Physiology&Behavior,Vol. 52, pp. 885-888, 1992
0031-9384/92 $5.00 + .00 Copyright © 1992 PergamonPress Ltd.
Printed in the USA.
Effect of Hepatic Portal Infusion of Water on Water Intake by Water-Deprived Rats MOTOI
KOBASHI
AND AKIRA ADACHI l
Department of Physiology, Okayama University Dental School, Okayama, 700, Japan R e c e i v e d 22 M a y 1992 KOBASHI, M. AND A. ADACHI. Effect of hepatic portal infusion of water on water intake by water-deprivedrats. PHYSIOL BEHAV 52(5) 885-888, 1992.--To determine whether or not hepatoportal osmoreceptive (or sodium-receptive) signals participate in the control of drinking, we examined the effects of portal infusion of water, 0.9% saline, and 1.8% saline on water intake by water-deprived rats. Infusion was started 0.5 h prior to the end of the water deprivation period for 3.5 h at a rate of 52 ul/min through either a portal or a jugular catheter. After 24-h water deprivation, water intake was measured successively for 24 h without food. As a result of the water infusion tests, water intake of the portal infusion group was significantly less than that of the jugular infusion group during and after the infusion. Portal infusion of neither 0.9% nor 1.8% saline affected the water intake compared to similar infusion into the jugular vein. It is concluded that hypotonic stimulation of the hepatoportal osmoreceptor suppresses water intake in water-deprived rats. On the contrary, isotonic or hypertonic stimulation does not produce any change of water intake. Hepatic
Portal
Osmoreceptor
Drinking
Water intake
IT is well known that hepatoportal osmoreceptors (or sodium receptors) participate in body fluid homeostasis in regulating urine output (1,7). Possible participation of these osmoreceptors in the control of fluid ingestion has been studied by salt preference tests. Portal infusion of hypertonic saline suppressed saline intake but did not influence water intake in normal rats (11), water-deprived rats (4), or in sodium-deficit rats ( 15,16). Hepatic vagotomy attenuated saline intake by rats, but did not affect water intake (5). In addition, hepatic vagotomy did not affect water intake induced by intraperitoneal injection of hypertonic saline, when only water was supplied (5,14), nor in the case of prandial drinking (1,3). Thus, the hepatoportal osmoreceptors regulate saline intake, but not water intake, when water and saline are supplied simultaneously. However, hepatic vagotomy induced overingestion of water after water deprivation, but water intake induced by angiotensin or intraperitoneal injection of hypertonic saline was not affected by hepatic vagotomy (13). It seems that hepatoportal osmoreceptors regulate water intake induced by water deprivation alone. In the present experiments, to determine whether or not hepatoportal osmoreceptive signals participate in the control of water intake, we examined the effects of portal infusion of water, 0.9% saline, and 1.8% saline on water intake in water-deprived rats.
Infusion
Water deprivation
Rat
rats were housed in 25 x 25 X 40-cm Plexiglas cages, on a 12:12 light:dark cycle (light from 0600 to 1800), in a room at 2 3 - 2 4 ° C temperature. Rats were maintained in this cage on tap water, except when deprived, and normal chow until the end of the experiments.
