Brain Research, 109 (1976) 541-552
541
© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
EVIDENCE FOR SAR1-ALAS-ANGIOTENSIN CROSSING THE BLOOD CEREBROSPINAL FLUID BARRIER TO ANTAGONIZE CENTRAL EFFECTS OF ANGIOTENSIN II
W. E. H O F F M A N AND M. I. PHILLIPS
Neurobehavior Laboratory, Department of Physiology and Biophysics, University of Iowa, Iowa City, Iowa 52242 (U.S.A.) (Accepted November 6th, 1975)
SUMMARY
Injections of angiotensin II into the cerebral ventricles of the rat produces both a drinking and a pressor response. We have measured both responses simultaneously in conscious animals. The effect of saralasin acetate (P113), a specific angiotensin II competitive antagonist, has been studied on these angiotensin II induced responses. The results show that: (1) P113 given intravenously (i.v.) in doses of 500 ng/min or 1800 ng/min has no observable effect on 50 ng angiotensin given intraventricularly (IVT). At 72 #g/min, however, there was a 55 ~ reduction in the drinking and pressor responses to 50 ng angiotensin II (IVT); (2) 500 ng of P113 given IVT abolished the effects of 50 ng angiotensin II also given IVT and (3) P113 given i.v. at all doses antagonized the pressor effects of angiotensin II (i.v.) responses. The data indicate that both the drinking and pressor responses to angiotensin II (IVT) injections are centrally mediated and show that when a high enough dose of P113 is given peripherally the central effects of angiotensin II can be reduced. This suggests that a fraction of the P113 injected i.v. may pass across the blood-CSF barrier. Since P113 has a similar structure to angiotensin II the results have implications for studies in which high peripheral doses of angiotensin II are used.
INTRODUCTION
Angiotensin II (All) circulating in the blood stream increases blood pressure by inducing arteriole vasoconstriction2. This peptide has also been reported to produce a hypertensive response mediated by the central nervous system (CNS)X,15,2L Intravenous (i.v.) infusions of AII elicit drinking behaviorxl. Intracranially applied All also produces drinking but at much shorter latencies and with doses 1000 times smaller than by the i.v. route6,10,zs,a2. The dipsogenic action of All is therefore apparently under CNS control.
542 Exactly how vascularly circulating All activates a drinking response is unclear. There is currently a question whether peripherally administered All induces drinking by crossing the blood-cerebrospinal fluid (CSF) barrier and stimulating periventricular receptor sites in the ventricles19,33, or if the vascularly circulating peptide stimulated central receptors outside the blood-brain barrier9,27, zg. Although radioactivity can be detected inside the ventricles after i.v. infusions of tritiated 17 or 14C-labeled AII3:~. this radioactivity is not due to an immunologically active peptide and is apparently the result of inactive fragments 12. One way for All to reach brain sites would be at the circumventricular organs which have little or no blood-brain barrier (BBB) 2°,22,24,a4. AII has been shown to fire cells in the subfornical organ (SFO) 19,23 and low dose injections of AII into the SFO lead to drinking 28. Lesions of the SFO, however, do not abolish drinking completely or permanently 4,16,z9 and therefore the SFO cannot be the exclusive All receptor site. The question here would be whether All acts on cells within the SFO or other circumventricular organs exclusively or if it may also cross the blood-CSF barrier to act on other periventricular sites. A way to approach this problem of what brain sites peripheral All can reach is to use saralasin acetate (sad-alaS-AII, Pl13, Norwich Pharmacal), which is a structural analog and a competitive antagonist of A I P L Intravenous infusions of P113 have been shown to inhibit the pressor and drinking responses produced by i.v. All infusions 5,8,21. lntraventricular (IVT) infusions of P113 are likewise capable of blocking pressor and drinking responses produced by central infusions of AIIS,a,a°,3L Little work has been done on the interaction of central and peripheral systems, however. The purpose of this study was to investigate central AII induced drinking and pressor mechanisms and to determine the ability of the blood CSF barrier to exclude the All analog, P1 ! 3. We present evidence that All IVT injections produce drinking and pressor responses which are both centrally mediated. Further, very large dose peripheral infusions of P113 are capable of inhibiting these responses, suggesting that a small fraction of the angiotensin analog crosses into the CSF to interact with periventricular All receptor sites. METHODS
