J Neural Transm [GenSect] (1992) 87:113-124
m Journal o f Neural Transmission 9 Springer-Verlag 1992 Printed in Austria
Neurotransmitter-mediated inhibition of post-mortem human brain adenylyl cyclase A. Garlind 1, C. J. Fowler 2, I. Alafuzoff 3, B. Winblad 1, and R. F. Cowburn 1
1Department of Geriatric Medicine, Karolinska Institute, Huddinge University Hospital, 2Astra Research Centre AB, S6dertfilje, and 3Department of Pathology, Huddinge University Hospital, Huddinge, Sweden
Accepted September 3, 1991
Summary. The effects of a range of neurotransmitter agonists showing selectivity for receptor types inhibitorily coupled to adenylyl cyclase were compared in membrane preparations of hippocampus, frontal cortex and caudate nucleus / striatum from previously frozen post-mortem human and rat brain. Agonists were tested against basal and forskolin stimulated activities, forskolin being a potent activator of the catalytic sub-unit of the enzyme. Of those agonists tested, only somatostatin (100 gM) and neuropeptide Y (10gM) gave consistent inhibitions of basal and forskolin stimulated enzyme activities in all three regions of both human and rat brain. Somatostatin-mediated inhibition of human brain adenylyl cyclase was reduced in the absence of GTP and in the presence of the guanine nucleotide partial agonist, guanosine 5'-O-thiodiphosphate, consistent with a G-protein-linked receptor. No such GTP-dependence was found for the neuropeptide Y-mediated adenylyl cyclase inhibition. GTP-dependent somatostatin mediated inhibitions of human brain adenylyl cyclase activity were of highest magnitude in the thalamus, intermediate magnitude in the hippocampus and caudate nucleus and lowest magnitude in the frontal cortex. It is concluded that of a range of neurotransmitter receptor agonists tested, only somatostatin gives robust, GTP-dependent responses that are reproducible enough to be used with post-mortem tissue for the comparison of receptor function in human brain disorders.
Keywords: Adenylyl cyclase, post-mortem human brain, somatostatin, receptor neurotransmitters.
Introduction Neurochemical studies of the human brain in disease states have in general been restricted to measurement of presynaptic variables such as neurotrans-
114
A. Garlind et al.
mitter levels, uptake and release, with postsynaptic markers in general being restricted to determinations of receptor recognition site densities using radioligand binding techniques (Dodd et al., 1988; Cowburn et al., 1989; Fowler et al., 1990). However measurements of recognition site densities give little indication as to the functional state of the receptor (Fowler, 1984). Such functional measurements can be obtained by following the receptor-mediated modulation of key enzymes involved in signal transduction, providing that the receptor-enzyme system is stable postmortem. One such enzyme is adenylyl cyclase. Neuronal adenylyl cyclase is responsive to a variety of neurotransmitters that either stimulate or inhibit the enzyme, giving either a rise or a fall in the levels of cyclic adenosine 3'-5'-monophosphate (cAMP), the intracellular second messenger of the system. Receptor mediations of adenylyl cyclase activity are GTPdependent and work through two homologous membrane-associated GTPbinding proteins. These so-called "G"-proteins, termed Gs and Gi for stimulatory and inhibitorily coupled receptors respectively, serve to couple the receptor to the catalytic moiety of the enzyme. (Casey and Gilman, 1988). Previous studies from our laboratory have demonstrated that guanine nucleotide mediation of adenylyl cyclase activity can be reliably measured in postmortem human brain samples (Cowburn etal., 1991). Little is known about the feasibility of measuring neurotransmitter receptor-mediated responses of adenylyl cyclase in post-mortem human brain. Dopamine stimulation of adenylyl cyclase activity has been demonstrated in homogenates of the caudate nucleus from autopsy samples (Clement-Cormier etal., 1974; Carenzi etal., 1975; Shibuya etal., 1979) and vasointestinal peptide (VIP) and isoprenaline stimulations have been reported in cortical and hippocampal membrane preparations (Danielsson etal., 1988; Ohm etal., 1991). However, less is known concerning inhibitorily coupled receptors. NPY has been reported to inhibit basal and forskolin stimulated activity in homogenates of cerebral cortex (Westlind-Danielsson et al., 1987). In the present study, we have investigated adenylyl cyclase activity in post-mortem human brain in response to a number of neurotransmitter agonists, the receptors for which have been reported to be inhibitorily coupled to the enzyme in rat brain tissue. Enzyme activities in membrane preparations from previously frozen human and rat tissue were compared in response to activation of acetylcholine muscarinic, somatostatin, neuropeptide Y (NPY), GABAB, ~2-noradrenergic and 5-HT1A-receptor types. Neurotransmitter mediated responses were studied relative to basal activity and also to enzyme activity assayed in the presence of forskolin, a potent activator of the catalytic sub-unit of the enzyme (Daly et al., 1982).
Experimental procedures Materials
Somatostatin and NPY were purchased from Peninsula laboratories Inc. (Belmont, California, U.S.A.). All other chemicalswere obtained from the Sigma ChemicalCo. (St. Louis, Mo. U.S.A.). Forskolin was dissolvedin absolute alcohol to give a 10raM stock solution.
Adenylyl cyclase in human brain
115
Tissue Human brain tissue was collected at autopsy from individuals who had no clinical history or symptoms of neurological or psychiatric disease. The mean age of the cases studied was 77 4- 4 (range 74-83) years and the death autopsy interval 8.8 4- 5.5 (range 6-17) hours. All experiments with the exception of the regional distribution studies, were performed on tissue from the gyrus frontalis superior. Tissue for the regional distribution and doseresponse studies was dissected from the superior frontal cortex, caudate nucleus, ventral hippocampus, thalamus and white matter. Upon dissection, tissue was slowly frozen in 250 mg-1 g pieces for storage at - 70 ~ None of the cases used showed evidence of neuropathological abnormalities as assessed using standard histological tests on fixed tissue. Rat brain tissue was from 200-300 g Sprague-Dawley animals. These animals were killed by decapitation and the brains removed. Tissue from the cerebral cortex, striatum and hippocampus was dissected and then frozen and stored as for the human brain samples.
