Regional distribution of sulfonylurea receptors in the brain of rodent and primate

Neuroscience Vol. 55, No. 4, pp. 1085-1091, 1993 Printed in Great Britain

0306-4522/93 $6.00 + 0.00 Pcrgamon Press Ltd 0 1993 IBRO

REGIONAL DISTRIBUTION OF SULFONYLUREA RECEPTORS IN THE BRAIN OF RODENT AND PRIMATE S. ZINI, E. lkmnu~,*

H. POLLARD, J. MOREAU and Y.

Bar-Am

Laboratoire de Neurobiologie et Physiopathologie du dkveloppement, INSERM U29, 123 Boulevard de Port-Royal, 75014 Paris, France Almtraet-Glibenclamide, one of the most potent antidiabetic sulfonylureas, inhibits the activity of ATP-sensitive K+ channels in the pancreas as well as in the brain through its binding to specific receptors. Quantitative autoradiography was used to localize such receptors in the brain of rat, mouse, guinea-pig and marmoset, using [ 3H]glibenclamide as radioligand. In all four species, specific glibenclamide binding sites were found to be heterogeneously distributed. The highest densities were in the cerebral cortex, the molecular layer of the cerebellar cortex, the thalamus and the caudateputamen. The globus pallidus and the substantia nigra were highly labelled in rat and mouse but poorly labelled in guinea-pig and marmoset. The distribution of glibenclamide binding sites in the hippocampus was different between the rodents and marmoset; in rodents, most binding sites were distributed in the fascia dentata and the CA3-CA4 fields of Ammon’s horn, contrasting with a very homogeneous distribution in all subfields of the marmoset hippocampus. In conclusion, we demonstrate that primate brain contains specific binding sites for [ 3H]glibenclamide with a distribution not exactly similar to that in rodent brain.

sulfonylureas such as glibenclamide have been used clinically for a long time in the treatment of non-insulin-dependent diabetes mellitus, even though their mechanism of action has only emerged recently. In pancreatic /?-cells, sulfonylureas mimic the effect of glucose by inhibiting ATPsensitive K+ channels2’*23*28 which induce a membrane depolarization and the activation of voltagedependent Ca *+ channels, leading to an increase of Ca*+ influx and insulin secretion.” The presence of specific high-affinity sulfonylurea binding sites has been demonstrated in the membranes of pancreatic /I-cells ‘using [‘H]glibenclamide as the radioThe relative potency with which various ligand. 8,Lo*22,26 sulfonylureas inhibit [ ‘H&libenclamide binding to )Y-cell membrane receptors correlates well with their efficacy in blocking ATP-sensitive K+ channels, which strongly suggests that sulfonylurea receptors are tightly linked to these channels.‘24 It is not established, however, whether the sulfonylurea receptor could be an associated protein or an integral part of the channel. Characterization of sulfonylurea photolabelling using receptors by either [ 3Hjglibenclamidezs~‘s or a radioiodinated derivative of glibenclamideM shows several photolabelled polypeptides of 140,000, 65,000, 55,000 and 30,000 mol. wt. The 140,ooO mol. wt. protein was identified as a high-afIlnity receptor (& in the nM range), whereas the three other proteins were assessed as low-affinity receptors (K,, in the PM rangem). The presence of sulfonylurea receptors has been reported in various tissues including cardiac muscle,4’ Hypoglycemic

*To whom correspondence should be addressed.

Autoradiosmooth muscle”*33 and brain.‘0@*25~35~36 graphic studies in rat brain have revealed a heterogeneous distribution, with the highest density of receptors in the motor neocortex, the molecular layer of the cerebellum, the substantia nigra and the globus pallidus.v*‘7*18~29 Although an endogenous peptide ligand for sulfonylurea receptors has been recently discovered in ovine brain,31J2 the function of these receptors in the brain remains unknown. It has been proposed that, as in pancreatic /I-cells, the receptors could be associated with the activity of ATP-sensitive K+ channels, providing a link between the metabolic state of the neuron and its excitability.*,” However, only few studies concerning the regional distribution of sulfonylurea receptors in the brain have been made in the rat and the mouse.‘8~1v~2v~M Notably, there is no information yet about the presence of such receptors in the brain of other rodent species or primates. Using [ 3H]glibenclamide as radioligand, we have performed a comparative in vitro autoradiographic study of glibenclamide binding sites in the brain of three different rodent species, the rat, the mouse and the guinea-pig, and of one primate species, the marmoset. EXPERIMENTAL PROCEDURES