Surgery All rats received surgery to catheterize the hepatic portal vein and the right jugular vein. No infusion control (sham infusion) group was prepared. Each catheter consisted of a Silastic tube (0.50 m m i.d., 1.00 m m o.d., Dow-Corning) connected to a polyethylene tube (PE-50, 0.58 m m i.d., 0.97 m m o.d., Intramedic) with a 23-gauge stainless steel tube. Rats were anesthetized with 50 mg/kg body weight sodium pentobarbital. The hepatic portal catheter was inserted into the ileocolic vein near the cecum through a midline abdominal incision. The catheter, a Silastic tube, was inserted toward the liver for about 4 cm, and fixed to the vessel with two sutures. The right jugular vein was exposed for catheterization. The end of a Silastic tube was inserted toward the atrium about 3 cm, and fixed with two sutures. The external ends of the portal and jugular catheters were tunneled under the skin and brought to the surface between the shoulder blades. After all incisions were closed, the catheters were passed through a 5-cm long rubber tube attached by dental cement to three screws in the skull. The catheters were protected by a 35-cm long stainless steel spiral spring and then joined to a dual channel swivel (375 D/22, Instech). After all surgery, the
METHOD
Subjects and Maintenance Eight-week-old male Sprague-Dawley rats (Charles River) weighing 255-330 g were used in all experiments. After surgery
Requests for reprints should be addressed to Dr. Akira Adachi, Department of Physiology, Okayama University Dental School, Shikata-cho, Okayama, 700, Japan. 885
886
KOBASHI A N D A D A C H I
without infusing the solution. Thus, rats were divided into seven groups: sham infusion (nine rats), portal infusion of water (10 rats), jugular infusion of water (nine rats), portal infusion of 0.9% saline (seven rats), jugular infusion of 0.9% saline (eight rats), portal infusion of 1.8% saline (nine rats), and jugular infusion of 1.8% saline (seven rats). Food intake from the start of the dark period until the start of infusion was measured daily for 3 days. Water intake was measured at all times after surgery. Before the test, daily food and water intakes were not significantly different among the seven groups. After all experiments, the animals were sacrificed by overdoses of anesthesia, and the integrity of the catheters and locations of their tips were confirmed. If any catheters appeared to leak, those data were omitted.
',OHT 17:00
I
I
18:00
21:00
17:,3o I I '. ~ i
I ! i i
I
I
I I
Infusion
(3.5 h)
~' vi
:,, Food Deprivation (25 h) - -
:~ ,~
Water Deprivation (24 h)
~;~
Measurement (24 h)---
I
FIG. 1. Schematic representation of experimental protocol.
rats were moved into the Plexiglas cages and maintained on normal chow and tap water until the termination of the tests.
Data Analysis Water intake was expressed in ml by multiplying the number of drops by 0.067. Cumulative intake was measured 0.5, 1, 2, 3, 6, 9, 12, and 24 h after presentation of tap water. Each 3-h intake in the dark period and 12-h intake in the light period are presented as bar graphs (Fig. 3). Data are presented as mean + SE. Results were analyzed using Student's t-tests.
Test Procedure All rats were tested in a water-deprived state. Eight or 9 days after surgery, rats were deprived of water from 1800 (at the beginning of the dark period). One h before the start of the next dark period (1700), food was removed. Infusion was started at 1730 for 3.5 fi at a rate of 52 ul/min. At 1800, tap water was supplied again. Water intake without food was measured for 24 h by a drop counter and recorded on a pen recorder for later analysis. Figure 1 shows the experimental protocol. Each rat received either portal or jugular infusion. Infusate was either deionized water, 0.9% saline, or 1.8% saline. Each rat was tested only one time. Sham infusion was tested by the same procedure
RESULTS
Portal and Jugular Infusion of Water Cumulative water intake during infusion of water through the portal or the jugular vein is presented in the left-hand graph of Fig. 2. Portal infusion of water suppressed water intake sig-
0.9% NaCI
Water
1.8% NaCI
50
~E 40
Sham Infusion (N=9) Portal Infusion (N=10) Jugular Infusion (N=9)
+ --i--
J r"
--o.=
Portal Infusion (N=7) Jugular Infusion (N=8)
+ +
Portal Infusion (N=9) Jugular Infusion (N=7)
I
N
a0
/
•~_~ 20
E 0 ~o
I .
•
%*
°.
:l
,.
,:
,.':
:*
12
~--,24 0
o*
0 0
~ 3
6
9
Time (Hours)
II ;
6
9
Time (Hours)
1)
I] 24
0
3
6
;
12
24
Time (Hours)
FIG. 2. Effects of the infusion of water (left), 0.9% saline (middle), and 1.8% saline (right) on water intake. Cumulative water intakes are shown. *p < 0.05, **p < 0.01, ***p < 0.001 when water intakes induced by portal infusion of water are compared with those of sham infusion; ,p < 0.05, --p
< 0.005, - - p < 0.00l when water intakes induced by portal infusion of water are compared with those of jugular infusion of water. Portal infusion of 0.9% and 1.8% saline did not produce any change compared to jugular infusion of the same solution.
