Drug modification of chloride transport in the choroid plexus- cerebrospinal fluid system of the rat

The mechanisms of anion and cation transport in the choroid plexus-cerebrospinal fluid (CP-CSF) system are a part of the blood-brain barrier system which provides a homeostatically regulated environment for the brain. The characteristics of CI transport in the CP-CSF system have been investigated both in vivo and in vitro. The following hypotheses have been tested in these studies: 1) that the primary carrier mechanism for CI transport in the CP is the CI-HCO3 antiport system which exists at both poles of the epithelial cell; and 2) that CI transport plays an important role in the CSF secretory process. The disulfonic stilbene (DIDS), a specific inhibitor of CI-HCO-3 exchange, was used to study the location and the mechanism of CI transport. Sprague-Dawley rats were used in all studies. Animals were nephrectomized under ether anesthesia and ketamine (80 mg/kg) was administered prior to placement of the animal in the stereotaxic instrument. Drugs were injected into the left lateral ventricle, and 36-Cl and 22-Na (100 microcuries/kg) were administered intraperitoneally (IP.) 30 min after drug treatment. The ventriculo-cistemal perfusion technique was used to measure the rate of formation of CSF during the control and drug treatment (DIDS and PGE2) periods. The quantity of 36-Cl and 22-Na in CSF, medulla and CPs was expressed as the percent of the tissue volume that would be occupied by the radioactivity if it existed at the same concentration as in the plasma water, i.e., Vd = 100 X (dpm/g tissue / dpm/g plasma H-2-O). Tissue electrolytes were extracted into a 0.02 N HNO-3-O.OI5 N Li solution and Na and K were measured by flame photometry. The results suggest that: 1) the CI-HCO-3 transport systems exist on both sides of the epithelial membrane of the CP; 2) DIDS inhibits CI and Na transport in the CP-CSF system; 3) as a result of inhibition of both CI and Na transport by DEDS, the rate of formation of CSF is decreased; and 4) PGE-2 stimulates the entry of CI into both CP and CSF which may account for its effect to increase the rate of formation of CSF. These observations support the hypothesis that CI transport may be one of the driving forces of CSF formation.

DRUG MODIFICATION OF CHLORIDE TRANSPORT IN TIIE CHOROID PLEXUS-CEREBROSPINAL FLUID SYSTEM OF TIIE RAT by Deng Quan-Sheng A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Pharmacology The Uni versity of Utah June 1986 THE UNIVERSITY OF CTAH GRADCATE SCHOOL SUPERVISORY COMwllTTEE APPROVAL of a dissertation submitted by Deng Quan-sheng This dissertation has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory. Chairman: Dona 1 Reed Ralph Karler C·_ C. Dean Withrow THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of The University of Utah: I have read the dissertation of Deng Quan-sheng m ItS final form and have found that (1) its format, citations. and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures. tables. and charts are in place; and (3) the final manuscript is satisfactory to the Supervis­ ory Committee and is ready for submission to the Graduate School. Approved for the Major Department D�J. Reed Chairman I Dean Approved for the Graduate Council Dean of The Graduate School Copyright© 1986 Deng Quan-Sheng All rights reserved ABSTRACT The mechanisms of anion and cation transport in the choroid plexus-cerebrospinal fluid (CP-CSF) system are a part of the blood-brain barrier system which provides a homeostatical1y regulated environment for the brain. The characteristics of C1 transport in the CP-CSF system have been investigated both in vivo and in vitro. The following hypotheses have been tested in these studies: 1) that the primary carrier mechanism for C1 transport in the CP is the C1-HC03 antiport system which exists at both poles of the epithelial cell; and 2) that Cl transport plays an important role in the CSF secretory process. The disulfonic stilbene (DIDS), a specific inhibitor of C1-HC03 exchange, was used to study the location and the mechanism of C1 transport. Sprague-Dawley rats were used in all studies. Animals were nephrectomized under ether anesthesia and ketamine (80 mg/kg) was administered prior to placement of the animal in the stereotaxic instrument. Drugs were injected into the left lateral ventricle, and 36(:1 and 22Na (100 microcuries/kg) were administered intraperitoneally (IP.) 30 min after drug treatment. The ventriculocisternal perfusion technique was used to measure the rate of formation of CSF during the control and drug treatment (DIDS and PGE2) periods. The quantity of 36Cl and 22Na in CSF, medulla and CPs was expressed as the percent of the tissue volume that would be occupied by the radioactivity if it existed at the same concentration as in the plasma water, i.e., Vd = 100 x (dpm/g tissue I dpm/g plasma H 20). Tissue electrolytes were extracted into a 0.02 N HN03-0.015 N Li solution and Na and K were measured by flame photometry . The results suggest that: 1) the CI-HC0 3 transport systems exist on both sides of the epithelial membrane of the CP; 2) DIDS inhibits CI and N a transport in the CP-CSF system; 3) as a result of inhibition of both CI and Na transport by DIDS, the rate of fonnation of CSF is decreased; and 4) PGE2 stimulates the entry of CI into both CP and CSF which may account for its effect to increase the rate of formation of CSF. These observations support the hypothesis that CI transport may be one of the driving forces of CSF formation. v TABLE OF CONTENTS ABS1RACT................................................................................... iv LIST OF FIGURES .......................................................................... ' vii ACKNOWlEOOMmTS ...................................................................... ix rnmODUCTION ............................................................................... 1 ME11IODS AND MA1ERIALS ................................................................4 Abbreviations ........................................................................ 4 Animals ............................................................................... 4 Materials .............................................................................. 4 General techniques .................................................................. 5 Analytical techniques ................................................................6 Experiments .......................................................................... 7 Calculations .......................................................................... 9 Statistics .............................................................................10 RESULTS ..................................................................................... 11 Permeability study ................................................................. 11 Time-course of the DIDS and PGE2 effects on Cl transport. ................ .11 Dose-response curves ............................................................. 11 Studies on the localizations of Cl-HC03 transport in the CP................. 22 Effects ofDIDS and PGE2 on Cl transport ................................... 22 Effects ofDIDS and PGE2 on Na transport................................... 33 Ventriculo-cistemal perfusion.................................................... 33 DISCUSSION ................................................................................. 39 Effects of DIDS, SITS and PGE2 on Cl transport. ........................... 39 Effects ofDIDS and PG~ on Na transport.................................... 40 Ventriculo-cistemal perfusion.................................................... 41 Possible mechanisms of action of DIDS and PGE2 on Cl transport......... 42 Model for the transport of Cl across CPo ....................................... 43 Summary ............................................................................46 REFERENCES ................................................................................ 47 VITA......................................................................................... 53 LIST OF FIGURES 1. Permeability of the blood-CSF barrier to DIDS ............................ 12 2. Time course of the effects of intraventricularly administered DIDS (12.3 J.lg in 25 J.lI artificial CSF/rat) or PGE2 (8.8 J.lg in 25 J.lI artificial CSF/rat) on 36CI transport from plasma into CSF ............................ 