Neuroseience Letters, [43 (1992) 140 15t~ ~ 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92!$ 05.00
146
NSL 08878
Cyclic AMP alteration of chloride transport into the choroid plexus-cerebrospinal fluid system Q.-S.
Deng ~ and C.E. Johanson
Program in Neurosurgery, Department of Clinical Neurosciences, Brown University~Rhode Island Hospital, Providence, R102903 (USA) (Received 17 March 1992; Revised version received 11 May 1992; Accepted 20 May 1992)
Key words: cAMP; Theophylline; Forskolin; Ouabain; Acetazolamide; BIood-CSF barrier The permeation of 36C1from blood into choroid plexus (CP) and cerebrospinal fluid (CSF) was studied in adult rats intraventricularly injected with either dibutyryl (db)-cAMP, theophylline with db-cAMP, or forskolin. All 3 treatments significantly enhanced the 15-min distribution (lid) of 36C1 into CSF, by an average of 17-33%. In contrast, ouabain and acetazolamide (inhibitors of CSF production) reduced 36C1uptake into CSF by 34% and 13%, respectively. A correlative analysis of the various treatments revealed that the volume of distribution of 36C1 in CSF varied inversely with that in the choroid plexus. This inverse relationship suggests drug effects on CI movement across the apical (CSF-facing) membrane of CP.
The role of cAMP as a second messenger to mediate hormonal action has been demonstrated in many secretory tissues. For mammalian choroid plexus (CP), however, there is no unitary model characterizing second messenger role in modulating cerebrospinal fluid (CSF) secretion. Furthermore, there is no consensus on how augmented levels of cAMP in CP or ventricular fluid affect CSF production [8, 17, 20]. This study was prompted by the observation that PGE2, which causes a build-up of cAMPiin CP [8], also stimulates C1 transport from blood to CSF [4]. CSF formation rate is directly proportional to net transchoroidal movement of C1 and Na [3, 13, l 8] from blood into ventricles. C1 transport and distribution in mammalian and amphibian CPs have been extensively analyzed[l, 5, 15, 16, 23, 27, 30-32, 34]. Investigations of in vitro bullfrog CP by Saito and Wright [25-27] furnished evidence for cAMP-activated conductance for C1 and HCO3 in the apical membrane. We postulated that cAMP administered to the CSF side of CP epithelium would enhance CI movement across rat blood-CSF barrier. Evidence is presented for a facilitating effect of intraventricularly administered cAMP on C1 transport across CP. ~Present address: Laboratory of Neuroscience, National Institute on Aging, Bethesda, MD 20892, USA. Correspondence:C.E. Johanson, Department of Clinical Neurosciences, Rhode Island Hospital, 593 Eddy Street, Providence, RI 02903, USA. Fax: (1) (401) 444-8260.