EFFECT OF PORTAL INFUSION ON WATER INTAKE
887
nificantly after 24-h water deprivation. Compared to the jugular infusion of water, the cumulative water intake was reduced at 3 h after presentation of tap water (t = 2.682, p < 0.05), at 24 h (t = 3.593, p < 0.005), and at all times between. Water intake in each 3-h interval in the dark period and in the 12-h light period is shown in the left histogram of Fig. 3. Water intake was reduced in every period (t = 4.249, p < 0.001, 3-6 h; t = 2.421, p < 0.05, 6-9 h; t = 2.879, p < 0.05, 9-12 h; t = 2.969, p < 0.01, 12-24 h). As shown in the left-hand graph of Fig. 2, the cumulative water intake after presentation of tap water was significantly reduced by the portal infusion of water from 1 h (t = 2.463, p < 0.05) to 24 h (t = 3.138, p < 0.01), compared to sham infusion. Three-h intake from 0 h to 3 h (t = 4.176, p < 0.001), from 3 h to 6 h (t = 3.242, p < 0.005), and 12-h water intake in the light period (t = 2.435, p < 0.05) were significantly reduced, as shown in the left-hand graph of Fig. 3.
duced by 24-h water deprivation. Rats treated by portal infusion of water drank less than those treated by jugular infusion. This fact shows that the hepatoportal region is involved in the suppression of water intake. The portal water infusion group also drank less than the portal isotonic 0.9% saline infusion group (t = 6.193, p < 0.001, 0-3 h; t = 2.297, p < 0.05, 3-6 h; t = 0.021, NS, 6-9 h; t = 2.585, p < 0.05, 9-12 h; t = 0.669, NS, 12-24 h). Osmoreceptors (or sodium receptors) in the hepatoportal region participate in this suppression. Hepatoportal osmoreceptive afferents project to the median preoptic nucleus (9) and to the lateral hypothalamic area (8), which is closely related to water intake. The suppression appeared during the infusion and lasted for a relatively long time. The suppression during the infusion is easily understood to be a function of the neural system; the hepatic branch of the vagus nerve conveys osmotic information about the portal blood. However, it is difficult to explain the long-lasting suppression following the end of the infusion. In our experiments, the portal infusion of water or 1.8% saline changed the plasma osmolarity 1-2%, estimating the portal blood flow as 5-10 ml/min. Calculating from a previous report (2), when rats weighing 300 g received injections of 15 ml water into the stomach, the absorption rate was 660 #l/min during the first 15 min and 100/A/min in the subsequent 45 min. Therefore, the infusion rate used in this experiment (52 #l/min) was thought to be within the range of physiological fluctuation. Previous works showed that hypertonic stimulation of the hepatoportal osmoreceptors changed saline intake without affecting water intake in the normal state (11), and in waterdeprived (4) and sodium-deficit rats (l 5,16). On the other hand, we observed suppression of water intake by hypotonic stimulation of the hepatoportal osmoreceptors, and hypertonic stimulation produced no change in water intake. It is suggested that hypotonic stimulation of the hepatoportal osmoreceptors should suppress water intake, and their hypertonic stimulation should suppress saline intake via independent neuronal circuits in the central nervous system (CNS). That is, the hepatic branch of the vagus nerve that conveys hepatoportal osmoreceptive signals to the CNS includes two types of neurons: one is activated by increment of osmolarity in the portal vein; the other is activated by decrement of osmolarity (1). Thus, there are
Portal and Jugular Infusion of 0.9% Safine The portal infusion of 0.9% saline did not significantlychange water intake compared to jugular infusion of 0.9% saline and sham infusion (middle graphs of Figs. 2 and 3).
Portal Infusion and Jugular Infusion of 1.8% Saline As illustrated in the right-hand graph of Fig. 2, the portal infusion of 1.8% saline did not change the cumulative water intake compared to the jugular infusion of 1.8% saline. However, the right-hand graph of Fig. 3 shows that 0-3-h water intake was facilitated in the portal infusion group compared to the sham infusion group (t = 2.124, p < 0.05), but not the jugular infusion group (t = 1.777, NS). The 6-9-h water intake was suppressed in the portal infusion group (t = 2.293, p < 0.05) but not significantly in the jugular infusion group (t = 1.405, NS). The 12h water intake in the light period was suppressed in the portal (t = 2.159, p < 0.05) and the jugular (t = 2.597, p < 0.05) infusion groups, compared to the sham group.
