inhibition of acetylcholine release

2 downloads 0 Views 1MB Size Report
Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06510 ..... of substance B as a tool for the study of the biochemical basis.
Proc. Nati. Acad. Sci. USA Vol. 83, pp. 7979-7983, October 1986

Neurobiology

Substance B: An endogenous brain factor that reverses presynaptic inhibition of acetylcholine release (neurotransmitter/nerve terminals /modulation)

L. BRUCE PEARCE*, CHRISTINA G. BENISHINt,

AND

JACK R. COOPERt

Department of Pharmacology, Yale University School of Medicine, New Haven, CT 06510

Communicated by Pasko Rakic, June 23, 1986

Evidence is presented of the existence of a ABSTRACT substance in brain (substance B) that reverses presynaptic inhibition of evoked acetylcholine (AcCho) release. Addition of guinea pig brain synaptosomal incubation medium or whole brain extracts to assays in which 1,1-dimethyl-4-phenylpiperazinium-evoked release of [3H]AcCho from guinea pig myenteric plexus synaptosomes reversed purinergic, muscarinic, and a2-adrenergic agonist inhibition. However, the extracts did not increase basal or evoked release of AcCho. Results of chromatography on Bio-Gel P-2 of a concentrated whole brain extract suggest this factor has an Mr of 700. Substance B was resistant to boiling under acid or alkaline conditions as well as incubation with phospholipase C and various proteases; ashing, however, completely destroyed activity. This endogenous factor was also found to antagonize agonist-mediated inhibition of electrically evoked contractions of myenteric plexus-longitudinal muscle strips from the guinea pig ileum. The demonstration of substance B with its ability to reverse presynaptic inhibition of AcCho release reveals one mode of presynaptic modulation.

date, convincing evidence for the existence of presynaptic adenosine receptors that modulate the release of AcCho from brain synaptosomes has been lacking, with some authors presenting negative findings (24). These observations are consistent with the fact that we were unable to reproducibly demonstrate effects of purinergic agonists on the release of AcCho from brain synaptosomes evoked by potassium, veratridine, or ouabain. Proceeding on the assumption that electrophysiological as well as receptor binding studies accurately reflect the existence of presynaptic purinergic receptors associated with neurons in brain, we investigated the possibility that a factor(s) was present in brain synaptosomal preparations that masked the existence of presynaptic purinergic receptors on cholinergic nerve terminals. In this report, we present evidence of the existence of a neuromodulator in brain that is capable of antagonizing the inhibitory action of a variety of receptor agonists on neurotransmitter release. This factor has tentatively been assigned the name substance B (from brain).

MATERIALS AND METHODS

The concept of presynaptic inhibition of neurotransmitter release was introduced by electrophysiologists in the 1950s (1). The elegant studies of Dudel and Kuffler on the crayfish neuromuscular junction (NMJ) (2) and Eccles and colleagues working on spinal cord reflexes in the cat (3, 4) provided the first clear description of this phenomenon. Dudel and Kuffler proposed that presynaptic -aminobutyric acid receptors exist on nerve terminals at the crayfish NMJ, which, when activated, caused a decrease in the number of quanta of neurotransmitter released by depolarization (2, 5). Since these pioneering studies, many laboratories have presented observations consistent with the existence of receptors on nerve terminals that either enhance or inhibit the release of neurotransmitter (6-10). Our laboratory has been involved in studying the nature and function of presynaptic receptors localized to neurons of the guinea pig ileum myenteric plexus that regulate the release of acetylcholine (AcCho). We have found that synaptosomes isolated from the myenteric plexus provide a model system for the study of presynaptic receptor control of AcCho release. With this preparation we have identified four presynaptic receptors-muscarinic, nicotinic, adrenergic, and purinergic (11-14). Recently, we have turned our attention to presynaptic receptors in brain synaptosomes, focusing initially on the actions of adenosine, a neuromodulator with broad biologic significance that is thought to exert some of its actions by means of presynaptic purinergic receptors (15, 16). Receptor binding, tissue slice, and electrophysiological studies have provided good evidence for the existence of purinergic receptors in brain that perturb neuronal activity (17-23). To