14 3. Time course of the effects of intraventricularly administered DIDS (12.3 J.lg in 25 J.lI artificial CSF/rat) or PGE2 (8.8 J.lg in 25 J.lI artificial CSF/rat) on 36CI transport from plasma to A) 4-CP and B) med......................................................................... 16 4. Time course of the effects of intraventricularly administered DIDS (12.3 J.lg in 25 J.lI artificial CSF/rat) or PGE2 (8.8 J.lg in 25 J.lI artificial CSF/rat) on 36CI transport from plasma into A) CP-L and B) CP-R ....................................................................... 18 5. Dose-response curves for the stimulation of 36CI transport in A) CSF, B) 4-CP, C) med. and D) CP-L and CP-R by PGE2 (25 J.lllrat) given intraventricularly .................................................................. 21 6. Dose-response curves for DIDS inhibition of 36CI transport in A) CSF, B) 4-CP, C) med. and D) CPLand CP-R by DIDS given intraventricularly ................................. 23 7. Effects of intravenously administered DIDS on the Vd of 36CI in CSF and various brain tissues ................................. 25 8. Effects of intraventricularly administered DIDS on the 36CI influx into various tissues from CSF................................. 27 9. Effects of intraventricularly administered DIDS or PGE2 on the Vd of 36CI in CSF and various brain tissues ..............................................................................30 10. Effects of various drug treatments on the intracellular concentration of CI in the CPo .................................... .32 11. Effects of intraventricularly administered DIDS (12.3 j.!g in 25 j.!l artificial CSF/rat) or PGE2 (8.8 j.!g in 25 j.!l artificial CSF/rat) on the Vd of 22Na in various regions of the CNS ..................................................... 34 12. Effects of DIDS or PGE2 in the ventriculo-cistemal perfusion fluid on the rate of formation of CSF ............................... 36 13. Schematic representation of the choroid plexus transport systems and possible sites of drug action on ion exchange in the adult rat CPo ............................................ 45 viii ACKNOWLEDGMENTS I would like to express my gratitude to all the members of the Department of Pharmacology for their contributions to my graduate training. I am particularly grateful to Drs. Conrad E. Johanson and Donal J Reed for their scientific guidance and friendly support throughout the course of my education at the University of Utah. Special thanks are expressed to Drs. Ralph Karler, John F. Ash and C. Dean Withrow for their encouragement and thoughtful criticisms while serving as members of my thesis committee. I also wish to thank Chak Tan for his technical assistance and professional typing of this manuscript Finally, a special note of appreciation I give to my loving wife, Yan Ya-ning, for her patience, support and understanding while I labored toward the doctoral degree. This work was supported by NIH grants NS 13988 and OM 07579. INTRODUCTION The high concentration of chloride in cerebrospinal fluid, relative to that in plasma, has been recognized for over 80 years (Nobecourt and Viosin, 1903). Since it was realized that a value of ReI (i.e., CSF [Cll I plasma [CIl) greater than unity is consistent with a Donnan distribution, the high [CIl in CSF was cited as evidence that CSF is a dialysate of plasma. It has been suggested that passive distribution can account for the observed concentrations of CI in the two fluids (Held et al., 1964). However, Hogben et al.(1960) reported that CI transport into CSF may be active rather than passive in the dogfish Squalus acanthias. As in other vertebrate species, the CSF [Cll in the dogfish is greater than that in plasma, but the electrical gradient between CSF and plasma is reversed (the CSF is negative rather than positive with respect to tissue extracellular fluids). The concept of a blood-CSF barrier is based on the observation that exchange of many materials between plasma and CSF is slower and more selective than exchange between plasma and extracellular fluid of peripheral tissues, e.g., muscle. The blood-brain barrier and blood-CSF barrier are frequently regarded as components of one system. Thus, the function of the blood-CSF barrier is similar to that of the blood-brain barrier, namely, to provide a controlled environment for the brain. Some circulating substances enter the brain via transport through the blood-CSF barrier rather than across cerebral capillaries (Pardridge et al., 1981). The choroid plexuses project into the ventricular cavities of the brain. The surface of each choroid plexus (CP) is covered by fine fronds consisting of tiny villous processes. Each villus is covered by a single layer of cuboidal epithelium and has a central core consisting of a capillary around which lies a small amount of loose connective tissue. The choroid plexuses produce approximately 50-70% of the CSF (Cserr, 1971). Smith (1980) demonstrated that CI is accumulated in the rat choroid 2 plexus at a concentration 3 times that expected from equilibrium distribution at a membrane potential of -50 mY, which implies that Cl is actively transported by the CPo Since the main anion in CSF is Cl, it is possible that active Cl transport in the CP-CSF system is responsible for some CSF production. Therefore, studies of how Cl is translocated across the CP-CSF system could give additional insights into how the blood-CSF bamer operates and how CSF is secreted by the CPo Cl transport can occur by a N a-Cl symport mechanism and/or by anion antiport, e.g., CI-HC0 3 exchange. Saito and Wright (1983) have proposed CI-HC0 3 antiport and Na-Cl cotransport in the basolateral membrane of frog CPo Johanson (1984) has postulated that CI-HC03 exchange occurs across both the apical and basolateral membranes of rat CP; there is currently no solid evidence for Na-Cl cotransport on either side of the rat CPo If CI-HC03 antiport exists, it may be located in the basolateral membrane, the apical membrane, or in both. DIDS or SITS (disulfonic stilbenes) may provide a means for localizing Cl-HC0 3 transport and furosemide and bumetanide can be used as tools for testing for the presence of Na-Cl cotransport. The volume of distribution (Vd) of various radioactive ions has been used as a measure of the transport function of the CP-CSF system (Davson, 1955). In the present study, effects on the Vd of Cl were assumed to represent changes in Cl transport in various regions of the CNS. The isothiocyano derivative of stilbene disulfonic acid (SITS), developed by Maddy (1964) as a covalent bonding agent for amino groups of the cell surface, has been used as a specific anion transport (CI-HC0 3) blocker in the following systems: erythrocyte (Cabantchik and Rothstein, 1972), turtle bladder (Ehren speck and Brodsky, 1976), snail neurons (Thomas, 1977), Ehrlich ascites cells (Aull et al., 1977), kidney (Hong et al., 1978), squid axons (Russell and Boron, 1976), astroglial cells (Kimelberg et al., 1979), and small intestine (White, 1980). The in vitro steady-state studies with SITS (Smith et al., 1985) and anion ratio analysis (Johanson et al., 1985) led to the conclusion that ClHC03 exchange exists in the choroid plexus system. 