NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

De Luca LA Jr, Menani JV, Johnson AK, editors. Neurobiology of Body Fluid Homeostasis: Transduction and Integration. Boca Raton (FL): CRC Press/Taylor & Francis; 2014.

Cover of Neurobiology of Body Fluid Homeostasis

Neurobiology of Body Fluid Homeostasis: Transduction and Integration.

Show details

Chapter 6Separating Thirst from Hunger

and .


Classical thirst stimuli can be modified by other ongoing physiologic conditions. In addition, although food intake and water drinking are traditionally seen as contemporaneous behaviors, stimuli for food intake can be separated from water drinking and vice versa. In this chapter, we address two issues: independence of the two behaviors and the impact of non-thirst–related stimuli on water drinking behaviors.

Do all stimuli for food intake induce concomitant water drinking? Although pattern analysis of ad libitum food and water consumption definitively demonstrated the coordination of those two behaviors (de Castro 1988; Fitzsimons and LeMagnen 1969; Kissileff 1969), it is clear that water drinking can occur under physiologic conditions in the absence of food intake (Zorilla et al. 2005), and even under conditions of water restriction some food intake does occur (Fitzsimons and LeMagnen 1969; Zorilla et al. 2005). However, a more physiologic approach to the concordance of the two behaviors is the examination of the intervals of water drinking during 24 h, ad libitum periods, and under these conditions, significant amounts of water are consumed independent of food intake (Zorilla et al. 2005). This may seem obvious when one considers the importance of osmotic and volemic stimuli for drinking; however, when one considers water drinking in the presence of food intake, a certain necessity for the coincident behaviors overrides their independence. This is in no small part due to the lubricative function of fluid taken with solid food, and the resulting osmotic stimuli following solute absorption in the stomach and intestines. But do all stimuli for food intake also stimulate drinking? Evidence from the Daniels laboratory (Mietlicki et al. 2009) provides a clear, negative answer.


One of the most powerful stimuli for food intake is ghrelin, acting in hypothalamic centers to stimulate food intake (Wren et al. 2000). However, when studied under conditions more suitable for the examination of drinking behavior, such as pharmacologic (angiotensin II) or physiologic (hyperosmolar) challenges, quite the opposite is observed. Indeed, ghrelin administration, while stimulating food intake, inhibits angiotensin II–induced drinking, as well as drinking in response to hyperosmolar challenge (Mietlicki et al. 2009). Similarly, when drinking behavior is the primary focus, food intake can be dissociated. Indeed, dipsogenic doses of angiotensin II inhibit rather than stimulate food intake (Porter and Potratz 2004).


Can decreased drinking behavior be the cause of decreased food intake? We believe so. Obestatin was originally identified as a posttranslational product of the ghrelin preprohromone (Zhang et al. 2005). It was named on the basis of its apparent ability to inhibit feeding, an activity that was almost immediately challenged by other researchers (Seone et al. 2006). In our hands, central administration of obestatin did reduce food intake to a slight but not significant degree; however, more impressive was the effect to inhibit water drinking in those same animals for up to 24 h after central administration before the onset of the dark phase (Samson et al. 2006). This inhibition of water, but not food, intake resulted in a significant loss of body weight over the 24-h period. The fact that neither food intake nor open field locomotor behaviors were altered by these same doses of obestatin suggested to us that the action of the peptide was selective for thirst. To test this possibility more directly, we examined the action of obestatin on pharmacologically driven thirst and observed a significant inhibitory effect of obestatin on angiotensin II–induced water drinking (Samson et al. 2006). Water drinking in response to a hypovolemic challenge also was inhibited by intracerebroventicularly (icv) administered obestatin (Samson et al. 2008a). Finally, we identified the subfornical organ as the potential site of the action of obestatin (Samson et al. 2006), again suggesting a unique action of the peptide on thirst mechanisms, independent of any significant action on food intake.