Male Wistar rats (n = 4) Swiss OF1 mice (n = 4) and Hartley guinea-pigs (n = 4) were purchased from If&t-Credo (France). Male adult common marmosets (Coltitlrix &cchur, n = 3) originated from the breeding colony from INSERM U36 (Paris, France), and were kindly provided by Dr J. B. Mi&el. Animals were am@hetizul before decapitation. Brains ume removed and immediately frozen in liquid monochlorodifluoromethane. Coronal sections (20 pm) were cut with a cryostat (- 22°C) mounted on to

1085

S. ZIM et al.

1086

gelatin-coated slides and stored at -20°C until used. For experiments, thawed sections were incubated for I h at room temoerature in 50mM Tris-HC1 buffer (OH 7.4) containing i20 mM NaCl, 5 mM KCl, 1 mM “MgC$ 2.5mM C&f,. Slices were then incubated for 2 h at room tempe&re in the same buffer containing 1.5 nM 13~~~~de (50.4~/~01. CBA. Saclav. France) in t& absence (total bind&g) or- in the~pnxen& (nonspecific binding) of 1OpM unlabelled glibenclamide (Hoecht-Roussel Pharmaceutical, Paris, France). Thereafter, sections were washed three times in cold buffer for 20s and then dipped twice in cold water. After drying, sections were exposed for six weeks to ‘H-sensitive hypertilms (Amersham, France) in parallel with internal 3H-micro-scales standards (Amersham, France); these matter from for central gray were calibrated 155-73260 d.p.m./mg tissue (which conesponds to 1.4-655 fmol [?H~~ncl~&/rng tissue). ~ntifi~tion of the autoradiographs were performed relying on the optical density of the standards with a computer assisted image analyser (Samba 2005/Alcatel TITN Answare, Grenoble, France). Histological determination of the brain areas was performed on sections used for autoradiography or adjacent sections with conventional Nissl staining. Visualization of the mossy fibres throughout the marmoset hippocampus was performed with the Timm’s sulphide silver method,‘* after ~~trnent of sections for 15min in 0.1 mM phosphate buffer (pH 7.2) containing 1% sodium sutphide. Rl?sULTs

We have previously shown that [3H]glibenclamide binds to specific highalkity (4 = 0.3 nM) and lowaffinity (X, I; 100 nM) receptors in membranes of rat

Table 1. Regional distribution of specific fH~~ncla~de binding sites in the brain of rodents

Brain structures

Specific binding (fmol/mg tissue) Rat Mouse Guinea-pig

Neocortex

Frontal cortex Frontoparietai cortex Cl&p&al cortex Midbrain and basal forebrain Substantia nigra Globus pailidus Caudate putamen Thalamus Hypothalamus Pons

44 41 40

55 44 41 39 15 13

47 58 49

z 54 50 20 12

57 63 66

31 25 41 55 22 I5

Hi~rnp~ Fascia dentata Ammon’s horn: CA4 field CA3 field CA1 field

44 44 44 34

42

69 74 61 44

Cerebellum Molecular layer Granular layer

47 21

61 33

61 32

64 :

Binding experiments were carried out with 1.5 mM ~~~~~, nonspecik binding was evaluated in the pmsence of 10 PM of unlabelled gl&enclamide. Each value represents the mean of four to eight independent determinations from four animals. Standard errors are not included in the table since they never exceeded 2% of the mean value in all cases.

