Circadian rhythms have evolved to anticipate and adapt animals to the constraints of the earth’s 24-hour light cycle1. Although the molecular processes that establish periodicity in clock neurons of the suprachiasmatic nucleus (SCN) are well understood, the mechanisms by which axonal projections from the central clock drive behavioural rhythms are unknown2,3,4. Here we show that the sleep period in mice (Zeitgeber time, ZT0–12) is preceded by an increase in water intake promoted entirely by the central clock, and not motivated by physiological need. Mice denied this surge experienced significant dehydration near the end of the sleep period, indicating that this water intake contributes to the maintenance of overnight hydromineral balance. Furthermore, this effect relies specifically on the activity of SCN vasopressin (VP) neurons that project to thirst neurons in the OVLT (organum vasculosum lamina terminalis), where VP is released as a neurotransmitter. SCN VP neurons become electrically active during the anticipatory period (ZT21.5–23.5), and depolarize and excite OVLT neurons through the activation of postsynaptic VP V1a receptors and downstream non-selective cation channels. Optogenetic induction of VP release before the anticipatory period (basal period; ZT19.5–21.5) excited OVLT neurons and prompted a surge in water intake. Conversely, optogenetic inhibition of VP release during the anticipatory period inhibited the firing of OVLT neurons and prevented the corresponding increase in water intake. Our findings reveal the existence of anticipatory thirst, and demonstrate this behaviour to be driven by excitatory peptidergic neurotransmission mediated by VP release from central clock neurons.
Subscribe to Journal
Get full journal access for 1 year
only $3.83 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Mills, J. N. Human circadian rhythms. Physiol. Rev. 46, 128–171 (1966).
Mohawk, J. A., Green, C. B. & Takahashi, J. S. Central and peripheral circadian clocks in mammals. Annu. Rev. Neurosci. 35, 445–462 (2012).
Hastings, M. H., Maywood, E. S. & O’Neill, J. S. Cellular circadian pacemaking and the role of cytosolic rhythms. Curr. Biol. 18, R805–R815 (2008).
Antle, M. C. & Silver, R. Orchestrating time: arrangements of the brain circadian clock. Trends Neurosci. 28, 145–151 (2005).
Johnson, R. F., Beltz, T. G., Thunhorst, R. L. & Johnson, A. K. Investigations on the physiological controls of water and saline intake in C57BL/6 mice. Am. J. Physiol. Regul. Integr. Comp. Physiol. 285, R394–R403 (2003).
Spiteri, N. J. Circadian patterning of feeding, drinking and activity during diurnal food access in rats. Physiol. Behav. 28, 139–147 (1982).
Bourque, C. W. Central mechanisms of osmosensation and systemic osmoregulation. Nat. Rev. Neurosci. 9, 519–531 (2008).
Barney, C. C. & Folkerts, M. M. Thermal dehydration-induced thirst in rats: role of body temperature. Am. J. Physiol. 269, R557–R564 (1995).
Fitzsimons, J. T. Angiotensin, thirst, and sodium appetite. Physiol. Rev. 78, 583–686 (1998).
Johnson, A. K. & Buggy, J. Periventricular preoptic-hypothalamus is vital for thirst and normal water economy. Am. J. Physiol. 234, R122–R129 (1978).
McKinley, M. J., Denton, D. A. & Weisinger, R. S. Sensors for antidiuresis and thirst—osmoreceptors or CSF sodium detectors? Brain Res. 141, 89–103 (1978).
Ramsay, D. J., Thrasher, T. N. & Keil, L. C. The organum vasculosum laminae terminalis: a critical area for osmoreception. Prog. Brain Res. 60, 91–98 (1983).
McKinley, M. J., Denton, D. A., Oldfield, B. J., De Oliveira, L. B. & Mathai, M. L. Water intake and the neural correlates of the consciousness of thirst. Semin. Nephrol. 26, 249–257 (2006).
Farrell, M. J. et al. Cortical activation and lamina terminalis functional connectivity during thirst and drinking in humans. Am. J. Physiol. Regul. Integr. Comp. Physiol. 301, R623–R631 (2011).
Hollis, J. H., McKinley, M. J., D’Souza, M., Kampe, J. & Oldfield, B. J. The trajectory of sensory pathways from the lamina terminalis to the insular and cingulate cortex: a neuroanatomical framework for the generation of thirst. Am. J. Physiol. Regul. Integr. Comp. Physiol. 294, R1390–R1401 (2008).
Stephan, F. K. Circadian rhythms in the rat: constant darkness, entrainment to T cycles and to skeleton photoperiods. Physiol. Behav. 30, 451–462 (1983).
Satinoff, E. & Prosser, R. A. Suprachiasmatic nuclear lesions eliminate circadian rhythms of drinking and activity, but not of body temperature, in male rats. J. Biol. Rhythms 3, 1–22 (1988).