3 DIDS (4,4-diisothiocyano stilbene-2,2'-disulfonate), synthesized by Cabantchik and Rothstein (1974), has been demonstrated to be a potent inhibitor of anion exchange (CI-HC03) in the following systems: red blood cells (Cabantchik et al., 1974), com root protoplasts (Lin, 1981), Chara (Keifer and Lucas, 1982), rectal salt glands (Phillips and Hanrahan, 1984), small intestine (White, 1980), and gastric mucosa (Machen, 1984). In the rat CP-CSF system, it has been shown that DIDS significantly inhibits 36(:1 transport in in vitro experiment (Deng and Johanson, 1984). However, the pharmacological effects ofDIDS on CI transport in the rat CP-CSF system in vivo have not been studied. Prostaglandins are synthesized from arachidonic acid in mammalian brain and many other tissues, and have been suggested as modulators of central neurotransmission (Wolfe, 1982). Prostaglandins have been reported to influence ion transport across red blood cell membranes. They may be involved in the action of some diuretic drugs (Braquet et al., 1985). Since PGE2 has been shown to stimulate CI transport in the following systems: canine tracheal epithelium (Leikauf et al., 1985), corneal epithelium (Beitch et al., 1974), and kidney (Stoeken et ai., 1979), it is also likely that the drug may stimulate CI transport in the CP-CSF system. If so, studies of the effects of PG~ on Cl transport may give additional information concerning the nature of CI transport across the blood-CSF barrier. The purposes of this study were to determine: 1) some of the pharmacological effects ofDIDS and PGE2 on Cl transport in the CP-CSF system, i.e., time course, doseresponse curve; 2) whether CI-HC03 exchange exists at either or both poles of the rat CP epithelial cell; and 3) whether Cl and Na transport are correlated with CSF secretion. METHODS AND MATERIALS Abbreviations CP-R = right lateral ventricle choroid plexus PIa = plasma CP-L = left lateral ventricle choroid plexus Med = medulla 4-CP = fourth ventricle choroid plexus CSF = cisternal cerebrospinal fluid Cor = cortex Animals The experimental animals were Sprague-Dawley, male rats (Holtzman, Madison, WI), 120 to 250 gm body weight. They were housed in a temperature- and humiditycontrolled facility. Temperature was maintained at 220 C and humidity at 40%. Animals were housed 4-6 per cage and had access to food (purina Rat Chow) and water ad libitum. Rats were acclimatized for a minimum of 4 days before they were used. Experiments were routinely performed between 8:00 a.m. and 3:00 p.m. Materials 3H mannitol (0.123 mCi/g), 3H inulin (389 mCi/g), 22Na-NaCI and Biofluor emulsifier cocktail were obtained from New England Nuclear (Boston,MA). 36CI-NaCI (12.7 mCi/g) was purchased from ICN Chemical and Radioisotope Division (Cleveland, OH). Heparin sodium (1000 units/ml) was acquired from Elkins-Sinn, Inc. (Cherry Hill, NJ). Ketamine hydrochloride (100 mg/ml) was obtained from Bristol Laboratories (Syracuse, NY). 4-Acetamide-4'-isothiocyano-stilbene-2, 2'-disulfonic acid (SITS) was purchased from ICN pharmaceutical, Inc. (Plainview, NY). 4,4'-diisothiocyano-2, 2'disulfonic acid stilbene (DIDS) and other drugs were supplied by Sigma Chemical 5 Company, (St. Louis, MO) except PGE2 which was a generous gift from Dr. W. Jee (Division of Radiobiology, University of Utah School of Medicine, Salt Lake City, UT). General techniques Anesthesia. Initially each animal was anesthetized with diethyl ether for 2 to 3 min at the start of the experiment while both renal pedic1es were ligated. This was done to avoid the renal excretion of radioisotopes during the experiment. Subsequently, ketamine HCl (80 mg/kg body weight) was administered IP to provide continuous anesthesia during ventricular cannulation. Intraventricular injection. A small animal stereotaxic instrument (David Kopf Instrument, Tujuanga, CA) was used for all needle placements in the ventricles. The skin on the top of the head was incised and a hole was drilled though the skull at the proper coordinates according to Pellegrino and Cushman (1967). Injection of DIDS (12.3 Ilg in 25 III artificial CSF/rat) or PGE2 (8.8 Ilg in 25 III artificial CSF/rat) was made from a 100~Jll syringe (Hamilton) by a portable infusion pump (Model 1100, Harvard Apparatus, Millis, MA) at a rate of 5 Ill/min. In a typical experiment, the rat was nephrectomized and then put in the stereotaxic apparatus. Drugs were administered intraventricularly (left ventricle). Thirty min later, 36Cl or 22Na (0.1 IlCi/g body weight) was given IP. After 15 min, blood and CSF were removed and the rat was killed. The plexuses (CP-R, CP-L, and 4-CP) were excised from the brain and weighed on a Cahn electrobalance (Model 4700). Following CP digestion for 24 hr, radioactivity was measured by liquid scintillation. Ventriculo-cisternal perfusion. Ventriculo-cistemal perfusions were carried out on male rats that weighed 120-250 g. The perfusion technique was a modification of the method of Johanson et al . (1974). The ventricular system was perfused with artificial CSF at the rate of 18 Ill/min through a 27-gauge needle, the tip of which was located in the lateral ventricle of the left cerebral hemisphere. The outflow during 15-min intervals was 6 collected from a 22-gauge needle whose tip was in the cisterna magna. 3H-inulin (3 flCi/ml) and drugs were added to the artificial CSF perfusate. Steady-state conditions were usually achieved after one hour of perfusion. Radioactivity was determined by liquid scintillation. Sampling offluids and tissues. Blood (3-4 ml) was removed from the abdominal aorta with a heparinized syringe 45 min. The CSF (0.05-0.1 ml) was drawn from the cisterna magna with a tapered 0.1 ml glass micropipette. Each animal was exsanguinated by severing the aorta. The CP-L, CP-R, 4-CP and Med were excised from the brain with fine forceps and placed on tared aluminum foil containers to dry at room temperature (22 0 C). Analytical techniques Radioactivity. To measure the radioactivity of the 3H-Iabelled mannitol and inulin, and the 36CI-Iabelled NaCI, samples of plasma (0.1 ml) , CSF (0.05-0.1 ml) and dried CP (with foil) were placed in plastic scintillation vials to which 0.7 ml of piperidine (1 M) was added to solubilize the samples. After digestion for 24 hr at room temperature (22 0C), Biofluor Emulsifier Cocktail (12 ml) was pipetted into each vial. Following agitation with a mixer and cooling for 0.5 hr, the radioactivity of the samples was measured with a Beckman LS 7500 Liquid Scintillation Counter (Fullerton, CA). For measuring 22Na, samples were placed in 12 x 75 mm capped tubes and counted with a Beckman Biogamma Gamma Counter (Fullerton, CA). Electrolytes. To measure total electrolyte concentration, CSF (0.02 ml), PIa (0.025 ml) and Med (25 mg) were placed in 12 x 75 mm tubes that contained 5 ml of an aqueous solution of 7.5 mM Li2S04 and 0.02 N RN03 for extracting of electrolyte; the CP was extracted in 2 ml of the solution. After 48 hr, the Na and K concentrations of the extracts were measured with an Instrument Laboratories 443 Flame Photometer 7 (Lexington, MA). Concentrations of Na and K are reported as mmoles/kg water in fluids or mmol/kg wet tissue. Fluorescence. The fluorescence ofDIDS was measured with an Aminco-Bowman Spectrofluorometer (American Instrument Co., Silver Spring, MI) at room temperature with an excitation wavelength of 350 nm and emission wavelength of 450 nm. Water content. CP wet weight was determined by the method of Johanson et al.(1976) with a Cahn (Model 4700) Automatic Electrobalance (Cerritos,CA). Experiments Permeability study. Thirty min after the intravenous injection of DIDS (25 mg/kg), CSF samples ( 100-150 Jll/rat) were removed with a glass micropipette and transferred to cuvettes that contained 3 ml standard DIDS solution (0.17 Jlg/ml). The fluorescence of DIDS was measured. Time-course. DIDS or PG~ was administered as described in general techniques. Thirty min later 36Cl (0.1 JlCilg body weight) was given IP. Then, the samples (blood, CSF and plexuses) were taken at either 5, 10, 15, 35, 45, or 60 min. Dose-response. DIDS or PG~ (either 10-5, 10-6, 10-7, or 10-8 M in a volume of 25 Jlllrat) was administered intraventricularly (left ventricle). Thirty min later 36Cl (0.1 JlCi/g body weight) was given IP. Then, the samples (blood, CSF and plexuses) were taken after 15 min. Studies on the localizations of Cl-HCOJ transport in the CP. i) Intravenous injection: DIDS (25 mg/kg) was injected into the vene cava. Thirty min after the injection, 36Cl (0.1 JlCi/g body weight) was given IP. Samples were taken 15 min later. ii) Intraventricular injection: 36(:1(3 JlCilrat), 3H-inulin (3 JlCi/rat) and DIDS (12.3 Jlg Irat) in 25 Jll artificial CSF were given intraventricularly at a rate of 5 Jlllmin. Thirty min later all the samples were taken and analyzed. 