Central administration of benzodiazepines alters water absorption by the rat ileum in vivo

GASTROENTEROLOGY 1987;93:330-4

Central Administration of Benzodiazepines Alters Water Absorption by the Rat Ileum In Vivo R. FOGEL, G. MICHELSON, and T. GAGINELLA

T. SENLER,

D. MARSHALL,

Department of Medicine and Department of Surgery, University Kentucky and Department of Pharmacology, Hoffmann-LaRoche

Two types of benzodiazepine receptors have been identified in the central nervous system. The aim of these experiments was to determine if ligands for these receptors alter basal water absorption by rat ileum in vivo after central administration. Specifically, the effects on net water flux of the systemic and central administration of diazepam and the central administration of RO 5-4864, a “peripheral” receptor agonist, and of the “central” receptor agonists clonazepam and lorazepam were determined. Diazepam increased absorption at 4.3 mg/250 g body wt i.p. but not at 430 pgl250 g body wt. Intracerebroventricular diazepam (28 pg) increased water absorption. Larger doses had a greater effect. Intracerebroventricular RO 5-4864 (100 pg) increased net water absorption; intracerebroventricular lorazepam (50 or 100 pg) or clonazepam (100 pg) reduced basal water qbsorption. Systemic atropine (2 mg/kg iv.) abolished the effect of lorazepam (100 pg i.c.v.). To evaluate the possibility that diazepam and RO 5-4864 have effects similar to those of calcium channel antagonists, nifedipine, nitrendipine, and diltiazem were administered intracerebroventriculariy. The dihydropyridine calcium channel antagonists nifedipine and nitrendipine increased basal water absorption. Diltiazem, a benzothiazepine comReceived June 13, 1986. Accepted March 2, 1987. Address requests for reprints to: Ronald P. Fogel, M.D., Division of Gastroenterology, Ambulatory Care Building, 550 South Jackson Street, University of Louisville, Louisville, Kentucky 40292. This work was supported by a grant from Hoffmann-LaRoche, Inc. A portion of this work was published in abstract form (Gastroenterology 1985;88:1385) and presented at the annual meeting of the American Gastroenterological Association, May 1985. The authors thank Sheila Carpenter for secretarial assistance. Dr. Gaginella’s present address is: Division of Gastroenterology, University Hospital, N-249, Doan Hall, Ohio State University, Columbus, Ohio. 0 1987 by the American Gastroenterological Association 0016-5065/87/$3.50

T. BROWN,

of Louisville, Louisville, Inc., Nutley, New Jersey

pound, did not alter basal water absorption. We conclude that the binding of benzodiazepine agonists to receptors located in the central nervous system alters net water absorption by the rat ileum. Agonists of the central benzodiazepine receptor reduce basal water absorption via a choiinergic neural pathway. Peripheral agonists increase net water absorption. In this model, diazepam behaves as a peripheral receptor agonist. This study provides further evidence of a role for the central nervous system in the regulation of intestinal absorption of water and ions. Ijenzodiazepines (BZs) have been used extensively for the treatment of anxiety states and sleep disorders (1). The clinical use of BZs in the treatment of gastrointestinal disorders was based on the assumption that a reduction in anxiety would improve gastrointestinal symptoms or alter the perception of the symptoms. Indeed, File and Pearce (2) found that chlordiazepoxide and lorazepam prevented cold water-induced gastric ulceration in restrained rats. Oral or parenteral diazepam reduced basal gastric acid output in humans (3). An action in the central nervous system rather than a systemic effect was thought to be causal. Benzodiazepine receptors have been described in the brain (4-6). The classical “central” receptor, located on neurons, is coupled to the y-aminobutyric acid (GABA) receptor and binds diazepam, clonazepam, and lorazepam with high affinity (4,6-8). The binding of a BZ agonist potentiates GABA inhibition of central neurons (8). A second receptor type has been recognized with (a) low affinity for clonazepam, (b) high affinity for RO 5-4864, and (c) no interaction with the GABA receptor (9). This was designated as Abbreviations used in this paper: BZ, benzodiazepine; y-aminobutyric acid; i.c.v., intracerebroventricular.