Table 2. Regional distribution of [3HJglibenclamide binding sites in the marmoset brain Brain structures Neocortex Frontal cortex Anterior cingulate cortex Visual cortex, area striata Subgenualis cortex Insular cortex Temporal cortex Entorhinal region

Specifix binding (fmol/mg tissue) Layers I-III IV-VI 55 40 54 39 45 25 43 43 40 36 37 37 36 28

Basal forebrain Putamen nucleus Caudate nucleus Olfactory tubercule Accumbens nucleus Globus pallidus

39 38 36 32 22

Amygdaloid complex

35

Tbaiamus Anterodorsal nuc1eu.s Anterior nucleus of ventral thalamus Paraventricular nucleus Dorsomedial nucleus Dorsolateral nucleus Reuniens nucleus Medial geniculate nucleus Ventral nucleus of anterior thalamus Ventrolateral nucleus

36 34 34 31 30 30 27 22 22

Hy~thaIamus Anterior h~th~~ic area ~em~lIar nucleus Paraventricular nucleus

27 27 27

Hippocampal formation Fascia dentata Ammon’s horn subfields: CA4 CA1 CA3 Subiculum

34 34 34 33 32

Midbrain and hindbrain Interpeduncular nucleus Central gray Substantia nigra compacta Substantia nigra diffusa Superior colliculus Superior central nucleus Pontine reticular nucleus

36 29 27 27 20 16 15

Cerebellum Molecular layer Granular layer

36 I9

Binding experiments were carried out with 1.5nM [3H]glibenclamide; nonspecific binding was evaluated in the presence of 10 PM of unlabelled glibenclamide. Each value represents the mean of four to six independent determinations from three animals, Standard errors are not included in the table since they never exceeded 2% of the mean value.

brain.35v36The low concentration of [3H]glibencIamide (1.5 n&i) used here, as in a previous autoradiographic study,3o only labels the high-affinity receptors. Under these conditions, the level of specific binding deter-

Sulfonylurea receptors in rodent and primate brain mined from the autoradiographs was found to be of the same order of magnitude for rat, mouse, guineapig and marmoset, varying from 20 to 75 fmol/mg

AAH AC

AD AmC APM AS AV CA1 CA3 CA4 cc CCA Cd CFr CIn Cl co cos CPU cs CSg CTP DAH DB Er FD LC

anterior hypoU&mk area accumbens mlakus anterodd thalamic nucleus amygdaloid oomplex

tissue (Tables 1,2). However, the densities of binding sites were usually lower in rat and mafmoset than in mouse and guinea-pig. The level of nonspecific

Abbreviations used in the figures GM GP Gr

medial preoptic area visual cortex, area striata anterior nucleus of the ventral tbalamus field CA1 of Amman’s horn field CA3 of Aaxnon’s horn field CA4 of Armnon’s horn corpus callosum anterior cingulate cortex caudate nucleus frontal cortex insular cortex claustrum occipital cortex superior colliculus caudate-putamen superior central nucleus subgenualis cortex temporal cortex dorsal anterior hypothalamic nucleus fascicular nucleus of the Broca’s diagonal band entorhinal region fascia dentata granular cells of the fascia dentata central gray

1087

Hip L”D MD mf Mol NP OB Pm Pu PVH PVT Py pyr Re RTP SN SNC SND

Sub Th TO VA VL

medial genioulate nucleus globus pailiius granular layer of the cerebellum ~ppocampus interpeduncular nucleus dorsolateral thalamic nucleus domomedial tbalamic nucleus mossy fibres molecular layer pontine nucleus olfactory bulb premammillary nucleus putamen nucleus paraventricular hypothalamic nucleus paraventricular thalamic nucleus pyramidal tract pyramidal cell layer of Ammon’s horn Reuniens nucleus pontine reticular nucleus substantia nigra substantia nigra compacta substantia n&a diffusa subiiulum thalamus olfactory tubercule ventral nucleus of anterior thalamus ventrolateral thalamic nucleus.

Fig. 1. &atom&a&& iWtrat@ tBp on of [9&ibilk&g *an npmrcatative sag&al sections of rodent brain. (A) Rat, (B) mouse, (C) @nea-pig. Binding experiments were carried out as indicated in the methods, with 1.5 nM [3H]gliben&nide. ._.l_:.. _?M_. wmre. _--I_ ~caflc.L__ oar = -mm.