8 Effects of DIDS and PGE2 on Cl transport. i) In vivo study: DIDS or PGE2 was administered as described in general techniques. Thirty min later 36Cl (0.1 JlCi/g body weight) was given IP. Then, the samples (blood, CSF and plexuses) were taken at 15 min. ii) In vitro study: Immediately prior to sacrifice, each animal was anesthetized with diethyl ether. The animal was then exsanguinated by severing the abdominal aorta. After removal of the brain from the cranial cavity, CP-R and CP-L were dissected from the brain and transferred to a preincubation tube that contained 2 ml of artificial CSF where it remained for 15 min. The artificial CSF was maintained at 270C and was equilibrated with humidified 95 % 02 and 5 % C02. After the 15-min preincubation period, each CP was transferred to another tube of artificial CSF that contained 3H-mannitol, as a measure of extracellular space, and 36Cl-NaCl. After the transfer, each CP was incubated for 60 sec. At the end of the incubation, the CP was quickly removed, blotted by drawing it along a 5 centimeter length of dry plate glass, and placed on aluminum foil (2 mg) for weighing. Furosemide (10-3 M), bumetanide (10-3 M), SITS (10-3 M) and DIDS (10-3 M) were used in this study. Effects of DIDS and PGE2 on Na transport. DIDS (12.3 Jlg in 25 Jll artificial CSF/rat) or PGE2 (8.8 Jlg in 25 Jll artificial CSF/rat) was administered intraventricularly (left ventricle). Thirty min later 22Na (0.1 JlCi/g body weight) was given IP.. Then, the samples (blood, CSF and plexuses) were taken at 15 min. Ventriculo-cisternal perfusion. The ventricular system was perfused with artificial CSF at a rate of 18 Jll/min through a 27-gauge needle. The outflow during 15-min intervals was collected from a 22-gauge needle whose tip was in the cisterna magna. The perfusate contained 3H-inulin (3 JlCi/ml), an impermeable indicator as a measure of CSF production, and DIDS (10-4 M) or PGE2 (10-4 M). Steady-state conditions were usually achieved after one hour of perfusion. Radioactivity was determined with a liquid scintillation counter. 9 Calculations The volumes of distribution (Vd) in vivo, i.e., the sample content of 36CI, 22Na or 3H-Iabelled mannitol and inulin, were calculated by the following formula: Vd (%) = (dpm/g tissue or fluid / dpm/g plasma H 20 x D.F.) x 100, (1) where D.F. is the appropriate Donnan Factor, i.e., 1.05 for a monovalent anion, 0.95 for a monovalent cation, and 1.00 for an uncharged molecule. Vd of 36(:1 and 3H mannitol in vitro were determined by the formula: Vd (%) == (dpm/g CP I dpm/g artificial CSF) x 100. (2) Choroid plexus percent water was calculated by: H 20 t (%) = (wet weight -dry weight I wet weight)x 100. (3) CI concentration in intracellular H 20 of choroid plexus [Cl] was determined by the formula: (4) where [CI]csfis the concentration of stable CI in CSF medium, H200 is Vd of 3H-mannitol in extracellular fluid, and the Vd of cell CI is the difference between the Vd of total tissue 36(:1 and 3H-mannitol. For brevity, mM is used for mmol/kg cell H 20. The rates of CSF formation (Vf) were estimated by: (5) 10 where Vi is the rate of perfusion, Ci and Co are radioactivities of the indicator (inulin in this case) in the inflow fluid and cisternal effluent, respectively. Statistics The results are expressed as means ± S.E. The two-tailed Student's t-test was employed for comparing two means. One-way analysis of variance and the two-tailed Dunnett's multiple comparison test were used to compare the individual means if three or more means were analyzed (Dunnett, 1955; Randolph et al., 1985). The P<0.05 level of probability was chosen as significant. RESULTS Permeability study The data on the penneability of the blood-brain barrier to DIDS are presented in Fig. 1. They indicate that DIDS cannot easily penetrate the blood-brain barrier, since the relative fluorescence intensity of the CSF from the DIDS-treated animals was not different from the control values. Time-course of the DIDS and PGE2 effects on CI transport Fig. 2. shows the time course of the effects ofDIDS (12.3 Jlg/rat) and ( 8.8 Jlg/rat) on CI transport into CSF, compared with the control group. PG~ PGE2 significantly increased the 15-min Vd of 36CI in CSF to 40.5 ± 2.1 and DIDS decreased it to 20.7 ± 1.3 compared with the control value of 29.7 ± 1.0. At 25 min, PGE 2 elevated the Vd of 36CI to 55.3 ± 0.7 and DIDS reduced it to 35.5 ± 2.9, compared with the control 46.4 ± 3.5. The characteristics of the effects of DIDS and PGE2 on Cl transport in 4-CP, Med, CP-R and CP-L are similar to those in CSF (Fig. 3. and 4). Dose-response curves The dose-response relations for PGE2 are shown in Fig. 5. In nephrectomized rats, intraventricular injection of PGE2 (10-7, 10-6, 10-5 M) increased the Vd of 36CI in CSF after 45 min by 6, 23, and 36%, respectively. Fig. 5 also shows the typical responses to increasing doses ofPGE2 by 4-CP (Fig. 5 B), Med (Fig.5 C) CP-R and CPL (Fig.5 D). After the intraventricular injection of 10.7 , 10-6, 10- 5 M of DIDS the Vd of 36Cl in CSF when measured 45 min after drug administration showed a dose-dependent decrease 12 Fig.1. Permeability of the blood-CSF barrier to DIDS. A) Standard curve of DIDS measured by fluorescence spectrophotometer. The concentrations of DIDS are: A=control (H20); B= 0.17 }lg/m1; C= 0.33 }lg/m1; D= 0.5 }lg/m1; 0.83 Jlg/ml; F= 1.16 Jlg/ml. B) The concentration of DIDS in CSF 45 min after the intravenous injection (25 mg/kg). A= control (H20); B= 0.1 7 Jlg/ml as a standard; C and D = experimental animals. 13 A. 60 .-c>--- 50 0) .-Q) 40 c Q) > c Q) 30 20 0:: 10 0 300 350 400 450 500 550 500 550 Wavel enQth-Nanometers B. 60 .-- >-- 50 0) c Q) 40 c Q) 30 Q) 20 > c 0:: 10 0 300 350 400 450 Wave 1ength-Nanometers 14 Fig. 2. Time course of the effects of intraventricularly administered DIDS (12.3 Jlg in 25 Jll artificial CSF/rat) or PGE2 (8.8 Jlg in 25 Jll artificial CSF/rat ) on 36Cl transport from plasma into CSF. DIDS and PGE2 were given intraventricularly whereas 36Cl was administered IP. Ordinate: Vd in percent. Abscissa: time in min. Values are the means ± S.E. (n=5) for each experiment. 15 T'lme course of DI DS and PGE2 90 ,-... 80 ......... 70 ;:-.: LL. (/') u ....c (OU I"") 60 50 40 ..... 0 "C > 20 10 0 50 60 16 Fig. 3. Time course of the effects of intraventricu1arly administered DIDS (12.3 Ilg in 25 III artificial CSF/rat) or PGE2 (8.8 Ilg in 25 III artificial CSF/rat ) on 36C1 transport from plasma to A) 4-CP, and B) med. compared with controls. DIDS and PGE2 were given intraventricu1ar1y whereas 36C1 was administered IP. Ordinate: Vd in percent. Abscissa: time in min. Values are the means ± S.E. (n=5) for each experiment. 17 A. c.. U I "It"' .....C ..... U <.0 ,..., \t- o C > 15 20 I I 30 50 Time (min) B. "C Q) ~ .....c ..... U <.0 ,..., \t- o C > 10 20 30 Time (min) 50 60 18 Fig.4. Time course of the effects of intraventricularly administered DIDS (12.3 f.lg in 25 f.ll artificial CSF/rat ) or PGE2 (8.8 f.lg in 25 f.ll artificial CSF/rat ) on 36CI transport from plasma into A) CP-L, and B) CP-R, compared with controls. DIDS and PGE2 were given intraventricularly whereas 36CI was administered IP. Ordinate: Vd in percent. Abscissa: time in min. Values are the means ± S.E. (n=5) for each experiment. 19 A. -I I a.. u .-1: u (0 M ~ o "C > Time (min) B. I a::: I a.. u T rI~T?/!~I~ .1: u (0 M ~ o "C > yl/ I 30 Time (min) 60 20 Fig. 5. Dose-response curves for the stimulation of 36CI transport in A) CSF, B) 4-CP, C) Med. and D) CP-L and CP-R by PGE 2 ( 25 J,1l/rat ) given intraventricularly. 36CI was administered IP. Ordinate: Vd in percent. Abscissa: doses of PGE2 on a logarithmic scale. Values are the mean ± S.E. (n=4) for each experiment. 21 A. B. 42 - :--.: ....., u.. 3S - 4 :--.: ....., 3 0... U I V" (J') u c 3 .-u .c 3 ..u <.0 <.0 M M ~ 34 33 32 31 30 ~ 0 0 3 "0 "0 > > 2 29 28 oI J c. D. 42 7. :--.: ....., - 6. :--.: ....., "0 (f) 0... Q) u c 34 ..... ::E c ''- ..- ..- u u <.0 <.0 M M ~ ~ 0 0 "0 "0 > > or I 10- 7 I 10- 6 Dose of PGE2 (M ) I 10- 5 rnJ • CP-L 36 32 30 28 26 or I I I 10- 7 10- 6 10- 5 Doseof PGE2 (M) 22 with a maximum decreases of 7, 13, and 30%, respectively (Fig. 6 A). The Vd of 36CI in fourth ventricle CP was lowered in a dose-dependent manner after 10-7 and 10-6 M DIDS, with maximum decreases of 7 and 14%, respectively (Fig.6 B). In Med, CP-R and CP-L, the dose-dependent decreases of the Vd of 36CI after intraventricular administration ofDIDS are shown in Fig. 6 C and D. Studies on the localizations of CI-HC0 3 transport in the CP As shown in Fig. 7, the intravenous injection of DIDS (25 mg/kg) yielded Vd of 36CI that were significantly smaller than those in control animals except in the Med. DIDS decreased the Vd of CI by 26, 28, 26, 28 and 21 % in CSF, cortex, CP-R, CP-L, and 4CPo These data are what would be expected if CI-HC03 exchange occurs on the plasmafacing basolateral membrane of rat choroid plexus. Ratios of 3H-inulin I 36CI ,as a measure of 36CI movement from CSF into various CNS regions, are displayed in Fig. 8. DIDS (12.3 flg/rat ) decreased the ratio by 27, 31, 72, 49, and 46 % in CSF, in Med, CP-R, CP-L, and 4-CP, respectively. Ratios were used to minimize the effects of variations in extracellular space. Effects of DIDS or PGE 2 on CI transport In vivo experiments. Fig. 9 illustrates the effects of DIDS or PGE2 administered intraventricularly on 36CI transport. DIDS (12.3 flg/rat) decreased the Vd of 36CI by 30, 14, 13, and 11 % in the CSF, Med, CP-R, and CP-L, respectively, with no effect in the 4-CP. In the same study, PGE2 (8.8 flg/rat) increased the Vd of 36Cl by 36, 33, 33, and 20% in eSF, CP-R, CP-L, and 4-CP, respectively, with no effect in the Med. In vitro experiments. The effects of furosemide, bumetanide, SITS and DIDS on Cl distribution in the CP-L are shown in Fig. 10. The CP-L was preincubated in 2 ml synthetic CSF for 15-min followed by a 6O-sec incubation in 2 ml synthetic CSF which 23 Fig. 6. Dose-response curves for DIDS inhibition of 36Cl transport in A) CSF, B)4-CP, C) med. and D) CP-L and CP-R by DIDS given intraventricularly. administered IP. Values are the mean ± S.E. (n=4) . 