Neurotransmitter Release Assay. Synaptosomes (P2 fraction) isolated from guinea pig ileum myenteric plexuslongitudinal muscle strips as described (11) were suspended in Krebs/Ringer bicarbonate buffer (pH 7.4) (KRB buffer) in the presence of 5 AM choline containing tracer amounts of [3H]choline and incubated at 37°C for 30 min under 1 atmosphere (1 atm = 101.3 kPa) of 95% 02/5% CO2. Synaptosomes were washed by centrifugation at 4°C and 3000 X g for 10 min, followed by resuspension in KRB buffer containing 10 ,uM eserine and centrifugation twice more to remove exogenous labeled choline. Loaded and washed synaptosomes were resuspended in ice-cold KRB buffer by gentle homogenization (glass/Teflon), added to assay tubes maintained on ice, and then incubated for 10 min at 37°C under 1 atmosphere of 95% 02/5% C02. Experiments were conducted so that the release of [3H]AcCho was examined in the presence of substance B under several conditions: in the absence of added drugs, in the presence of evoking agent, and in the presence of evoking tgent plus inhibitory drug. These results were compared against those obtained in the absence of substance B. In this manner effects of the brain extract on basal, evoked release and inhibition of evoked release were determined. Addition of the brain extract to neurotransmitter release assays did not significantly increase either basal or evoked release (although more concentrated extracts often inhibited basal and evoked neurotransmitter release). In Abbreviations: AcCho, acetylcholine; (-)-PIA, (-)-phenylisopropyladenosine; DMPP, 1,1-dimethyl-4-phenylpiperazinium. *Present address: Department of Pharmacology, Harvard Medical School, 250 Longwood Avenue, Boston, MA 02115. tPresent address: Department of Physiology, University of Maryland School of Medicine, Baltimore, MD 21201. tTo whom reprint requests should be addressed.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

7979

7980

Neurobiology: Pearce et al.

experiments in which (-)-phenylisopropyladenosine [(-)PIA] was used as the inhibitory modulator, all assays contained 0.1% dimethyl sulfoxide. The quantitative analysis of [3H]AcCho release was determined as described (11). Synaptosomal Supernate Preparation. Guinea pig brain synaptosomes (P2 fraction) were incubated in the presence of unlabeled choline chloride and washed by centrifugation to remove exogenous choline as described above for myenteric plexus synaptosomes. The choline-loaded and washed synaptosomes were resuspended in KRB buffer by gentle homogenization and then incubated for 10 min at 370C under 1 atmosphere of 95% 02/5% C02. The incubation mixture was chilled in an ice-water bath for 60 sec and then centrifuged at 3000 x g for 10 min and the supernate was retained. This synaptosomal supernate was briefly purged with 95% 02/5% CO2 to maintain the correct pH prior to addition to assays in which the effect of modulating agents on nicotinic-evoked release of [3H]AcCho from myenteric plexus synaptosomes was determined. Brain synaptosomal fractions were prepared by the method of Cotman (25) and extracts of these synaptosomes were obtained by tissue mizing and centrifugation as described below for whole brain preparations. Brain Extract Preparation. Whole guinea pig or rabbit brain (without cerebellum) was homogenized (glass/Teflon) in 5 vol of ice-cold distilled water; the homogenate was then tissue mized at maximal speed for 60 sec on ice using a Tekmor tissue mizer. The resulting broken cell preparation was centrifuged at 100,000 x g for 1 hr and the supernatant solution was ultrafiltered under nitrogen [50 psi (1 psi = 6.89 kPa)] through an Amicon UM-10 (10,000 Mr cutoff) filter. The clear colorless filtrate was lyophilized and stored at -20°C for future use. There was no loss of activity for at least 2 weeks. Extracts of brain prepared in this manner were reconstituted in Krebs/Ringer buffer (pH 7.4) prior to use. Intact Longitudinal Muscle Strip Preparation. The guinea pig ileum longitudinal muscle strip preparation was prepared for recording isometric contractions as described by Paton and Zar (26). Male Hartley guinea pigs (350-450 g) were sacrificed by cervical dislocation and the ileum, excluding the distal 10-12 cm, was removed. The longitudinal muscle layer (with the attached myenteric plexus) was gently teased away from the underlying muscle. Muscle strips (doubled-over) -5-6 cm were suspended in a 5-ml organ bath containing 5 ,uM choline and 25 mM Hepes/Krebs/Ringer buffer (pH 7.4) continuously gassed with 95% 02/5% CO2 with 0.5 g of tension applied to the muscle strip. Following equilibration for 2-3 hr, during which time the buffer within the organ bath was frequently changed, maximal contractions were elicited by electrically stimulating the tissue with square wave pulses at 10-12 V/cm with a duration of 1 msec and frequency of 0.1 Hz. Drugs examined were added directly to the organ bath and washed out by perfusion with fresh KRB buffer. Gel Chromatography. An extract from three guinea pig brains was reconstituted by suspension in 2 ml of Krebs/ Ringer buffer containing 25 mM Hepes at pH 7.4. This suspension was added to a 1.6 x 90 cm Bio-Gel P-2 column preequilibrated with the same buffer and 4.7-ml fractions were collected at 0.5 ml/min. The column void volume was determined using bovine serum albumin and calibrated using various low molecular weight compounds that were detected either by their absorbance at 280 nm or the elution of associated radiolabel. All operations were carried out at 4°C. Materials. (-)-PIA and adenosine deaminase were purchased from Boehringer Mannheim. Choline chloride, ATP, ATP:choline phosphotransferase (EC 2.7.1.32; choline kinase), 1,1-dimethyl-4-phenylpiperazinium (DMPP) iodide, tetraphenylboron, Hepes, and phospholipase C were obtained from Sigma. Formula 963 scintillation cocktail, [14C]glucose, [14C]sucrose, and [methyl-3H]choline chloride were obtained from New England Nuclear. Physostigmine