Kalsbeek, A. et al. SCN outputs and the hypothalamic balance of life. J. Biol. Rhythms 21, 458–469 (2006).
Moore, R. Y., Speh, J. C. & Leak, R. K. Suprachiasmatic nucleus organization. Cell Tissue Res. 309, 89–98 (2002).
Abrahamson, E. E. & Moore, R. Y. Suprachiasmatic nucleus in the mouse: retinal innervation, intrinsic organization and efferent projections. Brain Res. 916, 172–191 (2001).
Buijs, R. M. & Kalsbeek, A. Hypothalamic integration of central and peripheral clocks. Nat. Rev. Neurosci. 2, 521–526 (2001).
Chen, T. W. et al. Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature 499, 295–300 (2013).
Serradeil-Le Gal, C. et al. Nonpeptide vasopressin receptor antagonists: development of selective and orally active V1a, V2 and V1b receptor ligands. Prog. Brain Res. 139, 197–210 (2002).
Gunaydin, L. A. et al. Ultrafast optogenetic control. Nat. Neurosci. 13, 387–392 (2010).
Han, X. et al. A high-light sensitivity optical neural silencer: development and application to optogenetic control of non-human primate cortex. Front. Syst. Neurosci. 5, 18 (2011).
Serradeil-Le Gal, C. et al. An overview of SSR149415, a selective nonpeptide vasopressin V(1b) receptor antagonist for the treatment of stress-related disorders. CNS Drug Rev. 11, 53–68 (2005).
Hu, S. B. et al. Vasopressin receptor 1a-mediated negative regulation of B cell receptor signaling. J. Neuroimmunol. 135, 72–81 (2003).
Prager-Khoutorsky, M. & Bourque, C. W. Anatomical organization of the rat organum vasculosum laminae terminalis. Am. J. Physiol. Regul. Integr. Comp. Physiol. 309, R324–R337 (2015).
Trudel, E. & Bourque, C. W. A rat brain slice preserving synaptic connections between neurons of the suprachiasmatic nucleus, organum vasculosum lamina terminalis and supraoptic nucleus. J. Neurosci. Methods 128, 67–77 (2003).
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Franklin, K. B. & Paxinos, G. The Mouse Brain in Stereotaxic Coordinates. Third edn (Elsevier, 2008).
Altstein, M. & Gainer, H. Differential biosynthesis and posttranslational processing of vasopressin and oxytocin in rat brain during embryonic and postnatal development. J. Neurosci. 8, 3967–3977 (1988).
Alstein, M., Whitnall, M. H., House, S., Key, S. & Gainer, H. An immunochemical analysis of oxytocin and vasopressin prohormone processing in vivo. Peptides 9, 87–105 (1988).
Zaelzer, C. et al.. ΔN-TRPV1: A molecular co-detector of body temperature and osmotic stress. Cell Reports 13, 23–30 (2015).
This work was supported by a Foundation Grant from the Canadian Institutes of Health Research (FDN 143337), a James McGill Chair, and a Studentship awarded by the Research Institute of the McGill University Health Centre (RIMUHC). The RIMUHC receives generous funding from the Fonds de Recherche Québec Santé. We thank M. Bouvier for human V1aR; H. Gainer for anti-VP neurophysin primary antibodies; and N. Cermakian, K-F. Storch, and M. Prager-Khoutorsky for comments on an early draft of the manuscript.
The authors declare no competing financial interests.
Nature thanks C. Colwell, H. de la Iglesia, H. Okamura and the other anonymous reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
a, Micrograph shows fluorescence (excitation 650 nm, emission 700 nm) in a 50-μm-thick coronal section taken through the anterior hypothalamus of a mouse injected intravenously with 1% Evans blue. The section corresponds to a plane lying 0.55 mm rostral to the bregma as defined elsewhere31. b, Schematic illustrating the outline of the OVLT and surrounding structures including the anterior commissure (AC), ventral diagonal band of Broca (vDBB), and horizontal diagonal band of Broca (hDBB). Scale bar applies to a and b. c, Panels show brightfield images (left), Evans blue fluorescence (middle), and schematics (right) in consecutive coronal sections spanning the entire rostro-caudal extent of the OVLT. Position relative to bregma is indicated on brightfield panels.
