- Email: [email protected]
Postingestional modulation of drinking induced in rats by angiotensin II: Intragastric infusion and sham drinking studies
Physiology & Behavior, Vol. 40, pp. 539-543. Copyrighte PergamonJournals Ltd., 1987. Printedin the U.S.A.
0031-9384/87$3.00 + .00
Postingestional Modulation of Drinking Induced in Rats by Angiotensin II: Intragastric Infusion and Sham Drinking Studies J U A N I T A J. S A L I S B U R Y , N E I L E. R O W L A N D 2 A N D M E L V I N J. F R E G L Y
Departments o f Psychology and Physiology University o f Florida, Gainesville, FL 32611 R e c e i v e d 25 S e p t e m b e r 1986 SALISBURY, J. J., N. E. ROWLAND AND M. J. FREGLY. Postingestional modulation of drinking induced in rats by angiotensin H: lntragastric infusion and sham drinking studies. PHYSIOL BEHAV 40(4) 539--543, 1987.--The dipsogenic potency of angiotensin II (ANGII, 200 ~g/kg, SC) was examined after the removal of various postingestional factors. In the first experiment, rats with intragastric catheters were injected SC with ANGII and allowed water to drink. During the induced drinking, either NaCI (1.5 M) or water was injected into the stomach via the catheter at a rate of 0.1 ml for each ml water ingested orally. The water intake was identical in the two conditions. In a second experiment, rats, fitted with gastric fistulas, were administered ANGII and subsequent water intake with fistulas open was compared to that occurring with the fistulas closed. Rats drank more water during the first trial with fistulas open than with fistulas closed. Water intake during subsequent trials with an open fistula rose above that observed on the first trial. In a third experiment, rats with gastric fistulas were offered 0.15 M NaCI to drink. Intake was greater when the fistula was open than when it was closed. Intake of 0.15 M NaCI increased during the second trial with fistula open. Rats drank more 0.15 M NaCI during the first trial with an open fistula in Experiment 3 than those rats given water to drink on their first trial with fistula open in Experiment 2. These data suggest both oropharyngeal and postingestional factors interact in the control of ANGII-induced fluid intake. Angiotensin II Water intake Termination of drinking
NaCI intake
Intragastric infusions
DEPLETION of plasma volume is associated with both an increase in plasma angiotensin II (ANGII) concentrations and a compensatory fluid intake known as extracellular dehydration drinking. Injection of ANGII into water-replete animals also induces them to drink and it is therefore believed that ANGII mediates, at least in part, the fluid intake induced by extracellular dehydration [5, 11, 19]. ANGII may also act additively with other stimuli, such as intracellular dehydration, to initiate drinking. It is believed that separate neural pathways carry these signals, and that their signals add at an integrator and activate a final common pathway for drinking [5, 11, 21]. Much less is known about the factors that influence the amount and type of fluid consumed in response to elevated concentrations of ANGII in plasma. Studies using long-term infusion have revealed that the bulk of the fluid intake (80%) is consumed during the first hour, and that only small amounts of fluid are consumed thereafter [4, 7, 16]. This is
Sham drinking
not due to tachyphylaxis to ANGII because other studies have shown that either withholding water during the first hour of administration of ANGII or offering 0.15 M NaC1 solution to drink instead of water, produces just as much drinking in the second hour as during the first hour [8,10]. Rats injected with ANGII will often select an isotonic mixture if offered both NaCI and water to drink (see [8] for a recent summary). These studies show that ingestion of water, but not 0.15 M NaCI, potently inhibits ANGIIinduced drinking despite the continuing presence of ANGII. The termination of drinking to a SC injection of ANGII seems to be the result of a similar active inhibitory process [10]. Consistent with these conclusions, either intraperitoneal (IP) or intragastric (IG) preloads of water inhibited drinking in response to SC injection of ANGII and other dipsogens that activate the renin-angiotensin cascade, but preloads of 0.15 M NaCI were much less effective [9]. Most of these data can be explained according to the fol-
1portions of these data were presented at the April 1986 meeting of the Eastern Psychological Association (New York) and the October 1986 meeting of the American Physiological Society (New Orleans). This research was supported by NIH grant HL14526, and was also approved by the University of Florida All-University Committee for Use and Care of Lab Animals. ZRequests for reprints should be addressed to Dr. Neil E. Rowland, Department of Psychology, Unversity of Florida, Gainesville, FL 32611.
539
540
SALISBURY, R O W L A N D AND FREGLY A N G I O T E N S I N TI- = 200)Jg/kg SC PLUS I N T R A G A S T R I C I N F U S I O N S
E~
= WATER 0 . 1 5 M NaCI (n=11)
Implantation of Intragastric ('atheters
30-
¢~ =i 0 Z:
25$:S:S:3:S:S
O4
"~" LU
20-
v
,< I,-
z
Y///,4 15-
GI D
.../
"
10-
Y////, 5/////i
1.5M NaCI
WATEF
mesh cages in a vivarium with a controlled 12:12 light/dark cycle (lights on 0700-1900) and temperature (23_+ I°C). All rats were provided with Purina Laboratory Chow pellets (No. 5001) and distilled water to drink except where noted.