Fifty-nine male Sprague-Dawley rats weighing 400-500 g were used as subjects. Three days before the start of the test, 31 subjects were implanted with one lateral ventricular cannula and 8 rats with bilateral ventricular cannulae. The guide cannulae (14 mm, 22 g tubing) were implanted to extend through the lateral ventricles (coordinates with respect to bregma: 0.5 mm posterior, 1.0 mm lateral, 6.0 mm deep from dura, fiat skull) for ventricular application of the peptide as in the method of Johnson and Epstein is. A 30 g stylus maintained the guide cannuta patent during recovery. On the morning of the test day each subject was catheterized with a femoral artery catheter for blood pressure recording and a femoral vein catheter for drug infusion. The rats were anesthetized with ether and PE 50 tubing catheters filled with
543 heparinized saline were inserted into the femoral vessels. The femoral artery catheter was tapered so that when it was fitted into the vessel the tip lay in the vicinity of the abdominal aorta. Both catheters were led subcutaneously through a small incision in the back. They were held in place by dental acrylic cement attached to a wound clip on the back. The animals were allowed 2 h to recover. During testing, animals were unanesthetized and completely unrestrained. During testing all subjects were in excellent condition, able to move freely around their cages and drink when given All IVT.
Procedure To study the effects of P113 on All responses the following 3 experiments were performed: (1) PI13 (i.v.) on All (IVT) pressor and drinking responses; (2) Pl13 (i.v.) on the pressor response of A l l (i.v.) infusions and (3) P113 (IVT) infusions on the drinking and pressor effects of AII (IVT) injections. All procedures with the doses involved are shown with the results in Table I. For intraventricular injections, the stylus in the guide cannula was removed and a 30-g injection cannula inserted. The injection cannula sat flush with the end of the guide cannula and was connected by PE 10 tubing to a 50-#1 Hamilton syringe mounted in a microburette outside the home cage. For blood pressure recording the arterial catheter was connected to an extended piece of PE 50 tubing which led outside the cage to a Statham P23Gb blood pressure transducer. Mean blood pressure, pulse pressure, and heart rate were recorded on a Beckman R411 dynograph recorder. For data analysis results are expressed as mean and standard error and comparisons were made with paired t-tests.
(1) Effect of P l l 3 i.v. on AH I V T responses Three doses of i.v. P113 were used in these tests. The first group of 5 animals (group A) received 500 ng/min Pl13. In the second group of 7 animals (group B) 1800 ng/min P113 was administered. In group C (n -----9) 72,000 ng/min P113 was the dosage. In all groups 50 ng AII in a 5-/zl CSF vehicle given over 15 sec was the IVT test injection. The first lateral ventricular injection to be given was 5 /~1 of artificial CSF (Elliot's B Solution, Travenol Laboratories) infused over 15 sec. One hour later the response to an IVT injection of 50 ng AII in 5/~1 artificial CSF was recorded. After a further interval of 60 min, an i.v. infusion of isotonic saline at a rate of 18/zl/min was begun. Seven minutes later the rat received a 50-ng AII microinjection delivered to the lateral ventricles over 15 sec. Seven minutes after this AII microinjection, the saline infusion was terminated. One hour after the end of the preceding test, an i.v. infusion of either 500 ng/ min, 1800 ng/min or 72,000 ng/min of Pl13 given in a saline vehicle, was started. Seven minutes after the start of this infusion the animal received an injection of 50 ng/5 #1 AII applied to the lateral ventricles. Seven minutes after the AII microinjection the 500 ng/min and 1800 ng/min Pl13 infusions were terminated by turning off the infusion pump. The 72,000 ng/min P113 infusion was terminated 1 min after the AII
544 (IVT) injection. The drinking and pressor responses in all tests were recorded for 15 rain. This time period was sufficiently long to allow the response to appear and dissipate. To monitor the effect of the i.v. saline infusions on central All responses, 10 animals were prepared with one lateral ventricle cannula and a femoral artery catheter as described above. After a control (5 #1 CSF) injection each subject received two 50 ng/5 #1 All (IVT) injections. These tests were separated by 1 h.