Membrane preparation Membranes were prepared from human and rat tissue as described by Cowburn et al. (1991). Briefly, crude synaptic membrane fractions were homogenised in 20 volumes (w/v) of icecold buffer (10 mM Tris-acetate/2mM EGTA, pH 7.4) and then centrifuged at 1000 g for 10 rains at 4 ~ The supernatants obtained were centrifuged at 20 000 g for 20 rains and resultant pellets resuspended in 20 volumes of fresh buffer prior to sonication and recentrifugation at 20 000 g. This stage was repeated and the final pellet resuspended in 4 volumes of buffer, and then slowly frozen in aliquots and stored at - 2 0 ~
Adenylyl cyclase assay For assay of adenylyl cyclase, the membrane aliquots were thawed on ice and diluted in ice-cold Tris-EGTA buffer. The samples were then preincubated on ice for 15rain in Eppendorf tubes containing 200 gl reaction mixture, which consisted of 80 mM Tris acetate buffer (pH7.4), 101aM GTP, l mM theophylline, 0.3% bovine serum albumin, 0.5raM EGTA, a nucleoside triphosphate regenerating system (5 mM phosphocreatine and 50 units/ ml creatine phosphokinase), 2 mM MgSO4 (unless otherwise stated, Table 2), 100 gM NaC1 and where appropriate, forskolin and the test agonists. After preincubation for 15 minutes on ice the reactions were started by the addition of ATP (0.5 mM) and the samples incubated for 10 rain at 30 ~ Reactions were stopped by boiling the tubes for 3 min. After cooling, the tubes were centrifuged at 13 000 g for 5 min at 4 ~ Duplicate 50 gl aliquots of supernatant were assayed for cAMP content using an Amersham cAMP assay kit (TRK 432, Amersham International, Amersham, U.K.).
Protein determination Membrane protein content of the samples was determined using a modification of the standard Lowry procedure, with bovine serum albumin as the standard (Markwell et al., 1978).
Presentation of data In Fig. 1, data are shown as pmol cAMP produced. (rag protein)- 1. rain- 1 to indicate the level of adenylyl cyclase activity found in post-mortem human brain samples and significance of effect was determined using a one-way ANOVA. A large inter-sample variation in the adenylyl cyclase activities was found for these preparations, consistent with other studies from our laboratory demonstrating that basal and stimulated adenylyl cyclase activities covary (Cowburn et al., unpublished results). Such inter-sample variations in basal and
116
A. Garlind et al.
forskolin-stimulated adenylyl cyclase activity will mask modest inhibitory effects upon cAMP formation produced by stimulation of inhibitorily-coupled receptors. This problem can be circumvented, however, by expression of data as % inhibition. In consequence, for the data shown in the Tables and in Figs. 2 and 3, results were expressed in this manner and the significance of the inhibitory tests were determined using a one-sample t-test vs 0% inhibition. Results
Forskolin stimulation of adenylyl cyclase activity The optimal conditions for the assay o f adenylyl cyclase in post m o r t e m h u m a n brain samples have been described elsewhere ( C o w b u r n et al., 1991). Using these conditions as described in experimental procedures, forskolin gave a concent r a t i o n - d e p e n d e n t stimulation o f basal adenylyl cyclase activity in h u m a n frontal cortical samples, basal activity being defined as the p r o d u c t i o n o f c A M P in the presence o f 10 g M G T P (Fig. 1). In all further experiments with forskolin, a c o n c e n t r a t i o n giving effective but not m a x i m a l stimulation over basal activity (1 g M ) was used, unless otherwise stated.
Modulation of basal adenylyl cyclase activity by neurotransmitter receptor agonists In b o t h rat a n d h u m a n , a n d in all three brain regions studied, basal adenylyl cyclase activity was significantly reduced by s o m a t o s t a t i n (100 g M ) a n d N P Y
I - -
lO'SM GTP
I
600 '7
._= E
3.. 20,.
400 -
o3
g & 200-
7-+--1
-6 E
0
0
10-810-710-610-s10-4 Forskolin (M)
Fig. 1. Stimulation of human frontal cortical adenylyl cyclase activity by forskolin. Data are means + s.e.m., n = 4 for the forskolin and GTP concentrations shown. At a [GTP] of 10gM, one-way ANOVA indicated significant effect of forskolin (F5,18 = 18.4, p < 0.001). **p < 0.01, Dunnett's t-test vs [forskolin] = 0 at 10gM GTP. +0.1 > p < 0.05, two-tailed t-test for 0 vs 10gM GTP in the absence of forskolin
Adenylyl cyclase in human brain
117
(10 I~M), (Fig. 2). The effect of somatostatin on basal adenylyl cyclase activity was unaltered by the presence of bacitracin (data not shown) suggesting little proteolytic breakdown of the peptide under the assay conditions used. The muscarinic agonists carbachol, acetylcholine and oxotremorine significantly reduced basal adenylyl cyclase activity in the rat striatum, but were without consistent effect in the human samples at the concentration tested (100gM) (Fig. 2). The GABAB-agonist baclofen (100 gM) gave significant reductions in rat frontal cortex and human hippocampus. The 0h-adrenoceptor agonist clonidine and the 5-HT1A-receptor agonist 8-OH DPAT, respectively, did not produce consistent inhibitions of basal activity in either rat or human samples at the concentration tested (100 ~tM) (Fig. 2).
Modulation of forskolin-stimulated adenylyl cyclase activity by neurotransmitter receptor agonists Neurotransmitter receptor agonist effects were further studied in the same three brain regions in the presence of 1 gM forskolin. The effects of the receptor agonists on forskolin-stimulated adenylyl cyclase activity mirrored their effects on basal activity, with the exceptions of baclofen and 8-OH-DPAT, which gave significant inhibitions of forskolin-stimulated adenylyl cyclase activity in rat
CORTEX I-
(5 X 0 >' o
Z _~
Z
m
=
(.3
o.. Z
o
.< rn
o
_J 0
~
r, 0
d X 0
=
(.3
p > 0.05