0.15M NaCI
INFUSION
FIG. 1. Mean (_+SE) intakes (ml/2 hr) of either water or 0.15 M NaCI by rats injected with ANGII and given various intragastric infusions concurrent with the drinking responses (0.1 ml infused per ml ingested). *p<0.01 differs from other two conditions.
lowing hypothesis. In a fluid-replete animal, ingestion of water will produce osmotic dilution of plasma and tissues, and this may provide an inhibitory signal which overrides an ongoing A N G I I stimulus. Whether the concomitant ingestion of NaC1 represents a true NaC1 appetite, a strategy to avoid plasma dilution, or is simply related to palatability is not clear [20]. The present series of experiments attempts to test some aspects of this hypothesis, by means of independent manipulation of oropharyngeal and postingestional factors associated with drinking. The two techniques used are: (1) intragastric infusion during oral ingestion of water in order to produce isotonic fluid in the stomach while the rat tastes only water, and (2) sham-drinking when either most or all of the ingested fluids drain out through a gastric fistula. E X P E R I M E N T 1. M O D U L A T I O N O F P O S T I N G E S T I V E OSMOTIC FACTORS In this experiment, using rats with indwelling intragastric (IG) catheters, we compared the intake of distilled water by rats injected with A N G I I and in which the effective osmolality of the ingested fluid was increased to isotonicity by concurrent IG infusion of NaCI, with the intake when isotonic NaC1 was the fluid available to drink. METHOD
Animals and Housing The subjects were l l male Sprague-Dawley rats, purchased from Zivic-Miller Labs, Allison Park, PA. and weighing 300-400 g. They were housed individually in hanging wire
Rats were anesthetized with 2.5 ml Equithesin/kg IP and, using procedures described previously [14], were fitted with indwelling IG catheters which entered through the stomach wall and were secured by a purse string suture, then exteriorized at a pedestal on the skull. The rats were given prophylactic penicillin and were allowed at least 5 days for recovery. Pl'ocdduF(' The dipsogenic response of each rat to SC injection of 200 /~g ANGII/kg ('~'Ileu-ANGII, Sigma) was measured. Food was removed at the time of the injection, and water intake was measured from graduated burettes attached to the front of their cages for 2 hr after injection of ANGII. The main phase of the experiment commenced the next day. The gastric catheters of each rat were attached by polyethylene tubing (PE-60) to remote infusion syringes and, on counterbalanced trials on different days, either 1.5 M NaCI or water was infused. These infusions were made by hand and consisted of 0.1 ml of infusate each time the rat had spontaneously ingested 1 ml of water from the burette. This continued until the rat stopped drinking. Thus, the 1.5 M NaCI infusions were diluted approximately 10-fold in the stomach to a final concentration close to isotonicity. The infusions of water simply increased the ingested volume by 10% and served as a control. In a final test, A N G I I was injected SC and 0.15 M NaC1 was offered instead of water as the drinking fluid. In this case 0.15 M NaC1 was also infused IG (0.1 ml/ml ingested). Food was returned after each daily test.
Statistics The data were examined by analyses of variance for repeated measures. Comparison between some treatments were made with paired t-tests. RESULTS The results are shown in Fig. I. The quantity of water consumed, and the time-course of drinking, by the rats infused with water were, as expected, approximately the same as for the rats receiving no infusion. When the stomach of each rat was infused with 1.5 M NaC1, the water intake was not significantly different from that observed following gastric infusion with water, t(10)=0.61, p>0.2. In contrast, when 0.15 M NaCl was offered as the drinking fluid, the intake was approximately 5-fold higher than it was for water, t(16)=6.2, p<0.01. DISCUSSION In this experiment, the postingestional consequences of ingested water were modified such that they would then provide osmotic dilution (water infusion) or no osmotic dilution (NaCl infusion). The water intakes were the same in these two conditions. It seems reasonable to assume that, within a few minutes, complete mixing of the ingested water and infused NaC1 will have occurred. The present results thus indicate that factors related to dilution are not directly controlling the quantity of water consumed. In agreement
P O S T I N G E S T I O N A L FACTORS A N D A N G I I - I N D U C E D D R I N K I N G
541
ANGIOTENSIN "rr" = 2 0 0 . , u g / k g S C A N D S H A M DRINKING
SO-
[]
"
WATER O. 15M NaCI
50.
to
40-
O -r
30" .(
20" ::) _t u,.
10"
•
1
2 CLOSED
3
4
4
5 OPEN
FISTULA
5
FISTULA
FIG. 2. Mean (-SE) intakes (ml/3 hr) of either water or 0.15 M NaCI by rats fitted with a gastric fistula and injected with ANGII. The five trials with the fistula closed are shown on the left, and the five trials with the fistula open are shown on the right (open and closed trials were alternated in the protocol). *p<0.05 intake with fistula open differs from intake with fistula closed.
with these data, Fitts and Simpson [4] have recently reported that intravenous infusions ofhypertonic NaCI that prevented plasma dilution from occurring as a result of water intake induced by central infusions of A N G I I did not increase (or disinhibit) further water intake. These two sets of results were not expected based on our previous findings that the water intake induced by A N G I I is inhibited more potently by IG or IP preloads of water than isotonic saline [9]. However, when the animals in this study were allowed to drink isotonic saline, the intake was greatly enhanced compared to the intake of water. This suggests that the taste of salt has a major facilitatory role in drinking induced by A N G I I , and Lhat osmotic dilution alone may not be the only factor affecting satiation. E X P E R I M E N T 2. A N G I I - I N D U C E D W A T E R I N T A K E AND SHAM DRINKING In this experiment, we used a sham-drinking procedure to evaluate the effects of complete, or near complete, removal of postingestional effects. METHOD
stomach, and a small piece of polyester mesh (RM54 Mersilene, Ethicon Inc.) was used to facilitate adhesion to the stomach and peritoneal walls. Rats were given postoperative penicillin, housed in plastic cages with wood shavings, and were allowed one week for recovery. They continued to appear in good health and to gain weight throughout the testing period.
Procedure After recovery, and to familiarize rats with the procedure, the fistula was opened and the gastric contents washed out with isotonic saline. The fistula was then closed and A N G I I (200 ~g/kg, SC) was administered, Distilled water was presented in graduated tubes and water intake was recorded at the end of 3 hr. Following this initial trial, tests were carried out four times with the fistula closed and four times with the fistula open. Under the latter condition (the sham drinking situation), water was observed to drain freely from the fistula within 1-2 sec of the onset of drinking. The order of the open and closed trials was alternated. At the end of the above series of tests, the rats were again injected with A N G I I and were offered 0.15 M NaCI to drink instead of water, once with the fistula open and once with the fistula closed.
