Neuroscience Letters, 127 (1991) 21-24
21
~/1991 Elsevier Scientific Publishers Ireland Ltd. 0304-3940/91/$ 03.50 ADONIS 030439409100259A NSL 07775
Autoradiographic study of the cellular localization of [3H]glibenclamide binding sites in the rat hippocampus Evelyne Tremblay, Sylvie Zini and Yehezkel Ben-Ari Laboratoire de Neurobiologie et Physiopathologie du d?veloppement, INSERM U29, Paris (France) (Received 31 January 1991; Revised version received 6 March 1991; Accepted 6 March 1991)
Key words: ATP-sensitive potassium channel; Glibenclamide; Autoradiography; Lesion; Rat hippocampus [3H]Glibenclamide, a potent ATP-sensitive K ÷ channel blocker, is a specific ligand for the identification of the sulfonylureas receptors which are closely related to ATP-sensitive K ÷ channels. In order to determine the pre- or postsynaptic localization of these receptors in the rat hippocampus, we studied the effects of selective kainate or colchicine-induced lesions on the regional distribution of [3H]glibenclamide binding. A decreased binding was found in the following conditions: in CA3, after destruction of the mossy fiber terminals but not of the CA3 cells; in CAI, after destruction of the CA I cells but not of the Schaffer collaterals; in the fascia dentata, after destruction of the granule cells. These results suggest that glibenclamide binding sites are mainly localized on the mossy fibers in CA3, and on the granular and pyramidal neurons in the fascia dentata and CA 1, respectively.
Among the broad diversity of potassium channels identified in every cell type, ATP-sensitive potassium (KATP) channels are distinguishable from the others by their sensitivity to the metabolic state of the cell [3]. These channels, which are inhibited by intracellular ATP, are present in several peripheral tissues like cardiac muscle [15, 17, 23], skeletal muscle [22] as well as in pancreatic fl-cells [12] where they are involved in the control of insulin release [4]. Hypoglycaemic sulfonylureas, like glibenclamide, are potent blockers of these channels, and their therapeutic effect is likely due to the existence of specific binding sites on the fl-cells [21]. KAT P channels and specific binding sites for sulfonylureas are also present in the central nervous system [5, 16]. In the substantia nigra, which is enriched in glibenclamide binding sites, KAT P channels are sensitive to the extracellular glucose concentration and modulate the release of y-aminobutyric acid (GABA) [2, 20]. In the hippocampus, a structure particularly sensitive to the deleterious effects of ischemia, sulfonylureas binding sites are also present [16], notably in the CA3 region where KAT P channels may exercise a presynaptic modulation of transmitter release during anoxia or hyperglycaemia [6, 7]. If the sulfonylurea receptor is the KATPchannel or a structure closely related to that channel which controls its activity [20, 24], it would be of interest to determine its cellular Correspondence: E. Tremblay, Laboratoire de Neurobiologie et Physiopathologie du d+veloppement, INSERM U29, 123 Boulevard de Port-Royal, 75014 Paris, France.
localization in the hippocampus. In this aim, we have examined, with an autoradiographic method, the distribution and the density of [3H]glibenclamide binding sites in the rat hippocampus after selective lesions, by kainate or colchicine injections [10, 13 and see below], of the major pre- and/or postsynaptic elements. For lesion experiments, male Wistar rats were anesthetized with chloral hydrate (300 mg/kg, i.p.). One group received two injections of 1.5 pg of colchicine (dissolved in 0.6 pl of physiological saline) into the right hippocampus, at the following coordinates according to the atlas of Albe-Fessard et al. [1]: A 2.9; L 3.3; H 6.4; and A 5.4; L 1.3; H 7.0. The other group received kainic acid (KA) injections (1.2 pg dissolved in 0.3 /A of 50 mM phosphate buffer, pH 7.4) into the right amygdala (A 6.0; L 4.0; H 3.0). Eight to 10 days later, these animals as well as normal non-treated ones (n--3 in each group) were sacrificed, the brains rapidly removed and frozen at - 5 0 ° C in isopentane. Coronal sections (20 pm) of the dorsal hippocampus were cut with a cryostat (-20°C) and mounted onto gelatin-coated slides for binding experiments. In order to control the extent of the lesions, histological procedures were made on alternate sections. These include classical Nissl staining of body cells and specific Timm staining of the mossy fibers [14]. The autoradiographic study of [3H]glibenclamide binding was performed according to Mourre et al. [16] with minor modifications. Briefly, the sections were incubated for 2 h in 50 mM Tris-HC1 buffer (pH 7.4) with 3 nM [3H]glibenclamide (CEA, Saclay, France: 50.4 Ci/
22
mmol) in the absence (total binding) or presence (non specific binding) of 10 -5 M unlabelled glibenclamide. After rinsing and drying, the sections were exposed for 4 months to [3H]-sensitive films (Amersham) simultaneously with internal radioactive standards. Quantification of the autoradiographs were performed relying on