All experiments were done with male Sprague-Dawley rats (120-200 g) from Holtzman (Madison, WI). Animals were acclimatized in a temperature- and humiditycontrolled vivarium for at least 4 days after their arrival, and were consistently used in experimentation between 08.00 h and 15.00 h. Ether anesthesia was utilized for 4-5 min while both renal pedicles were ligated in order to maintain a stable level of isotope in plasma. Prior to stereotaxic placements, ketamine HC1 (80 mg/kg) was given intraperitoneally. The skull was exposed with a midline incision, and a hole was drilled 5.3 mm posterior to the bregma and 4 mm lateral (left) to the sagittal suture. For the intracerebroventricular (i.c.v.) administration of test substances, a 27-gauge needle was introduced through the burr hole to a depth 2 mm from the dural surface. A polyethylene catheter, filled with the solution to be administered, was attached to the needle. A 100-pl syringe (Hamilton) was used for the injection, at a rate of 5 pl/min by means of an infusion pump (Harvard Apparatus, Model 1100). The injection volume was 25/~1. No untoward effects on respiration or femoral arterial pressure were observed when the injected artificial CSF (aCSF) solutions, with or without cAMP, were delivered into the lateral ventricle. Evans blue dye was infused after some experiments to verify that the injectate was reaching the choroid plexuses. Forty-five minutes before the end of the experiment, and 30 min before the i.p. injection of 36C1,each drugwas administered i.c.v, at a concentration of 1 mM in the injectate. Drug concentration in CSF, after allowing for
147
30 min o f post-injection distribution in the ventricular system, was estimated to be between 0.1 and 0.01 m M ; this was deduced f r o m the ~ 10-fold dilution o f ~4C-inulin, injected intraventricularly in separate animals. D r u g s were generally dissolved in aCSF, except for forskolin which was solubilized in a 1.6% solution o f ethanol in aCSF. Ouabain, acetazolamide, dibutyryl c A M E theophylline and forskolin were obtained f r o m Sigma Chemical. The a p p r o a c h was to evaluate effects o f test substances (dibutyryl ( d b ) - c A M P and agents p r o m o t i n g c A M P accumulation) and to c o m p a r e these induced alterations with those b r o u g h t a b o u t by standard inhibitors o f C S F formation, i.e.. ouabain and acetazolamide. Each animal received 0.1/aCi o f 36C1 per gram b o d y weight. After each experiment, samples were taken f r o m blood (abdominal aorta), C S F (cisterna magna), and choroid plexus (lateral ventricle). Each C P was analyzed for water content [12]. All tissue and C S F samples were assayed for 36C1 activity by liquid scintillation in a Beckman LS 7500 counter. Rate o f CI transport from blood into both C P and C S F was assessed by volume o f distribution (V d) o f 36C1 [15, 31, 32]. Vd is equal to 100x dpm/g C P (or CSF) d p m / g plasma H~O. Various factors were considered in selecting the 15-min sample time to estimate rate o f permeation o f CI into the C S F system [5]. Fifteen minutes is long enough to permit significant accumulation o f C1 in C P p a r e n c h y m a , and yet short e n o u g h to allow evaluation o f the fast c o m p o n e n t o f C S F uptake (which is generally accepted to reflect transport across C P [13]). The arterial integral o f plasma [36C1] vs. time was not significantly different in controls vs. treated animals. The mean volume o f distribution of 36Cl in C S F o f controls at 15 min was 30% (Fig. 1, top). O u a b a i n and acetazolamide caused inhibition o f 36C1 transport into CSF, whereas d b - c A M P (and agents p r o m o t i n g c A M P production) enhanced m o v e m e n t o f 36C1 into C S F from blood. D b - c A M P increased 36C1 permeation by 32%. Smaller increases were observed when combining theophylline with db-cAMP, and when giving forskolin alone. 36C1 distribution in C P varied considerably a m o n g the 6 treatments (Table I and Fig. 1, bottom). Values for Vd o f 36C1, calculated on wet tissue weight, are presented in Table 1. Because the water content o f the tissue (and thus its wet weight) affects the calculated value for 36C1 Vd, it is essential to consider changes in tissue water content when there are large alterations in the latter. M a r k e d retention o f water by tissue was f o u n d after ouabain, whereas the greatest loss o f water was observed in the t h e o p h y l l i n e / d b - c A M P treatment (Table I). Due to such experimental variations in tissue water
(D iiI
Ii
03 U cI
50 40 30 20 10
IID
0
.-.