36CI was 24 A. B. 34 ".... 32 32 ".... ;-.: ;-.: ......" ......" 30 c.. LL U I "It' V) u c '''u CD M 28 c ...- 26 u CD M 24 \0- \0- 0 "0 22 ol "0 I I 10- 8 10- 7 I 10- 6 Q) 27 26 ol I 10- 5 5.5 ;-.: 5.4 ......" 27 0') c.. - 5.2 - 26 \0- 5.0 \0- 25 4.9 "0 '''- u CD M 0 "0 > I 10- 6 ".... u c I 10- 7 28 5.3 ~ I 10- 8 D. 5.6 "0 28 25 C. ......" 29 :>- 20 ;-.: 30 0 :>- ".... 31 c '''- u 5.1 CD M 0 CP-R • CP-L I > 4.8 OT 0 1 24 I 10- 8 I 10- 7 Dose of DIDS (M) I 10- 6 ol I 10- 7 I 10- 6 Dose of DIDS (M) I 10- 5 25 Fig. 7. Effects of intravenously administered DIDS on the Vd of 36CI in CSF and various brain tissues. Thirty min after the intravenous injection of DIDS (25 mglkg), 36CI was given IP. Tissue samples were obtained 15 min after 36CI adminstration. Statistical comparisons were done with paired Student's t test. Each bar represents the mean ±S.E. for 5 animals. 26 Effects of intravenous 0 I OS on 36C1 distribution o a 40 Con DIDS u CD M 'o 20 "C :> 10 CSf Med Cor CP-R CP-L 4-CP 27 Fig. 8. Effects of intraventricularly administered DIDS on the 36Cl influx into various tissues from CSF. 36cl (3 JlCi/rat), 3H-inulin (3 JlCi/rat) and DIDS (12.3 Jlg/rat) were given intraventricularly. Samples were removed at 30 min. The results are expressed as the ratio of the 3H-inulin Vd to the 36Cl Yd. Statistical comparisons were done with paired Student's t test. Each bar represents mean ± S.E. for 5 animals. 28 36 C1 and 3 H -inul in infl ux exp. ( Ratio anal ysis: 3 HI 36 C1 ) 7 [J Control C8l 6 (JJ o c Co f) (JJ c 'r- 0 C Co - ~ 'r- f) U :J:CD M It- 0 .-......o 4 ..- I M 5 3 0 ...... 2 C a:: Pla CSF Med CP-R CP-L DIDS 29 Fig. 9. Effects of the intraventricularly administered DIDS or PGE2 on the Vd of 36Cl in CSF and various brain tissues. Each bar represents mean ±S.E. for 6 animals. *P<O.05; **P<O.Ol compared with the control group by Dunnett's multiple range test. 30 Effects of 0 I OS and PGE2 on 36 C1 distribution in vivo o ~ PGE2 ~ OIOS so ** ** Con 40 U 30 ... ,.....;r.;...r...JI"'~ CD M '0 20 "0 > 10 o CSF Med CP-R CP-L 4-CP 31 Fig. 10. Effects of various drug treatments on the intracellular concentration of CI in the CPo All preincubations were 15 min. All incubations were 60 sec at 27 °C. Each bar represents mean ± S.E. for 4-8 animals. Con = control. Fur = 10-3 M furosemide. Bum 10- 3 M bumetanide. SIT and DIDS refer to 10-3 M disulfonic stilbenes. *P<0.05; **P<O.Ol by Dunnett's multiple range test. 32 Effects of drugs on Cl in vitro CP 80 ,..... '- 70 Q,) 0.. .... U C .- -- :) C • o Q,) 0 C 00.::£ 0 ........ 60 SO 40 30 uo 20 E 10 ~ ......." O~--~~~~~~~~~~~L.~~~ Con Fur Bum SITS DIDS 33 contained 36CI and either furosemide or bumetanide. Neither treatment changed the Cl concentration 72.4 ± 0.5 mmollkg cell water (n=8) and 74.3 ± 1.6 mmollkg cell water (n=5), respectively. However, SITS (10- 3 M) and DIDS (10- 3 M) decreased the concentration of CI in the CP to 54 ± 0.7 mM (n=6) and 43 ± 0.4 mM (n=4) compared with the control value of 72.5 ± 2.6 mM (n=8 ). Effects of DIDS or PGE z on Na transport Fig. 11 illustrates the effects of DIDS or PGE 2 on 22Na transport in vivo. DIDS (12.3 J..lg/rat) decreased the Vd of 22Na by 29, 36, 14, 18, and 20% in CSF, Med, CP-R, CP-L, and 4-CP, respectively. However, PGE2 (8.8 J..lg/rat) did not significantly alter the Vd of 22Na in any of the tissues. Ventriculo-cisternal perfusion Fig. 12 shows the effects of DIDS or PGE2 on the rate of formation of CSF. DIDS reduced the rate of formation of CSF by 30% to 3.7 ± 0.15 J..lllmin whereas PGE2 increased formation of CSF by 30% to 6.8 ± 0.28 J..lllmin, from the control rate, 5.25 ± 0.15 J..lllmin. The effects of DIDS or PGE2 on N a and K concentrations in various regions are shown in Table 1. It indicates that DIDS significantly decreased the concentration of Na in CSF without affecting the concentration of K and that PGE2 increased the concentration of K in plasma. 34 Fig. 11. Effects of intraventricularly administered DIDS (12.3 J,lg in 25 J,ll artificial eSF/rat) and PGE2 (8.8 J,lg in 25 J,ll artificial eSF/rat) on the Vd of 22Na in various regions of the eNS. Each bar represents mean ± S.E. for 4-5 animals. *P<O.05; **P<O.OI compared with control group by Dunnett's multiple range test. 35 Effects of 0 I OS end PGE2 on 22Ne distribution in vivo 40 _ o Con rIJ PGE2 IQ1 DIDS 30 :--: ........ o ~ N 20 \t- o "'C ::> 10 o CSF Med CP-R CP-L 4-CP 36 Fig. 12. Effects of DIDS or PGE2 in the ventriculo-cistemal perfusate on the rate of formation of CSF. The control (Con.) perfusion fluid was artificial CSF (n=5). DIDS (10- 4 M, n=4) significantly decreased the rate of CSF formation (P<O.Ol) while PGE2 (10-4 M, n=4) significantly inceased it (P<0.01). 37 v. c. """ c: E ....... -- --- :::1. ......, 0 c 'c: 0 c E L. 0 '- **.... 7 x X 6 X)e I- T )eX 5-1- o ** .,. 4 ....l'Sa 2i~)( 1. i)e . (f) 0 Con )( )( )c ~~~x0 DIDS Con [ffi) DIDS )() 3 L1U perfus ion experiment . PGE2 38 TABLE 1 Na and K concentrations Na and K concentrations in various regions of the brain 45 min after the intraventricular administration of either DIDS or PGE2. Each value is the mean ± S.E. for 6 animals. Statistical differences were determined by two-tailed Dunnett's multiple range test at levels of significance,*P<0.05 and **P<O.Ol, as compared to control. PIa CSF Med CP-R CP-L 4-CP (mM / kg wet tissue) ------------------------------------------------------------------------------------------------------------ [Na] Con. 149.2 ±2.5 39.4 ±2.3 64.8 ±2.9 62.0 ±3.5 66.0 ±2.6 143.6 ±1.6 154.2 ±3.1 ** 126.2 ±3.7 DIDS 40.6 ±1.4 60.4 ±3.1 58.5 ±3.9 59.9 ±1.2 PGE2 140.4 ±0.6 147.8 ±0.8 44.3 ±1.6 72.3 ±1.6 65.4 ±2.3 70.3 ±1.4 ------------------------------------------------------------------------------------------------------------ [K] Con. 5.6 ±0.3 3.3 ±0.2 90.3 ±4.6 91.6 ±3.7 90.4 ±1.6 84.0 ±2.0 DIDS 6.8 ±0.3 ** 8.0 ±0.4 3.2 ±0.1 92.2 ±1.9 94.5 ±1.8 87.9 ±1.9 85.7 ±1.9 3.4 ±0.1 99.2 ±1.4 79.6 ±2.2 84.5 ±1.5 81.1 ±2.4 PGE2 DISCUSSION Effects of DIDS, SITS and PGE 2 on Cl transport DIDS and PGE2 had no significant effect on the Vd of 36Cl in CSF during the first 10 min after intraventricular administration. After 10 min, DIDS significantly decreased, and PGE2 increased transport. The lack of effect before 10 min probably reflects the fact that an effective concentration of the drug had not been achieved at the site of Cl transport. PGE2' in concentrations below 10-7 M, had little effect on the Vd of 36Cl in CSF, Med, 4-CP, and CP-Rand CP-L, compared to control. At higher concentrations, PGE2 appears to have an effect which is proportional to the dose over the range studied. Similarly, DIDS, in low concentration (10- 8 M), had little effect on the Vd of 36(:1 in any of the tissues studied. At higher concentrations, a dose-dependent inhibitory effect on CI transport was observed. It is worth noting that the Vd of 36(:1 in the left lateral ventricle CP is lower than that in right lateral ventricle CP after intraventricular injection of either PGE2 or DIDS. This may be due to the fact that the injections were made into the left ventricle and caused greater interference with CI uptake by the left lateral ventricle CP. A major finding in this kinetic study with DIDS and SITS is that they both inhibit a part of Cl transport in the CP-CSF system. DIDS is a potent inhibitor of anion transport (Cl-HC03) and it binds covalently and irreversibly to molecules on the outer surface of the membrane of erythrocytes. It does not penetrate the membrane of human red cells (Cabantchik and Rothstein, 1974) and probably does not penetrate the blood-CSF barrier, based on the data from the fluorescence spectrophotometry experiments (Fig. 1). The data for the effects of DIDS confirm the existence of CI-HC03 antiport in CP and demonstrate the location of this transport system on both membranes of the rat CP epithelium. This is based on the following evidence: 1) 40 the Vd of 36CI is reduced in the CP and CSF following the injection of DIDS and 36CI on the blood side, Le., DIDS inhibits CI influx; 2) the Vd of 36CI is reduced in CP and Med following the injection ofDIDS and 36CI into the CSF, i.e., DIDS inhibits CI efflux from the ventricles; and 3) both the in vitro and in vivo experiments show that DIDS decreases the Vd of 36CI in CPo PGE2 has been demonstrated to stimulate CI transport in the following systems: canine tracheal epithelium (Leikauf et aI., 1985), corneal epithelium (Beitch et al., 1974), and kidney (Stocken et al., 1979). In the present study, it was observed that PGE2 also stimulates CI transport in the CP-CSF system. It has been shown that of a series of prostaglandins, only PGE2 increased cellular cyclic AMP concentration due primarily to adenylate cyclase activation in rat CP (Feldman et aI., 1979). In an in vitro study, Chaudhari and Kirschenbaum (1985) provided evidence for specific PGE2 receptors in the rat glomerulus. Furthermore, Bito (1972 a, b) and Bito et al. (1976 a, b) have demonstrated that prostaglandins are transported across the blood-brain and blood-CSF barriers by facilitated or active transport. If the evidence cited above, obtained from a variety of tissues and species, is applicable to the CP, it may be that PGE2 combines with PGE2 receptors, stimulates adenylate cyclase, increases the cyclic AMP levels, and enhances CI transport in the CP-CSF system. Effects of DIDS and PGE 2 on Na transport CI is quantitatively the primary anion and Na the main cation in CSF. The active transport of N a across the CP into CSF has long been thought to be the driving force for CSF production. Yates et al. (1964) found that the CP Na-K ATPase system is involved in the formation of CSF through the active secretion of Na ions into the ventricle. Na transport from blood into CSF may involve two mechanisms: Na-H antiport (Maren, 1972; Sachs et aI., 1982) and Na-K exchange (Zeuthen et aI., 1978). In the present study, DIDS significantly reduced the Vd of 22Na in CSF, Med and CP which suggests that DIDS 41 inhibits CSF fonnation by affecting both Cl and Na transport. Since DIDS apparently does not affect K concentration in CP and CSF, it is concluded that DIDS inhibits some transport system other than N a-K exchange. In the present study, the Vd of 36Cl was increased by PGE2' In contrast, the Vd of 22N a was slightly reduced (Fig. 11 ) and the stable N a concentration in CSF was also decreased in CSF (Table 1). These observations agree with other reports (Lino et al., 1978; Al-Bazzaz et al., 1981; Al-Awqati et ai., 1972; and Leyssac et al., 1975) in which PGE2 or PGE 1 inhibited Na transport. However, PGE2 stimulates Na transport in toad bladder (Lipson et ai., 1971) and frog skin (Fassina et ai., 1976). Leyssac et ai., (1974) proposed that prostaglandins inhibit N a transport in epithelia with low electrical resistance. Since the CP is an epithelium with low electrical resistance (Wright, 1972), the finding that PGE2 reduced Na transport in the present study is in accord with the above hypothesis. In the present study, it was also shown that PGE2 does not seem to affect K movement, as estimated from the concentration of [K] and [Na] in CSF and CPo This suggests that PGEz inhibits Na transport primarily by an effect on Na-H antiport (Fig. 13). The present study demonstrated that DIDS not only inhibited Cl-HC03 transport, but it also may inhibit NaHC03 cotransport in rat CP-CSF system. Ventriculo-cisternal perfusion The ventriculo-cistemal perfusion technique is widely used and is a reliable method for quantitative determination of the bulk formation of CSF in the ventricular system (Segal et al., 1977). Ventriculo-cisternal perfusion was used to ascertain whether the two drugs, DIDS and PGE2, affect the rate of CSF production. The results show that DIDS does inhibit CSF production (30%) by decreasing Cl and Na transports and that PGE2 stimulates CSF production (30%) by accelerating Cl transport in the CP-CSF system. Furthermore, the data suggest that drugs which affect either CI or both CI and Na transport in this system may alter CSF production. 42 The data show that PGE2 increases [K] in blood. The mechanisms to explain this observation remain to be established. Possible mechanisms of action of DIDS and PGE2 on CI transport It is known that the mechanism of action ofDIDS in the red blood cell is associated with the polypeptides of band 3 obtained from the predominant protein of erythrocyte membranes which apparently contains the putative anion-exchange protein of erythrocytes (Rothstein, 1975). Band 3 consists of three distinct domains: 1) a hydrophilic N-terminal segment of about 38,000 daltons (Bender et aI., 1971; Cabantchik and Rothstein, 1974; Drickamer, 1976; Passow and Zaki, 1978; Fukuda et aI., 1978), 2) a transmembrane segment of 17,000 daltons (Steck et aI., 1976) and 3) a membrane-associated C-terminal segment of 38,000 daltons. The third domain that contains most of the carbohydrate of band 3, appears to have the most important role in the process of anion transport (Jenkins and Tamer, 1977; and Fukuda et al., 1978;). The finding that Cl transport in the CP-CSF system was reduced by DIDS suggests that the mechanism of action of DIDS in CP may involve the same mechanism as in red blood cells. The properties of the Cl-HC03 carrier in CP may be similar to that of the band 3 in erythrocytes. However, the exact mechanism of Cl-HC03 transport in CP remains to be delineated. In in vitro studies, it has been shown that high affinity and specific PGE2 binding sites are present in the isolated intact rat kidney glomerulus. Such studies establish the existence of PGE2 receptors in rat tissues (Chandhai et al., 1985). Prostaglandins modulate fluid movement across the toad bladder (Flores et al., 1975), and the tubular cells of the kidney (Beck et aI., 1971; and Grantham and Orloff, 1968) through a cyclic AMPmediated mechanism. Of 5 prostaglandins studied, only PGE2 resulted in an increase in cAMP accumulation in rat choroid plexus (Feldman et aI., 1979). Saito and Wright (1983) found that HC03 transport across CP is increased by agents which enhance cellular cAMP. 43 The present results with PGE2, plus evidence listed above, support the concept that this nucleotide may be involved in the mechanism of action of PGE2. Thus, it is postulated that in the CP-CSF system, PGE2 1) binds to PGE2 receptors; 2) stimulates accumulation of cAMP; 3) accelerates CI-HC03 exchange, and 4) increases CSF formation rate. Model for the transport of Cl across CP A model for CP ion transport based on the present experiments and the data of others is presented in Fig. 13. It is well-known that the Na-K pump, located predominantly on the apical side of the CP, actively pumps CSF K into the cell as it extrudes intracellularNa (Vates etal., 1964; Zeuthen etal., 1978; and Johanson etal., 1974). The following characteristics of the CI-HC03 transport system have been demonstrated: 1) ratio analysis ([CI] I [HC03]) has shown that CI-HC03 are transported one to one (Johanson et al., 1985); 2) SITS and DIDS, disulfonic stilbene inhibitors of CIHC03 exchange, reduced cell [CI] and Vd of 36CI (see text); 3) furosemide and bumetanide, inhibitors of Na-CI cotransport, (Imai et at., 1977; Frizzel et al., 1979 and Warnock et al., 1982) do not decrease [CI] in CP (Figure 12). This suggests that CI transport in rat CP may occur primarily by CI-RC03 transport rather than Na-CI cotransport. Based on the strong correlation between transmembrane [H] gradient and 22Na uptake into the CP (r2 =0.99), Murphy (1984) demonstrated that a significant portion ofNa uptake from blood into CP is due to Na-H exchange. Previous evidence has shown that SITS blocks the increase of HC03 in CSF in hypercapnia (Javaheri and Weyne, 1983) and that DIDS inhibits RC03 transport in frog CP (Saito and Wright, 1983). The present experiments show that DIDS reduces the Vd of CI in CSF, Med. and CPo Such evidence may indicate that DIDS or SITS interacts with