the optical density of the standards, with a computer assisted image analyzer (IMSTAR). In contralateral non-lesioned hippocampi (following colchicine or KA treatment) as in normal non-treated rats, the distribution of [3H]glibenclamide binding sites was similar to that previously described [16]. As shown
A
Control C
B
D
E
KA
F
* tit
~,Y
,t
G
oo,c. Fig. 1. Autoradiographs illustrating the labelling by [3H]glibenclamide of the hippocampus following KA or colchicine-induced lesions. C: control non-lesioned side. E,G: sides lesioned by KA or colchicine (Colc.) injections respectively. Histological stainings obtained with alternate sections of non-lesioned and lesioned hippocampi are illustrated in A, B, D and F. A,D: Nissl stain; arrowheads in D indicate the neuronal loss of pyramidal cells and gliosis in CA3 after KA injection. B,F: Timm stain of the mossy fibers; note in F the loss of Timm staining of these fibers after intrahippocampal colchicine. CAI and CA3, subfields of the Ammon's horn; Fd, fascia dentata; lu, stratum lucidum; p and g, pyramidal and granular cell layers, respectively•
23 TABLE I C O M P A R E D DENSITIES O F SPECIFIC [3H]GLIBENCLAMIDE B I N D I N G SITES IN CONTROL A N D LESIONED RAT HIPPOCAMPI Unilateral hippocampal lesions after KA (n = 7) or colchicine (n = 7) injections were made as described in the text. Binding experiments were performed with 3 nM [3H]glibenclamide. Non-specific binding represented 13 % of the total binding. Measurements were made in the stratum lucidum of CA3, in the cellular layer and adjacent stratum radiaturn of CA1. The values found in the cellular and outer molecular layers of the fascia dentata (Fd) were not significantly different and thus pooled. Each value represents the mean + S.E.M. (in fmol/mg tissue) of 6-10 determinations for each case. CA3
CA1
Fd
Colchicine group Control side Lesioned side
82 _ 3 55 +__4**
78 _+3 54 _ 8*
79__+2 41 __+6**
K A group Control side Lesioned side
86-t- 1 90 + 3
76___ l 76 +__2
84+ 1 84_ 1
Statistically significant differences from control group: *P
21
~/1991 Elsevier Scientific Publishers Ireland Ltd. 0304-3940/91/$ 03.50 ADONIS 030439409100259A NSL 07775
Autoradiographic study of the cellular localization of [3H]glibenclamide binding sites in the rat hippocampus Evelyne Tremblay, Sylvie Zini and Yehezkel Ben-Ari Laboratoire de Neurobiologie et Physiopathologie du d?veloppement, INSERM U29, Paris (France) (Received 31 January 1991; Revised version received 6 March 1991; Accepted 6 March 1991)
Key words: ATP-sensitive potassium channel; Glibenclamide; Autoradiography; Lesion; Rat hippocampus [3H]Glibenclamide, a potent ATP-sensitive K ÷ channel blocker, is a specific ligand for the identification of the sulfonylureas receptors which are closely related to ATP-sensitive K ÷ channels. In order to determine the pre- or postsynaptic localization of these receptors in the rat hippocampus, we studied the effects of selective kainate or colchicine-induced lesions on the regional distribution of [3H]glibenclamide binding. A decreased binding was found in the following conditions: in CA3, after destruction of the mossy fiber terminals but not of the CA3 cells; in CAI, after destruction of the CA I cells but not of the Schaffer collaterals; in the fascia dentata, after destruction of the granule cells. These results suggest that glibenclamide binding sites are mainly localized on the mossy fibers in CA3, and on the granular and pyramidal neurons in the fascia dentata and CA 1, respectively.
Among the broad diversity of potassium channels identified in every cell type, ATP-sensitive potassium (KATP) channels are distinguishable from the others by their sensitivity to the metabolic state of the cell [3]. These channels, which are inhibited by intracellular ATP, are present in several peripheral tissues like cardiac muscle [15, 17, 23], skeletal muscle [22] as well as in pancreatic fl-cells [12] where they are involved in the control of insulin release [4]. Hypoglycaemic sulfonylureas, like glibenclamide, are potent blockers of these channels, and their therapeutic effect is likely due to the existence of specific binding sites on the fl-cells [21]. KAT P channels and specific binding sites for sulfonylureas are also present in the central nervous system [5, 16]. In the substantia nigra, which is enriched in glibenclamide binding sites, KAT P channels are sensitive to the extracellular glucose concentration and modulate the release of y-aminobutyric acid (GABA) [2, 20]. In the hippocampus, a structure particularly sensitive to the deleterious effects of ischemia, sulfonylureas binding sites are also present [16], notably in the CA3 region where KAT P channels may exercise a presynaptic modulation of transmitter release during anoxia or hyperglycaemia [6, 7]. If the sulfonylurea receptor is the KATPchannel or a structure closely related to that channel which controls its activity [20, 24], it would be of interest to determine its cellular Correspondence: E. Tremblay, Laboratoire de Neurobiologie et Physiopathologie du d+veloppement, INSERM U29, 123 Boulevard de Port-Royal, 75014 Paris, France.