200
i
180
0
A
C
D
T
F
0
A
C
D
T
F
v
160 r-
•-
140 120
>
100
Fig. 1. Top: CSF uptake of a('C1from blood of adult rats treated with drugs injected intraventricularly. At the time of 3~'C1administration, the drug concentration in ventricular CSF was 0,1 0.01 mM; see text. O, ouabain; A, acetazolamide; C, control (artificial CSF vehicle); D, dibutyryl cAMP: T, Theophylline plus db-cAMP: F, forskolin. There was no significant difference in CI uptake between controls receiving aCSF and those vehicle controls getting 1.6% ethanol in aCSF (tbr forskolin dissolution) therefore, all of the control values were combined to yield the mean presented as the unfilled bar. Each bar is the mean _+S.E.M. for 4 6 rats. 1~. volume of distribution (%) of ~'C1in CSF 15 min alter injection of tracer. *P < 0.05, treatment vs. control, by multiple range test. Bottom: volume of distribution (15 min) of ~C1 in lateral ventricle CP of rats subjected to various treatments. The net accumulation of 3~C1by CP is a function of basolateral (blood-side) uptake and apical (CSF-side) efflux. Due to the experimental differences in tissue water content induced by the various drugs (Table 1). it was advantageous to express the ~CI uptake on a dry weight basis. Bars arc means +_S.E.M. for 4 6 animals. The abcissa letter symbols for drug treatment, i.e, compositions of the i.c.v, injectate, are the same as those described in the top part of the tigure. Treatments with ouabain and acetazolamide are significantly different from control, and from treatments with dbcAMP (with or without theophylline) and forskolin. *P < 0.05 by multipie range test.
content, we have expressed the 3('C1 uptake by CP on a dry weight basis (Fig. 1, bottom). This effectively eliminates water content as a variable, and thus provides a clearer picture o f the a m o u n t of CI per unit weight o f C R O u a b a i n and acetazolamide b r o u g h t about marked accumulation o f 3{'C1 in CP: in comparison, c A M P and its enhancers led to significantly less Cl remaining in tissue (Fig. 1, bottom). The correlation between ~¢'C1in C S F and that in C P is
148 LD _..1
y-
r
45 T
v
2
= 0.82
D
0
Ii
O3 t.)
35
¢-
,oU
25
15 ~>
1 0
Ell
i
i
i
i
i
130
140
150
160
170
2
S >-
i
120
r
= 0.81
0
5.0
tD n" W
4.0 n
I-
146
NSL 08878
Cyclic AMP alteration of chloride transport into the choroid plexus-cerebrospinal fluid system Q.-S.
Deng ~ and C.E. Johanson
Program in Neurosurgery, Department of Clinical Neurosciences, Brown University~Rhode Island Hospital, Providence, R102903 (USA) (Received 17 March 1992; Revised version received 11 May 1992; Accepted 20 May 1992)
Key words: cAMP; Theophylline; Forskolin; Ouabain; Acetazolamide; BIood-CSF barrier The permeation of 36C1from blood into choroid plexus (CP) and cerebrospinal fluid (CSF) was studied in adult rats intraventricularly injected with either dibutyryl (db)-cAMP, theophylline with db-cAMP, or forskolin. All 3 treatments significantly enhanced the 15-min distribution (lid) of 36C1 into CSF, by an average of 17-33%. In contrast, ouabain and acetazolamide (inhibitors of CSF production) reduced 36C1uptake into CSF by 34% and 13%, respectively. A correlative analysis of the various treatments revealed that the volume of distribution of 36C1 in CSF varied inversely with that in the choroid plexus. This inverse relationship suggests drug effects on CI movement across the apical (CSF-facing) membrane of CP.
