Brain Research, 54~ (1991)322 .325 Elsevier A DONIS I)(Xl68993912463t)Y

322

BRES 24639

Inhibition of Na+/Ca 2+ exchange enhances delayed neuronal death elicited by glutamate in cerebetlar granule cell cultures N. A n d r e e v a 1, B. Khodorov 2, E. Stelmashook 1, E. Cragoe Jr. 3 and I. Victorov ~ 1Brain Research Institute, and 21nstitute of General Pathology and Pathophysiology, Academy of Medical Sciences, Moscow (U. S. S.R.) and 3p.o. Box 1548, Nacogdoches, TX 75963-1548 (U.S.A.) (Accepted 22 January 1991) Key words: Na+/Ca2÷ exchange; Glutamate; 3",4"-Dichlorobenzamil;5-(N-4-chlorobenzyi)-2",4"-dimethylbenzamil;Cerebellar granule cell

Experiments have been carried out on the primary cerebellar granule cell cultures from 7- to 8-day-old Wistar rats. To study a possible contribution of Na+/Ca2+ exchange to the toxic effect of glutamate, two amiloride derivatives, 3",4"-dichlorobenzamil (DCB) and 5-(N-4-chlorobenzyl)-2",4"-dimethylbenzamil(CBDMB), known to be the potent inhibitors of this exchange system, were used. Addition of DCB or CBDMB (at 30 and 10/~M, respectively) to a 25/~M glutamate solution dramatically enhanced the delayed neuronal death observed during the 4 h after termination of glutamate treatment. Similar but insignificantly smaller effects were obtained when these agents were added to the cultures in the post-glutamate period. Removal of Na÷ (by substituting for choline chloride) from the external Mg2+-free solution in the post-glutamate period also enhanced a delayed neuronal damage. The data obtained suggest that Na+/Ca2+ exchanger does not constitute the route for Ca2÷ entry during the post-glutamate period but, on the contrary, attenuates glutamate neurotoxicity providing Ca2+ extrusion from the cells under the conditions of a sustained Ca2÷ influx.

Over the last few years progress has been made in studies of cellular mechanisms by which an 'acute' (several minutes) exposure of nerve cells to the excitatory transmitter glutamate can produce delayed neuronal injury 3-6'9'14. Thus, it is established that the cascade of processes leading to delayed nerve cell death begins with the 'fast' influx of Ca 2+ and Na ÷ via glutamate-activated ionic channels 1'4, and includes stable activation of protein kinase C associated with a sustained increase in the cytosolic concentration of free Ca2+(Cai) 14. This elevation of [Ca2+]i is apparently due to an enhanced influx of Ca/+ . The ion transport system responsible for this process has not yet been determined. Among different possible Ca 2÷ transporters that could provide this sustained Ca 2÷ influx, the Na+/Ca 2+ exchange system was recently considered by Manev et al. TM. This idea motivated the present study. In order to block Na+/Ca 2÷ exchange in the cerebellar granule cells during glutamate treatment and/or in the post-glutamate period, two derivatives of amiloride with this property w e r e used 12"13. The results using these two compounds, 3",4"-dichlorobenzamil (DCB) and 5-(N4-chlorobenzyl)-2",4"-dimethylbenzamil (CBDMB), are seen in Fig. 1. In addition, we have examined the effects of Na ÷ removal (by substituting choline chloride) during the post-glutamate period.

The results obtained clearly show that blockade of the Na+/Ca 2÷ exchange system strongly enhances the development of glutamate-induced delayed neuronal death, suggesting the involvement of the Na+/Ca 2+ exchanger in Ca 2÷ extrusion both during glutamate treatment and after its termination. Primary cultures o f cerebellar granule cells. Dissociated cultures were prepared from the cerebella of 7- to 8-day-old Wistar rats. The cerebella were washed with Ca 2÷- and Mg2÷-free salt solution (CMF) 16 and incubated in CMF containing 0.05% trypsin, 0.02% EDTA, 0.0004% DNAase, 3 mg/ml BSA and 0.8% glucose (15 min, 37 °C). After incubation the tissue was washed twice with a solution containing 0.3% BSA and 1% fetal calf serum and dissociated by repeated pipetting in a nutrient medium of the following composition: human placental serum (5%), fetal calf serum (5%), minimal essential medium (Eagle) (90%), glucose (0.8%), glutamine (2 mM), insulin (0.1 U/ml) and HEPES (10 mM). After mild centrifugation (1000 rpm, 1 min) the cells were resuspended in the required volume of the nutrient medium. Then 0.1 ml of cell suspension (106-5 x 10 6 cells/ml) was applied to poly-L-lysine- or collagen-coated glass coverslips 11 and cultured in plastic Petri dishes (40 mm in diameter) placed in CO2-incubator (5% COz, 95%

Correspondence." B. Khodorov, Institute of General Pathology and Pathophysiology,Baltiiskaya St., 8, Moscow, 125315, U.S.S.R.