(2) Effect of P l l 3 i.v. on A H i.v. pressor response The purpose of this test was to show the effectiveness of peripheral P l I 3 on the pressor response to peripheral AII. The drinking response to All by this route was not recorded. Fourteen animals were each implanted with a femoral artery catheter and two femoral vein catheters. These rats then received an i.v. dose of 180 ng/min of AI1 in an 18 #l/min saline vehicle for 5 min. The blood pressure response was recorded. Following a 1-h recovery period an i.v. infusion of either 500 ng/min (n = 5), 1800 ng/min (n = 7) or 72,000 ng/min of Pt 13 in two rats was started via the other i.v. catheter. Seven minutes after the beginning of the PI13 infusion, an i.v. infusion of 180 ng/min of AII was simultaneously infused through the first i.v. catheter. Concurrent infusions of P113 and A I I were continued for 5 min and then terminated, Five of the animals which had been tested for the effect of i.v. PI13 on AII (IVT) were also used in this test. The effect of subsequent P113 infusions on the i.v. A | I pressor responses were the same on both naive and pre-tested rats. Two hours following completion of the AII (IVT) testing procedure previously described, these 5 animals were tested for peripheral AII pressor responses. (3) Central P113 and A l l infusions on pressor and drinking responses The ability of central infusions of P113 to block the pressor and drinking responses induced by AII (IVT) injections was tested in 8 animals. These animals had bilateral lateral ventricular cannulae implanted. One hour after control (5/A CSF) IVT injections the dipsogenic and pressor response to 50 ng of AII was tested in one lateral ventricular cannula. One hour later an infusion of Pl13 (100 ng//~l/min) was started in the other ventricular cannula. Five minutes later, after a dose of 500 ng of P l l 3 had been delivered, this infusion was stopped and 50 ng/5 #1 AII was injected into the previous AII test site. The effect of the peptide injection on blood pressure and water intake was recorded for 15 rain. In order to interpret the effect of volume of IVT infusions in this experiment 6 control animals were run. The first AII test was a 50 ng/5/,1 IVT injection as above. One hour later, 5/,l of artificial CSF was infused over a 5-min period into the second bilateral ventricular cannula. At the end of the 5-min infusion 50 ng AII was again infused into the first ventricular cannula to test for drinking and pressor responses. RESULTS
(1) Effect of P l l 3 (i.v.) on A H (IVT) pressor and drinking responses The control level for pressor and drinking responses to AII (IVT) in 21 rats was
8 5- 1
21 5- 2
18 5- 2
17 5- 1
17 5- 1
16 ± 2
* All i.v. infusions were at a rate o f 18/~l/min.