Autoradiographic

grains are seen as

B

Fig. 2.

Sulfonylurea receptors in rodent and primate brain

1089

Fig. 3. Histological representation of neuronal cells (A) and mossy fibres (B) of the marmoset hippocampus. Stainings were performed on coronal sections adjacent to the section illustrated in Fig. 2D. (A) Nissl staining; (B) Timrn’sstaining illustrating that mossy fibres terminate in the suprapyramidal layer of the CA3XA4 pyramidal cells. Scale bar = 200 mm.

binding was found to be similar in the four species and calculated to be 30-50% of the total binding, depending on the region (data not shown). Consistent with previous autoradiographic studies performed on the brain of rat and mouse,‘**19,29 we found a heterogeneous ~s~bution of [3H&libenclamide binding sites in these two species (Fig. lA,B) as well as in the brain of the any-pig {Fig. IC!) and of the marmoset (Fig. 2). In rodent brain, the highest density of specific binding sites was found in the cerebral cortex (motor and sensory areas), the molecular layer of the cerebellar cortex, the caudate-putamen and the thalamus. Substantial labelling was also found in the hippocampus where the highest densities of binding sites were found in the fascia dentata and the CA3-CA4 fields of Ammon’s horn, whereas the CA 1 field clearly showed less labelling (Table 1). The main differences between the three rodent species concerned the substantia nigra and the globus pallidus which showed greater labelling in rat and mouse (Fig. lA,B, Table 1) than in guinea-pig (Fig. lC, Table 1). As a whole, the other midbrain and hindbrain areas were poorly labelled (Fig. 1). The results obtained in the marmoset brain are illustrated in Fig. 2 and Table 2. The highest densities of specific binding sites were found in the external layers of the frontal, cingulate, and visual cortices (Fig. 2A, B, C). The subgenualis, insularis, temporal and entorhinal cot&es also contained high densities of binding sites (Fig. 2A, B, D), as well as the

putamen and the caudate nucleus (Fig. 2A, B), the olfactory tubercules (Fig. 2A), the interpeduncular nucleus (Fig. 2D) and the molecular layer of the cerebellum (Fig. 2F). The thalamus was moderately and heterogeneously labelled with the highest densities in the medial and dorsolateral nuclei (Fig. ZB, C). In the hippocampus, the labelling was moderate and homo~neo~; in contrast to rodents, the fascia dentata and the CA3-CA4 fields of Ammon’s horn did not contain higher densities of binding sites than the CA1 field (Fig. 2D). The density of binding sites in the hypothalamus was moderate and also very homogeneous (Fig. 2B, C). As in guinea-pig, the substantia nigra was also moderately labelled (Fig. 2D). As a whole, the pontine areas were moderately to poorly labelled (Fig. 2E). Labelling of the fibre tracts such as the corpus callosum or the pyramidal tract did not differ from background. DISCUSSION

This study describes for the first time the presence of specific sulfonylurea binding sites in the brain of the guinea-pig and of a primate, the marmoset, The regional distribution of glibenclamide binding sites in the brain of these two species appears to be close to that in the rat and the mouse brain. Thus, in the four species we found a common high density of binding sites in the sensorimotor cortical areas, the caudateputamen nuclei and the molecular layer of the cerebelfar cortex. However, slight differences exist

Fig. 2. Au~r~o~~s dilating the ~~bution of 13~~~l~e binding sites in the marmoset brain on representative rostra1 (A) to caudal (F) coronal sections and on visual cortex (G), Binding experiments were carried out as indicated in the methods with 1.5 nM [3~i~cl~de. Autoradiographic grains are seen as white. Brain areas were determined according to the stereotaxic atlas of Stepban ef aL2’ Scale bar = HI0mm.