GABA,

August 1987

a “peripheral” receptor because it was described initially in noncentral nervous system tissues, e.g., kidney, heart, adrenal cortex, and liver (9). Recently, the peripheral receptor has been identified in rat brain (10-12). The central BZ receptor mediates the anxiolytic effects of BZs (13). The role of the peripheral receptor located in the brain is unknown. Recent evidence suggests that agonists of the peripheral BZ receptor change nerve activity by inhibiting calcium-dependent neuronal processes. Blockade of calcium channels and inhibition of brain membrane calcium-calmodulin protein kinase activity have been reported (14~15). The hypothesis that BZs might alter intestinal water and ion absorption resulted from the finding that the central administration of a GABA agonist (muscimol) decreased basal water and ion absorption by the rat ileum in vivo, whereas a GABA antagonist (bicuculline) increased absorption (16). The interaction between GABA and BZ receptors (8) caused us to postulate that central BZ receptor ligands might alter water and ion absorption by the rat ileum in vivo. In the present studies we investigated the effects of intracerebroventricular injection of agonists of the central and peripheral BZ receptor on basal water and ion absorption by the rat ileum in vivo.

Materials and Methods In vivo transport studies using the single-pass perfusion technique were performed as described previrats (Laboratory Supply, ously (16).Male Sprague-Dawley Indianapolis, Ind.), weighing 200-300 g, were maintained on a standard laboratory rat diet with free access to water. Animals were anesthetized with sodium pentobarbital (65 mgikg, Butler Co., Columbus, Ohio]. Ileal loops were lo-15 cm long and ended 10 cm proximal to the cecum. The loops were washed with warm saline and were cannulated. Balanced electrolyte solution was perfused at a constant temperature (37°C) and rate (0.5 ml/min) with a Wiz perfusion pump (ISCO Inc., Lincoln, Neb.). Body temperature was maintained with a thermocouple-controlled heating lamp (Thermostemp Temperature Controller, Yellow Springs Instrument Co., Yellow Springs, Ohio). The balanced electrolyte perfusing solution contained (mM): 115 NaCl, 25 NaHCO,, 2.4 K2HP04, 0.4 KHZP04, 1.2 CaCl,, and 1.2 MgCl*. Unlabeled polyethylene glycol (mol wt 4000; 2 g/L) and [14C]polyethylene glycol (New England Nuclear, Boston, Mass.) were used as nonabsorbable volume markers. The solution was constantly percolated with 95% O,-5% CO2 to maintain pH 7.4. After an initial 80-min steady-state period, basal intestinal transport was measured during three 20-min periods. Diazepam was injected intraperitoneally and transport was measured for 120 min. In some animals, BZs or calcium channel antagonists, or both, were injected into the lateral ventricles of the brain.

