the study advances the understanding of central mechanisms regulating body fluid homeostasis and the adaptive capacity of neural circuits under sustained osmotic challenge. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. The authors declare no conflicts of interest. G.A.O. and P.M.G.: Conception or design of the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work. The researchers are supported by São Paulo State Research Foundation (FAPESP), #2023/04290-5 (GAO) and #2024/16941-3 (PMG).