1090

s.&NI et al.

among q&es, notably in the hfppoeampus, the globus pallidus and the subatantia nigra. In the of the three rodents, the distribution ~~~~ binding sites was identical, w&h high densities in the fascia dentata and the CA3-CA4 &&ls of Ammo&s horn and a lower density in the CA1 field. Previous lesion studies in the rat hippocampus have demonstrated that this striking contrast could be explained by a preferential location of gliben&mide binding sites on the granular cells of the fascia dent&a and on the mossy fibre terminals innervating the CA3-CA4 pyramidal c&is of Ammon’s horn.= The great homology of the autoradiographs obtained in the mouse and the guinea-pig hippocampi suggests that glibenclamide binding sites are located similarly in the granule cells and the mossy fibre terminals, In contrast, the labelling of glibenclamide binding sites throughout the marmoset hip~mpus was moderate and very homogeneous as compared with that obtained in rodents. This discrepancy can not be attributed to a differential distribution of hippocampal mossy fibres between rodent and primate brain. Thus, Fig. 3 illustrates with the selective Timm’s staining that the mossy fibres innervate the same hippocampal subfields in the marmoset brain as in rodents,‘2 i.e. the suprapyramida1 layer of the CA3-CA4 neurons. This could marmoset hip~c~pus, that, in indicate glibenclamide binding sites are not preferentially located on granular cells and on mossy fibre terminals but could also be present on the other cells. From previous studies in the rat, we have suggested that the presynaptic location of glibenclamide binding sites in the CA3 field could reflect the presence of ATP-sensitive K+ channels from which the activation during a metabo~c stress, such as anoxia, could prevent the excessive release of ~utarnate.~,~.~ This might partly explain why pyramidal cells in CA3 are more resistant than those in CA1 to the deleterious effect of anoxia. Taking into account this hypothesis, the absence of a preferential location of glibenclamide binding sites on mossy fibre terminals in the mar-

moset hippocampus could indicate that the amount of presynaptic glibenclamide receptors in the CA3 field would not be a alit factor for protection against anoxia-induoed mezonal damage. The globus pallidus and the substantia nigra of the rat and the mouse are very enriched with binding sites, in agreement with previous data,18,19,29 Furthermore, it has been demonstrated that glibenclamide binding sites in the rat substantia uigra are presynaptically located on terminals of GAaQergic pallidonigral neurons where they regulate the release of GABA, likely through an inhi~tion of ATPsensitive K+ channels.” Taking into account that the pallidonigral system of the guinea-pig and the marmoset contain a low density of binding sites, it would be of interest to examine whether this regulation still occurs in these two species. In this context, it is interesting to note that previous experiments in guinea-pig demonstrated that ATP-sensitive K+ channels are functionally operative in a neuronal subpopulation of the substantia nigra and can be inhibited by the sulfonylurea tolbutamide,” despite the small number of sulfonylurea binding sites demonstrated in this study. CONCLUSION

The presence of ~i~ncla~de binding sites in the primate brain suggests the existence of an endogenous ligand for these receptors which could be similar to that recently isolated from ovine brain” and strengthens the hypothesis that these receptors may play an important functional role in the central nervous system. Like in rodent species, these receptors could be associated to the activity of some type of K+ channels and affect neurotrans~ssion by modulating the release of some neurom~iato~. Acknowledgements-We thank G. Ghilini for technical assistance and S. Guidasci for illustrations. We are grateful to Hoecht-Roussel Pharmaceutical for its generous gift of glibenclamide. Financial support for this work was provided by the Association Franpise contre les Myopathies.

I. Aguilar-Bryan L., Nichols C. G., Rajan A. S., Parker C. and Bryan J. (1992) Co-expression of sulfonylurea receptors and I(-ATF’ channels in hamster insuhnoma tumor (HIT) cells-evidence for direct association of the receptor with the channel. J. biol. G%em.267, 14,934-14,940. 2. AguiIar-Bryan L., Nelson D. A., Vu Q. A., Humphrey M. B. and Boyd A. P. (1990) Photoaffinity labeling and partial purification of the cell sulfonylurea receptor using a novel, biologically active glyburide analog, J. biol. Chem. 265, 8218-8224. 3. Amoroso S., Schmidt-Antomamhi M., Foss& M. and Lazdunski M. (1990) Glucose, sulfonyhueas and neurotransmitter reIease: role of AI’P-sensitive K+ channels. Science 247,852--8X 4. Ben-Ari Y. (1990) Galanin and ~~~~~de mod&ate the anoxic release of glutamate in rat CA3 hip~mp~ neurons. Eur. f. Neurosci. 2, 62-68. 5. Bemardi H., Fosset M. and Lazdunski M. (1988) Characterization, purification, and affinity labeling of the brain [‘Hlglibenclamide-binding protein, a putative neuronal ATP-regulated K+ channel. Proc. mtn. Acud. Sci. @, 98 16-9820. 6. French J. F., Riera L. C., Mullins U. L. and Sarmiento J. G. (1991) Modulation of [3H]glibenclamide binding to cardiac and insulinoma membranes. Eur. J. Pharmac. 207, 23-28.