BENZODIAZEPINES AND WATER ABSORPTION 331

After the measurement of basal transport, rats were placed in a stereotaxic frame. Intracerebroventricular injection (i.c.v.) was performed using the coordinates of Pellegrino et al. (17). The junction of the saggital and coronal sutures was the zero point. The ventricle was punctured at a point 2 mm lateral to the zero point. At a depth of 3.5 mm, cerebrospinal fluid was aspirated before each injection. The total volume injected was 10 ~1. Another group of animals received systemic atropine (Z mg/kg) before i.c.v. lorazepam (100 pg). Atropine, alone, did not alter basal transport (13.5 rfI 3.6 vs. 14.2 t 3.2 @l/cm per 20 min before and after atropine, respectively, n = 6). Ileal transport was measured for three 20-min periods after injection. Water and electrolyte transport were calculated using standard formulas and are expressed as microliters or milliequivalents per centimeter length of loop per 20 min. Net absorption from the lumen was expressed as a positive value. Mean values were obtained by averaging the results from the multiple collection periods for each perfusion. Drugs were dissolved in dimethyl sulfoxide or saline. Neither saline nor dimethyl sulfoxide, alone, alter transport during the hour after i.c.v. injection (14.8 2 2.1 vs. 18.4 i 3.4 @cm per 20 min, before and after saline, respectively, n = 8; 17.1 t 3.8 vs. 19.1 ‘-c3.9 @cm per 20 min before and after dimethyl sulfoxide, respectively, n = 12). At the completion of the perfusion studies, the length of the loop was measured under 15 g of tension. The recovery of polyethylene glycol was determined using a scintillation counter (Packard Tri Carb, Packard Instruments Inc., Downers Grove, Ill). Periods in which polyethylene glycol recovery was not 100% ? 5% were discarded. This occurred in ~3% of the periods. Animals were not included if more than two periods were discarded. Quench correction was performed by the method of external standards. Sodium ion concentration was measured with an ion-selective electrode (Technicon C800, Technicon Instrument Corp., Tarrytown, N.Y.).

Drugs Diazepam, RO 5-4864, lorazepam, clonazepam, nifedipine, nitrendipine, and diltiazem were obtained from Hoffmann-LaRoche, Nutley, N.J.

Statistical

Analysis

Statistical analyses were performed using twotailed Student’s t-test for paired or unpaired data (18). All results are expressed

as the mean -t SEM.

Results The effects of systemic and of i.c.v. diazepam on water absorption are shown in Figures 1 and 2, respectively. Intraperitoneal administration of diazepam at 430 pgl250 g body wt did not change net water transport. A lo-fold larger dose (4.3 mg/250 g i.p.) increased water absorption. That effect was seen in the first hour after diazepam administration and was constant during the 2 h of study. The alteration of intestinal absorption was attributed to a central

332

GASTROENTEROLOGY Vol. 93, No. 2

FOGEL ET AL.

effect of diazepam, because doses of diazepam that were ineffective if administered systemically incre:!sed water absorption after i.c.v. injection. The lowest central dose of diazepam that increased water absorption was 28 pg. Diazepam (100 pg) significantly increased water (Figure 2) and sodium absorption (2.0 ? 0.3 vs. 3.1 ? 0.3 pEq/cm per 20 min, before and after diazepam, respectively; n = 5, p < 0.05). To determine if the diazepam effect was due to an action at the central or the peripheral receptor, we studied the effects of the central receptor agonists lorazepam and clonazepam, and of the peripheral BZ receptor agonist RO 5-4864. Lorazepam (100 pg) and clonazepam (100 pg) reduced net water absorption (Figure 3). Lorazepam (50 pg) also reduced net water absorption (16.6 2 2.4 vs. 8.6 IfI 2.0 pllcm per 20 min, before and after lorazepam, respectively; n = 11, p < 0.01). The magnitude of the reduction in Diazepam

Diazepam

430 pg IP N.S.

4.3

mg IP

p co.01 15

ia

-

Figure

0

Pre Drug

a

Post

Drug

1. Effect of intraperitoneal (IP) diazepam on net intestinal water transport in the rat ileum. Intestinal transport was measured before and after intraperitoneal bolus administration of diazepam. The dose of diazepam was per 250 g body wt. The number of animals studied is indicated in the bars. A t-test of water transport before and after drug administration in the same loop provided the data for the p values shown. N.S., not significant.

dose tcs)

-2-1

Figure 2. Effect on net intestinal water transport in the rat ileum of diazepam administered into the lateral cerebral ventricle of the brain. Intestinal transport was measured before and after intracerebroventricular bolus administration of diazepam. The change in net water transport was the difference between the rate of transport after diazepam and the rate before diazepam. Numbers in parentheses represent the number of animals studied. Each point represents the mean 2 SEM. p < 0.05 for all doses studied, except 2.8 pg.