DISCUSSION Our present results clearly showed that, during and after the infusion, portal infusion of water suppressed water intake in-
2o
0 . 9 % NaCI
Water I p
0031-9384/92 $5.00 + .00 Copyright © 1992 PergamonPress Ltd.
Printed in the USA.
Effect of Hepatic Portal Infusion of Water on Water Intake by Water-Deprived Rats MOTOI
KOBASHI
AND AKIRA ADACHI l
Department of Physiology, Okayama University Dental School, Okayama, 700, Japan R e c e i v e d 22 M a y 1992 KOBASHI, M. AND A. ADACHI. Effect of hepatic portal infusion of water on water intake by water-deprivedrats. PHYSIOL BEHAV 52(5) 885-888, 1992.--To determine whether or not hepatoportal osmoreceptive (or sodium-receptive) signals participate in the control of drinking, we examined the effects of portal infusion of water, 0.9% saline, and 1.8% saline on water intake by water-deprived rats. Infusion was started 0.5 h prior to the end of the water deprivation period for 3.5 h at a rate of 52 ul/min through either a portal or a jugular catheter. After 24-h water deprivation, water intake was measured successively for 24 h without food. As a result of the water infusion tests, water intake of the portal infusion group was significantly less than that of the jugular infusion group during and after the infusion. Portal infusion of neither 0.9% nor 1.8% saline affected the water intake compared to similar infusion into the jugular vein. It is concluded that hypotonic stimulation of the hepatoportal osmoreceptor suppresses water intake in water-deprived rats. On the contrary, isotonic or hypertonic stimulation does not produce any change of water intake. Hepatic
Portal
Osmoreceptor
Drinking
Water intake
IT is well known that hepatoportal osmoreceptors (or sodium receptors) participate in body fluid homeostasis in regulating urine output (1,7). Possible participation of these osmoreceptors in the control of fluid ingestion has been studied by salt preference tests. Portal infusion of hypertonic saline suppressed saline intake but did not influence water intake in normal rats (11), water-deprived rats (4), or in sodium-deficit rats ( 15,16). Hepatic vagotomy attenuated saline intake by rats, but did not affect water intake (5). In addition, hepatic vagotomy did not affect water intake induced by intraperitoneal injection of hypertonic saline, when only water was supplied (5,14), nor in the case of prandial drinking (1,3). Thus, the hepatoportal osmoreceptors regulate saline intake, but not water intake, when water and saline are supplied simultaneously. However, hepatic vagotomy induced overingestion of water after water deprivation, but water intake induced by angiotensin or intraperitoneal injection of hypertonic saline was not affected by hepatic vagotomy (13). It seems that hepatoportal osmoreceptors regulate water intake induced by water deprivation alone. In the present experiments, to determine whether or not hepatoportal osmoreceptive signals participate in the control of water intake, we examined the effects of portal infusion of water, 0.9% saline, and 1.8% saline on water intake in water-deprived rats.
Infusion
Water deprivation
Rat
rats were housed in 25 x 25 X 40-cm Plexiglas cages, on a 12:12 light:dark cycle (light from 0600 to 1800), in a room at 2 3 - 2 4 ° C temperature. Rats were maintained in this cage on tap water, except when deprived, and normal chow until the end of the experiments.
Surgery All rats received surgery to catheterize the hepatic portal vein and the right jugular vein. No infusion control (sham infusion) group was prepared. Each catheter consisted of a Silastic tube (0.50 m m i.d., 1.00 m m o.d., Dow-Corning) connected to a polyethylene tube (PE-50, 0.58 m m i.d., 0.97 m m o.d., Intramedic) with a 23-gauge stainless steel tube. Rats were anesthetized with 50 mg/kg body weight sodium pentobarbital. The hepatic portal catheter was inserted into the ileocolic vein near the cecum through a midline abdominal incision. The catheter, a Silastic tube, was inserted toward the liver for about 4 cm, and fixed to the vessel with two sutures. The right jugular vein was exposed for catheterization. The end of a Silastic tube was inserted toward the atrium about 3 cm, and fixed with two sutures. The external ends of the portal and jugular catheters were tunneled under the skin and brought to the surface between the shoulder blades. After all incisions were closed, the catheters were passed through a 5-cm long rubber tube attached by dental cement to three screws in the skull. The catheters were protected by a 35-cm long stainless steel spiral spring and then joined to a dual channel swivel (375 D/22, Instech). After all surgery, the
METHOD
Subjects and Maintenance Eight-week-old male Sprague-Dawley rats (Charles River) weighing 255-330 g were used in all experiments. After surgery
Requests for reprints should be addressed to Dr. Akira Adachi, Department of Physiology, Okayama University Dental School, Shikata-cho, Okayama, 700, Japan. 885
886
KOBASHI A N D A D A C H I
without infusing the solution. Thus, rats were divided into seven groups: sham infusion (nine rats), portal infusion of water (10 rats), jugular infusion of water (nine rats), portal infusion of 0.9% saline (seven rats), jugular infusion of 0.9% saline (eight rats), portal infusion of 1.8% saline (nine rats), and jugular infusion of 1.8% saline (seven rats). Food intake from the start of the dark period until the start of infusion was measured daily for 3 days. Water intake was measured at all times after surgery. Before the test, daily food and water intakes were not significantly different among the seven groups. After all experiments, the animals were sacrificed by overdoses of anesthesia, and the integrity of the catheters and locations of their tips were confirmed. If any catheters appeared to leak, those data were omitted.