323 0

II

TABLE I

(DCB)

H2N~NLNH 2 NH2

Effects of Na+/Ca2+ exchange blockade by amiloride derivatives or removal of Na +from external solution

--\Cl

In all cases, P < 0.001 when compared with corresponding group treated with 25/tM glutamate only. The number of dead cells in untreated control cultures was 9.0 + 2.6%. B

0 " N=,] "' " -- NH- - C 2~__> H ~ z-'C H 3 CI . . ~ N y C-

(CBDMB)

H

Fig. 1. Structure of amiloride derivatives, blockers of Na+/ea 2+ exchanger. A: 3",4"-dichlorobenzamil (DCB). B: 5-(N-4-chlorobenzyl)-2",4"-dimethylbenzamil (CBDMB).

air and 98% relative humidity, 35.5 + 0.5 °C) for 7-8 days. To prevent proliferation of non-neuronal cells, arabinosine monocytoside (10 a M ) was added to the cell cultures on the second day in vitro. A m i l o r i d e analogs. T h e amiloride analogs, DCB and CBDMB, were synthesized for this study by the general methods described earlier 7. Culture treatment with glutamate and amiloride derivatives. Experiments were carried out with 7- to 8-day-old

cultures. The initial culture medium was collected and the monolayers were washed once with Mg2+-free balanced salt solution. Mg 2÷ was omitted from all glutamate-containing solutions in order to potentiate the contribution of N M D A channels to the glutamate effect 17. The cultures were divided into 3 groups: the control cultures were incubated in the MgE+-free solution for 15 min at room temperature, the second group of cultures were exposed to glutamate (25 a M ) for 15 min, and the third experimental group of cultures was exposed to glutamate-containing solution with addition of 30 a M DCB or 10/~M CBDMB for 15 min at room temperature. After the exposure to glutamate and an amiloride derivative, the cultures were washed twice with Mg 2+free solution to remove glutamate, returned to the initial culture medium and maintained for 4 h in a CO 2incubator. In the other experimental approach, the cultures were treated with DCB or CBDMB just after glutamate exposure. In this instance, the cultures were challenged with glutamate (25 a M , 15 min) then washed twice with Mg2+-free solution and incubated for 4 h with 30 a M DCB or 10 a M CBDMB which were added either to the culture medium or to MgZ+-free solution. No significant difference has been observed in the number of dead cells observed in the cultures incubated in the initial culture medium with DCB or CBDMB and those incubated in Mg2+-free solution with these drugs for a 4-h period. In

Condition

% of dead cells Application Application during glumduring postmate treatment glutamate period

Glutamate 25pM 35.2 + 4.4 plus DCB 30pM 90.1 + 1.7 plus CBDMB 10/~M 88.6 + 3.2 minus Na ÷ -

70.7 + 6.6 62.3 + 4.3 75.2 + 4.6

Application during glummate treatment and in post-glutamateperiod

89.6 + 3.9 96.6 + 0.6 -

some experiments, DCB or CBDMB were added to glutamate-containing Mg2+-free solution, as well as to the culture medium during the post-glutamate period. Since DCB and CBDMB were dissolved in dimethyl sulfoxide, equal concentrations of this solvent were added to all the cultures (control and glutamate-treated) during the appropriate periods of incubation. Mg2+-free salt solution had the following composition (in mM): NaCI 137, KCI 5.0, Na2HPO 4 0.035, N a H C O 3 12, CaCI 2 2.3, glucose 11; pH 7.6. In the next series of experiments, glutamate-treated (25 a M during 15 min) cultures were washed with Na ÷and Mg2÷-free solution (in mM: choline 120, KCI 5.4, CaCI 2 2.3, glucose 15, Tris-Cl 25; pH 7.4-7.6) and incubated in this medium for 4 h in a CO2-incubator. After the 4-h period of incubation sufficient for the development of a delayed neuronal death, all the cultures were fixed with an ethanol-formaldehyde-acetic acid (7:2:1) mixture and stained with vanadium hematoxiline is. The reliability of the results obtained by this method was confirmed by the parallel use in some experiments of the fluorescent staining method with propidium iodide and fluorescein diacetate 9. The percent of damaged neurons was determined by counting the intact and pyknotic nuclei of the granule cells in 20 view fields (common surface 2.8 mm 2) chosen at random from each monolayer. The significance of differences was estimated by Student's t-test. In the intact 7- to 8-day-old cultures, the cerebellar granule cells are easily distinguished in phase contrast due to their small size (6-9 a m in diameter), round or oval shape and highly refracticle cell bodies 2. During glutamate exposure, acute cell swelling and the appearance of dark granules in neuronal somata were observed. Two hours after termination of glutamate exposure bright