The role of cAMP as a second messenger to mediate hormonal action has been demonstrated in many secretory tissues. For mammalian choroid plexus (CP), however, there is no unitary model characterizing second messenger role in modulating cerebrospinal fluid (CSF) secretion. Furthermore, there is no consensus on how augmented levels of cAMP in CP or ventricular fluid affect CSF production [8, 17, 20]. This study was prompted by the observation that PGE2, which causes a build-up of cAMPiin CP [8], also stimulates C1 transport from blood to CSF [4]. CSF formation rate is directly proportional to net transchoroidal movement of C1 and Na [3, 13, l 8] from blood into ventricles. C1 transport and distribution in mammalian and amphibian CPs have been extensively analyzed[l, 5, 15, 16, 23, 27, 30-32, 34]. Investigations of in vitro bullfrog CP by Saito and Wright [25-27] furnished evidence for cAMP-activated conductance for C1 and HCO3 in the apical membrane. We postulated that cAMP administered to the CSF side of CP epithelium would enhance CI movement across rat blood-CSF barrier. Evidence is presented for a facilitating effect of intraventricularly administered cAMP on C1 transport across CP. ~Present address: Laboratory of Neuroscience, National Institute on Aging, Bethesda, MD 20892, USA. Correspondence:C.E. Johanson, Department of Clinical Neurosciences, Rhode Island Hospital, 593 Eddy Street, Providence, RI 02903, USA. Fax: (1) (401) 444-8260.
All experiments were done with male Sprague-Dawley rats (120-200 g) from Holtzman (Madison, WI). Animals were acclimatized in a temperature- and humiditycontrolled vivarium for at least 4 days after their arrival, and were consistently used in experimentation between 08.00 h and 15.00 h. Ether anesthesia was utilized for 4-5 min while both renal pedicles were ligated in order to maintain a stable level of isotope in plasma. Prior to stereotaxic placements, ketamine HC1 (80 mg/kg) was given intraperitoneally. The skull was exposed with a midline incision, and a hole was drilled 5.3 mm posterior to the bregma and 4 mm lateral (left) to the sagittal suture. For the intracerebroventricular (i.c.v.) administration of test substances, a 27-gauge needle was introduced through the burr hole to a depth 2 mm from the dural surface. A polyethylene catheter, filled with the solution to be administered, was attached to the needle. A 100-pl syringe (Hamilton) was used for the injection, at a rate of 5 pl/min by means of an infusion pump (Harvard Apparatus, Model 1100). The injection volume was 25/~1. No untoward effects on respiration or femoral arterial pressure were observed when the injected artificial CSF (aCSF) solutions, with or without cAMP, were delivered into the lateral ventricle. Evans blue dye was infused after some experiments to verify that the injectate was reaching the choroid plexuses. Forty-five minutes before the end of the experiment, and 30 min before the i.p. injection of 36C1,each drugwas administered i.c.v, at a concentration of 1 mM in the injectate. Drug concentration in CSF, after allowing for
147
30 min o f post-injection distribution in the ventricular system, was estimated to be between 0.1 and 0.01 m M ; this was deduced f r o m the ~ 10-fold dilution o f ~4C-inulin, injected intraventricularly in separate animals. D r u g s were generally dissolved in aCSF, except for forskolin which was solubilized in a 1.6% solution o f ethanol in aCSF. Ouabain, acetazolamide, dibutyryl c A M E theophylline and forskolin were obtained f r o m Sigma Chemical. The a p p r o a c h was to evaluate effects o f test substances (dibutyryl ( d b ) - c A M P and agents p r o m o t i n g c A M P accumulation) and to c o m p a r e these induced alterations with those b r o u g h t a b o u t by standard inhibitors o f C S F formation, i.e.. ouabain and acetazolamide. Each animal received 0.1/aCi o f 36C1 per gram b o d y weight. After each experiment, samples were taken f r o m blood (abdominal aorta), C S F (cisterna magna), and choroid plexus (lateral ventricle). Each C P was analyzed for water content [12]. All tissue and C S F samples were assayed for 36C1 