Proc. Natl. Acad. Sci. USA 83 (1986)

sulfate (eserine) was obtained from ICN and butyronitrile was from Aldrich. Glycylglycine, trypsin, and pronase were obtained from Calbiochem-Behring. Bio-Gel P-2 polyacrylamide gel (400 mesh) was obtained from Bio-Rad.

RESULTS The existence of a factor that reverses receptor-activated inhibition of AcCho release was observed in experiments in which incubation medium from guinea pig brain synaptosomes was added to assays in which the effect of the purinergic agonist (-)-PIA on the release of labeled AcCho from myenteric plexus synaptosomes was examined. When the supernate obtained following centrifugation to remove the synaptosomes was added to these assays, inhibition of DMPP-evoked AcCho release produced by 10 ,M (-)-PIA was completely reversed, whereas the evoked release of [3H]AcCho was unaffected (Table 1). Subsequently, using an extract prepared from whole brain, the inhibition of DMPPevoked release of [3H]AcCho from ileal synaptosomes occurring in the presence of the muscarinic agonist oxotremorine (100 AM), the purinergic agonist (-)-PIA (10 MM), and the adrenergic agonist clonidine (100 MM) was dramatically antagonized (Fig. 1). With this preparation it was necessary to use an unusually high concentration of clonidine. In subsequent experiments using the intact ileal strips as a bioassay, 100 nM clonidine was sufficient to inhibit electrically stimulated contractions and this inhibition was reversed by the addition of substance B (27). Extracts prepared from rat and guinea pig brain synaptosomes purified on sucrose gradients as well as extracts from whole rabbit brain were also found to reverse presynaptic inhibition. Extracts alone did not result in an increase in either basal or evoked release of [3H]AcCho. To effect a more efficient and less laborious assay for this endogenous neuromodulator (substance B), a more classical approach was taken whereby the intact guinea pig ileal longitudinal muscle strip preparation was employed. The results of experiments in which the effect of increasing concentrations of whole brain extracts on the inhibition by 200 nM 2-chloroadenosine of electrically evoked contractions are shown in Fig. 2. Addition of brain extracts alone did not have a significant effect on the force of concentrations, whereas increasing concentrations of the brain extract added to the tissue bath resulted in a concentration-dependent reversal of inhibition produced by the purinergic agonist. It should be noted that desensitization to the inhibitory effect of 2-chloroadenosine does not occur over this same period. Since results of studies conducted in our laboratory and that of others have shown that purinergic agonists affect electriTable 1. Effect of brain synaptosomal incubation medium on purinergic inhibition of AcCho release % inhibition of n [3H]AcCho release Experimental condition 10 104 + 10 (-)-PIA (10 MM) + brain synaptosomal 10 14 + 11 incubation medium Brain synaptosomes (P2 fraction) suspended in Krebs/Ringer buffer (pH 7.4) were preincubated for 10 min at 37°C under 1 atmosphere of 95%02/5% CO2. Synaptosomes were removed from the incubation medium by centrifugation and a volume (200 ,l) of the resultant supernate was added to assays in which DMPP (10 ,uM)-evoked release of [3H]AcCho from ileal synaptosomes was measured. Assays (400 ul) were run for 10 min at 37°C under 1 atmosphere of 95%O2/50% CO2 in Krebs/Ringer buffer (pH 7.4) containing 10 uM eserine and in the presence or absence of 10 ,M (-)-PIA. Assays were terminated by incubation on ice for 60 sec followed by centrifugation. Three-hundred-microliter samples of the supernate were assayed for [3H]AcCho as described (11).