the effect on b o t h basal a n d f o r s k o l i n - s t i m u l a t e d adenylyl cyclase activity were studied. F o r s k o l i n was used in these experiments as a m e a n s o f presensitizing the enzyme to inhibitory agonist responses (Daly et al., 1982). I n h i b i t i o n of rat adenylyl cyclase activity was f o u n d with the muscarinic agonists, s o m a t o s t a t i n a n d N P Y (Figs. 2 a n d 3), a finding in a g r e e m e n t with the literature (Olianas etal., 1982; W e s t l i n d - D a n i e l s s o n etal., 1987; M c K i n n e y et al., 1988; M a r k s t e i n etal., 1989; N a g a o etal., 1989; Schettini etal., 1989). T h e G A B A B - a g o n i s t baclofen a n d the 5-HT1A-agonist 8 - O H D P A T also p r o d u c e d inhibitory effects. These were m o r e p r o n o u n c e d in the presence o f forskolin consistent with animal studies ( D u m u i s et al., 1987; O k s e n b e r g et al., 1988; N i s h i k a w a et al., 1 9 8 9 ) b u t were n o t seen in all three regions (Figs. 2 a n d 3). N o significant inhibition was seen with the ~2-adrenoceptor agonist clonidine, with stimulatory effects being seen in the rat h i p p o c a m p u s (Figs. 2 a n d 3). These inconsistent effects of clonidine on adenylyl cyclase activity are similar to those r e p o r t e d in the literature, with b o t h stimulatory (Pilc a n d E n n a , 1985) a n d inhibitory ( K i t a m u r a etal., 1985; K u n o et al., 1990) effects having been shown. W i t h respect to the h u m a n samples, no consistent effects o f the muscarinic agonists u p o n basal or forskolinstimulated adenylyl cyclase were seen (Figs. 2 a n d 3). In addition, the effects o f carbachol u p o n a given m e m b r a n e p r e p a r a t i o n varied n o t only with the assay c o n d i t i o n used, b u t also f r o m assay to assay for the same c o n d i t i o n
Adenylyl cyclase in human brain
121
indicating that this parameter was not stable in post-mortem brain tissue. Similarly the effects of baclofen, clonidine and 8-OH-DPAT upon both basal and forskolin-stimulated adenylyl cyclase activity were not sufficiently robust in the human samples to be of practical use for the functional study of these receptor systems in post-mortem tissue. In contrast to the above, robust inhibitory effects of the neuropeptides somatostatin and NPY, were found in both human and rat samples. The NPY effect on basal and forskolin-stimulated activity is in agreement with the study of Westlind-Danielsson et al. (1987), using rat and human cortical samples. A relatively high concentration of somatostatin (100 gM) was required to produce inhibition. The same concentration was used by Schettini etal. (1989) in their study on somatostatin inhibition of rat brain adenylyl cyclase although lower concentrations have been found to be effective in the rat cerebral cortex when freshly dissociated cells are used (Colas et al., 1990). The somatostatin analogue [D-Trp, D-Cys]-somatostatin was inactive at the highest concentration tested (10 gM) (Table 2), a finding that contrasts with that of Chneiweiss et al. (1987) who found the somatostatin analogue to be a more potent inhibitor than somatostatin. However, two lines of evidence would suggest that the effect of somatostatin is receptor-mediated rather than a non-specific inhibitory effect. Firstly, the GTP dependence of the inhibitory effect (Table 1) would suggest involvement of a G-protein in the response. This finding is consistent with binding studies on post-mortem human brain tissue which have shown that the binding of somatostatin to its recognition site is affected by GTP in a manner consistent with the involvement of a G-protein (Bergstr6m etal., 1991). In addition Moser and Cramer (1990) have shown that somatostatin acts through G-proteins on dopaminergic adenylyl cyclase in rat brain. Furthermore, Nagao etal. (1989) and Schettini etal. (1989) have shown that somatostatin-reduced cAMP production in the rat brain occurs via a G-protein coupled to the adenylyl cyclase. Secondly there appeared to be a regional variation in the magnitude of the inhibitory effects of somatostatin, with large effects being found in thalamus, hippocampus and nucleus caudatus, smaller effects in the frontal cortex and no significant effect being seen in white matter (Table 1). This variation, plus the lack of inhibitory effect of somatostatin (1 gM) on basal adenylyl cyclase activity in primary cultures of mouse embryonic glial cells reported by Chneiweiss et al. (1985) would argue against a non-specific inhibitory effect of the neuropeptide. The relatively low effect of somatostatin upon frontal cortical adenylyl cyclase and the robust effect seen in the thalamus is at variance with studies measuring somatostatin receptor recognition site densities, where the reverse is found (Beal etal., 1986; Whitford etal., 1986). However in a recent study of post-mortem human brain (Bergstr6m et al., 1991) we have seen no correlation between somatostatin receptor site densities and somatostatin-induced adenylyl cyclase inhibition. This discrepancy may reflect different receptor coupling efflciencies in the different regions. Such regional variations in receptor coupling efficiency have been found for other receptor systems, such as, for
122
A. Garlind et al.
example, muscarinic receptors coupled to phosphoinositide-specific phospholipase C (Fisher and Snider, 1987). With respect to NPY, the involvement of a G-protein-coupled receptor is less clear, since the inhibitory effect of adenylyl cyclase did not show GTP dependence (Table 1). This finding differs from radioligand binding data, where it has been shown that the binding of N-[propionyl-3H]NPY to human frontal and temporal cortex is reduced in the presence of GTP and the stable GTP analogue Gpp[NH]p, but not by ATP or App[NH]p (Westlind-Danielsson et al., 1987). One possible explanation for this discrepancy is that NPY not only interacts with the G-protein-coupled receptor but also directly affects the Gprotein-adenylyl cyclase complex. Although such an explanation is speculative, evidence is emerging that neuropeptides can act in this manner: thus in rat peritoneal mast cells, substance P has been suggested to act not only via its receptors but also through a direct activation of G-proteins (Mousli et al., 1990). In conclusion, the present study has investigated the feasibility of measuring the modulation in human brain post-mortem samples of adenylyl cyclase by a number of inhibitorily-coupled receptor systems. Of the agonists tested, only the neuropeptides somatostatin and NPY produced robust inhibitory responses. The effect of somatostatin was GTP-dependent and showed a regional variation in efficacy, consistent with an action via a G-protein-coupled receptor rather than via a non-specific effect on the adenylyl cyclase catalytic activity. These data would suggest that measurement of somatostatin-mediated inhibition of adenylyl cyclase can provide a useful measure of somatostatinergic receptor function in human post-mortem brain samples that can be applied to the study of the neurochemical pathology of this receptor system.
Acknowledgements We would like to thank the Swedish MRC (grant number B90.12X.05664), Axelsson Johnsons Foundation, Hedlunds Foundation, Stohnes Foundation, Gamla Tj/inarinnor Foundation and the Loo and Hans Ostermans Foundation for financialsupport. The award of a WellcomeTrust Travelling Fellowship (R.F.C) is also gratefully acknowledged.