activity by liquid scintillation in a Beckman LS 7500 counter. Rate o f CI transport from blood into both C P and C S F was assessed by volume o f distribution (V d) o f 36C1 [15, 31, 32]. Vd is equal to 100x dpm/g C P (or CSF) d p m / g plasma H~O. Various factors were considered in selecting the 15-min sample time to estimate rate o f permeation o f CI into the C S F system [5]. Fifteen minutes is long enough to permit significant accumulation o f C1 in C P p a r e n c h y m a , and yet short e n o u g h to allow evaluation o f the fast c o m p o n e n t o f C S F uptake (which is generally accepted to reflect transport across C P [13]). The arterial integral o f plasma [36C1] vs. time was not significantly different in controls vs. treated animals. The mean volume o f distribution of 36Cl in C S F o f controls at 15 min was 30% (Fig. 1, top). O u a b a i n and acetazolamide caused inhibition o f 36C1 transport into CSF, whereas d b - c A M P (and agents p r o m o t i n g c A M P production) enhanced m o v e m e n t o f 36C1 into C S F from blood. D b - c A M P increased 36C1 permeation by 32%. Smaller increases were observed when combining theophylline with db-cAMP, and when giving forskolin alone. 36C1 distribution in C P varied considerably a m o n g the 6 treatments (Table I and Fig. 1, bottom). Values for Vd o f 36C1, calculated on wet tissue weight, are presented in Table 1. Because the water content o f the tissue (and thus its wet weight) affects the calculated value for 36C1 Vd, it is essential to consider changes in tissue water content when there are large alterations in the latter. M a r k e d retention o f water by tissue was f o u n d after ouabain, whereas the greatest loss o f water was observed in the t h e o p h y l l i n e / d b - c A M P treatment (Table I). Due to such experimental variations in tissue water
(D iiI
Ii
03 U cI
50 40 30 20 10
IID
0
.-.
200
i
180
0
A
C
D
T
F
0
A
C
D
T
F
v
160 r-
•-
140 120
>
100
Fig. 1. Top: CSF uptake of a('C1from blood of adult rats treated with drugs injected intraventricularly. At the time of 3~'C1administration, the drug concentration in ventricular CSF was 0,1 0.01 mM; see text. O, ouabain; A, acetazolamide; C, control (artificial CSF vehicle); D, dibutyryl cAMP: T, Theophylline plus db-cAMP: F, forskolin. There was no significant difference in CI uptake between controls receiving aCSF and those vehicle controls getting 1.6% ethanol in aCSF (tbr forskolin dissolution) therefore, all of the control values were combined to yield the mean presented as the unfilled bar. Each bar is the mean _+S.E.M. for 4 6 rats. 1~. volume of distribution (%) of ~'C1in CSF 15 min alter injection of tracer. *P < 0.05, treatment vs. control, by multiple range test. Bottom: volume of distribution (15 min) of ~C1 in lateral ventricle CP of rats subjected to various treatments. The net accumulation of 3~C1by CP is a function of basolateral (blood-side) uptake and apical (CSF-side) efflux. Due to the experimental differences in tissue water content induced by the various drugs (Table 1). it was advantageous to express the ~CI uptake on a dry weight basis. Bars arc means +_S.E.M. for 4 6 animals. The abcissa letter symbols for drug treatment, i.e, compositions of the i.c.v, injectate, are the same as those described in the top part of the tigure. Treatments with ouabain and acetazolamide are significantly different from control, and from treatments with dbcAMP (with or without theophylline) and forskolin. *P < 0.05 by multipie range test.
content, we have expressed the 3('C1 uptake by CP on a dry weight basis (Fig. 1, bottom). This effectively eliminates water content as a variable, and thus provides a clearer picture o f the a m o u n t of CI per unit weight o f C R O u a b a i n and acetazolamide b r o u g h t about marked accumulation o f 3{'C1 in CP: in comparison, c A M P and its enhancers led to significantly less Cl remaining in tissue (Fig. 1, bottom). The correlation between ~¢'C1in C S F and that in C P is
148 LD _..1
y-
r
45 T
v
2
= 0.82
D
0
Ii
O3 t.)
35
¢-
,oU
25
15 ~>
1 0
Ell
i
i
i
i
i
130
140
150
160
170
2
S >-
i
120
r
= 0.81
0
5.0
tD n" W
4.0 n
I-