324

w

|

|

e Q g

O ~-

° a

-Am

b

Fig. 2. Cerebellar granule cells in monolayer culture. Seven days in vitro. Vanadium hematoxilin staining. Bar = 20 #M. a: control culture, nuclei of the intact cells (NN) are seen. b: the culture exposed to 25 pM glutamate. Both intact (NN) and damaged ceils (arrows) are determined, c: culture exposed to 25 ¢tM glutamate and 30 pM DCB. Arrows indicate the pyknotic nuclei of the dead cells.

and small (3.0-3.5 p M in diameter) pyknotic nuclei began to appear. In hematoxilin-stained preparations glutamate-induced neuronal death manifested itself by the appearance of dark colored compact pyknotic nuclei (Fig. 2). The ratio of the number of these nuclei to the number of intact and damaged cells was considered as the index of glutamate-induced neurotoxicity. The percent of pyknotic nuclei in the cultures exposed to 25 p M glutamate did not exceed 50% of all granule cells (mean 35%, Table I).

(1) Effect of DCB and CBDMB on glutamate-treated granule cells. In our experiments, DCB and CBDMB were employed at concentrations (30 p M and 10 pM, respectively) corresponding to Ko.5 for blockage of Na+/Ca 2÷ exchange. When added to glutamate-containing Mg2+-free solution, both these drugs effectively enhanced the neurocytotoxicity of glutamate. DCB increased the number of damaged neurons to 90% and the same result was observed upon addition of 10 /~M CBDMB to the glutamate solution (Table I). In the other series of experiments, DCB and CBDMB were applied to the cultures after termination of glutamate treatment. The results of these experiments proved to be similar to the former ones: both DCB and CBDMB greatly enhanced glutamate-induced neuronal death. When DCB or CBDMB were added to the cultures both during glutamate treatment and in the post-glutamate period, the number of dead cells was insignificantly increased as compared to when the cul-

tures were treated with DCB or CBDMB only in the glutamate period (Table I). (2) Effect of Nat-removal after glutamate treatment. In all these experiments, cell cultures were treated for 15 min with 25 /~M glutamate in normal Nat-containing Mg2+-free solution and then transferred to Na t - and Mg2+-free 'post-glutamate solution'. The results of a 4-h incubation in this Nat-free solution after glutamate treatment was the same as that with the blockers of Na+/Ca 2+ exchange. In each instance, up to 75% of the neurons were dead by this time. It should be noted that in control experiments, all the solutions (with or without 30/~M DCB and 10 p M CBDMB) were well tolerated by cerebellar granule cells. An enhancement of glutamate neurotoxicity by removal of Na + from the external solution was also observed by Mattson et al. t5 in experiments with hippocampal cell cultures. However, in these experiments glutamate was present in the Mg2+-containing solution during 6 h, and, thus, the results of these experiments hardly can be directly compared with ours. Our data are in a good agreement with those of Hartley and Choi 1°, obtained in the experiments with cortical cultures treated by NMDA. The results of the experiments with two Na+/Ca 2÷ exchange inhibitors, DCB and CBDMB, as well as the experiments with Na t omission from post-glutamate solution, show that the inhibition of Na+/Ca 2+ exchange both during the period of glutamate application and in