Animals and Implantation of Fistula The animals and general conditions of housing were the same as in Experiment 1. Nine rats weighing between 300 and 400 g were anesthetized with 2.5 ml Equithesin/kg and fitted with an indwelling, stainless steel, gastric fistula with a stainless steel, screw-in plug. The fistulas weighed 4.5 g, and the inner and outer diameters were 7.0 and 8.5 mm, respectively. Each end of the fistula was formed into a flange 12.5 mm in diameter to facilitate holding the fistula in place. Fistulas were implanted through a stab wound in the
RESULTS The results are shown in Fig. 2. Water intake during the first four trials with fistula closed was constant. Water intake with the fistula open was greater than that observed with the fistula closed, F(1,17)= 19.74, p<0.01, and about twice that on the first trial with fistula closed (p<0.05). Water intakes with the fistula open increased with each successive trial, F(3,54)=7.1, p<0.01, and by the fourth trial water intake was about four times that when the fistula was closed. When
542
SALISBURY. ROWLAND AND FREGLY
isotonic saline was offered to drink during the 5th trial, intakes were increased by about 20 ml compared with the 4th water trial in both closed and open conditions (trial 4 vs. trial 5: F(1,12)=24.6, p<0.01).
ANGIOTENSIN11TM 200.ug/k9 SC
100-
DISCUSSION
The removal of postingestional factors associated with drinking clearly facilitated the intake in response to administration of ANGII. Thus, in contrast to the results of the previous experiment, it appears that osmotic, volumetric, or other factors play an active role in termination of ANGIIinduced drinking. It is also evident that postingestional factors interact with oropharyngeal factors in the termination of ANGIl-induced drinking because the rats drank more isotonic saline than water during the fifth trial, when isotonic saline was offered to drink instead of water. If osmotic postingestional stimuli alone terminated drinking, then the quantities of isotonic saline and water sham-drunk should be identical because the postingestional consequences of these fluids would be the same. Whether the increased NaC1 intake reflects previous experience with ANGII from the earlier trials or was due to the taste of salt is not clear from these data. Finally, while fluids were observed to drain freely from the fistulas, we have not made formal measures of the amount so lost by the animal. However, previous studies indicate that absorption should be minimal [18,22], but may contribute to the eventual cessation of sham drinking after about 2 hr.
.EE
z
AND SHAM
DRINKING
--OPEN FISTULA - CLOSED FISTULA •-TRIAL 1 o-TRIAL 2 (n=6)
80-
.o-----
60-
- o
/
z /- / 40-
/ O
/
_--"
0 ~ /
,0
.-0-
" ~
......
•
0
1-0
//;/ 0 ~ I - •
l
~
0
~
0
/
/
ii
i
go
io
go
'
120
'
150
~;o
TIME (rain)
FIG, 3. Cumulative mean intake of 0.15 M NaCl by rats injected with A N G I ] . Data from two trials with fistula closed and two with fistula open are shown; open and closed trials were alternated in the protocol. The intake on both trials with fistula open exceeded that of
the trials with fistula closed at all times (p<0.05).
In this experiment, sham-drinking was examined in rats drinking 0.15 M NaCI instead of water.
NaCI on trial 5 of Experiment 2 was not simply an artifact of previous testing of those rats, because it occurred in naive rats in the present study. Intake of NaCI was increased by about 20 ml during the first trial with the fistula open, and increased still more on the subsequent trial.
METHOD
GENERAL DISCUSSION
E X P E R I M E N T 3: A N G I I - I N D U C E D I N T A K E O F 0.15 M NACL AND SHAM-DRINKING
Animals. Surgery and Procedare Six rats weighing between 300 and 400 g were fitted with gastric fistulas in the manner described for Experiment 2. Immediately after implantation of the fistula, subjects were given ad lib access to both 0.15 M NaCI solution and distilled water for one week. At the end of this time, the animals were injected with A N G I I (200/zg/kg, SC) and given only 0.15 M NaCI to drink. Intakes were recorded every 15 min for 3 hr. On the first day, half of the rats were tested with their fistulas open and the other half with their fistulas closed. The rats were tested every 2 days for a total of four trials, two with the fistulas open, and two with them closed. The order of the trials was alternated between subjects. RESULTS
The results are plotted in Fig. 3 as cumulative intakes during the 3 hr tests. Intake of isotonic saline was similar during the trials with the fistula closed. Intake during the trials with the fistula open was higher, F(I,11)=34.4, p<0.01, than with the fistula closed. It was also greater on the second trial than the first, t(5)=6.43, p<0.01. These differences were apparent in the first 30 rain of the trial, F(I,11)= 12.6, p<0.05. It is also evident that ingestion continued for about 2.5 hr with the fistula open. DISCUSSION
These present findings verify that the effect of isotonic
The present series of experiments investigated the mechanisms of termination of ANGII-induced drinking in rats. The results suggest that the quantity of fluid consumed depends on an interaction of oropharyngeal and postingestional factors. In Experiment 1, it was expected that when ingested water was made isotonic in the stomach, water intake would be enhanced. This expectation was based on findings that preloads of isotonic saline are less effective in inhibiting ANGIl-induced water intake than are water preloads [9,17]. The results did not support this prediction; rats drank the same amount of water irrespective of the concurrent IG infusion. However, when isotonic saline was offered to drink, rats drank more of this fluid. These results indicate the importance of gustatory factors, but the experimental design did not allow us to determine whether the animals were responding either to taste alone or to a combination of taste and postingestional effects. The removal of most or all postingestional consequences in the sham-drinking paradigm allowed us to compare more readily ANGII-induced water and saline intakes. It was clear from Experiments 2 and 3 that postingestional and taste factors interacted in a complex way to inhibit fluid intake. During the first trial with the fistula open, fluid intake of rats sham-drinking water after administration of A N G I I was twice that occurring during the first trial with the fistula closed, indicating that gastric distention and/or fluid absorption inhibit further fluid intake when the fistula is d o s e d .