References Bergstr6m L, Garlind A, Nilsson L, Alafuzoff I, Fowler CJ, Winblad B, Cowburn RF (1991) Regional distribution of somatostatin receptor binding and modulation of adenylyl cyclase activity in Alzheimer's disease brain. J Neurol Sci (in press) Beal MF, Tran VT, Mazurek MF, Chattha G, Martin JB (1986) Somatostatin binding sites in human and monkey brain: localizationand characterization. J Neurochem 46: 359-365 Carenzi A, Gillin JC, Guidotti A, SchwartzMA, Trabucci M, Wyatt RJ (1975) Dopaminesensitive adenylate cyclase in human caudate nucleus: a study in control subjects and schizophrenic patients. Arch Gen Psychiatry 32:1056-1059 Casey PJ, Gilman AG (1988) G protein involvementin receptor-effectorcoupling. J Biol Chem 263:2577-2580 ChneiweissH, GlowinskiJ, Pr~mont J (1985) Modulation by monoaminesof somatostatin-
Adenylyl cyclase in human brain
123
sensitive adenylate cyclase on neuronal and glial cells from the mouse brain in primary cultures. J Neurochem 44:1825-1831 Chneiweiss H, Bertrand Ph, Epelbaum J, Kordon C, Glowinski J, Premont J, Enjalbert A (1987) Somatostatin receptors on cortical neurones and adenohypophysis: comparison between specific binding and adenylate cyclase inhibition. Eur J Pharm 138:249-255 Clement-Cormier YC, Kebabian JW, Petzold GL, Greengard P (1974) Dopamine-sensitive adenylate cyclase in mammalian brain: a possible site of action of antipsychotic drugs. Proc Natl Acad Sci 71:1113-1117 Colas B, Lopez Ruiz MP, Prieto JC, Arilla E (1990) Somatostatin inhibition of VIP- and isoproterenol-stimulated cyclic AMP accumulation in dissociated cells from rat cerebral cortex. Neuropeptides 15:235-239 Cowburn RF, Hardy JA, Roberts PJ (1989) Transmitter deficits in Alzheimer's disease. In: Davies DC (ed) Alzheimer's disease: towards an understanding of the aetiology and pathogenesis. John Libbey, London, pp 9-32 Cowburn RF, Garlind A, O'Neill C, Alafuzoff I, Winblad B, Fowler CJ (1991) Characterization and regional distribution of adenylyl cyclase activity from human brain. Neurochem Int 18:389-398 Daly JW, Padgett W, Seamon KB (1982) Activation of cyclic AMP-generating systems in brain membranes and slices by the diterpene forskolin: augmentation of receptormediated responses. J Neurochem 38:532-544 Danielsson E, Eckern/is S-A, Westlind-Danielsson A, Nordstr6m O, Bartfai T, Gottfries C-G, Wallin A (1988) VIP-sensitive adenylate cyclase, guanylate cyclase, muscarinic receptors, choline acetyltransferase and acetylcholinesterase in brain tissue afflicted by Alzheimer's disease/senile dementia of the Alzheimer type. Neurobiol Aging 9: 153162 Dodd PR, Hambley JW, Cowburn RF, Hardy JA (1988) A comparison of methodologies for the study of functional transmitter neurochemistry in human brain. J Neurochem 50:1333-1345 Dumuis A, Sebben M, Bockaert J (1988) Pharmacology of 5-hydroxytryptamine-lA receptors which inhibit cAMP production in hippocampal and cortical neurons in primary culture. Mol Pharmacol 33:178-186 Fisher SK, Snider RM (1987) Differential receptor occupancy requirements for muscarinic cholinergic stimulation of inositol lipid hydrolysis in brain and in neuroblastomas. Mol Pharmacol 32:81-90 Fowler CJ (1984) Receptor binding studies: yet more cause for caution. Trends Pharmacol Sci 5:498-499 Fowler C J, Cowburn RF, Hardy JA, Wester P, Winblad B (1990) Neurotransmitter function in post-mortem human brain: an overview. In: Bunney WE, Hippius H, Laakman G, Schmauss M (eds) Neuropsychopharmacology. Springer, Berlin Heidelberg New York Tokyo, pp 668-674 Kitamura Y, Nomura Y, Segawa T (1985) Possible involvement of inhibitory GTP binding regulatory protein in a2-adrenoceptor-mediated inhibition of adenylate cyclase in cerebral cortical membranes of rats. J Neurochem 5:1504-1508 Kuno N, Kamisaki Y, Itoh T (1990) Inhibition of cyclic AMP accumulation by az-adrenoceptors in the rat cerebral cortex. Eur J Pharmacol 176:281-287 Markstein R, St6ckli KA, Reubi JC (1989) Differential effects ofsomatostation on adenylate cyclase as functional correlate for different brain somatostatin receptor subpopulations. Neurosci Lett 104:13-18 Markwell MAK, Haas SM, Bieber LL, Tolbert NE (1978) A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal Biochem 87:206-210 McKinney M, Anderson D, Vella-Rountree L (1988) Different agonist-receptor active
124
A. Garlind etal.: Adenylyl cyclase in human brain
conformations for rat brain M1 and M2 muscarinic receptors that are separately coupled to two biochemical effector systems. Mol Pharmacol 35:39-47 Moser A, Cramer H (1990) Somatostatin acts through G-proteins on dopaminergic adenylate cyclase in the caudate-putamen of the rat. Neurochem Res 15:1085-1088 Mousli M, Bueb J-L, Bronner C, Rouot B, Landry Y (1990) G protein activation: a receptorindependent mode of action for cationic amphiphilic neuropeptides and venom peptides. Trends Pharmacol Sci 11:358-362 Nagao M, Sakamoto C, Matozaki T, Nishizaki H, Konda Y, Nakano O, Baba S (1989) Coupling of inhibitory GTP binding protein to somatostatin receptors on rat cerebrocortical membranes. Folia Endocrinol 65:1357-1366 Nishikawa M, Kuriyama K (1989) Functional coupling of cerebral 7-aminobutyric acid (GABA)~ receptor with adenylate cyclase system: effect of phaclofen. Neurochem Int 14:85-90 Ohm TG, Bohl J, Lemmer B (1991) Reduced basal and stimulated (isoprenaline, Gpp(NH)p, forskolin) adenylate cyclase activity in Alzheimer's disease correlated with histopathological changes. Brain Res 540:229-236 Okada F, Tokumitso Y, Nomura Y (1989) Pertussis toxin attenuates 5-hydroxytryptamine1A receptor mediated inhibition of forskolin-stimulated adenylate cyclase activity in rat hippocampal membranes. J Neurochem 52:1566-1569 Oksenberg D, Peroutka S (1988) Antagonism of 5-hydroxytryptaminelA (5-HT1A) receptormediated modulation of adenylate cyclase activity by pindolol and propranolol isomers. Biochem Pharmacol 37:3429-3433 Olianas C, Onali P, Neff NH, Costa E (1983) Adenylate cyclase activity of synaptic membranes from rat striatum. Inhibition by muscarinic receptor agonists. Mol Pharmacol 23:393-398 Pilc A and Enna SJ (1985) Synergistic interaction between s-and 13-adrenergic receptors in rat brain slices: possible site for antidepressant drug action. Life Sci 37:1183-1194 Rasenick MM, Hughes JM, Wang N (1989) Guanosine-5'-0-thiodiphosphate functions as a partial agonist for the receptor-independent stimulation of neural adenylate cyclase. Brain Res 488:105-113 Schettini G, Florio T, Meucci O, Landolfi E, Grimaldi M, Ventra C, Marino A (1989) Somatostatin inhibition of adenylate cyclase activity in different brain areas. Brain Res 492:65-71 Shibuya M (1987) Dopamine-sensitive adenylate cyclase activity in the striatum in Parkinson's disease. J Neural Transm 44:287-295 Westlind-Danielsson A, Und6n A, Abens J, Andell S, Bartfai T (1987) Neuropeptide Y receptors and the inhibition of adenylate cyclase in the human frontal and temporal cortex. Neurosci Lett 74:237-242 Whitford CA, Bloxham CA, Snell CR, Candy JM, Hirst BH (1986) Regional distribution of high-affinity [3 H]somatostatin binding sites in the human brain. Brain Res 398: 141-147 Authors' address: Dr. A. Garlind, Department of Geriatric Medicine, Huddinge Sjukhus, S-141-86, Huddinge, Sweden. Received May 31, 1991
m Journal o f Neural Transmission 9 Springer-Verlag 1992 Printed in Austria
Neurotransmitter-mediated inhibition of post-mortem human brain adenylyl cyclase A. Garlind 1, C. J. Fowler 2, I. Alafuzoff 3, B. Winblad 1, and R. F. Cowburn 1
1Department of Geriatric Medicine, Karolinska Institute, Huddinge University Hospital, 2Astra Research Centre AB, S6dertfilje, and 3Department of Pathology, Huddinge University Hospital, Huddinge, Sweden
Accepted September 3, 1991
Summary. The effects of a range of neurotransmitter agonists showing selectivity for receptor types inhibitorily coupled to adenylyl cyclase were compared in membrane preparations of hippocampus, frontal cortex and caudate nucleus / striatum from previously frozen post-mortem human and rat brain. Agonists were tested against basal and forskolin stimulated activities, forskolin being a potent activator of the catalytic sub-unit of the enzyme. Of those agonists tested, only somatostatin (100 gM) and neuropeptide Y (10gM) gave consistent inhibitions of basal and forskolin stimulated enzyme activities in all three regions of both human and rat brain. Somatostatin-mediated inhibition of human brain adenylyl cyclase was reduced in the absence of GTP and in the presence of the guanine nucleotide partial agonist, guanosine 5'-O-thiodiphosphate, consistent with a G-protein-linked receptor. No such GTP-dependence was found for the neuropeptide Y-mediated adenylyl cyclase inhibition. GTP-dependent somatostatin mediated inhibitions of human brain adenylyl cyclase activity were of highest magnitude in the thalamus, intermediate magnitude in the hippocampus and caudate nucleus and lowest magnitude in the frontal cortex. It is concluded that of a range of neurotransmitter receptor agonists tested, only somatostatin gives robust, GTP-dependent responses that are reproducible enough to be used with post-mortem tissue for the comparison of receptor function in human brain disorders.