325 the post-glutamate p e r i o d strongly enhances the delayed neuronal d e a t h induced by glutamate. In the past, it has been assumed that N a ÷ / C a 2+ exhange in glutamatet r e a t e d neurons was a possible route of Ca 2÷ entry responsible for d e l a y e d cell d a m a g e TM. H o w e v e r , o u r d a t a exclude this possibility. By contrast, N a + / C a 2+ exchange provides extrusion of Ca 2÷ from the cells under conditions of an e n h a n c e d Ca 2+ influx triggered by glutamate. Earlier, it had b e e n shown that the sustained Ca 2+ influx, as well as a d e l a y e d neuronal injury cannot be p r e v e n t e d by post-glutamate t r e a t m e n t either with voltage-dependent Ca2+-channel blockers (nitrendipine, verapamil,

diltiazem, or cobalt ions) or with different drugs and conditions known to block different subtypes of glutamate-activated cationic channels (OL-2-amino-5-phosphonovalerate, phencyclidine, or MK-801) 14. A t present, the N a + / C a 2÷ exchanger system also can be excluded from the list of these Ca 2÷ t r a n s p o r t i n g systems. T h e r e f o r e , in future studies, attention can be focused on two o t h e r possible routes for Ca 2÷ entry: non-specific leakage across the m e m b r a n e 14 and specific Ca2+-channels acti-

1 Bouchelouche, P., Belhage, B., Frandsen, A., Drejer, I. and Schousboe, A., Glutamate receptor activation in cultured cerebellar granule cells increases cytosolic free Ca 2÷ by mobilization of cellular Ca 2÷ and activation of Ca 2÷ influx, Exp. Brain Res., 76 (1989) 281-291. 2 Castilio, L., Andreeva, N. and Victorov, I., Dissociated nervous cell cultures from rat cerebellum. Morphological characterization, Int. J. Neurosci., 48 (1989) 293. 3 Choi, D.W., Ionic dependence of glutamate neurotoxicity, J. Neurosci., 7 (1987) 369-379. 4 Choi, D.W., Glutamate neurotoxicity and diseases of the nervous system, Neuron, 1 (1988) 623-634. 5 Choi, D.W., Koh, J.-Y. and Peters, S., Pharmacology of glutamate neurotoxicity in cortical cell culture: attenuation by NMDA antagonists, J. Neurosci., 8 (1988) 185-196. 6 Choi, D.W., Maulucci-Gedde, M. and Kriegstein, A.R., Glutamate neurotoxicity in cortical cell culture, J. Neurosci., 7 (1987) 357-368. 7 Cragoe Jr., E.J., Woltersdorf Jr., O.W., Bicking, J.B., Kwong, S.E and Jones, J.H., Pyrazine Diuretics. II. N-Amidino3-amino-5-substituted-6-halopyrazinecarboxamides, J. Med. Chem., 10 (1967) 66-75. 8 Downes, C.P., Inositol phosphates: a family of signal molecules, Trends Neurosci., 11 (1988) 336-339. 9 Favaron, M., Manev, H., Alho, H., Bertolino, M., Ferret, B., Guidotti, A. and Costa, E., Gangliosides prevent glutamate and kainate neurotoxicity in primary neuronal cultures of neonatal rat cerebellum and cortex, Proc. Natl. Acad. Sci. U.S.A., 85

(1988) 7351-7355. 10 Hartley, D.M. and Choi, D.W., Delayed resque of N-methylD-aspartate receptor-mediated neuronal injury in cortical culture, J. Pharmacol. Exp. Therapeutics, 250 (1989) 752-758. 11 Hawrot, E., Cultured sympathetic neurons: effects of cellderived and synthetic substrata on survival and development, Dev. Biol., 77 (1980) 136-151. 12 Kim, D. and Smith, T.W., Inhibition of multiple trans-sarcolemmal cation flux pathways by dichlorobenzamil in cultured chick heart cells, Mol. Pharmacol., 30 (1986) 164-170. 13 Kleyman, T.R. and Cragoe Jr., E.J., Amiloride and its analogs as tools in the study of ion transport, J. Memb. Biol., 105 (1988) 1-21. 14 Manev, H., Favaron, M., Guidotti, A. and Costa, E., Delayed increase of Ca 2+ influx elicited by glutamate: role in neuronal death, Mol. Pharmacol., 36 (1989) 106-112. 15 Mattson, M.P., Gunthrie, P.B. and Kater, S., A role of Na+-dependent Ca 2+ extrusion in protection against neuronal excitotoxicity, FASEB, 3 (1989) 2519-2526. 16 Moscona, A.A., Cell aggregation: properties of specific cell ligands and their role in the formation of muiticellular system, Dev. Biol., 18 (1968) 250-277. 17 Nowak, L. and Ascher, P., Divalent cation effects on NMDAactivated channels can be described as Mg-like or Ca-like, Soc. Neurosci. Abstr., 11 (1985) 953. 18 Victorov, I.V., Total staining method of monolayer cell cultures with vanadium hematoxilin, Bull. Exp. Biol. Med., 109, N6 (1990) 612-613 (in Russian).

vated from inside the cell by the second messengers IP3 and IP48.