',OHT 17:00
I
I
18:00
21:00
17:,3o I I '. ~ i
I ! i i
I
I
I I
Infusion
(3.5 h)
~' vi
:,, Food Deprivation (25 h) - -
:~ ,~
Water Deprivation (24 h)
~;~
Measurement (24 h)---
I
FIG. 1. Schematic representation of experimental protocol.
rats were moved into the Plexiglas cages and maintained on normal chow and tap water until the termination of the tests.
Data Analysis Water intake was expressed in ml by multiplying the number of drops by 0.067. Cumulative intake was measured 0.5, 1, 2, 3, 6, 9, 12, and 24 h after presentation of tap water. Each 3-h intake in the dark period and 12-h intake in the light period are presented as bar graphs (Fig. 3). Data are presented as mean + SE. Results were analyzed using Student's t-tests.
Test Procedure All rats were tested in a water-deprived state. Eight or 9 days after surgery, rats were deprived of water from 1800 (at the beginning of the dark period). One h before the start of the next dark period (1700), food was removed. Infusion was started at 1730 for 3.5 fi at a rate of 52 ul/min. At 1800, tap water was supplied again. Water intake without food was measured for 24 h by a drop counter and recorded on a pen recorder for later analysis. Figure 1 shows the experimental protocol. Each rat received either portal or jugular infusion. Infusate was either deionized water, 0.9% saline, or 1.8% saline. Each rat was tested only one time. Sham infusion was tested by the same procedure
RESULTS
Portal and Jugular Infusion of Water Cumulative water intake during infusion of water through the portal or the jugular vein is presented in the left-hand graph of Fig. 2. Portal infusion of water suppressed water intake sig-
0.9% NaCI
Water
1.8% NaCI
50
~E 40
Sham Infusion (N=9) Portal Infusion (N=10) Jugular Infusion (N=9)
+ --i--
J r"
--o.=
Portal Infusion (N=7) Jugular Infusion (N=8)
+ +
Portal Infusion (N=9) Jugular Infusion (N=7)
I
N
a0
/
•~_~ 20
E 0 ~o
I .
•
%*
°.
:l
,.
,:
,.':
:*
12
~--,24 0
o*
0 0
~ 3
6
9
Time (Hours)
II ;
6
9
Time (Hours)
1)
I] 24
0
3
6
;
12
24
Time (Hours)
FIG. 2. Effects of the infusion of water (left), 0.9% saline (middle), and 1.8% saline (right) on water intake. Cumulative water intakes are shown. *p < 0.05, **p < 0.01, ***p < 0.001 when water intakes induced by portal infusion of water are compared with those of sham infusion; ,p < 0.05, --p
< 0.005, - - p < 0.00l when water intakes induced by portal infusion of water are compared with those of jugular infusion of water. Portal infusion of 0.9% and 1.8% saline did not produce any change compared to jugular infusion of the same solution.