P O S T I N G E S T I O N A L FACTORS AND ANGII-INDUCED D R I N K I N G Removal of postingestional factors, however, did not render water functionally equivalent to 0.15 M NaC1. In Experiment 3, sham-consumption of saline on the first trial was much higher than sham-intake of water in Experiment 2. Our data indicate that the taste of NaCI is an important facilitator of intake in response to administration of ANGII. The sham-consumption of either drinking fluid increased with successive sham-drinking trials. This effect has been observed in sham-ingestion paradigms after water deprivation [1,22]. Whether this increase reflects a learning phenomenon has not been clarified. In the present study, intakes in the alternating trials with the fistula closed did not change, so the basis for the increase during successive trials with the fistula open must reside in the lack of fluid retention rather than any other situational cue. In this regard, it would be of interest to examine the effects of discriminative stimuli (e.g., odors or locations) associated with sham versus real drinking (or feeding) on the acquisition of sham-ingestion. While the present experiments were in progress, another study was published using similar methods to address the issue of termination of ANGII-induced drinking [15]. These investigators examined sham-drinking in pigeons with esophageal fistulas after an IV bolus of ANGII. Their findings that both the quantity and duration of water intake in-
543
creased in the sham-drinking situation agree with the present data. The onset of the drinking response to ANGII appears to be related to the rate at which ANGII is infused, and hence, to circulating ANGII levels [3,12]. It has not been possible to equate quantity of fluid consumed over an extended period with circulating ANGII levels [19]. Furthermore, it is clear that the dipsogenic effect of ANGII after SC administration far outlasts its short half-life [10]. Taken together, these findings suggest that the termination of ANGIIinduced drinking is not related to the removal of the initial ANGII signal. Indeed, plasma levels of ANGII which are increased by water deprivation actually rise still further for at least 4 hr after access to water is restored, even though most of the drinking occurs within the first hour [2]. This again is a situation in which elevated plasma levels of ANGII may not be dipsogenic because of plasma dilution or other consequences of ingestion. In summary, the termination of ANGII-induced drinking, and thereby the quantity of fluid consumed in response to a given dose or concentration of ANGII, depends upon oral, gastric and postgastric factors. Osmotic dilution alone does not appear to be an adequate explanation (Experiment 1 and [4]). Further studies will be needed to define the critical factor(s).
REFERENCES
1. Blass, E. M. and W. G. Hall. Drinking termination: Interactions among hydrational, orogastric, and behavioral controls in rats. Psychol Rev 83: 365-374, 1976. 2. DiNicolantonio, R. and F. A. Mendelsohn. Plasma renin and angiotensin in dehydrated and rehydrated rats. Am J Physiol 250: R838-R901, 1986. 3. Epstein, A. N. and S. Hsiao. Angiotensin as a dipsogen. In: Control Mechanisms of Drinking, edited by G. Peters, J. T. Fitzsimons and L. Peters-Haefeli. Berlin: Springer-Verlag, 1975. 4. Fitts, D. A. and J. B. Simpson. Drinking and natriuresis during volume expansion and intracranial angiotensin or carbachol. Am J Physiol 251: R381-R387, 1986. 5. Fitzsimons, J. T. The Physiology Of Thirst And Sodium Appetite. Cambridge: Cambridge University Press, 1979. 6. Fitzsimons, J. T., J. Kucharzcyk and G. Richards. Systemic angiotensin-induced drinking in the dog. A physiological phenomenon. J Physiol London 276: 435-448, 1978. 7. Fitzsimons, J. T. and B. J. Simons. The effect on drinking in the rat of intravenous infusions of angiotensin, given alone or in combination with other stimuli of thirst. J Physiol London 203: 45-57, 1969. 8. Fregly, M. J., O. E. Lockley and N. E. Rowland. Water versus NaCI intake by rats administered certain dipsogens acutely. Brain Res Bull 18: 245-251, 1987. 9. Fregly, M. J. and N. E. Rowland. Effect of intragastric and intraperitoneal water and saline loads on the pharmacologic induction of drinking in rats. Brain Res Bull 16: 407-414, 1986. 10. Fregly, M. J. and N. E. Rowland. Do peripheral and cerbroventricular injections of angiotensin II act at the same site? Studies on the additivity of drinking. Bruin Res Bull 16: 249-257, 1986.
11. Greenleaf, J. E. and M. J. Fregly. Dehydration-induced drinking: Peripheral and central aspects. Fed Proc 41: 2507-2510, 1982. 12. Johnson, A. K., J. E. Mann, R. Wolfgang, J. E. Johnson and D. Ganten. Plasma angiotensin II concentrations and experimentally induced thirst. Am J Physiol 240: R229-R234, 1981. 13. Mook, D. Oral and postingestional determinants of intake of various solutions in rats with esophageal fistulas. J Comp Physiol Psychol 56: 645-659, 1963. 14. Nicolaidis, S. and N. E. Rowland. Metering of fluid intake and determinants of ad libitum drinking in rats. Am J Physiol 231: 1-8, 1976. 15. Normile, H. J. and R. A. Barraco. Peripheral factors involved in the control of angiotensin-induced drinking in the pigeon tColumba livia). Pharmacol Res Commun 18: 111-127, 1986. 16. Radio, G. J., J. E. Summy-Long, A. Daniele-Severs and W. B. Severs. Hydration changes produced by central infusion of angiotensin II. Am J Physiol 223: 1221-1226, 1972. 17. Rolls, B. J. and D. J. McFarland. Hydration releases inhibition of feeding produced by intracranial angiotensin. Physiol Behav ll: 881-884, 1973. 18. Sclafani, A. and J. Nissenbaum. Is gastric sham feeding really sham feeding? Am J Physiol 248: R387-R390, 1986. 19. Stricker, E. M. The renin-angiotensin system and thirst: Some unanswered questions. Fed Proc 37: 2704--2710, 1978. 20. Stricker, E. M. Thirst and sodium appetite after colloid treatment in rats. J Comp Physiol Psychol 95: 1-25, 1981. 21. Toates, F. M. A physiological control theory of the hungerthirst interaction. In: Hunger Models, edited by D. A. Booth. London: Academic Press, 1978. 22. Waldbillig, R. J. and W. C. Lynch. Oroesophageal factors in the patterning of drinking. Physiol Behav 22: 205-209, 1979.