Keywords: Adenylyl cyclase, post-mortem human brain, somatostatin, receptor neurotransmitters.
Introduction Neurochemical studies of the human brain in disease states have in general been restricted to measurement of presynaptic variables such as neurotrans-
114
A. Garlind et al.
mitter levels, uptake and release, with postsynaptic markers in general being restricted to determinations of receptor recognition site densities using radioligand binding techniques (Dodd et al., 1988; Cowburn et al., 1989; Fowler et al., 1990). However measurements of recognition site densities give little indication as to the functional state of the receptor (Fowler, 1984). Such functional measurements can be obtained by following the receptor-mediated modulation of key enzymes involved in signal transduction, providing that the receptor-enzyme system is stable postmortem. One such enzyme is adenylyl cyclase. Neuronal adenylyl cyclase is responsive to a variety of neurotransmitters that either stimulate or inhibit the enzyme, giving either a rise or a fall in the levels of cyclic adenosine 3'-5'-monophosphate (cAMP), the intracellular second messenger of the system. Receptor mediations of adenylyl cyclase activity are GTPdependent and work through two homologous membrane-associated GTPbinding proteins. These so-called "G"-proteins, termed Gs and Gi for stimulatory and inhibitorily coupled receptors respectively, serve to couple the receptor to the catalytic moiety of the enzyme. (Casey and Gilman, 1988). Previous studies from our laboratory have demonstrated that guanine nucleotide mediation of adenylyl cyclase activity can be reliably measured in postmortem human brain samples (Cowburn etal., 1991). Little is known about the feasibility of measuring neurotransmitter receptor-mediated responses of adenylyl cyclase in post-mortem human brain. Dopamine stimulation of adenylyl cyclase activity has been demonstrated in homogenates of the caudate nucleus from autopsy samples (Clement-Cormier etal., 1974; Carenzi etal., 1975; Shibuya etal., 1979) and vasointestinal peptide (VIP) and isoprenaline stimulations have been reported in cortical and hippocampal membrane preparations (Danielsson etal., 1988; Ohm etal., 1991). However, less is known concerning inhibitorily coupled receptors. NPY has been reported to inhibit basal and forskolin stimulated activity in homogenates of cerebral cortex (Westlind-Danielsson et al., 1987). In the present study, we have investigated adenylyl cyclase activity in post-mortem human brain in response to a number of neurotransmitter agonists, the receptors for which have been reported to be inhibitorily coupled to the enzyme in rat brain tissue. Enzyme activities in membrane preparations from previously frozen human and rat tissue were compared in response to activation of acetylcholine muscarinic, somatostatin, neuropeptide Y (NPY), GABAB, ~2-noradrenergic and 5-HT1A-receptor types. Neurotransmitter mediated responses were studied relative to basal activity and also to enzyme activity assayed in the presence of forskolin, a potent activator of the catalytic sub-unit of the enzyme (Daly et al., 1982).
Experimental procedures Materials
Somatostatin and NPY were purchased from Peninsula laboratories Inc. (Belmont, California, U.S.A.). All other chemicalswere obtained from the Sigma ChemicalCo. (St. Louis, Mo. U.S.A.). Forskolin was dissolvedin absolute alcohol to give a 10raM stock solution.
Adenylyl cyclase in human brain
115
Tissue Human brain tissue was collected at autopsy from individuals who had no clinical history or symptoms of neurological or psychiatric disease. The mean age of the cases studied was 77 4- 4 (range 74-83) years and the death autopsy interval 8.8 4- 5.5 (range 6-17) hours. All experiments with the exception of the regional distribution studies, were performed on tissue from the gyrus frontalis superior. Tissue for the regional distribution and doseresponse studies was dissected from the superior frontal cortex, caudate nucleus, ventral hippocampus, thalamus and white matter. Upon dissection, tissue was slowly frozen in 250 mg-1 g pieces for storage at - 70 ~ None of the cases used showed evidence of neuropathological abnormalities as assessed using standard histological tests on fixed tissue. Rat brain tissue was from 200-300 g Sprague-Dawley animals. These animals were killed by decapitation and the brains removed. Tissue from the cerebral cortex, striatum and hippocampus was dissected and then frozen and stored as for the human brain samples.