water absorption after 50 or 100 pg of lorazepam was the same (8.1 ? 2.0 vs. 8.1 ? 2.1 @cm per 20 min reduction, due to 50 and 100 pg, respectively). Lorazepam, 10 pg, did not alter water absorption (24.1 + 4.9 vs. 22.7 + 4.6 pi/cm per 20 min, before and after lorazepam, respectively; n = 5, p > 0.05). The lorazepam (100 pg) effect was prevented by systemic atropine (2 mg/kg) (11.3 + 1.5 vs. 14.9 + 5.1 @cm per 20 min before and after lorazepam/atropine respectively; n = 5, p > 0.05). Central injection of the peripheral BZ agonist R05-4864 (100 pg) enhanced water absorption (4.1 + 1.1 vs. 10.7 ? 2.1 &cm per 20 min, before and after RO 5-4864, respectively; n = 7, p < 0.05). The dihydropyridine compounds nifedipine and nitrendipine enhanced basal water absorption after i.c.v. administration (Figure 4). In contrast, diltiazem, a benzothiazepine calcium channel antagonist, did not alter water transport (14.7 -C 3.2 vs. 12.6 ? 3.8 pi/cm per 20 min, before and after diltiazem, respectively; n = 7, p > 0.05). To determine if the effects of diazepam and nifedipine were additive, 2.8 pg of diazepam, a subthreshold dose, was administered intracerebroventricularly with 100 pg of nifedipine. Intestinal water absorption was increased (17.1 t 1.6 vs. 22.8 -+ 1.8 &cm per 20 min, before and after diazepam/nifedipine, respectively; n = 9, p < 0.01) but the magnitude of the change was not different from nifedipine alone (5.7 +_ 1.7 vs. 6.2

August 1987

BENZODIAZEPINES AND WATER ABSORPTION

p< 3

LORAZEPAM

lOOJIgICV

0.01

0

PRE DRUG

m

POST DRUG

CLONAZEPAM 1OOrg ICV

p < 0.0 1

Figure 3. Effect of intracerebroventricular (XV) lorazepam and clonazepam on net water transport in the rat ileum. Basal water absorption was measured for 60 min before and after intracerebroventricular injection. Bars represent mean 2 SEM of the animals studied. The number of animals studied is indicated in the bars. A t-test of the water transport before and after drug administration in the same loop provided the data for the p values shown.

333

in nonneuronal tissue, but subsequently it was found in the central nervous system ($10-12). The peripheral receptor binds RO 5-4864 with high affinity; clonazepam binds with low affinity. Our results suggest that the central and peripheral BZ receptor agonists have different effects on mucosal absorptive processes. Clonazepam and lorazepam, high affinity agonists of the central receptor, reduce basal water absorption. The mechanism by which these central agonists alter water absorption is unknown. Clonazepam and lorazepam bind to the complex GABA-BZ receptor and enhance the effect of GABA (8).Atropine prevented the reduction in net water absorption due to lorazepam. Previously we reported that atropine prevented intestinal secretion due to muscimol, a GABA agonist (16).It can be postulated that cholinergic nerves mediate the effects of both central BZ receptor agonists and muscimol. Whether lorazepam promotes GABA release or directly alters cholinergic innervation of the intestine has not been studied. RO 5-4864, a peripheral receptor agonist, enhances absorption. In this study, diazepam appears to act as a peripheral, rather than a central, receptor Nifedipine (100

Nitrendipine

pg)

p < 0.00

1

Pre

Drug

25

(100

pg)

D < 0.01

t 2.2 pllcm per 20 min, for diazepam/nifedipine and nifedipine alone, respectively, n = 9, p > 0.05).