Extended Data Figure 2 Preparation of angled slices of mouse hypothalamus that retain the OVLT and SCN.
a, Schematic diagram showing the relative positions of the OVLT, SCN and other structures in the sagittal plane as defined elsewhere31. Median eminence (ME), median preoptic nucleus (MnPO), medial septum (MS), optic chiasma (OC), posterior pituitary (PP), third ventricle (3V). The brain slice was obtained at an angle of 34° relative to the horizontal plane. b, Brightfield image of a horizontal slice (plane as shown in a) obtained from a mouse injected intravenously with 1% Evan’s blue. c, Evans blue fluorescence observed in the same slice in the small rectangular region identified in b. d, Schematic diagram illustrating various structures retained in the slice preparation (area shown by large rectangle in b).
a, Top, consecutive coronal bright-field micrographs showing the site of microsphere injection in one of three mice tested. Note the presence of material (orange) in three of the sections. Bottom, schematics illustrating the location of the microspheres (orange) within the area encompassing the OVLT. b, Representative sections from four brain areas containing VP-immunoreactive neurons (red) obtained from the brain shown in a: SCN, supraoptic nucleus (SON), bed nucleus of the stria terminalis (BNST) and the hypothalamic paraventricular nucleus (PVN). Note that only the SCN contains retrogradely labelled VP neurons (open arrows), whereas neurons containing retrogradely transported beads in BNST and PVN are VP-negative (arrows). c, Schematic diagram illustrating the positions of all 46 VP-positive neurons projecting to the OVLT for the brain shown in a. d, Table listing other brain areas containing neurons projecting to the OVLT for the brain shown in a. Only coronal sections positioned between +0.14 and −0.7 mm relative to Bregma were analysed (sites identified according to ref. 31).
Extended Data Figure 4 Transgenic mice expressing optogenetic probes and fluorescent reporters in VP neurons.
a, Schematic illustrating the strategies used to produce mice in which VP neurons selectively express ChETA and tdTomato or ArchT and eGFP. b, Panels show the presence of VP (red, upper) and eGFP (green, lower) in an immunolabelled section through the SCN. Note that neurons containing eGFP are VP-positive (arrows), but that not all VP neurons contain eGFP. An analysis of five sections indicated that 41% of the VP SCN neurons (93/225 cells) expressed the fluorescent reporter. c, Left, live VP neurons identified by the presence of eGFP in the SCN of a hypothalamic slice prepared from an ArchT mouse. One cell is being targeted with a patch clamp micropipette. Middle, bright field; right, merge. d, Patch clamping of SCN VP neurons identified by the expression of tdTomato in ChETA mice (layout as in c). e, VP containing fibres are visible by immunolabelling in the wild-type OVLT (upper panel; VP in green, NeuN in blue), and by eGFP expression in the OVLT of ArchT mice (lower) or tdTomato expression in ChETA mice (not shown).
a, Effects of bath-applied VP (concentrations shown below) on GCaMP6m fluorescence in HEK293 cells transfected with GCaMP6m alone, or co-transfected with GCaMP6m and the human V1aR. Note that VP has no effect in the absence of V1aR, but that dose-dependent increases were observed in the VP sensors co-transfected with GCaMP6m and V1aR. b, Bar graphs show mean ± s.e.m. values of fluorescence changes (relative to baseline) caused by 1 μM VP in both types of cells (****P < 0.0001; NS, not significant; paired t-test; n shown in brackets). c, Bar graphs show mean ± s.e.m. values of fluorescence (relative to baseline) induced by different concentrations of VP in HEK293 cells transfected with GCaMP6m and V1aR (n shown in brackets). d, Examples of GCaMP6m fluorescence in VP sensor cells treated with 10 nM VP in the absence (control) or presence of 10 μM SR49059. e, Bar graphs show mean ± s.e.m. values of GCaMP6m fluorescence (relative to baseline) induced by 10 nM VP cells in the absence and presence of 10 μM SR49059. Cells were first tested in the presence of SR49059, then SR49059 was washed out and cells were retested in the absence of SR49059 (**P < 0.01; two-way RM ANOVA and Holm-Sidak post-hoc test).
a, Schematic diagram illustrates the configuration of the experiment. HEK293 cells transfected with GCaMP6m and V1aR (VP sensors) were plated over the slice and GCaMP6m fluorescence was imaged over various regions (OVLT illustrated here). b, Upper panels show GCaMP6m fluorescence in VP sensors overlying the OVLT before (baseline) and after electrical stimulation of the SCN (SCN STIM; 10 Hz, 30 s) in a slice of mouse hypothalamus. The lower graph plots the time course of changes in fluorescence (ΔF; expressed as per cent of basal fluorescence, % Change) induced by SCN stimulation (bar) in a number of cells. c, Bar graphs show mean ± s.e.m. values of fluorescence changes induced by SCN stimulation in slices of mouse hypothalamus relative to baseline (***P < 0.001; n shown in brackets, paired t-test). d, Photo of a horizontal rat hypothalamic slice depicting four regions where VP sensor fluorescence was measured after SCN stimulation: OVLT (blue), insular cortex (black), nucleus accumbens (grey) and SCN (magenta). e, Bar graphs show mean ± s.e.m. GCaMP6m fluorescence before and after SCN stimulation in five representative sensor cells lying over the OVLT, and all cells imaged over the other areas sampled. Note significant release in SCN and OVLT, but not in the other regions (paired t-test used for analysis of OVLT and SCN, and Wilcoxon test used for other regions; *P < 0.05; NS, not significant).