0031-9384/87$3.00 + .00
Postingestional Modulation of Drinking Induced in Rats by Angiotensin II: Intragastric Infusion and Sham Drinking Studies J U A N I T A J. S A L I S B U R Y , N E I L E. R O W L A N D 2 A N D M E L V I N J. F R E G L Y
Departments o f Psychology and Physiology University o f Florida, Gainesville, FL 32611 R e c e i v e d 25 S e p t e m b e r 1986 SALISBURY, J. J., N. E. ROWLAND AND M. J. FREGLY. Postingestional modulation of drinking induced in rats by angiotensin H: lntragastric infusion and sham drinking studies. PHYSIOL BEHAV 40(4) 539--543, 1987.--The dipsogenic potency of angiotensin II (ANGII, 200 ~g/kg, SC) was examined after the removal of various postingestional factors. In the first experiment, rats with intragastric catheters were injected SC with ANGII and allowed water to drink. During the induced drinking, either NaCI (1.5 M) or water was injected into the stomach via the catheter at a rate of 0.1 ml for each ml water ingested orally. The water intake was identical in the two conditions. In a second experiment, rats, fitted with gastric fistulas, were administered ANGII and subsequent water intake with fistulas open was compared to that occurring with the fistulas closed. Rats drank more water during the first trial with fistulas open than with fistulas closed. Water intake during subsequent trials with an open fistula rose above that observed on the first trial. In a third experiment, rats with gastric fistulas were offered 0.15 M NaCI to drink. Intake was greater when the fistula was open than when it was closed. Intake of 0.15 M NaCI increased during the second trial with fistula open. Rats drank more 0.15 M NaCI during the first trial with an open fistula in Experiment 3 than those rats given water to drink on their first trial with fistula open in Experiment 2. These data suggest both oropharyngeal and postingestional factors interact in the control of ANGII-induced fluid intake. Angiotensin II Water intake Termination of drinking
NaCI intake
Intragastric infusions
DEPLETION of plasma volume is associated with both an increase in plasma angiotensin II (ANGII) concentrations and a compensatory fluid intake known as extracellular dehydration drinking. Injection of ANGII into water-replete animals also induces them to drink and it is therefore believed that ANGII mediates, at least in part, the fluid intake induced by extracellular dehydration [5, 11, 19]. ANGII may also act additively with other stimuli, such as intracellular dehydration, to initiate drinking. It is believed that separate neural pathways carry these signals, and that their signals add at an integrator and activate a final common pathway for drinking [5, 11, 21]. Much less is known about the factors that influence the amount and type of fluid consumed in response to elevated concentrations of ANGII in plasma. Studies using long-term infusion have revealed that the bulk of the fluid intake (80%) is consumed during the first hour, and that only small amounts of fluid are consumed thereafter [4, 7, 16]. This is
Sham drinking
not due to tachyphylaxis to ANGII because other studies have shown that either withholding water during the first hour of administration of ANGII or offering 0.15 M NaC1 solution to drink instead of water, produces just as much drinking in the second hour as during the first hour [8,10]. Rats injected with ANGII will often select an isotonic mixture if offered both NaCI and water to drink (see [8] for a recent summary). These studies show that ingestion of water, but not 0.15 M NaCI, potently inhibits ANGIIinduced drinking despite the continuing presence of ANGII. The termination of drinking to a SC injection of ANGII seems to be the result of a similar active inhibitory process [10]. Consistent with these conclusions, either intraperitoneal (IP) or intragastric (IG) preloads of water inhibited drinking in response to SC injection of ANGII and other dipsogens that activate the renin-angiotensin cascade, but preloads of 0.15 M NaCI were much less effective [9]. Most of these data can be explained according to the fol-
1portions of these data were presented at the April 1986 meeting of the Eastern Psychological Association (New York) and the October 1986 meeting of the American Physiological Society (New Orleans). This research was supported by NIH grant HL14526, and was also approved by the University of Florida All-University Committee for Use and Care of Lab Animals. ZRequests for reprints should be addressed to Dr. Neil E. Rowland, Department of Psychology, Unversity of Florida, Gainesville, FL 32611.
539
540
SALISBURY, R O W L A N D AND FREGLY A N G I O T E N S I N TI- = 200)Jg/kg SC PLUS I N T R A G A S T R I C I N F U S I O N S
E~
= WATER 0 . 1 5 M NaCI (n=11)
Implantation of Intragastric ('atheters
30-
¢~ =i 0 Z:
25$:S:S:3:S:S
O4
"~" LU
20-
v
,< I,-
z
Y///,4 15-
GI D
.../
"
10-
Y////, 5/////i
1.5M NaCI
WATEF
mesh cages in a vivarium with a controlled 12:12 light/dark cycle (lights on 0700-1900) and temperature (23_+ I°C). All rats were provided with Purina Laboratory Chow pellets (No. 5001) and distilled water to drink except where noted.
0.15M NaCI
INFUSION
FIG. 1. Mean (_+SE) intakes (ml/2 hr) of either water or 0.15 M NaCI by rats injected with ANGII and given various intragastric infusions concurrent with the drinking responses (0.1 ml infused per ml ingested). *p<0.01 differs from other two conditions.