S~fonylur~

receptors in rodent and primate brain

1091

7. French J. F., Riera L. C. and Sarmiento J. G. (1990) Identifleation of high and low (GTP-sensitive) afiinity ~3~i~ncl~& binding sites in cardiac ventricular membranes. &?&em. biophys. Res. Commun. 167, 1400-1405. 8. Gaines K. L., Hamilton S. and Boyd A. E. (1988) Characterization of the sulfonylurea receptor on beta cell membranes. J. biol. Chem. 2.43, 25892592. 9. Gehlert D., Mais D., Gaekenheimer S., Krushinski J. and Robertson D. (1990) Localization of ATP sensitive potassium channels in the rat brain using a novel radioligand, [12SI]iodoglibenclamide. Eur. J. Pharmac. l&j, 373-375. 10. Geisen K., Hitzel V., (Ikomonopoulos R., Ptinter J., Weyer R. and Summ H. D. (1985) Inhibition of [3Hjglibenclamide binding to sulfonylurea receptors by oral antidiabetics. Arzeim. Forsch. Drug Res. 35, 707-712. 11. Gopalakrishnan M., Johnson D. E., Janis R. A. and Triggle D. J. (1991) Characterization of binding of the ATP-sensitive potassium channel ligand, [‘Hlglyburide, to neuronal and muscle preparations. J. P/tarmac. exp. Ther. 2S7, 116221171. 12. Haug F. (1973) A light microscopic study in the rat with Timm’s sulphide silver method. Methodological considerations and cytological and regional staining patterns. Adu. Anat. Embryoi. Ceil Biot. 47, l-71. 13. Hiiuser M. A., de Weille J. R. and Lazdunski M. (1991) Activation by cromakalim of pm- and post-synaptic A~-motive K+ channels in subs~ntia nigra. Biochern. biophys. Res. Commun. 174, 909-914. 14. Kerry P., Murphy J. and Greenfield S. (1992) Neuronal selectivity of ATP-sensitive potassium channels in guinea-pig substantia nigra revealed by responses to anoxia. J. Physiol. 453, 167-183. 15. Kramer W., Gekonomopulos R., Piinter J. and Summ H. D. (1988) Direct phototinity labeling of the putative sulfonylurea receptor in rat /J-cell tumor membranes by [‘Hlglibenclamide. Fe& Eur. biochem. Sots Z.&t. 229,355-359. 16. Lupo B. and Bataille D. (1987) A binding site for [rH]glipizide in the rat cerebral cortex. Eur. J. Pharmac. 140,157-169. 17. Miller R. J. (1990) Glucose-regulated potassium channels are sweet news for neurobiologists. Trends Neurosci. 13, 197-199. 18. Mourre C., Ben Ari Y., Bemardi H., Fosset M. and Lazdunski M. (1989) Antidiabetic sulfonylureas: localization of binding sites in the brain and effects on the hyperpolarization induced by anoxia in hippocampal slices. Bruin Res. 486, 159-164. 19. Mourre C., Widmann C. and Lazdunski M. (1990) Sulfonylurea binding sites associated with ATP-regulated K+ channels in the central nervous system: autoradiographic analysis of their distribution and ontogenesis, and of their localization in mutant mice cerebellum. Bruin Res. 519, 2943. 20. Nelson D. A., Aguilar-Bryan L. and Bryan J. (1992) Specificity of photola~ling of /&eel1membrane proteins with an ‘2Sf-labeled glybuiide analog. J. biol. C&m. 267, 14,928-14,933. 21. Niki I., Kelly R. P., Ashcroft S. J. H. and Ashcroft F. M. (1989) ATP-sensitive K-channels in HIT TlS @cell studied by patch-clamp methods, 86Rb efflux and glibenclamide binding. Eur. J. Pharmac. 