the carrier, probably CI-HC03 ATPase (Jose et al., 1983; Humphreys et al., 1979; DePont etal., 1972; Van Amelsvoort et al., 1977; DeRenzis et al., 1977), of CI-HC03 transport and inhibits CI and HC03 exchange. Since DIDS also decreased the Vd of 22Na without 44 Fig. 13. Schematic representation of the choroid plexus transport systems and possible sites of drug action on ion exchange in the adult rat CP. Values for electrolyte concentrations (mM) in blood, CSF, and CP are those reported by Smith, 1980 and Johanson, 1984. On the left side of the figure, the events at the basolateral membrane of the CP cell are depicted, and on the right side are shown the events at the apical membrane of the CPo Evidence for Na+-K+ and Na-H antiport systems has been furnished by Zeuthen et al. (1978), Sachs (1982), Smith and Johanson (1980), and Murphy (1984). ClHC0 3 exchange has been demonstrated by Kimelberg (1982) and Johanson (1985). The present data confirm that the Cl-HC03 system exists on both sides of the CPo Anion exchange can be in either direction (from blood to CSF or vice versa) depending upon the relative concentration of Cl and HC03 in the cell and the CSF. Normally, there is net secretion of N a and Cl (from plasma to CSF) and net reabsorption of K and H (from CSF to plasma). This schema also illustrates the probable sites of action ofDIDS and PGE2: 1) DIDS may inhibit Cl-HC03 and Na-H transport on both sides of the cell; as a result, it would decrease the Vd of Cl and Na without altering [K]; and 2) PGE2 binds to PGE2 receptors in the CP membrane, activates the adenyl ate cyclase of the membrane, increases the rate of formation of cAMP and protein kinases, and thus accelerates Cl exchange and reduces N a transport in the membrane of CP. 45 Blood CP CSF 114 143 67 120 ( mMoll Kg H20) 46 155 4 150 10 3 20 20 -5 ; ;: ;':'I: : I I1I~I~I~I/I~I~I/'I\I':I 'lIl1i':I':~I~I J>IJ'I Basol ateral ';~:~" :.; Membrane,..;':': (;; ::: ~ N ......: '.', Api cal Membrane ~8 ,:. ',' :.: '1 I. :.:. I' :.: Tight Junction ·: Ii· · · I'· ·I I I I I I I I I I I Ii· · ·'i· ·'~ ~.:.:::~ '.'.':':', ~ '.','. ",r (:~{ ,'.' .. .. ,':' .':' '.' ' , •• ,•• ;.;,) .. ,~, ........ , ) + - Microvill i 46 an effect on [K] in CSF, this implies that DIDS inhibits a Na transport system (perhaps NaHC03 contransport) in the choroid plexus other than the Na-K system. It is hypothesized that PGE2 interacts with PGE2 receptors (Chaudhari and Kirschenbaum, 1985), increases cAMP levels, and stimulates CI-HC03 transport. This is supported by the findings that PGE2 increase cAMP in the rat CP (Feldman et al., 1979) and that cAMP and theophylline stimulate HC03 transport in frog CP (Saito and Wright, 1983). Since the CP is a low electrical resistance epithelia (Wright, 1972), PGE2 also decreased the Vd of 22Na and Na concentration without alteration of CSF [K]. This implies that PGE2 inhibits Na-HC03 transport rather than the Na-K system in the choroid plexus. Summary For the first time, evidence has been presented to support the hypotheses that: 1) the CI-HC03 transport systems exist on both sides of the epithelial membrane of the CP; and 2) CI transport is one of the driving forces of CSF formation. It was demonstrated that: l)DIDS inhibits CI and N a transport in the CP-CSF system; 2) as a result of inhibition of both CI and Na transport by DIDS, the rate of formation of CSF is decreased; and 3) PGE2 stimulates the entry of CI into both CP and CSF which may account for its effect to increase the rate of formation of CSF. Based on the present experiments and the data of others, it is postulated that DIDS inhibits CI transport by affecting Cl-HC0 3 ATPase and PGE2 increases CI transport by affecting the PGE2 -receptor-cAMP system in the rat CPo REFERENCES AI-AWQATI, W. and GREENOUGH, W.B.: Prostaglandins inhibit intestinal sodium transport. Nature New BioI. 238: 26-27, 1972. AI-BAZZAZ, F., YADAVA, V.P. and WESTENFELDER, C.: Modification ofNa and CI transport in canine tracheal mucosa by prostaglandins. Am. I. PhysioI. 240: F101-F105, 1981. AULL, F., NACHBAR, M.S., and OPPENHEIM, J.D.: Chloride self-exchange in ehrlich ascites cells inhibition by furosemide and 4-Acetamido-4'-isothiocyano2,2'-disulfonic acid. Biochim. Biophys. Acta 471: 341-347, 1977. BECK, N.P., KANEKO, T., ZOR, U., FIELD, J.B. and DAVIS, B.B.: Effects of vasopressin and prostaglandin E on the adenyl cyclase-cyclic 3', 5'-adenosine monophosphate systems of the renal medulla of the rat. J. Clin. Invest. 50: 2461-2465, 1971. BEITCH, B.R., BEITCH, 1. and ZADUNAISKY, J.A.: The stimulation of chloride transport by prostaglandins and their interaction with epinephrine, theophylline and cyclic AMP in the corneal epithelium. I. Membr. BioI. 19: 381-396, 1974. BENDER, W.W., GARAN, H., and BERG, H.C.: Protein of human erythrocyte membrane as modified by pronase. J. Mol. BioI. 58: 783-797, 1971. BITO, L.Z.: Accumulation and apparent active transport of prostaglandins by some rabbit tissues in vitro. I. PhysioI. 221: 371-387, 1972a. BITO, L.Z.: Comparative study of concentrative prostaglandin accumulation by various tissues of mammals and marine vertebrates and invertebrates. Compo Biochem. Physiol. 43A: 65-82, 1972b. BITO, L.Z., DA VSON, H. and HOLLINGSWORTH, J.R.: Facilitated transport of prostaglandins across the blood-cerebrospinal fluid and blood-brain barriers. J. Physiol. 256: 273-285, 1976a. BITO, L.Z., DAVSON, H. and SALV AOOR, E.V.: Inhibition of in vitro concentrative prostaglandin accumulation by prostaglandins, prostaglandin analogues and by some inhibitors of organic anion transport. I. Physiol. 256: 257-271, 1976b. BRAQUET, P. GAEA Y, R.P., FROLICHh, I.C. and NICOSIA, S.: Prostaglandins and membrane ion transport, Raven Press, New York, 1985. CABANTCHIK, Z.I. and ROTHSTEIN, A.: Membrane proteins related to anion permeability of human red blood cells. II. Effects of proteolytic enzymes on disulfonic stilbene sites of surface proteins. J. Membr. BioI. 15: 227-248, 1974. 48 CABANTCHIK, Z.l. and ROTIISTEIN, A.: The nature of the membrane sites controlling anion permeability of human red blood cells as determined by studies with disulfonic stilbene derivatives. 1. Membrane. BioI. 10: 311-330, 1972. CHAUDHARI, A. and KIRSCHENBAUM, M.A.: Specific prostagladin E2 binding sites in isolated rat glomeruli: evidence for glomerular PGE receptors. Prostaglandins Leuketrienes and Medicine 20: 55-68, 1985. CSERR, H.F.: Physiology of the choroid plexus. PhysioI. Rev. 51: 273-311, 1971. DAVSON, H.: A comparative study of the aquous humour and cerebrospinal fluid in the rabbit. J. Physiol. 129: 111-133, 1955. DENG, Q.S. and JOHANSON, C.E.: Effects of different temperature pH and pharmacological agents on chloride transport in rat choroid plexus. Fed. Proc.: 43-4697, . DEPONT, J.J., HANSEN, T. and BONTING, S.L.: An anion-sensitive ATPase in lizard gastric mucosa. Biochim. Biophys. Acta 274: 189-200, 1972. DERENZIS, G., and BORNANCIN, M.: A CI-/HC03- ATPase in the gills of carassius auratus: its inhibition by thiocyanate. Biochim. Biophys. Acta 467: 192-207, 1977. DRICKAMER, L.K.: Fragmentation of the 95,000 dalton transmembrane polypeptide in human erythrocyte membranes. Arrangement of the fragments in the lipid bilayer. J. BioI. Chern. 251: 5115-5123, 1976. DUNNETT, C.W.: A multiple comparison procedure for comparing several treatments with a control. 1. Am. Statist. Ass. 50: 1096-1121,1955. EHRENSPECK, G. and BRODSKY, W.A.: Effects of 4-Acetamido-4'-isothiocyano2,2'-disulfonic stilbene on ion transport in turtle bladders. Biochim. Biophys. Acta 419: 555-558, 1976. FASSINA, G., CARPENEOO, F. and SANTI, R.: Effect of prostaglandin El on isolated short-circuit frog skin. Life Sci. 8: 181-187, 1976. FELDMAN, A.M., EPSTEIN, M.H. and BRUSILOW, S.W.: Effect of cholera toxin and prostaglandins on the rat choroid plexus in vitro. Brain Research 167: 119-128, 1979. FLORES, J., WITKUM, P.A., BECKMAN, B. and SHARP, G.W.G.: Stimulation of osmotic water flow in toad bladder by prostaglandin El: Evidence for different compartment of cAMP. J. Clin. Invest. 56: 256-262, 1975. FRIZZELL, R.A., FIELD, M. and SCHULTZ, S.G.: Sodium-coupled chloride transport by epithelial tissues. Am. 1. Physiol. 236: FI-F8, 1979. FUKUDA, M., ESHDAT, Y., TARONE, G., and MARCHESI, V.T.: Isolation and characterization of pep tides derived from the cytoplasma segment of band 3, the predominant intrinsic membrane protein of the human erythrocyte. J. BioI. Chern. 