Membrane preparation Membranes were prepared from human and rat tissue as described by Cowburn et al. (1991). Briefly, crude synaptic membrane fractions were homogenised in 20 volumes (w/v) of icecold buffer (10 mM Tris-acetate/2mM EGTA, pH 7.4) and then centrifuged at 1000 g for 10 rains at 4 ~ The supernatants obtained were centrifuged at 20 000 g for 20 rains and resultant pellets resuspended in 20 volumes of fresh buffer prior to sonication and recentrifugation at 20 000 g. This stage was repeated and the final pellet resuspended in 4 volumes of buffer, and then slowly frozen in aliquots and stored at - 2 0 ~
Adenylyl cyclase assay For assay of adenylyl cyclase, the membrane aliquots were thawed on ice and diluted in ice-cold Tris-EGTA buffer. The samples were then preincubated on ice for 15rain in Eppendorf tubes containing 200 gl reaction mixture, which consisted of 80 mM Tris acetate buffer (pH7.4), 101aM GTP, l mM theophylline, 0.3% bovine serum albumin, 0.5raM EGTA, a nucleoside triphosphate regenerating system (5 mM phosphocreatine and 50 units/ ml creatine phosphokinase), 2 mM MgSO4 (unless otherwise stated, Table 2), 100 gM NaC1 and where appropriate, forskolin and the test agonists. After preincubation for 15 minutes on ice the reactions were started by the addition of ATP (0.5 mM) and the samples incubated for 10 rain at 30 ~ Reactions were stopped by boiling the tubes for 3 min. After cooling, the tubes were centrifuged at 13 000 g for 5 min at 4 ~ Duplicate 50 gl aliquots of supernatant were assayed for cAMP content using an Amersham cAMP assay kit (TRK 432, Amersham International, Amersham, U.K.).
Protein determination Membrane protein content of the samples was determined using a modification of the standard Lowry procedure, with bovine serum albumin as the standard (Markwell et al., 1978).
Presentation of data In Fig. 1, data are shown as pmol cAMP produced. (rag protein)- 1. rain- 1 to indicate the level of adenylyl cyclase activity found in post-mortem human brain samples and significance of effect was determined using a one-way ANOVA. A large inter-sample variation in the adenylyl cyclase activities was found for these preparations, consistent with other studies from our laboratory demonstrating that basal and stimulated adenylyl cyclase activities covary (Cowburn et al., unpublished results). Such inter-sample variations in basal and
116
A. Garlind et al.
forskolin-stimulated adenylyl cyclase activity will mask modest inhibitory effects upon cAMP formation produced by stimulation of inhibitorily-coupled receptors. This problem can be circumvented, however, by expression of data as % inhibition. In consequence, for the data shown in the Tables and in Figs. 2 and 3, results were expressed in this manner and the significance of the inhibitory tests were determined using a one-sample t-test vs 0% inhibition. Results
Forskolin stimulation of adenylyl cyclase activity The optimal conditions for the assay o f adenylyl cyclase in post m o r t e m h u m a n brain samples have been described elsewhere ( C o w b u r n et al., 1991). Using these conditions as described in experimental procedures, forskolin gave a concent r a t i o n - d e p e n d e n t stimulation o f basal adenylyl cyclase activity in h u m a n frontal cortical samples, basal activity being defined as the p r o d u c t i o n o f c A M P in the presence o f 10 g M G T P (Fig. 1). In all further experiments with forskolin, a c o n c e n t r a t i o n giving effective but not m a x i m a l stimulation over basal activity (1 g M ) was used, unless otherwise stated.
Modulation of basal adenylyl cyclase activity by neurotransmitter receptor agonists In b o t h rat a n d h u m a n , a n d in all three brain regions studied, basal adenylyl cyclase activity was significantly reduced by s o m a t o s t a t i n (100 g M ) a n d N P Y
I - -
lO'SM GTP
I
600 '7
._= E
3.. 20,.
400 -
o3
g & 200-
7-+--1
-6 E
0
0
10-810-710-610-s10-4 Forskolin (M)
Fig. 1. Stimulation of human frontal cortical adenylyl cyclase activity by forskolin. Data are means + s.e.m., n = 4 for the forskolin and GTP concentrations shown. At a [GTP] of 10gM, one-way ANOVA indicated significant effect of forskolin (F5,18 = 18.4, p < 0.001). **p < 0.01, Dunnett's t-test vs [forskolin] = 0 at 10gM GTP. +0.1 > p < 0.05, two-tailed t-test for 0 vs 10gM GTP in the absence of forskolin
Adenylyl cyclase in human brain
117
(10 I~M), (Fig. 2). The effect of somatostatin on basal adenylyl cyclase activity was unaltered by the presence of bacitracin (data not shown) suggesting little proteolytic breakdown of the peptide under the assay conditions used. The muscarinic agonists carbachol, acetylcholine and oxotremorine significantly reduced basal adenylyl cyclase activity in the rat striatum, but were without consistent effect in the human samples at the concentration tested (100gM) (Fig. 2). The GABAB-agonist baclofen (100 gM) gave significant reductions in rat frontal cortex and human hippocampus. The 0h-adrenoceptor agonist clonidine and the 5-HT1A-receptor agonist 8-OH DPAT, respectively, did not produce consistent inhibitions of basal activity in either rat or human samples at the concentration tested (100 ~tM) (Fig. 2).
Modulation of forskolin-stimulated adenylyl cyclase activity by neurotransmitter receptor agonists Neurotransmitter receptor agonist effects were further studied in the same three brain regions in the presence of 1 gM forskolin. The effects of the receptor agonists on forskolin-stimulated adenylyl cyclase activity mirrored their effects on basal activity, with the exceptions of baclofen and 8-OH-DPAT, which gave significant inhibitions of forskolin-stimulated adenylyl cyclase activity in rat
CORTEX I-
(5 X 0 >' o
Z _~
Z
m
=
(.3
o.. Z
o
.< rn
o
_J 0
~
r, 0
d X 0
=
(.3
p > 0.05