AII alone 24 5- 2 + i.v. saline* 14 5- 3 + 500 n g / m i n i.v. PI13 21 ± 1 +1800 ng/ m i n i.v. Pl13 5- 7 2 / ~ g / m i n i.v. P l 1 3 7.1 5- 0.8
3.6 5- 1.0
6.0 q- 0.9
Group A n=5
Group C n--9
Group A n=5
Group B n= 7
Drinking (ml)
Blood pressure increase (mm Hg)
Angiotensin II, 50 ng/5 t~l I V T
A I I INDUCED PRESSOR AND DRINKING RESPONSES
TABLE I
3.0 ± 0.6
4.0 5- 1.0
1.2 5- 0.5
3.0 5- 0.6
5.2 4- 1.4
5.3 5- 1.0
Group C n=9
Group B n = 7
105-2
46±5
53 4 - 3
255-1
Group F n= 7
Group E n=5
2±1
65 5 - 1 0
Group G n--2
Blood pressure increase (mm Hg)
Angiotensin 11, 180 ng/min i.v.*
546
120 "~ O
100
i u
80
~
60
z
541
© Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
EVIDENCE FOR SAR1-ALAS-ANGIOTENSIN CROSSING THE BLOOD CEREBROSPINAL FLUID BARRIER TO ANTAGONIZE CENTRAL EFFECTS OF ANGIOTENSIN II
W. E. H O F F M A N AND M. I. PHILLIPS
Neurobehavior Laboratory, Department of Physiology and Biophysics, University of Iowa, Iowa City, Iowa 52242 (U.S.A.) (Accepted November 6th, 1975)
SUMMARY
Injections of angiotensin II into the cerebral ventricles of the rat produces both a drinking and a pressor response. We have measured both responses simultaneously in conscious animals. The effect of saralasin acetate (P113), a specific angiotensin II competitive antagonist, has been studied on these angiotensin II induced responses. The results show that: (1) P113 given intravenously (i.v.) in doses of 500 ng/min or 1800 ng/min has no observable effect on 50 ng angiotensin given intraventricularly (IVT). At 72 #g/min, however, there was a 55 ~ reduction in the drinking and pressor responses to 50 ng angiotensin II (IVT); (2) 500 ng of P113 given IVT abolished the effects of 50 ng angiotensin II also given IVT and (3) P113 given i.v. at all doses antagonized the pressor effects of angiotensin II (i.v.) responses. The data indicate that both the drinking and pressor responses to angiotensin II (IVT) injections are centrally mediated and show that when a high enough dose of P113 is given peripherally the central effects of angiotensin II can be reduced. This suggests that a fraction of the P113 injected i.v. may pass across the blood-CSF barrier. Since P113 has a similar structure to angiotensin II the results have implications for studies in which high peripheral doses of angiotensin II are used.
INTRODUCTION
Angiotensin II (All) circulating in the blood stream increases blood pressure by inducing arteriole vasoconstriction2. This peptide has also been reported to produce a hypertensive response mediated by the central nervous system (CNS)X,15,2L Intravenous (i.v.) infusions of AII elicit drinking behaviorxl. Intracranially applied All also produces drinking but at much shorter latencies and with doses 1000 times smaller than by the i.v. route6,10,zs,a2. The dipsogenic action of All is therefore apparently under CNS control.
542 Exactly how vascularly circulating All activates a drinking response is unclear. There is currently a question whether peripherally administered All induces drinking by crossing the blood-cerebrospinal fluid (CSF) barrier and stimulating periventricular receptor sites in the ventricles19,33, or if the vascularly circulating peptide stimulated central receptors outside the blood-brain barrier9,27, zg. Although radioactivity can be detected inside the ventricles after i.v. infusions of tritiated 17 or 14C-labeled AII3:~. this radioactivity is not due to an immunologically active peptide and is apparently the result of inactive fragments 12. One way for All to reach brain sites would be at the circumventricular organs which have little or no blood-brain barrier (BBB) 2°,22,24,a4. AII has been shown to fire cells in the subfornical organ (SFO) 19,23 and low dose injections of AII into the SFO lead to drinking 28. Lesions of the SFO, however, do not abolish drinking completely or permanently 4,16,z9 and therefore the SFO cannot be the exclusive All receptor site. The question here would be whether All acts on cells within the SFO or other circumventricular organs exclusively or if it may also cross the blood-CSF barrier to act on other periventricular sites. A way to approach this problem of what brain sites peripheral All can reach is to use saralasin acetate (sad-alaS-AII, Pl13, Norwich Pharmacal), which is a structural analog and a competitive antagonist of A I P L Intravenous infusions of P113 have been shown to inhibit the pressor and drinking responses produced by i.v. All infusions 5,8,21. lntraventricular (IVT) infusions of P113 are likewise capable of blocking pressor and drinking responses produced by central infusions of AIIS,a,a°,3L Little work has been done on the interaction of central and peripheral systems, however. The purpose of this study was to investigate central AII induced drinking and pressor mechanisms and to determine the ability of the blood CSF barrier to exclude the All analog, P1 ! 3. We present evidence that All IVT injections produce drinking and pressor responses which are both centrally mediated. Further, very large dose peripheral infusions of P113 are capable of inhibiting these responses, suggesting that a small fraction of the angiotensin analog crosses into the CSF to interact with periventricular All receptor sites. METHODS