Discussion The central nervous system regulation of intestinal water and ion absorption has not been characterized extensively. The central administration of enkephalins enhanced basal water absorption (19). The GABA receptor agonist muscimol reduced water absorption, whereas the antagonist bicuculline en(16).We now report that BZ hanced absorption receptor agonists alter rat ileal basal water absorption in vivo after central injection. This conclusion is based on the findings that (a) less diazepam was required to alter water absorption after central injection than after systemic administration and (b) other BZ receptor agonists alter intestinal water absorption after central administration. Benzodiazepine receptors have been identified in the brain. The central neuronal BZ receptor binds diazepam, clonazepam, and lorazepam with high affinity, but has lower affinity for RO 5-4864 (4).The BZ receptor forms a larger complex with the GABA receptor (8).The anticonvulsant effect of BZs is attributed to ligand binding to this receptor complex and enhanced GABA neurotransmission (8).A peripheral BZ receptor originally was reported to occur

0

?? Post

Drug

Figure 4. Effect of intracerebroventricular injection of nifedipine and nitrendipine on net water transport in the rat ileum. Basal water absorption was measured for 60 min before and after intracerebroventricular injection. Bars represent mean ? SEM of the animals studied. The number of animals studied is indicated in the bars. A t-test of the water transport before and after drug administration in the same loop provided the data for the p values shown.

334

GASTROENTEROLOGY Vol. 93, No. 2

FOGEL ET AL.

agonist. Water absorption was increased after i.c.v. diazepam, as occurred after RO 5-4864,rather than being decreased, as it was after i.c.v. clonazepam or lorazepam. The actions of the peripheral BZ receptor agonists are not readily explained. We speculate that the “peripheral” agonists alter central neural activity, which changes neural outflow to the intestinal mucosa and thus mucosal water and ion absorption. Neurotransmitter release is a calcium-dependent process (20). RO 5-4864interferes with neurotransmission by (a) calcium-calmodulin protein kinase inhibition (15) and (b) inhibition of calcium uptake (14,21). Cantor et al. in synaptosomal preparations (22) reported that dihydropyridine-type calcium channel antagonists displaced RO 5-4864 from membrane receptors in brain, kidney, and heart; central receptor ligands flurazepam and clonazepam were not displaced. The observation that nifedipine and nitrendipine, dihydropyridine compounds, increased water absorption, similar to diazepam and RO 5-4864,suggests that these agents might function to interfere with a calcium-dependent process. Diltiazem did not alter water absorption, suggesting that calcium channel blockade may not be the mechanism of action, However, diltiazem does not displace RO 5-4864 from brain receptors whereas nifedipine does. The cell binding site may be important in producing the central effects of calcium channel antagonists on water absorption. Nifedipine has actions in addition to the blockade of slow calcium channels (23,24). The possibility that peripheral BZ receptor agonists interfere with these other actions, e.g., inhibition of calcium release from intracellular stores (23)or stimulation of calcium extrusion (24),has not been investigated. The simultaneous i.c.v. administration of nifedipine and diazepam did not provide additional information about the mechanism of action of diazepam. At 100 pg each, the combination of diazepam and nifedipine killed all the rats. A subthreshhold dose of diazepam (2.8pg)did not influence the nifedipine effect. In conclusion, this study provides additional evidence of central regulation of intestinal mucosal water absorption. Intracerebroventricular administration of BZs alters basal water absorption by the rat ileum, in vivo. Central BZ receptor agonists reduce basal water absorption by a cholinergic mechanism. RO 5-4864 and diazepam, acting as peripheral BZ agonists, increase water absorption. The mechanism of action for peripheral agonists is unknown. Additional studies with ligands for BZ receptors are warranted to determine their value for treatment of intestinal abnormalities of water and ion absorption.