Is the selective action of obestatin on thirst physiologically relevant? Because obestatin is processed from the same preprohormone as ghrelin, gene knockout (embryonic gene ablation) or translation compromise (antisense oligonucleotides, small interfering RNA, ribozymes) approaches would abrogate the production of both peptides. How then would you interpret any behavioral phenotype that resulted? Would it be caused uniquely by loss of obestatin or by loss of ghrelin, or even loss of both? Additionally, no consensus exists on the identity of the obestatin receptor (Lauwers et al. 2006; Moechars et al. 2006), and no selective obestatin receptor antagonist has been developed. Therefore, we turned to passive immunoneutralization as an approach to examine the question of physiological relevance. Indeed, pretreatment with a selective obestatin antiserum resulted in a highly significant increase in water drinking in ad libitum fed and watered rats (Samson et al. 2008a). Food intake was also elevated in the antiserum-treated animals, although that increase failed to reach statistical significance. These results together with our observation of direct actions of the peptide on subfornical organ neurons pointed to a physiologically relevant action of obestatin on fluid and electrolyte homeostasis. This became even more evident when we demonstrated that in addition to inhibiting water intake, obestatin, administered icv, inhibited vasopressin, but not oxytocin, secretion in overnight water–restricted animals and vasopressin secretion in response to central angiotensin II administration (Samson et al. 2008a). This action of obestatin also appeared to have physiologic relevance because central administration of anti-obestatin antiserum resulted in exaggerated vasopressin, but not oxytocin, secretion in response to overnight water restriction (Samson et al. 2008a). Here, then, is an example of an endogenous neuropeptide that controls fluid and electrolyte homeostasis by acting primarily to inhibit water drinking, and if any effects on food intake are observed, those are in all likelihood secondary to its antidipsogenic actions.


Is drinking behavior affected by non-thirst–related cues, in addition to those associated with food intake? Certainly, the answer is yes. Most obvious would be the effect of nauseogenic stimuli on drinking behavior. A second would be anxiety-related drinking, also known as psychogenic polydipsia. However, more subtle influences can also modify drinking behavior, not the least of which is ambient circulatory pressure. Several research groups have demonstrated the ability of changes in mean arterial pressures to modulate pharmacologically driven water drinking (Evered et al. 1988; Hosutt et al. 1978; Robinson and Evered 1987; Thunhorst and Johnson 1993; Thunhorst et al. 1993). It should make sense then that increased arterial pressure, via high- and low-pressure baroreceptive mechanisms, might not only reduce vasopressin secretion, but also the drive to consume fluids. In the opposite direction, hypotensive stimuli, in particular hypovolemia, stimulate both vasopressin release and water drinking, as well as increase autonomic outflow, resulting in restoration of the normovolemia. But can perfusion pressures drop even lower, past some threshold for drinking? Certainly work from Stricker’s group established that possibility (Hosutt et al. 1978). Is the effect of hypotension robust enough to alter ad libitum drinking behavior, such as that entrained to the onset of the dark phase?

6.4.1. Can Baroreflex Also Present Confusing Signals for Thirst and Hunger?

In 2008, our group discovered a novel neuropeptide encoded in the somatostatin preprohormone that when administered icv, unlike somatostatin, inhibited food and water intake in a significant and dose-dependent fashion (Samson et al. 2008b). The antidipsogenic and anorexigenic action of this peptide, named neuronostatin, could be blocked by pretreatment with a melanocortin receptor antagonist, SHU9119. This suggested an interaction of neuronostatin with pro-opiomelanocortin (POMC) producing neurons (Yosten and Samson 2010), something we established in collaboration with Alastair Ferguson’s assistance by electrophysiologic approaches (Samson et al. 2008b). At that point, it appeared that the effects of neuronostatin on water intake might be mainly a reflection of its anorexigenic activity because not only were POMC neurons activated by the peptide, NPY producing neurons were also inhibited (Samson et al. 2008b). But are those anorexigenic and antidipsogenic actions physiologically relevant? Because neuronostatin is a product of posttranslational processing of the somatostatin preprohormone, any attempt to compromise the peptide’s production would also compromise somatostatin production and thus the effects observed would be uninterpretable. In fact, the absence of an altered growth phenotype of, or growth hormone secretion in, somatostatin gene knockout animals proves this point (Low et al. 2001).