EFFECT OF PORTAL INFUSION ON WATER INTAKE
887
nificantly after 24-h water deprivation. Compared to the jugular infusion of water, the cumulative water intake was reduced at 3 h after presentation of tap water (t = 2.682, p < 0.05), at 24 h (t = 3.593, p < 0.005), and at all times between. Water intake in each 3-h interval in the dark period and in the 12-h light period is shown in the left histogram of Fig. 3. Water intake was reduced in every period (t = 4.249, p < 0.001, 3-6 h; t = 2.421, p < 0.05, 6-9 h; t = 2.879, p < 0.05, 9-12 h; t = 2.969, p < 0.01, 12-24 h). As shown in the left-hand graph of Fig. 2, the cumulative water intake after presentation of tap water was significantly reduced by the portal infusion of water from 1 h (t = 2.463, p < 0.05) to 24 h (t = 3.138, p < 0.01), compared to sham infusion. Three-h intake from 0 h to 3 h (t = 4.176, p < 0.001), from 3 h to 6 h (t = 3.242, p < 0.005), and 12-h water intake in the light period (t = 2.435, p < 0.05) were significantly reduced, as shown in the left-hand graph of Fig. 3.
duced by 24-h water deprivation. Rats treated by portal infusion of water drank less than those treated by jugular infusion. This fact shows that the hepatoportal region is involved in the suppression of water intake. The portal water infusion group also drank less than the portal isotonic 0.9% saline infusion group (t = 6.193, p < 0.001, 0-3 h; t = 2.297, p < 0.05, 3-6 h; t = 0.021, NS, 6-9 h; t = 2.585, p < 0.05, 9-12 h; t = 0.669, NS, 12-24 h). Osmoreceptors (or sodium receptors) in the hepatoportal region participate in this suppression. Hepatoportal osmoreceptive afferents project to the median preoptic nucleus (9) and to the lateral hypothalamic area (8), which is closely related to water intake. The suppression appeared during the infusion and lasted for a relatively long time. The suppression during the infusion is easily understood to be a function of the neural system; the hepatic branch of the vagus nerve conveys osmotic information about the portal blood. However, it is difficult to explain the long-lasting suppression following the end of the infusion. In our experiments, the portal infusion of water or 1.8% saline changed the plasma osmolarity 1-2%, estimating the portal blood flow as 5-10 ml/min. Calculating from a previous report (2), when rats weighing 300 g received injections of 15 ml water into the stomach, the absorption rate was 660 #l/min during the first 15 min and 100/A/min in the subsequent 45 min. Therefore, the infusion rate used in this experiment (52 #l/min) was thought to be within the range of physiological fluctuation. Previous works showed that hypertonic stimulation of the hepatoportal osmoreceptors changed saline intake without affecting water intake in the normal state (11), and in waterdeprived (4) and sodium-deficit rats (l 5,16). On the other hand, we observed suppression of water intake by hypotonic stimulation of the hepatoportal osmoreceptors, and hypertonic stimulation produced no change in water intake. It is suggested that hypotonic stimulation of the hepatoportal osmoreceptors should suppress water intake, and their hypertonic stimulation should suppress saline intake via independent neuronal circuits in the central nervous system (CNS). That is, the hepatic branch of the vagus nerve that conveys hepatoportal osmoreceptive signals to the CNS includes two types of neurons: one is activated by increment of osmolarity in the portal vein; the other is activated by decrement of osmolarity (1). Thus, there are
Portal and Jugular Infusion of 0.9% Safine The portal infusion of 0.9% saline did not significantlychange water intake compared to jugular infusion of 0.9% saline and sham infusion (middle graphs of Figs. 2 and 3).
Portal Infusion and Jugular Infusion of 1.8% Saline As illustrated in the right-hand graph of Fig. 2, the portal infusion of 1.8% saline did not change the cumulative water intake compared to the jugular infusion of 1.8% saline. However, the right-hand graph of Fig. 3 shows that 0-3-h water intake was facilitated in the portal infusion group compared to the sham infusion group (t = 2.124, p < 0.05), but not the jugular infusion group (t = 1.777, NS). The 6-9-h water intake was suppressed in the portal infusion group (t = 2.293, p < 0.05) but not significantly in the jugular infusion group (t = 1.405, NS). The 12h water intake in the light period was suppressed in the portal (t = 2.159, p < 0.05) and the jugular (t = 2.597, p < 0.05) infusion groups, compared to the sham group.
DISCUSSION Our present results clearly showed that, during and after the infusion, portal infusion of water suppressed water intake in-
2o
0 . 9 % NaCI
Water I p