Fifty-nine male Sprague-Dawley rats weighing 400-500 g were used as subjects. Three days before the start of the test, 31 subjects were implanted with one lateral ventricular cannula and 8 rats with bilateral ventricular cannulae. The guide cannulae (14 mm, 22 g tubing) were implanted to extend through the lateral ventricles (coordinates with respect to bregma: 0.5 mm posterior, 1.0 mm lateral, 6.0 mm deep from dura, fiat skull) for ventricular application of the peptide as in the method of Johnson and Epstein is. A 30 g stylus maintained the guide cannuta patent during recovery. On the morning of the test day each subject was catheterized with a femoral artery catheter for blood pressure recording and a femoral vein catheter for drug infusion. The rats were anesthetized with ether and PE 50 tubing catheters filled with
543 heparinized saline were inserted into the femoral vessels. The femoral artery catheter was tapered so that when it was fitted into the vessel the tip lay in the vicinity of the abdominal aorta. Both catheters were led subcutaneously through a small incision in the back. They were held in place by dental acrylic cement attached to a wound clip on the back. The animals were allowed 2 h to recover. During testing, animals were unanesthetized and completely unrestrained. During testing all subjects were in excellent condition, able to move freely around their cages and drink when given All IVT.
Procedure To study the effects of P113 on All responses the following 3 experiments were performed: (1) PI13 (i.v.) on All (IVT) pressor and drinking responses; (2) Pl13 (i.v.) on the pressor response of A l l (i.v.) infusions and (3) P113 (IVT) infusions on the drinking and pressor effects of AII (IVT) injections. All procedures with the doses involved are shown with the results in Table I. For intraventricular injections, the stylus in the guide cannula was removed and a 30-g injection cannula inserted. The injection cannula sat flush with the end of the guide cannula and was connected by PE 10 tubing to a 50-#1 Hamilton syringe mounted in a microburette outside the home cage. For blood pressure recording the arterial catheter was connected to an extended piece of PE 50 tubing which led outside the cage to a Statham P23Gb blood pressure transducer. Mean blood pressure, pulse pressure, and heart rate were recorded on a Beckman R411 dynograph recorder. For data analysis results are expressed as mean and standard error and comparisons were made with paired t-tests.
(1) Effect of P l l 3 i.v. on AH I V T responses Three doses of i.v. P113 were used in these tests. The first group of 5 animals (group A) received 500 ng/min Pl13. In the second group of 7 animals (group B) 1800 ng/min P113 was administered. In group C (n -----9) 72,000 ng/min P113 was the dosage. In all groups 50 ng AII in a 5-/zl CSF vehicle given over 15 sec was the IVT test injection. The first lateral ventricular injection to be given was 5 /~1 of artificial CSF (Elliot's B Solution, Travenol Laboratories) infused over 15 sec. One hour later the response to an IVT injection of 50 ng AII in 5/~1 artificial CSF was recorded. After a further interval of 60 min, an i.v. infusion of isotonic saline at a rate of 18/zl/min was begun. Seven minutes later the rat received a 50-ng AII microinjection delivered to the lateral ventricles over 15 sec. Seven minutes after this AII microinjection, the saline infusion was terminated. One hour after the end of the preceding test, an i.v. infusion of either 500 ng/ min, 1800 ng/min or 72,000 ng/min of Pl13 given in a saline vehicle, was started. Seven minutes after the start of this infusion the animal received an injection of 50 ng/5 #1 AII applied to the lateral ventricles. Seven minutes after the AII microinjection the 500 ng/min and 1800 ng/min Pl13 infusions were terminated by turning off the infusion pump. The 72,000 ng/min P113 infusion was terminated 1 min after the AII
544 (IVT) injection. The drinking and pressor responses in all tests were recorded for 15 rain. This time period was sufficiently long to allow the response to appear and dissipate. To monitor the effect of the i.v. saline infusions on central All responses, 10 animals were prepared with one lateral ventricle cannula and a femoral artery catheter as described above. After a control (5 #1 CSF) injection each subject received two 50 ng/5 #1 All (IVT) injections. These tests were separated by 1 h.