The neuronostatin receptor has not been identified and thus we could not compromise the peptide’s action by deletion of its receptor. Similarly, selective neuronostatin antagonists currently are not available; therefore, we sought to selectively compromise the action of neuronostatin in vivo by passive immunoneutralization. Adult male rats bearing lateral cerebroventricle cannulas were habituated to metabolic cages as previously described (Yosten and Samson 2010). Daily ad libitum food and water intakes were monitored for minimally 3 days before the administration of 3 μL preimmune rabbit serum (NRS) or 3 μL rabbit antirat neuronostatin antiserum (Phoenix Pharmaceuticals, Burlingame, CA, USA) 1 h before lights were turned off (lights on 0600–1800 hours). Food and water intakes were then monitored hourly until 2100 hours and again 24 and 48 h later. We had hypothesized that compromise of neuronostatin action (i.e., neutralization with antiserum) would result in exaggerated eating and drinking because the pharmacologic effect of the peptide was an inhibition of both. Surprisingly, the anti-neuronostatin treated rats actually ate less food and drank less water than NRS-treated controls (Figure 6.1). Because they responded with an appropriate startle response, we did not consider the decreased behavior to be related to a deficit in general motor function. No differences were observed between the two groups in terms of food and water intakes or body weight gains after 24 h. Thus, the inhibitory effects observed on the day of treatment were reversible.

FIGURE 6.1. Ad libitum food (a) and water (b) intakes in male rats following icv administration of normal rabbit serum (NRS) or anti-neuronostatin (NST) antiserum.


Ad libitum food (a) and water (b) intakes in male rats following icv administration of normal rabbit serum (NRS) or anti-neuronostatin (NST) antiserum. Intakes in NRS-treated rats did not differ significantly from those administered saline previously (more...)

We then hypothesized that the failure of the animals to eat or drink was not primarily due to loss of the action of endogenous neuronostatin on feeding and drinking circuits in the brain, but was instead secondary to a loss of the action of endogenous peptide on autonomic centers in the medulla. We have demonstrated that central administration of exogenous neuronostatin elevates mean arterial pressure by two mechanisms: increased sympathetic efferent tone and increased vasopressin release (Yosten et al. 2011). Was it possible then that in the absence of endogenous neuronostatin, arterial pressures were low enough to compromise baroreflex function and thus the rats were experiencing orthostatic hypotension when they rose to access the food and water? To test this hypothesis, we monitored mean arterial pressure in conscious, unrestrained rats before and after central administration of NRS or anti-neuronostatin antiserum. Indeed, arterial pressures fell significantly in the anti-neuronostatin treated rats, were unstable, and fell even more when the animals rose to move around the cages.

6.4.2. Does Loss of Baroreflex Alter Thirst and Hunger?

To test the hypothesis that orthostasis was the potential cause of the failure of the antiserum treated rats to eat and drink at the onset of the dark phase, we instrumented animals with lateral cerebroventricular cannulas, and 5 days later catheters were placed in the carotid artery and jugular vein, as previously described (Yosten and Samson 2010). On the following day, animals were habituated to the testing room, the carotid catheter was connected to a pressure transducer, and the jugular catheter was connected to an extension tube to facilitate intravenous administration of either sodium nitroprusside (2.5, 7.5, and 15 μg/kg body weight), to lower arterial pressures, or phenylephrine (5.0, 25, and 50 μg/kg body weight), to raise pressures. Heart rates and arterial pressures were monitored for at least 20 min before and continuously during testing. As can be seen in Figure 6.2, pretreatment with anti-neuronostatin antiserum resulted in a significant change in baroreflex function (comparison of slopes of the regression lines, tα(2),73 = 29.2, p < 0.001) such that those animals responded with less of a heart response to either a drop or a rise in mean arterial pressure. Thus, we conclude that the inhibitory effect of anti-neuronostatin pretreatment on food and water intake was in all likelihood secondary to a compromise in autonomic function that resulted in orthostasis and therefore a deficit in locomotor activity. These data, as well as current studies in our laboratory using molecular approaches to compromise neuronostatin function by abrogating production of the G protein-coupled receptor we are proposing to be the cognate receptor for the peptide, point to a physiologically relevant role for endogenous neuronostatin in the central control of autonomic function, but not a direct role in thirst or hunger.