the effect on b o t h basal a n d f o r s k o l i n - s t i m u l a t e d adenylyl cyclase activity were studied. F o r s k o l i n was used in these experiments as a m e a n s o f presensitizing the enzyme to inhibitory agonist responses (Daly et al., 1982). I n h i b i t i o n of rat adenylyl cyclase activity was f o u n d with the muscarinic agonists, s o m a t o s t a t i n a n d N P Y (Figs. 2 a n d 3), a finding in a g r e e m e n t with the literature (Olianas etal., 1982; W e s t l i n d - D a n i e l s s o n etal., 1987; M c K i n n e y et al., 1988; M a r k s t e i n etal., 1989; N a g a o etal., 1989; Schettini etal., 1989). T h e G A B A B - a g o n i s t baclofen a n d the 5-HT1A-agonist 8 - O H D P A T also p r o d u c e d inhibitory effects. These were m o r e p r o n o u n c e d in the presence o f forskolin consistent with animal studies ( D u m u i s et al., 1987; O k s e n b e r g et al., 1988; N i s h i k a w a et al., 1 9 8 9 ) b u t were n o t seen in all three regions (Figs. 2 a n d 3). N o significant inhibition was seen with the ~2-adrenoceptor agonist clonidine, with stimulatory effects being seen in the rat h i p p o c a m p u s (Figs. 2 a n d 3). These inconsistent effects of clonidine on adenylyl cyclase activity are similar to those r e p o r t e d in the literature, with b o t h stimulatory (Pilc a n d E n n a , 1985) a n d inhibitory ( K i t a m u r a etal., 1985; K u n o et al., 1990) effects having been shown. W i t h respect to the h u m a n samples, no consistent effects o f the muscarinic agonists u p o n basal or forskolinstimulated adenylyl cyclase were seen (Figs. 2 a n d 3). In addition, the effects o f carbachol u p o n a given m e m b r a n e p r e p a r a t i o n varied n o t only with the assay c o n d i t i o n used, b u t also f r o m assay to assay for the same c o n d i t i o n
Adenylyl cyclase in human brain
121
indicating that this parameter was not stable in post-mortem brain tissue. Similarly the effects of baclofen, clonidine and 8-OH-DPAT upon both basal and forskolin-stimulated adenylyl cyclase activity were not sufficiently robust in the human samples to be of practical use for the functional study of these receptor systems in post-mortem tissue. In contrast to the above, robust inhibitory effects of the neuropeptides somatostatin and NPY, were found in both human and rat samples. The NPY effect on basal and forskolin-stimulated activity is in agreement with the study of Westlind-Danielsson et al. (1987), using rat and human cortical samples. A relatively high concentration of somatostatin (100 gM) was required to produce inhibition. The same concentration was used by Schettini etal. (1989) in their study on somatostatin inhibition of rat brain adenylyl cyclase although lower concentrations have been found to be effective in the rat cerebral cortex when freshly dissociated cells are used (Colas et al., 1990). The somatostatin analogue [D-Trp, D-Cys]-somatostatin was inactive at the highest concentration tested (10 gM) (Table 2), a finding that contrasts with that of Chneiweiss et al. (1987) who found the somatostatin analogue to be a more potent inhibitor than somatostatin. However, two lines of evidence would suggest that the effect of somatostatin is receptor-mediated rather than a non-specific inhibitory effect. Firstly, the GTP dependence of the inhibitory effect (Table 1) would suggest involvement of a G-protein in the response. This finding is consistent with binding studies on post-mortem human brain tissue which have shown that the binding of somatostatin to its recognition site is affected by GTP in a manner consistent with the involvement of a G-protein (Bergstr6m etal., 1991). In addition Moser and Cramer (1990) have shown that somatostatin acts through G-proteins on dopaminergic adenylyl cyclase in rat brain. Furthermore, Nagao etal. (1989) and Schettini etal. (1989) have shown that somatostatin-reduced cAMP production in the rat brain occurs via a G-protein coupled to the adenylyl cyclase. Secondly there appeared to be a regional variation in the magnitude of the inhibitory effects of somatostatin, with large effects being found in thalamus, hippocampus and nucleus caudatus, smaller effects in the frontal cortex and no significant effect being seen in white matter (Table 1). This variation, plus the lack of inhibitory effect of somatostatin (1 gM) on basal adenylyl cyclase activity in primary cultures of mouse embryonic glial cells reported by Chneiweiss et al. (1985) would argue against a non-specific inhibitory effect of the neuropeptide. The relatively low effect of somatostatin upon frontal cortical adenylyl cyclase and the robust effect seen in the thalamus is at variance with studies measuring somatostatin receptor recognition site densities, where the reverse is found (Beal etal., 1986; Whitford etal., 1986). However in a recent study of post-mortem human brain (Bergstr6m et al., 1991) we have seen no correlation between somatostatin receptor site densities and somatostatin-induced adenylyl cyclase inhibition. This discrepancy may reflect different receptor coupling efflciencies in the different regions. Such regional variations in receptor coupling efficiency have been found for other receptor systems, such as, for
122
A. Garlind et al.
example, muscarinic receptors coupled to phosphoinositide-specific phospholipase C (Fisher and Snider, 1987). With respect to NPY, the involvement of a G-protein-coupled receptor is less clear, since the inhibitory effect of adenylyl cyclase did not show GTP dependence (Table 1). This finding differs from radioligand binding data, where it has been shown that the binding of N-[propionyl-3H]NPY to human frontal and temporal cortex is reduced in the presence of GTP and the stable GTP analogue Gpp[NH]p, but not by ATP or App[NH]p (Westlind-Danielsson et al., 1987). One possible explanation for this discrepancy is that NPY not only interacts with the G-protein-coupled receptor but also directly affects the Gprotein-adenylyl cyclase complex. Although such an explanation is speculative, evidence is emerging that neuropeptides can act in this manner: thus in rat peritoneal mast cells, substance P has been suggested to act not only via its receptors but also through a direct activation of G-proteins (Mousli et al., 1990). In conclusion, the present study has investigated the feasibility of measuring the modulation in human brain post-mortem samples of adenylyl cyclase by a number of inhibitorily-coupled receptor systems. Of the agonists tested, only the neuropeptides somatostatin and NPY produced robust inhibitory responses. The effect of somatostatin was GTP-dependent and showed a regional variation in efficacy, consistent with an action via a G-protein-coupled receptor rather than via a non-specific effect on the adenylyl cyclase catalytic activity. These data would suggest that measurement of somatostatin-mediated inhibition of adenylyl cyclase can provide a useful measure of somatostatinergic receptor function in human post-mortem brain samples that can be applied to the study of the neurochemical pathology of this receptor system.
Acknowledgements We would like to thank the Swedish MRC (grant number B90.12X.05664), Axelsson Johnsons Foundation, Hedlunds Foundation, Stohnes Foundation, Gamla Tj/inarinnor Foundation and the Loo and Hans Ostermans Foundation for financialsupport. The award of a WellcomeTrust Travelling Fellowship (R.F.C) is also gratefully acknowledged.