lowing hypothesis. In a fluid-replete animal, ingestion of water will produce osmotic dilution of plasma and tissues, and this may provide an inhibitory signal which overrides an ongoing A N G I I stimulus. Whether the concomitant ingestion of NaC1 represents a true NaC1 appetite, a strategy to avoid plasma dilution, or is simply related to palatability is not clear [20]. The present series of experiments attempts to test some aspects of this hypothesis, by means of independent manipulation of oropharyngeal and postingestional factors associated with drinking. The two techniques used are: (1) intragastric infusion during oral ingestion of water in order to produce isotonic fluid in the stomach while the rat tastes only water, and (2) sham-drinking when either most or all of the ingested fluids drain out through a gastric fistula. E X P E R I M E N T 1. M O D U L A T I O N O F P O S T I N G E S T I V E OSMOTIC FACTORS In this experiment, using rats with indwelling intragastric (IG) catheters, we compared the intake of distilled water by rats injected with A N G I I and in which the effective osmolality of the ingested fluid was increased to isotonicity by concurrent IG infusion of NaCI, with the intake when isotonic NaC1 was the fluid available to drink. METHOD
Animals and Housing The subjects were l l male Sprague-Dawley rats, purchased from Zivic-Miller Labs, Allison Park, PA. and weighing 300-400 g. They were housed individually in hanging wire
Rats were anesthetized with 2.5 ml Equithesin/kg IP and, using procedures described previously [14], were fitted with indwelling IG catheters which entered through the stomach wall and were secured by a purse string suture, then exteriorized at a pedestal on the skull. The rats were given prophylactic penicillin and were allowed at least 5 days for recovery. Pl'ocdduF(' The dipsogenic response of each rat to SC injection of 200 /~g ANGII/kg ('~'Ileu-ANGII, Sigma) was measured. Food was removed at the time of the injection, and water intake was measured from graduated burettes attached to the front of their cages for 2 hr after injection of ANGII. The main phase of the experiment commenced the next day. The gastric catheters of each rat were attached by polyethylene tubing (PE-60) to remote infusion syringes and, on counterbalanced trials on different days, either 1.5 M NaCI or water was infused. These infusions were made by hand and consisted of 0.1 ml of infusate each time the rat had spontaneously ingested 1 ml of water from the burette. This continued until the rat stopped drinking. Thus, the 1.5 M NaCI infusions were diluted approximately 10-fold in the stomach to a final concentration close to isotonicity. The infusions of water simply increased the ingested volume by 10% and served as a control. In a final test, A N G I I was injected SC and 0.15 M NaC1 was offered instead of water as the drinking fluid. In this case 0.15 M NaC1 was also infused IG (0.1 ml/ml ingested). Food was returned after each daily test.
Statistics The data were examined by analyses of variance for repeated measures. Comparison between some treatments were made with paired t-tests. RESULTS The results are shown in Fig. I. The quantity of water consumed, and the time-course of drinking, by the rats infused with water were, as expected, approximately the same as for the rats receiving no infusion. When the stomach of each rat was infused with 1.5 M NaC1, the water intake was not significantly different from that observed following gastric infusion with water, t(10)=0.61, p>0.2. In contrast, when 0.15 M NaCl was offered as the drinking fluid, the intake was approximately 5-fold higher than it was for water, t(16)=6.2, p<0.01. DISCUSSION In this experiment, the postingestional consequences of ingested water were modified such that they would then provide osmotic dilution (water infusion) or no osmotic dilution (NaCl infusion). The water intakes were the same in these two conditions. It seems reasonable to assume that, within a few minutes, complete mixing of the ingested water and infused NaC1 will have occurred. The present results thus indicate that factors related to dilution are not directly controlling the quantity of water consumed. In agreement
P O S T I N G E S T I O N A L FACTORS A N D A N G I I - I N D U C E D D R I N K I N G
541
ANGIOTENSIN "rr" = 2 0 0 . , u g / k g S C A N D S H A M DRINKING
SO-
[]
"
WATER O. 15M NaCI
50.
to
40-
O -r
30" .(
20" ::) _t u,.
10"
•
1
2 CLOSED
3
4
4
5 OPEN
FISTULA
5
FISTULA
FIG. 2. Mean (-SE) intakes (ml/3 hr) of either water or 0.15 M NaCI by rats fitted with a gastric fistula and injected with ANGII. The five trials with the fistula closed are shown on the left, and the five trials with the fistula open are shown on the right (open and closed trials were alternated in the protocol). *p<0.05 intake with fistula open differs from intake with fistula closed.
with these data, Fitts and Simpson [4] have recently reported that intravenous infusions ofhypertonic NaCI that prevented plasma dilution from occurring as a result of water intake induced by central infusions of A N G I I did not increase (or disinhibit) further water intake. These two sets of results were not expected based on our previous findings that the water intake induced by A N G I I is inhibited more potently by IG or IP preloads of water than isotonic saline [9]. However, when the animals in this study were allowed to drink isotonic saline, the intake was greatly enhanced compared to the intake of water. This suggests that the taste of salt has a major facilitatory role in drinking induced by A N G I I , and Lhat osmotic dilution alone may not be the only factor affecting satiation. E X P E R I M E N T 2. A N G I I - I N D U C E D W A T E R I N T A K E AND SHAM DRINKING In this experiment, we used a sham-drinking procedure to evaluate the effects of complete, or near complete, removal of postingestional effects. METHOD
stomach, and a small piece of polyester mesh (RM54 Mersilene, Ethicon Inc.) was used to facilitate adhesion to the stomach and peritoneal walls. Rats were given postoperative penicillin, housed in plastic cages with wood shavings, and were allowed one week for recovery. They continued to appear in good health and to gain weight throughout the testing period.
Procedure After recovery, and to familiarize rats with the procedure, the fistula was opened and the gastric contents washed out with isotonic saline. The fistula was then closed and A N G I I (200 ~g/kg, SC) was administered, Distilled water was presented in graduated tubes and water intake was recorded at the end of 3 hr. Following this initial trial, tests were carried out four times with the fistula closed and four times with the fistula open. Under the latter condition (the sham drinking situation), water was observed to drain freely from the fistula within 1-2 sec of the onset of drinking. The order of the open and closed trials was alternated. At the end of the above series of tests, the rats were again injected with A N G I I and were offered 0.15 M NaCI to drink instead of water, once with the fistula open and once with the fistula closed.