415, 47-55. 22. Niki I., Nicks J. L. and Ashcroft S. J. H. (1990) The B-cell glibenclamide receptor is an ADP-binding protein. Biochem. J. 268, 713-718. 23. S&mid-Antomarchi H., De Weille J., Fosset M. and Lazdunski M. (1987) The antidiabetic sulfonylurea glibenclamide is a potent blocker of the ATP-modulated K+ channel in insulin secreting cells. Biochem. biophys. Res. Commun. 146, 21-25. 24. Schmid-Antomarchi H., De Weille J., Fosset M. and Lazdunski M. (1987) The receptor for antidiabetic sulfonylureas controls the activity of the ATP-modulated K+ channel in insulin-secreting cells. J. biol. Chem. 262, 15,840-15,844. 25. Schwanstecher M., Behrends S., Brandt C. and Panten U. (1992) The binding properties of the solubilized sulfonylurea receptor from a pancreatic fi-cell line are modulated by Mg+ + complex of ATP. J. Phurmac. exp. Ther. 262,495502. 26. Schwanstecher M., L&r S., Rietze I. and Panten U. (1991) Phosphate and thiophosphate group donating adenine and guanine nucleotides inhibit glibenclamide binding to membranes from pancreatic islets. ~~y~-Sc~je~berg’s Arch. Pharmac. 343,83-89. 27. Stephan H., Baron G. and Schwerdtfeger W. K. (1980) The Brain of the Colon ~urmoset (Callithrix jacchus). Springer, Berlin. 28. Sturgess N. C., Ashford M. L. J., Cook D. L. and Hales C. N. (1985) The sulfonylurea receptor may be an ATP-sensitive K’ channel. L.ancet B, 474-475. 29. Treheme J. M. and Ashford M. L. J. (1991) The regional distribution of sulphonylurea binding sites in rat brain. Neuroscience 40, 523-53 1. 30. Tremblay E., Zini S. and Ben-Ari Y. (1991) Autoradiographic study of the localization of [‘H]glibenclamide binding sites in the rat brain. Neurosci. I.&t. 127, 21-24. 31. Virsolvy-Vergine A., B&k M., Dufour M., Cauvin A., Lupo B. and Bataille D. (1988) An endogenous ligand for the central sulfonylurea receptor. Fe& Eur. biochem. Sots Lett. 242, 65-69. 32. Virsolvy-Vergine A., Leray H., Kuroki S., Lupo B., Dufour M. and Bataille D. (1992) Endosulfine, an endogenous peptidic ligand for the sulfonylurea receptor-purification and partial characterization from ovine brain. Proc. natn. Acad. Sci. U.S.A. BQ, 6629-6633. 33. Zini S., Ben-Ari Y. and Ashford M. L. J. (1991) Characterization of sulfonylurea receptors and the action of potassium channel openers on cholinergic neuro~nsmission in guinea-pig isolated small intestine. J. Phurrnac. exp. Ther. 259, 566-573. 34. Zini S., Raisin M., Lange1 U., Bartfai T. and Ben-Ari Y. (1993) Galanin reduces release of endogenous excitatory amino-acids in the rat hippocampus. Eur. J. Pharmac. 245, l-7. 35. Zini S., Tremblay E., Roisin M. P. and Ben-Ari Y. (1991) Two binding sites for [‘H&libenclamide in the rat brain. Brain Res. 542, 151-154. 36. Zini S., Zini R. and Ben-Ari Y. (1993) Nucleotides modulate the low afhnity binding sites for [3H]glibenclamide in the rat brain. J. Phurmuc. exp. Ther. 264, 701-708. (Accepted 15 March 1993)