FIGURE 6.2. Intracerebroventricular administration of anti-neuronostatin (NST) antiserum alters baroreflex function in conscious, unrestrained male rats (slopes compared by linear regression analysis).


Intracerebroventricular administration of anti-neuronostatin (NST) antiserum alters baroreflex function in conscious, unrestrained male rats (slopes compared by linear regression analysis). Pressures were lowered or raised by intra-arterial administration (more...)


In summary, elevations in mean arterial pressure can buffer drinking responses to thirst stimuli, probably through baroreflex activation, and hypotension can similarly alter drinking behavior, when the thirst centers in the brain are not alerted to the drop in pressure because of a compromised baroreflex. Investigators fluent in the thirst literature are well aware of these potential pitfalls when data from feeding studies are reported, but can the same be said about those less well versed in that literature? Clearly, the literature on obestatin tells a cautionary tale (Zhang et al. 2007). Additionally, when novel peptides are first tested for potential effects on food and water intakes, experiments should be designed that potentially separate the two behaviors, and the possibility that either of those two behaviors are a result of changes in autonomic function must be examined. Our experiences certainly have taught us that valuable lesson.


Drs. Yosten and Samson are supported by the National Institutes of Health (HL66023) and the American Heart Association (10GRNT4470043).


  • de Castro J. A microregulatory analysis of spontaneous fluid intake in humans: Evidence that the amount of liquid ingested and its timing is mainly governed by feeding. Physiol Behav. 1988;3:705–14. [PubMed: 3237784]
  • Evered M.D, Robinson M.M, Rose P.A. Effect of arterial pressure on drinking and urinary responses to angiotensin II. Am J Physiol Regul Integr Comp Physiol. 1988;254:R69–74. [PubMed: 3337271]
  • Fitzsimons J.T, Le Magnen J. Eating as a regulatory control of drinking in the rat. J Comp Physiol Psychol. 1969;67:273–83. [PubMed: 5787378]
  • Hosutt J.A, Rowland N, Stricker E.M. Hypotension and thirst in rats after isoproterenol treatment. Physiol Behav. 1978;21:593–8. [PubMed: 740778]
  • Kissileff H.R. Food-associated drinking in the rat. J Comp Physiol Psychol. 1969;67:284–300. [PubMed: 5787379]
  • Lauwers E, Landuyt B, Arckens L, Schoof L, Luyten W. Obestatin does not activate orphan G protein-coupled receptor GPR39. BiochemBiophys Res Commun. 2006;351:21–25. [PubMed: 17054911]
  • Low M.J, Otero-Corchon V, Parlow A.F. et al. Somatostatin is required for masculinization of growth hormone – regulated hepatic gene expression but not growth. J Clin Invest. 2001;107:1571–80. [PMC free article: PMC200191] [PubMed: 11413165]
  • Mietlicki E.G, Nowak E.L, Daniels D. The effect of ghrelin on water intake during dipsogenic conditions. Physiol Behav. 2009;96:37–43. [PMC free article: PMC2635884] [PubMed: 18761029]
  • Moechars D, Depoortere I, Moreaux B. et al. Altered gastrointestinal and metabolic function in the GPR39-obestatin receptor knockout-mouse. Gastroenterology. 2006;131:1131–41. [PubMed: 17030183]