(2) Effect of P l l 3 i.v. on A H i.v. pressor response The purpose of this test was to show the effectiveness of peripheral P l I 3 on the pressor response to peripheral AII. The drinking response to All by this route was not recorded. Fourteen animals were each implanted with a femoral artery catheter and two femoral vein catheters. These rats then received an i.v. dose of 180 ng/min of AI1 in an 18 #l/min saline vehicle for 5 min. The blood pressure response was recorded. Following a 1-h recovery period an i.v. infusion of either 500 ng/min (n = 5), 1800 ng/min (n = 7) or 72,000 ng/min of Pt 13 in two rats was started via the other i.v. catheter. Seven minutes after the beginning of the PI13 infusion, an i.v. infusion of 180 ng/min of AII was simultaneously infused through the first i.v. catheter. Concurrent infusions of P113 and A I I were continued for 5 min and then terminated, Five of the animals which had been tested for the effect of i.v. PI13 on AII (IVT) were also used in this test. The effect of subsequent P113 infusions on the i.v. A | I pressor responses were the same on both naive and pre-tested rats. Two hours following completion of the AII (IVT) testing procedure previously described, these 5 animals were tested for peripheral AII pressor responses. (3) Central P113 and A l l infusions on pressor and drinking responses The ability of central infusions of P113 to block the pressor and drinking responses induced by AII (IVT) injections was tested in 8 animals. These animals had bilateral lateral ventricular cannulae implanted. One hour after control (5/A CSF) IVT injections the dipsogenic and pressor response to 50 ng of AII was tested in one lateral ventricular cannula. One hour later an infusion of Pl13 (100 ng//~l/min) was started in the other ventricular cannula. Five minutes later, after a dose of 500 ng of P l l 3 had been delivered, this infusion was stopped and 50 ng/5 #1 AII was injected into the previous AII test site. The effect of the peptide injection on blood pressure and water intake was recorded for 15 rain. In order to interpret the effect of volume of IVT infusions in this experiment 6 control animals were run. The first AII test was a 50 ng/5/,1 IVT injection as above. One hour later, 5/,l of artificial CSF was infused over a 5-min period into the second bilateral ventricular cannula. At the end of the 5-min infusion 50 ng AII was again infused into the first ventricular cannula to test for drinking and pressor responses. RESULTS
(1) Effect of P l l 3 (i.v.) on A H (IVT) pressor and drinking responses The control level for pressor and drinking responses to AII (IVT) in 21 rats was
8 5- 1
21 5- 2
18 5- 2
17 5- 1
17 5- 1
16 ± 2
* All i.v. infusions were at a rate o f 18/~l/min.
AII alone 24 5- 2 + i.v. saline* 14 5- 3 + 500 n g / m i n i.v. PI13 21 ± 1 +1800 ng/ m i n i.v. Pl13 5- 7 2 / ~ g / m i n i.v. P l 1 3 7.1 5- 0.8
3.6 5- 1.0
6.0 q- 0.9
Group A n=5
Group C n--9
Group A n=5
Group B n= 7
Drinking (ml)
Blood pressure increase (mm Hg)
Angiotensin II, 50 ng/5 t~l I V T
A I I INDUCED PRESSOR AND DRINKING RESPONSES
TABLE I
3.0 ± 0.6
4.0 5- 1.0
1.2 5- 0.5
3.0 5- 0.6
5.2 4- 1.4
5.3 5- 1.0
Group C n=9
Group B n = 7
105-2
46±5
53 4 - 3
255-1
Group F n= 7
Group E n=5
2±1
65 5 - 1 0
Group G n--2
Blood pressure increase (mm Hg)
Angiotensin 11, 180 ng/min i.v.*
546
120 "~ O
100
i u
80
~
60
z