Animals and Implantation of Fistula The animals and general conditions of housing were the same as in Experiment 1. Nine rats weighing between 300 and 400 g were anesthetized with 2.5 ml Equithesin/kg and fitted with an indwelling, stainless steel, gastric fistula with a stainless steel, screw-in plug. The fistulas weighed 4.5 g, and the inner and outer diameters were 7.0 and 8.5 mm, respectively. Each end of the fistula was formed into a flange 12.5 mm in diameter to facilitate holding the fistula in place. Fistulas were implanted through a stab wound in the
RESULTS The results are shown in Fig. 2. Water intake during the first four trials with fistula closed was constant. Water intake with the fistula open was greater than that observed with the fistula closed, F(1,17)= 19.74, p<0.01, and about twice that on the first trial with fistula closed (p<0.05). Water intakes with the fistula open increased with each successive trial, F(3,54)=7.1, p<0.01, and by the fourth trial water intake was about four times that when the fistula was closed. When
542
SALISBURY. ROWLAND AND FREGLY
isotonic saline was offered to drink during the 5th trial, intakes were increased by about 20 ml compared with the 4th water trial in both closed and open conditions (trial 4 vs. trial 5: F(1,12)=24.6, p<0.01).
ANGIOTENSIN11TM 200.ug/k9 SC
100-
DISCUSSION
The removal of postingestional factors associated with drinking clearly facilitated the intake in response to administration of ANGII. Thus, in contrast to the results of the previous experiment, it appears that osmotic, volumetric, or other factors play an active role in termination of ANGIIinduced drinking. It is also evident that postingestional factors interact with oropharyngeal factors in the termination of ANGIl-induced drinking because the rats drank more isotonic saline than water during the fifth trial, when isotonic saline was offered to drink instead of water. If osmotic postingestional stimuli alone terminated drinking, then the quantities of isotonic saline and water sham-drunk should be identical because the postingestional consequences of these fluids would be the same. Whether the increased NaC1 intake reflects previous experience with ANGII from the earlier trials or was due to the taste of salt is not clear from these data. Finally, while fluids were observed to drain freely from the fistulas, we have not made formal measures of the amount so lost by the animal. However, previous studies indicate that absorption should be minimal [18,22], but may contribute to the eventual cessation of sham drinking after about 2 hr.
.EE
z
AND SHAM
DRINKING
--OPEN FISTULA - CLOSED FISTULA •-TRIAL 1 o-TRIAL 2 (n=6)
80-
.o-----
60-
- o
/
z /- / 40-
/ O
/
_--"
0 ~ /
,0
.-0-
" ~
......
•
0
1-0
//;/ 0 ~ I - •
l
~
0
~
0
/
/
ii
i
go
io
go
'
120
'
150
~;o
TIME (rain)
FIG, 3. Cumulative mean intake of 0.15 M NaCl by rats injected with A N G I ] . Data from two trials with fistula closed and two with fistula open are shown; open and closed trials were alternated in the protocol. The intake on both trials with fistula open exceeded that of
the trials with fistula closed at all times (p<0.05).
In this experiment, sham-drinking was examined in rats drinking 0.15 M NaCI instead of water.
NaCI on trial 5 of Experiment 2 was not simply an artifact of previous testing of those rats, because it occurred in naive rats in the present study. Intake of NaCI was increased by about 20 ml during the first trial with the fistula open, and increased still more on the subsequent trial.
METHOD
GENERAL DISCUSSION
E X P E R I M E N T 3: A N G I I - I N D U C E D I N T A K E O F 0.15 M NACL AND SHAM-DRINKING
Animals. Surgery and Procedare Six rats weighing between 300 and 400 g were fitted with gastric fistulas in the manner described for Experiment 2. Immediately after implantation of the fistula, subjects were given ad lib access to both 0.15 M NaCI solution and distilled water for one week. At the end of this time, the animals were injected with A N G I I (200/zg/kg, SC) and given only 0.15 M NaCI to drink. Intakes were recorded every 15 min for 3 hr. On the first day, half of the rats were tested with their fistulas open and the other half with their fistulas closed. The rats were tested every 2 days for a total of four trials, two with the fistulas open, and two with them closed. The order of the trials was alternated between subjects. RESULTS
The results are plotted in Fig. 3 as cumulative intakes during the 3 hr tests. Intake of isotonic saline was similar during the trials with the fistula closed. Intake during the trials with the fistula open was higher, F(I,11)=34.4, p<0.01, than with the fistula closed. It was also greater on the second trial than the first, t(5)=6.43, p<0.01. These differences were apparent in the first 30 rain of the trial, F(I,11)= 12.6, p<0.05. It is also evident that ingestion continued for about 2.5 hr with the fistula open. DISCUSSION
These present findings verify that the effect of isotonic
The present series of experiments investigated the mechanisms of termination of ANGII-induced drinking in rats. The results suggest that the quantity of fluid consumed depends on an interaction of oropharyngeal and postingestional factors. In Experiment 1, it was expected that when ingested water was made isotonic in the stomach, water intake would be enhanced. This expectation was based on findings that preloads of isotonic saline are less effective in inhibiting ANGIl-induced water intake than are water preloads [9,17]. The results did not support this prediction; rats drank the same amount of water irrespective of the concurrent IG infusion. However, when isotonic saline was offered to drink, rats drank more of this fluid. These results indicate the importance of gustatory factors, but the experimental design did not allow us to determine whether the animals were responding either to taste alone or to a combination of taste and postingestional effects. The removal of most or all postingestional consequences in the sham-drinking paradigm allowed us to compare more readily ANGII-induced water and saline intakes. It was clear from Experiments 2 and 3 that postingestional and taste factors interacted in a complex way to inhibit fluid intake. During the first trial with the fistula open, fluid intake of rats sham-drinking water after administration of A N G I I was twice that occurring during the first trial with the fistula closed, indicating that gastric distention and/or fluid absorption inhibit further fluid intake when the fistula is d o s e d .