References Bergstr6m L, Garlind A, Nilsson L, Alafuzoff I, Fowler CJ, Winblad B, Cowburn RF (1991) Regional distribution of somatostatin receptor binding and modulation of adenylyl cyclase activity in Alzheimer's disease brain. J Neurol Sci (in press) Beal MF, Tran VT, Mazurek MF, Chattha G, Martin JB (1986) Somatostatin binding sites in human and monkey brain: localizationand characterization. J Neurochem 46: 359-365 Carenzi A, Gillin JC, Guidotti A, SchwartzMA, Trabucci M, Wyatt RJ (1975) Dopaminesensitive adenylate cyclase in human caudate nucleus: a study in control subjects and schizophrenic patients. Arch Gen Psychiatry 32:1056-1059 Casey PJ, Gilman AG (1988) G protein involvementin receptor-effectorcoupling. J Biol Chem 263:2577-2580 ChneiweissH, GlowinskiJ, Pr~mont J (1985) Modulation by monoaminesof somatostatin-
Adenylyl cyclase in human brain
123
sensitive adenylate cyclase on neuronal and glial cells from the mouse brain in primary cultures. J Neurochem 44:1825-1831 Chneiweiss H, Bertrand Ph, Epelbaum J, Kordon C, Glowinski J, Premont J, Enjalbert A (1987) Somatostatin receptors on cortical neurones and adenohypophysis: comparison between specific binding and adenylate cyclase inhibition. Eur J Pharm 138:249-255 Clement-Cormier YC, Kebabian JW, Petzold GL, Greengard P (1974) Dopamine-sensitive adenylate cyclase in mammalian brain: a possible site of action of antipsychotic drugs. Proc Natl Acad Sci 71:1113-1117 Colas B, Lopez Ruiz MP, Prieto JC, Arilla E (1990) Somatostatin inhibition of VIP- and isoproterenol-stimulated cyclic AMP accumulation in dissociated cells from rat cerebral cortex. Neuropeptides 15:235-239 Cowburn RF, Hardy JA, Roberts PJ (1989) Transmitter deficits in Alzheimer's disease. In: Davies DC (ed) Alzheimer's disease: towards an understanding of the aetiology and pathogenesis. John Libbey, London, pp 9-32 Cowburn RF, Garlind A, O'Neill C, Alafuzoff I, Winblad B, Fowler CJ (1991) Characterization and regional distribution of adenylyl cyclase activity from human brain. Neurochem Int 18:389-398 Daly JW, Padgett W, Seamon KB (1982) Activation of cyclic AMP-generating systems in brain membranes and slices by the diterpene forskolin: augmentation of receptormediated responses. J Neurochem 38:532-544 Danielsson E, Eckern/is S-A, Westlind-Danielsson A, Nordstr6m O, Bartfai T, Gottfries C-G, Wallin A (1988) VIP-sensitive adenylate cyclase, guanylate cyclase, muscarinic receptors, choline acetyltransferase and acetylcholinesterase in brain tissue afflicted by Alzheimer's disease/senile dementia of the Alzheimer type. Neurobiol Aging 9: 153162 Dodd PR, Hambley JW, Cowburn RF, Hardy JA (1988) A comparison of methodologies for the study of functional transmitter neurochemistry in human brain. J Neurochem 50:1333-1345 Dumuis A, Sebben M, Bockaert J (1988) Pharmacology of 5-hydroxytryptamine-lA receptors which inhibit cAMP production in hippocampal and cortical neurons in primary culture. Mol Pharmacol 33:178-186 Fisher SK, Snider RM (1987) Differential receptor occupancy requirements for muscarinic cholinergic stimulation of inositol lipid hydrolysis in brain and in neuroblastomas. Mol Pharmacol 32:81-90 Fowler CJ (1984) Receptor binding studies: yet more cause for caution. Trends Pharmacol Sci 5:498-499 Fowler C J, Cowburn RF, Hardy JA, Wester P, Winblad B (1990) Neurotransmitter function in post-mortem human brain: an overview. In: Bunney WE, Hippius H, Laakman G, Schmauss M (eds) Neuropsychopharmacology. Springer, Berlin Heidelberg New York Tokyo, pp 668-674 Kitamura Y, Nomura Y, Segawa T (1985) Possible involvement of inhibitory GTP binding regulatory protein in a2-adrenoceptor-mediated inhibition of adenylate cyclase in cerebral cortical membranes of rats. J Neurochem 5:1504-1508 Kuno N, Kamisaki Y, Itoh T (1990) Inhibition of cyclic AMP accumulation by az-adrenoceptors in the rat cerebral cortex. Eur J Pharmacol 176:281-287 Markstein R, St6ckli KA, Reubi JC (1989) Differential effects ofsomatostation on adenylate cyclase as functional correlate for different brain somatostatin receptor subpopulations. Neurosci Lett 104:13-18 Markwell MAK, Haas SM, Bieber LL, Tolbert NE (1978) A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal Biochem 87:206-210 McKinney M, Anderson D, Vella-Rountree L (1988) Different agonist-receptor active
124
A. Garlind etal.: Adenylyl cyclase in human brain
conformations for rat brain M1 and M2 muscarinic receptors that are separately coupled to two biochemical effector systems. Mol Pharmacol 35:39-47 Moser A, Cramer H (1990) Somatostatin acts through G-proteins on dopaminergic adenylate cyclase in the caudate-putamen of the rat. Neurochem Res 15:1085-1088 Mousli M, Bueb J-L, Bronner C, Rouot B, Landry Y (1990) G protein activation: a receptorindependent mode of action for cationic amphiphilic neuropeptides and venom peptides. Trends Pharmacol Sci 11:358-362 Nagao M, Sakamoto C, Matozaki T, Nishizaki H, Konda Y, Nakano O, Baba S (1989) Coupling of inhibitory GTP binding protein to somatostatin receptors on rat cerebrocortical membranes. Folia Endocrinol 65:1357-1366 Nishikawa M, Kuriyama K (1989) Functional coupling of cerebral 7-aminobutyric acid (GABA)~ receptor with adenylate cyclase system: effect of phaclofen. Neurochem Int 14:85-90 Ohm TG, Bohl J, Lemmer B (1991) Reduced basal and stimulated (isoprenaline, Gpp(NH)p, forskolin) adenylate cyclase activity in Alzheimer's disease correlated with histopathological changes. Brain Res 540:229-236 Okada F, Tokumitso Y, Nomura Y (1989) Pertussis toxin attenuates 5-hydroxytryptamine1A receptor mediated inhibition of forskolin-stimulated adenylate cyclase activity in rat hippocampal membranes. J Neurochem 52:1566-1569 Oksenberg D, Peroutka S (1988) Antagonism of 5-hydroxytryptaminelA (5-HT1A) receptormediated modulation of adenylate cyclase activity by pindolol and propranolol isomers. Biochem Pharmacol 37:3429-3433 Olianas C, Onali P, Neff NH, Costa E (1983) Adenylate cyclase activity of synaptic membranes from rat striatum. Inhibition by muscarinic receptor agonists. Mol Pharmacol 23:393-398 Pilc A and Enna SJ (1985) Synergistic interaction between s-and 13-adrenergic receptors in rat brain slices: possible site for antidepressant drug action. Life Sci 37:1183-1194 Rasenick MM, Hughes JM, Wang N (1989) Guanosine-5'-0-thiodiphosphate functions as a partial agonist for the receptor-independent stimulation of neural adenylate cyclase. Brain Res 488:105-113 Schettini G, Florio T, Meucci O, Landolfi E, Grimaldi M, Ventra C, Marino A (1989) Somatostatin inhibition of adenylate cyclase activity in different brain areas. Brain Res 492:65-71 Shibuya M (1987) Dopamine-sensitive adenylate cyclase activity in the striatum in Parkinson's disease. J Neural Transm 44:287-295 Westlind-Danielsson A, Und6n A, Abens J, Andell S, Bartfai T (1987) Neuropeptide Y receptors and the inhibition of adenylate cyclase in the human frontal and temporal cortex. Neurosci Lett 74:237-242 Whitford CA, Bloxham CA, Snell CR, Candy JM, Hirst BH (1986) Regional distribution of high-affinity [3 H]somatostatin binding sites in the human brain. Brain Res 398: 141-147 Authors' address: Dr. A. Garlind, Department of Geriatric Medicine, Huddinge Sjukhus, S-141-86, Huddinge, Sweden. Received May 31, 1991