a, Single-cell RT–PCR bands reveal the presence or absence of mRNA encoding VP V1aRs (upper panels) in seven (of 18) individual osmosensory OVLT neurons aspirated via patch pipette in a horizontal slice of rat hypothalamus. The positions of all sampled neurons were measured relative to the rostral part of the preoptic recess (POR) of the third ventricle, and mapped according to ref. 28. Most neurons were sampled in the core of the OVLT. b, OVLT neurons were identified as osmosensory neurons by the presence of PCR products reflecting expression of the channel gene Trpv1dn (lower panels)34. In total, 13 of 18 Trpv1dn-positive OVLT neurons (72%) were found to express V1aRs. c, Micrographs show expression of immunolabelled c-Fos in the OVLT of wild-type and V1aR−/− mice during the BP and AP. Note that the increase in c-Fos density observed in wild-type mice during the AP is absent in V1aR−/− mice.
a, Cell-attached recordings of spontaneous action potential firing in identified SCN VP neurons in slices prepared from ChETA and ArchT mice. Note how application of blue light (473 nm, blue bar) causes excitation of the ChETA neuron whereas application of yellow light (589 nm, yellow bar) inhibits the ArchT neuron. b, Bar graphs show mean ± s.e.m. values of firing rate (FR) measured in SCN VP neurons from both genotypes in the absence (white) and presence of light (blue for ChETA; yellow for ArchT; n shown in brackets; *P < 0.05; ChETA, paired t-test; ArchT, Wilcoxon test). c, Upper trace is a rate metre record (5-s bins) showing the effects of prolonged application of yellow light (bar) on firing rate of an SCN VP neuron from an ArchT mouse. Traces below show samples of neuronal activity recorded at the times indicated by the numbers. Note that the inhibitory effect of light on firing rate is sustained for >20 min but diminishes thereafter. d, Upper trace is a rate metre record (5-s bins) showing the effects of prolonged application of blue light (bar) on firing rate of an SCN VP neuron from a ChETA mouse. Traces below show samples of neuronal activity recorded at the times indicated by the numbers. Note that the excitatory effect of light on firing rate is sustained for >10 min but declines before the end of the stimulus.
a, Schematic diagram illustrating midline structures surrounding the mouse OVLT in the sagittal plane. Positions illustrated as described elsewhere31. Fibre-optic cannula attached to a slim titanium magnetic receptacle was implanted by insertion at an angle of 7° relative to the vertical plane. b, Photograph shows the slim titanium magnetic receptacle and fibre-optic cannula superimposed on a coronal brain slice at the level of the OVLT. c, Coronal section from the paraformaldehyde-fixed brain of a mouse that was implanted with a fibre-optic cannula. Inspection of the section revealed that the tip of the cannula had reached the most dorsal part (arrow) of the OVLT (dashed line). d, Plot shows the effect of blue light delivered for 1–2 min at different intensities on the firing rate of OVLT neurons in slices from a ChETA mouse in vitro. Each dot is a different cell. Note that threshold intensity is ~12.5 mW. e, Plot shows theoretical decay of light intensity through brain tissue as a function of distance using our specific parameters and the calculator module provided at http://optogenetic.org. Note that with a light output set at 22 mW (used in our in vivo experiments with ChETA mice), light intensity drops below threshold at a distance of ~125 μm from the tip of the fibre-optic probe. f, Coronal schematics illustrate implantation sites for all in vivo optogenetic experiments determined by post-hoc histological inspection (as in c). This image has been adapted from ref. 31. Distance from bregma in the rostro-caudal axis is shown below each panel. Note that experiments were only successful when the tip of the fibre-optic probe was placed directly into or above the OVLT.
Extended Data Figure 10 Transgenic ChETA and ArchT mice implanted with fibre-optic cannulae displayed increases in water intake during the AP.
Bar graphs show mean ± s.e.m. values of water intake during the BP (ZT19.5–21.5) and AP (ZT21.5–23.5) in groups of ChETA and ArchT mice (paired t-test, **P < 0.01; n shown in brackets).
About this article
Cite this article
Gizowski, C., Zaelzer, C. & Bourque, C. Clock-driven vasopressin neurotransmission mediates anticipatory thirst prior to sleep. Nature 537, 685–688 (2016). https://doi.org/10.1038/nature19756
Keeping time in the lamina terminalis: Novel oscillator properties of forebrain sensory circumventricular organs
The FASEB Journal (2020)
European Journal of Neuroscience (2020)
Current Opinion in Neurobiology (2020)
NMDA receptors potentiate activity‐dependent dendritic release of neuropeptides from hypothalamic neurons
The Journal of Physiology (2019)