P O S T I N G E S T I O N A L FACTORS AND ANGII-INDUCED D R I N K I N G Removal of postingestional factors, however, did not render water functionally equivalent to 0.15 M NaC1. In Experiment 3, sham-consumption of saline on the first trial was much higher than sham-intake of water in Experiment 2. Our data indicate that the taste of NaCI is an important facilitator of intake in response to administration of ANGII. The sham-consumption of either drinking fluid increased with successive sham-drinking trials. This effect has been observed in sham-ingestion paradigms after water deprivation [1,22]. Whether this increase reflects a learning phenomenon has not been clarified. In the present study, intakes in the alternating trials with the fistula closed did not change, so the basis for the increase during successive trials with the fistula open must reside in the lack of fluid retention rather than any other situational cue. In this regard, it would be of interest to examine the effects of discriminative stimuli (e.g., odors or locations) associated with sham versus real drinking (or feeding) on the acquisition of sham-ingestion. While the present experiments were in progress, another study was published using similar methods to address the issue of termination of ANGII-induced drinking [15]. These investigators examined sham-drinking in pigeons with esophageal fistulas after an IV bolus of ANGII. Their findings that both the quantity and duration of water intake in-
543
creased in the sham-drinking situation agree with the present data. The onset of the drinking response to ANGII appears to be related to the rate at which ANGII is infused, and hence, to circulating ANGII levels [3,12]. It has not been possible to equate quantity of fluid consumed over an extended period with circulating ANGII levels [19]. Furthermore, it is clear that the dipsogenic effect of ANGII after SC administration far outlasts its short half-life [10]. Taken together, these findings suggest that the termination of ANGIIinduced drinking is not related to the removal of the initial ANGII signal. Indeed, plasma levels of ANGII which are increased by water deprivation actually rise still further for at least 4 hr after access to water is restored, even though most of the drinking occurs within the first hour [2]. This again is a situation in which elevated plasma levels of ANGII may not be dipsogenic because of plasma dilution or other consequences of ingestion. In summary, the termination of ANGII-induced drinking, and thereby the quantity of fluid consumed in response to a given dose or concentration of ANGII, depends upon oral, gastric and postgastric factors. Osmotic dilution alone does not appear to be an adequate explanation (Experiment 1 and [4]). Further studies will be needed to define the critical factor(s).
REFERENCES
1. Blass, E. M. and W. G. Hall. Drinking termination: Interactions among hydrational, orogastric, and behavioral controls in rats. Psychol Rev 83: 365-374, 1976. 2. DiNicolantonio, R. and F. A. Mendelsohn. Plasma renin and angiotensin in dehydrated and rehydrated rats. Am J Physiol 250: R838-R901, 1986. 3. Epstein, A. N. and S. Hsiao. Angiotensin as a dipsogen. In: Control Mechanisms of Drinking, edited by G. Peters, J. T. Fitzsimons and L. Peters-Haefeli. Berlin: Springer-Verlag, 1975. 4. Fitts, D. A. and J. B. Simpson. Drinking and natriuresis during volume expansion and intracranial angiotensin or carbachol. Am J Physiol 251: R381-R387, 1986. 5. Fitzsimons, J. T. The Physiology Of Thirst And Sodium Appetite. Cambridge: Cambridge University Press, 1979. 6. Fitzsimons, J. T., J. Kucharzcyk and G. Richards. Systemic angiotensin-induced drinking in the dog. A physiological phenomenon. J Physiol London 276: 435-448, 1978. 7. Fitzsimons, J. T. and B. J. Simons. The effect on drinking in the rat of intravenous infusions of angiotensin, given alone or in combination with other stimuli of thirst. J Physiol London 203: 45-57, 1969. 8. Fregly, M. J., O. E. Lockley and N. E. Rowland. Water versus NaCI intake by rats administered certain dipsogens acutely. Brain Res Bull 18: 245-251, 1987. 9. Fregly, M. J. and N. E. Rowland. Effect of intragastric and intraperitoneal water and saline loads on the pharmacologic induction of drinking in rats. Brain Res Bull 16: 407-414, 1986. 10. Fregly, M. J. and N. E. Rowland. Do peripheral and cerbroventricular injections of angiotensin II act at the same site? Studies on the additivity of drinking. Bruin Res Bull 16: 249-257, 1986.
11. Greenleaf, J. E. and M. J. Fregly. Dehydration-induced drinking: Peripheral and central aspects. Fed Proc 41: 2507-2510, 1982. 12. Johnson, A. K., J. E. Mann, R. Wolfgang, J. E. Johnson and D. Ganten. Plasma angiotensin II concentrations and experimentally induced thirst. Am J Physiol 240: R229-R234, 1981. 13. Mook, D. Oral and postingestional determinants of intake of various solutions in rats with esophageal fistulas. J Comp Physiol Psychol 56: 645-659, 1963. 14. Nicolaidis, S. and N. E. Rowland. Metering of fluid intake and determinants of ad libitum drinking in rats. Am J Physiol 231: 1-8, 1976. 15. Normile, H. J. and R. A. Barraco. Peripheral factors involved in the control of angiotensin-induced drinking in the pigeon tColumba livia). Pharmacol Res Commun 18: 111-127, 1986. 16. Radio, G. J., J. E. Summy-Long, A. Daniele-Severs and W. B. Severs. Hydration changes produced by central infusion of angiotensin II. Am J Physiol 223: 1221-1226, 1972. 17. Rolls, B. J. and D. J. McFarland. Hydration releases inhibition of feeding produced by intracranial angiotensin. Physiol Behav ll: 881-884, 1973. 18. Sclafani, A. and J. Nissenbaum. Is gastric sham feeding really sham feeding? Am J Physiol 248: R387-R390, 1986. 19. Stricker, E. M. The renin-angiotensin system and thirst: Some unanswered questions. Fed Proc 37: 2704--2710, 1978. 20. Stricker, E. M. Thirst and sodium appetite after colloid treatment in rats. J Comp Physiol Psychol 95: 1-25, 1981. 21. Toates, F. M. A physiological control theory of the hungerthirst interaction. In: Hunger Models, edited by D. A. Booth. London: Academic Press, 1978. 22. Waldbillig, R. J. and W. C. Lynch. Oroesophageal factors in the patterning of drinking. Physiol Behav 22: 205-209, 1979.