Proc. Natl. Acad. Sci. USA 83 (1986)

Neurobiology: Pearce et al. 100

0

7981

60 0~~~~~~~

40

40 20

o0

OL

20

0

10 /LM

(-)-PIA Substance B -

-

+

100

AM

Oxotremorine _

+

-

0

100 AM Clonidine _

+

FIG. 1. Effect of brain extract presynaptic inhibition. Inhibition of [3H]AcCho release evoked by DMPP (10 MM) from guinea pig myenteric plexus synaptosomes was determined in the presence or absence of equal volumes (200 1l) of reconstituted brain extract. Assays (400 ,ul) were run for 10 min at 37TC under 1 atmosphere of 95% 02/5% CO2 in KRB buffer (pH 7.4) containing 10 AM eserine 10 M (-)-PIA,100 AM clonidine, and in the presence or absence of N and 100 MM oxotremorine. Results are expressed as percent inhibition of DMPP-evoked release of [3H]AcCho with evoked release amounting to -20% of total [3H]AcCho. Results are the mean of at least four experiments. on

somes.

Fig. 3 illustrates the results of experiments in which some of the chemical properties of the endogenous modulator were investigated. Substance B was not inactivated by boiling for 30 min at neutral pH or at either pH 1.0 or 12.5. Treatment with the enzymes Pronase, trypsin, and phospholipase C did not destroy the ability of the brain factor to reverse inhibition of contraction produced by 200 nM 2-chloroadenosine. However, complete loss of activity occurred upon ashing. Preliminary experiments indicate that this factor remains at the origin on paper electrophoresis (pH 5.3, 3 mA per strip, 2 hr) and is not retained on charcoal, ion-exchange resins, or tuBondapak C18 reverse-phase HPLC with 0.1% trifluoroacetic acid as the mobile phase. On TLC or paper chromatography, it migrates only with highly polar media. It is probably not a sugar since preincubation with periodate does not inactivate it. The results of gel chromatography of whole guinea pig brain extract shown in Fig. 4 suggest that this endogenous neuromodulator has an approximate Mr of 700,