  • Porter J.P, Potratz K.R. Effect of intracerebroventricular angiotensin II on body weight and food intake in adult rats. Am J Physiol Regul Integr Comp Physiol. 2004;287:R422–8. [PubMed: 15117728]
  • Robinson M.M, Evered M.D. Pressor action of intravenous angiotensin II reduces drinking response in rats. Am J Physiol Regul Integr Comp Physiol. 1987;252:R754–9. [PubMed: 3551636]
  • Samson W.K, White M.M, Price C, Ferguson A.V. Obestatin acts in brain to inhibit thirst. Am J Physiol Regul Integr Comp Physiol. 2006;292:R637–43. [PubMed: 16931650]
  • Samson W.K, Yosten G.L.C, Chang J.K, Ferguson A.V, White M.M. Obestatin inhibits vasopressin secretion: Evidence for a physiological action in the control of fluid homeostasis. J Endocrinol. 2008a;196:1–7. [PubMed: 18310451]
  • Samson W.K, Zhang J.V, Avsian-Kretchmer O. et al. Neuronostatin encoded by the somatostatin gene regulates neuronal, endocrine and metabolic functions. J Biol Chem. 2008b;283:31949–59. [PMC free article: PMC2581552] [PubMed: 18753129]
  • Seone L.M, Al-Massadi O, Pazoz Y, Pagotto U, Casaneuva F.F. Central obestatin administration does not modify either spontaneous or ghrelin-induced food intake in rats. J Endocrinol Invest. 2006;29:RC13–15. [PubMed: 17033253]
  • Thunhorst R.L, Johnson A.K. Effects of arterial pressure on drinking and urinary responses to intracerebroventricular angiotensin II. Am J Physiol. 1993;264:R211–7. [PubMed: 8430881]
  • Thunhorst R.L, Lewis S.J, Johnson A.K. Role of arterial baroreceptor input on thirst and urinary responses to intracerebroventricular angiotensin II. Am J Physiol. 1993;265:R591–95. [PubMed: 8214151]
  • Wren A.M, Small C.J, Ward H.L. et al. The novel hypothalamic peptide ghrelin stimulates food intake and growth hormone secretion. Endocrinology. 2000;141:4325–8. [PubMed: 11089570]
  • Yosten G.L, Samson W.K. The melanocortins, not oxytocin, mediate the anorexigenic and antidipsogenic actions of neuronostatin. Peptides. 2010;31:1711–4. [PMC free article: PMC2935629] [PubMed: 20600426]
  • Yosten G.L, Pate A.T, Samson W.K. Neuronostatin acts in brain to biphasically increase mean arterial pressure through sympatho-activation, followed by vasopressin secretion: The role of melanocortin receptors. Am J Physiol. 2011;300:R1194–9. [PMC free article: PMC3094032] [PubMed: 21325646]
  • Zhang J.V, Ren P.-G, Avsian-Kretchmer O. et al. Obestatin, a peptide encoded by the ghrelin gene, opposes ghrelin’s effect on food intake. Science. 2005;310:996–9. [PubMed: 16284174]
  • Zhang J.V, Klein C, Ren P.-G. et al. Response to comment on ‘Obestatin, a peptide encoded by the Ghrelin gene, opposes effects on food intake.’ Science. 2007;315:766. [PubMed: 17289961]
  • Zorilla E.P, Inoue K, Fekete E.M, Tabarin A, Valdez G.R, Koob G.F. Measuring meals: Structure of prandial food and water intake of rats. Am J Physiol. 2005;288:R1450–67. [PubMed: 15637168]
© 2014 by Taylor & Francis Group, LLC.
Bookshelf ID: NBK200963PMID: 24829995


  • PubReader
  • Print View
  • Cite this Page

Other titles in this collection

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Similar articles in PubMed

See reviews...See all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...