localization in the hippocampus. In this aim, we have examined, with an autoradiographic method, the distribution and the density of [3H]glibenclamide binding sites in the rat hippocampus after selective lesions, by kainate or colchicine injections [10, 13 and see below], of the major pre- and/or postsynaptic elements. For lesion experiments, male Wistar rats were anesthetized with chloral hydrate (300 mg/kg, i.p.). One group received two injections of 1.5 pg of colchicine (dissolved in 0.6 pl of physiological saline) into the right hippocampus, at the following coordinates according to the atlas of Albe-Fessard et al. [1]: A 2.9; L 3.3; H 6.4; and A 5.4; L 1.3; H 7.0. The other group received kainic acid (KA) injections (1.2 pg dissolved in 0.3 /A of 50 mM phosphate buffer, pH 7.4) into the right amygdala (A 6.0; L 4.0; H 3.0). Eight to 10 days later, these animals as well as normal non-treated ones (n--3 in each group) were sacrificed, the brains rapidly removed and frozen at - 5 0 ° C in isopentane. Coronal sections (20 pm) of the dorsal hippocampus were cut with a cryostat (-20°C) and mounted onto gelatin-coated slides for binding experiments. In order to control the extent of the lesions, histological procedures were made on alternate sections. These include classical Nissl staining of body cells and specific Timm staining of the mossy fibers [14]. The autoradiographic study of [3H]glibenclamide binding was performed according to Mourre et al. [16] with minor modifications. Briefly, the sections were incubated for 2 h in 50 mM Tris-HC1 buffer (pH 7.4) with 3 nM [3H]glibenclamide (CEA, Saclay, France: 50.4 Ci/
22
mmol) in the absence (total binding) or presence (non specific binding) of 10 -5 M unlabelled glibenclamide. After rinsing and drying, the sections were exposed for 4 months to [3H]-sensitive films (Amersham) simultaneously with internal radioactive standards. Quantification of the autoradiographs were performed relying on
the optical density of the standards, with a computer assisted image analyzer (IMSTAR). In contralateral non-lesioned hippocampi (following colchicine or KA treatment) as in normal non-treated rats, the distribution of [3H]glibenclamide binding sites was similar to that previously described [16]. As shown
A
Control C
B
D
E
KA
F
* tit
~,Y
,t
G
oo,c. Fig. 1. Autoradiographs illustrating the labelling by [3H]glibenclamide of the hippocampus following KA or colchicine-induced lesions. C: control non-lesioned side. E,G: sides lesioned by KA or colchicine (Colc.) injections respectively. Histological stainings obtained with alternate sections of non-lesioned and lesioned hippocampi are illustrated in A, B, D and F. A,D: Nissl stain; arrowheads in D indicate the neuronal loss of pyramidal cells and gliosis in CA3 after KA injection. B,F: Timm stain of the mossy fibers; note in F the loss of Timm staining of these fibers after intrahippocampal colchicine. CAI and CA3, subfields of the Ammon's horn; Fd, fascia dentata; lu, stratum lucidum; p and g, pyramidal and granular cell layers, respectively•
23 TABLE I C O M P A R E D DENSITIES O F SPECIFIC [3H]GLIBENCLAMIDE B I N D I N G SITES IN CONTROL A N D LESIONED RAT HIPPOCAMPI Unilateral hippocampal lesions after KA (n = 7) or colchicine (n = 7) injections were made as described in the text. Binding experiments were performed with 3 nM [3H]glibenclamide. Non-specific binding represented 13 % of the total binding. Measurements were made in the stratum lucidum of CA3, in the cellular layer and adjacent stratum radiaturn of CA1. The values found in the cellular and outer molecular layers of the fascia dentata (Fd) were not significantly different and thus pooled. Each value represents the mean + S.E.M. (in fmol/mg tissue) of 6-10 determinations for each case. CA3
CA1
Fd
Colchicine group Control side Lesioned side
82 _ 3 55 +__4**
78 _+3 54 _ 8*
79__+2 41 __+6**
K A group Control side Lesioned side
86-t- 1 90 + 3
76___ l 76 +__2
84+ 1 84_ 1
Statistically significant differences from control group: *P