References 1. Greenblatt DJ, Shrader RI, Abernethy DR. Current status of benzodiazepines (Part 2). N Engl J Med 1983;309:418-6. 2. File SE, Pearce JB. Benzodiazepines reduce gastric ulcers induced in rats by stress. Br J Pharm 1981;74:593-9. 3. Birnbaum D, Karmeli F, Tefera M. The effect of diazepam on human gastric secretion. Gut 1971;12:616-8. 4. Squires RF, Braestrup C. Benzodiazepine receptors in rat brain. Nature 1977;266:732-4. 5. Braestrup C, Squires RF. Specific benzodiazepine receptors in rat brain characterized by high-affinity [3H] diazepam binding. Proc Nat1 Acad Sci USA 1977;74:3805-9. 6. Mohler H, Okada T. Benzodiazepine receptor: demonstration in the central nervous system. Science 1977;198:849-51. 7. Braestrup C, Nielsen M, Biggio G, Squires RF. Neuronal localization of benzodiazepine receptors in cerebellum. Neurosci Lett 1979;13:219-24. 8. Tallman JF, Gallager DW. The GABA-ergic system: a locus of benzodiazepine action. Ann Rev Neurosci 1985;8:21-44. 9. Pellow S, File SE. Minireview behavioral actions of RO 5-4864: a peripheral-type benzodiazepine? Life Sci 1984; 35:229-40. 10. Anholt RRH, deSouza EB, Oster-Granite ML, Snyder SH. Peripheral-type benzodiazepine receptors: autoradiographic localization in whole body sections of neonatal rats. J Pharm Exp Ther 1985;233:517-26. 11. Schoemaker H, Bliss M, Yamamura HI. Specific high-affinity saturable binding of [3H] RO 5-4864 to benzodiazepine binding sites in the rat cerebral cortex. Eur J Pharmacol 1981; 71:173-5. 12. Schoemaker H, Boles RG, Horst D, Yamamura HI. Specific high-affinity binding sites for [3H] RO 5-4864 in rat brain and kidney. J Pharm Exp Ther 1983;225:61-9. 13. Tallman JF, Paul SM, Skolnick P, Gallager DW. Receptors for the age of anxiety: pharmacology of the benzodiazepines. Science 1980:207:274-81. 14. Taft WC, DeLorenzo RJ. Micromolar-affinity benzodiazepine receptors regulate voltage-sensitive calcium channels in nerve terminal preparations. Proc Nat1 Acad Sci USA 1984;81:3118-22. 15. DeLorenzo RJ, Burdette S, Holderness J. Benzodiazepine inhibition of the calcium-calmodulin protein kinase system in brain membrane. Science 1981;213:546-8. 16. Fogel R, Kaplan RB, Arbit E. Central action of y-aminobutyric acid ligands to alter basal water and electrolyte absorption in the rat ileum. Gastroenterology 1985;88:523-30. 17. Pellegrino LJ, Pellegrino AS, Cushman AJ. A stereotaxic atlas of the rat brain. New York: Plenum, 1979. 18. Snedecor GW, Cochran WG. Statistical methods, 6th ed. Ames, Iowa: Iowa State University Press, 1967. 19. Brown DR, Miller RJ. CNS involvement in the antisecretory action of (Met’) enkephalinamide on the rat intestine. Eur J Pharmacol 1983;90:441-4. 20. Blaustein MP. Effects of potassium, veratridine and scorpion venom on calcium accumulation and transmitter release by nerve terminals in vitro. J Physiol 1975;247:617-55. 21. Ferrendelli JA, Daniels-McQueen S. Comparative actions of phenytoin and other anticonvulsant drugs on potassium- and veratridine-stimulated calcium uptake in synaptosomes. J Pharm Exp Ther 1982;220:29-34. 22. Cantor EH, Kenessey A, Semenuk G, Spector S. Interaction of calcium channel blockers with non-neuronal benzodiazepine binding sites. Proc Nat1 Acad Sci USA 1984;81:1549-52, 23. Walus KM, Fondacaro JD, Jacobson ED. Effects of calcium and its antagonists on the canine mesenteric circulation. Circ Res 1981;48:692-700. 24. Church J, Zsoter TT. Calcium antagonistic drugs. Mechanism of action. Can J Physiol Pharmacol 1980;58:254-64.