253: 2419-2428, 1978. 49 GRANTHAM, J.J. and ORLOFF, J.: Effect of prostaglandin E1 on the permeability response of the isolated collecting tubule to vasopressin adenosine 3', 5'monophosphate and theophylline. J. Clin. Invest. 47: 1154-1161, 1968. HELD, D., FENCL, V. and PAPPENHEIMER, 1.R.: Electrical potential of cerebrospinal fluid. J. Neurophysiol. 27: 942-959, 1964. HOGBEN, C.A.M., WISTRAND, P. and MAREN, T.H.: Role of active transport of chloride in formation of dogfish cerebrospinal fluid. Am. J. Physiol. 199: 124126, 1960. HONG, S.K., GOLDINGER, J.M., SONG, Y.K., KOSCHIER, F.J. and LEE, S.H.: Effect of SITS on organic anion transport in the rabbit kidney cortical slice. Am. J. Physiol. 234: F302-F307, 1978. HUMPHREYS, M.H. and CHOU, L.Y.N.: Anion-stimulated ATPase activity of brush border from rat small intestine. Am. 1. Physiol. 236: E70-E76, 1979. IMAI, M.: Effect of bumetanide and furosemide on the thick ascending limb of Henle's loop of rabbits and rats perfused in vitro. Eur. 1. Pharmacol. 41: 409-416,1977. JAVAHERI, S. and WEYNE, J.: Effect of the anion blocker 'SITS' on cerebrospinal fluid [HC03-] in acute acid-base perturbations. In Central Neurone Environment, ed. by M.E. Schlafk:e, H.P. Koepchen, and W.R. See, pp. 22-28, Springer-Verlag, 1983. JENKINS, R.E. and TANNER, M.J.A.: The structure of the major protein of the human erythrocyte membrane. Characterization of the intact and major fragments. Biochem.1. 161: 139-147, 1977. JOHANSON, C.E.: Differential effects of acetazolamide, benzolamide and systemic acidosis on hydrogen and bicarbonate gradients across the apical and basolateral membranes of the choroid plexus. J. Pharmacol. Exp. Ther. 231: 502-511, 1984. JOHANSON, C.E., REED, D.J. and WOODBURY, D.M.: Active transport of sodium and potassium by the choroid plexus of the rat. J. Physiol. 241: 359-372, 1974. JOHANSON, C.E., FOLTZ, F.M. and THOMPSON, A.M.: The clearance of urea and sucrose from isotonic and hypertonic fluids perfused through the ventriculocisternal system. Exp. Brain Res. 20: 18-31, 1974. JOHANSON, C.E., PARANDOOSH, Z. and SMITH. Q.R.: CI-HC03 exchange in choroid plexus: analysis by the DMO method for cell pH. Am. J. Physiol. 249: F478-F484, 1985. JOSE, J.G and GASSNER, D.: Chloride and bicarbonate-stimulated ATPase activity in bovine lens epithelium. Ophthalmic Res. 15: 90-98, 1983. KEIFER, D.W. and LUCAS, W.J.: Potassium channels in chara corallina. Control and interaction with the electrogenic H+ pump. Plant Physiol. 69 S(65): 781-788, 1982. 50 KIMELBERG, H.K., BIDDLECOME, S. and BOURKE, R.S.: SITS-inhibitable Cltransport and Na+-dependent H+ production in primary astroglial cultures. Brain Research 173: 111-124,1979. LEIKAUF, G.D., UEKI, I.F., NADEL, J.A. and WIDDICOMBE, lH.: Bradykinin stimulates Cl secretion and prostaglandin E2 release by canine tracheal epithelium. Am. 1 Physiol. 248: F48-F55, 1985. LEYSSAC, P.P., BUKHAVE, K. and FREDERIKSEN, 0.: Inhibitory effect of prostaglandins on isosmotic fluid transport by rat gall bladder in vitro, and its modification by blockade of endogenous PGE biosynthesis with indomethacin. Acta Physiol. Scand. 92: 496-507, 1974. LEYSSAC, P.P., CHRISTENSEN, P., HILL, R. and SKINNER, S.L.: Indomethacin blockade of renal PGE synthesis: effect on total renal and tubular function and plasma renin concentration in hydropenic rats and on their response to isotonic saline. Acta Physiol. Scand. 94: 484-496, 1975. LIN, W.: Inhibition of anion transport in com root protoplasts. Plant Physiol. 68: 435538, 1981. LINO, Y and IMAI, M.: Effect of prostaglandins on Na transport in isolated collecting tubules. Pflugers Arch. 373: 125-132, 1978. LIPSON, L.C. and SHARP, G.W.G.: Effect of prostaglandin E1 on sodium transport and osmotic water flow in the toad bladder. Am. 1 Physiol. 220: 1046-1052, 1971. MACHEN, T.E. and FORTE, J.G.: Anion secretion by gastric mucosa. In Chloride transport coupling in biological membranes and epithelia, ed. by G.A. Gerencser, pp. 415-446, Elsevier, 1984. MADDY, H.: A fluorescent label for the outer components of the plasma membrane. Biochim. Biophys. Acta 88: 390-399, 1964. MAREN, T.H.: Bicarbonate formation in cerebrospinal fluid: role in sodium transport and regulation. Am. 1 Physiol. 222: 885-889, 1972. MURPHY, V.A.: Sodium-hydrogen exchange in the rat choroid plexus, University of Utah, 1984. Ph. D. Thesis. NOBECOURT, P. and VOISIN, R.: Les chlorures du liquide cephalorachidien. Arch. Gen. Med. 192: 3018-3025, 1903. PARDRIDGE, W.M., FRANK, H.lL., CORNFORD, E.M., BRAUN, L.D., CRANE, P.D. and OLDENDORF, W.H.: Neuropeptides and the blood-brain barrier. In Neurosecretion and brain peptides, ed. by J.B. Martin, S. Reichlin, and K.L. Bick, pp. 21-25, Raven Press, New York, 1981. PASSOW, H. and ZAKI, L.: Studies on the molecular mechanism of anion transport across the red blood cell membrane. In Molecular specialization and symmetry in membrane function, ed. by A.K Solomon and M. Karnovsky, Harvard Univ. Press, Cambridge, MA., 1978. 51 PELLEGRINO, L.J. and CUSHMAN, A. J.: A stereotaxic atlas of the rat brain, Meredith Publishing Company, New York, 1967. PHILLIPS, J.F. and HANRAHAN, J.W.: Chloride transport in insects: a novel Kstimulated electrogenic pump. In Chloride transport coupling in biological membranes and epithelia, ed. by G.A. Gerencser, pp. 151-182, Elsevier, 1984. RANDOLPH, L.K. and CIMINERA, J.L.: Statistics. In Remington's Pharmaceutical Sciences, ed. by A.R. Gennaro, pp. 104-139, Mack Publishing Company, 17th edition, 1985. ROTHSTEIN, A., CABANTCHIK, Z.I., and KNAUF, P.: Mechanisms of anion transport in red blood cells: Role of membrane protein. Fed. Proc., Fed. Am. Soc. Exp. BioI. 35: 3-10, 1976. ROTHSTEIN, A., CABANTCHIK, Z.I., BALSHIN, M., and JULIANO, R.: Enhancement of anion permeability in lecithin vesicles by hydrophobic proteins extracted fron red blood cell membranes. Biochem. Biophys. Res. Commun. 64: 144-150, 1975. RUSSELL, J.M. and BORON, W.F.: Role of chloride transport in regulation of intracellular pH. Nature 264: 73-74, 1976. SACHS, G., FALLER, L.D. and RABON, E.: Proton I hydroxy transport in gastric and intestinal epithelia. J. Membr. BioI. 64: 123-135, 1982. SAITO, Y. and WRIGHT, E.M.: Bicarbonate transport across the frog choroid plexus and its control by cyclic nucleotides. J. PhysioI. 336: 635-648, 1983. SEGAL, M.B. and POLLAY, M.: The secretion of cerebrospinal fluid. Exp. Eye. Res. 25(supp): 127-148, 1977. SMITH, Q.R.: Uptake of chloride-36 and sodium-22 by the choroid plexuscherebrospinal fluid-brain system of the immature and mature rat: evidence for active chloride transport by the choroid ependyma, Ph. D. Thesis, University of Utah, 1980. SMITH, Q.R. and JOHANSON, C.E.: Active transport of chloride by lateral ventricle choroid plexus of the rat. Am. J. Physiol. 249: F470-F477, 1985. SMITH, Q.R. and JOHANSON, C.E.: Effects of ouabain and potassium on ion concentrations in the choroidal epithelium. Am. J. PhysioI. 238: F399-F406, 1980. STECK, T.L., RAMOS, B., and STRAPAZON, E.: Proteolytic dissection of band 3, the predominant transmembrane polypeptide of the human erythrocyte membrane. Biochemistry 15: 1154-1161, 1976. STOCHES, J.B.: Effect of prostaglandin E2 on chloride transport across the rabbit thick ascending limb of Henle. J. Clin. Invest. 64: 495-502, 1979. THOMAS, R.C.: The role of bicarbonate, chloride and sodium ions in the regulation of intracellular pH in snail neurones. J. Physiol. 273: 317-338, 1976. 52 VAN AMELSVOORT, J.M., DEPONT, J.J. and BONTING, S.L.: Is there a plasma membrane-located anion-sensitive ATPase? Biochim. Biophys. Acta 466: 283301, 1977. VATES, T.S., BONTING, S.L. and OPPELT, W.W.: Na-K activated adenosine triphosphatase formation of cerebrospinal fluid in the cat. Am. J. Physiol. 206: 1165-1172, 1964. WARNOCK, D.G. and EVELOFF, J.: NaCl entry mechanisms in the luminal membrane of the renal tubule. Am. J. Physiol. 242: F561-F574, 1982. WHITE, J.F.: Bicarbonate-dependent chloride absorption in small intestine: ion fluxes and intracellular chloride activities. J. Membr.Biol. 53: 95-107, 1980. WOLFE, L.S.: Eicosanoids: prostaglandins, thromboxanes, leukotrienes, and other derivatives of carbon 20 unsaturated fatty acids. J. Neurochem. 38: 1-14, 1982. WRIGHT, E.M.: Mechanisms of ion transport across the choroid plexus. J. Physiol. 226: 545-571, 1972. ZEUTHEN, T. and WRIGHT, E.M.: An electrogenic Na+/K+ pump in choroid plexus. Biochim. Biophys. Acta 511: 517-522, 1978.