40 80 120 Substance B, Al

180

FIG. 3. Resistance of substance B to degradative enzymes. Lyophilized extracts were resuspended in Hepes/Krebs/Ringer buffer (pH 7.4) and incubated for 30 min at 370C in the presence or absence of large excesses of trypsin (n), adenosine deaminase (0), Pronase (o), and phospholipase C (o). Controls were boiled (.) and ashed (e). The incubates were then boiled for 30 min to destroy enzyme activity. The guinea pig ileum myenteric plexus-longitudinal muscle preparation was used as an assay. The ability of increasing amounts of substance B to reverse inhibition of electrically stimulated contraction by 200 mM 2-chloroadenosine is expressed as the percent of contraction height observed in the absence of agonist. The curves are the average of three experiments, with the exception of the curve obtained for the ashed preparation, which is the result of a

evoked AcCho release from guinea pig ileum longitudinal muscle strips (28), it is likely that the reversal of inhibition of contraction caused by 2-chloroadenosine reflects the disinhibition of AcCho release from the intact tissue preparation and corroborates our findings using synapto-

cally

20

single experiment.

assuming no interaction of the column with the factor. Acid hydrolysis (6 M HCl, 24 hr) of the partially purified preparation and subsequent amino acid analysis indicated the presence after hydrolysis of large amounts of glutamate and a trace of glycine. Whether this peptide represents substance B or a contaminant in the preparation remains to be determined.

DISCUSSION The identification of the endogenous neuromodulator tentatively referred to as substance B (from brain), which antagonizes presynaptic modulation, probably accounts for the difficulties encountered in our laboratory as well as that of others in consistently observing presynaptic modulation of AcCho release from brain, particularly synaptosomal preparations. Marchi et al. (29) have observed evidence of presynaptic muscarinic receptor-mediated inhibition of AcCho release using a superfused brain synaptosome preparation. An unexplained outcome of their study, however, was the observation that although striatum demonstrated the highest [3H]choline uptake and AcCho synthesis of the brain areas examined, evidence for autoreceptor regulation of

200 nM 2-Chloroadenosine

Substance B 10

40

ill

80 yl

160

Al

Al 20O4pIL

5 min

FIG. 2. Effect of substance B on the inhibition of electrically stimulated contractions of myenteric plexus muscle strips by 2-chloroadenosine. Doubled-over muscle strips (5 cm) were suspended in a 5-ml organ bath containing 5 AM choline and 25 MM Hepes/KRB buffer (pH 7.4) continuously gassed with 95%02/5% CO2 and maintained at 370C. Contractions were stimulated by square wave pulses at 10 V/cm with a duration of 1 msec and a frequency of 0.1 Hz. Lyophilized guinea pig whole brain extract was resuspended in the same buffer to give a 3.5-fold concentrate and increasing volumes were added to the organ bath.

7982

Neurobiology: Pearce et al.

Proc. Natl. Acad. Sci. USA 83 (1986) [3H]Choline [ \ 0 [14C]Glucose

IV

/\ e 0*5 0/ . 0.5

28

\0

t

24

~~0.41-iari

\

20

- [14C]Sucrose - AF iai A r. *Substance BX

>.

2.2

2.4

2.6

2.8

3.0

B1~~~~~~~2

0

0.-'4 __ ___ ___ ___ ___ ___ ___

16 Fraction

12

8

4

00

FIG. 4. Gel chromatography of guinea pig brain extract. Lyophilized extract from three guinea pig brains was suspended in 2 ml of KRB buffer containing 25 mM Hepes at pH 7.4. This suspension was added to a 1.6 x 90 cm Bio-Gel P-2 column preequilibrated in the same buffer and 4.7-ml fractions were collected at 0.5 ml/min and at 4°C. The column void volume was determined using bovine serum albumin and calibrated (see Inset) using various low molecular weight compounds that were detected either by their absorbance at 280 nm or the elution of associated radiolabel. Fractions containing substance B were assayed by addition of aliquots to guinea pig ileum myenteric plexus-longitudinal muscle preparation assay. The results are expressed as the percent reversal of inhibition of electrically stimulated contractions produced by 200 nM 2-chloroadenosine.

AcCho release was disappointing, with effects only seen at very high concentrations of agonist. Situations in which presynaptic modulation of AcCho release has been observed often require very high concentrations of receptor agonist in conjunction with low levels of stimulation of transmitter release. We have observed a functional competition between substance B and presynaptic agonist inhibition reflected by the fact that higher levels of presynaptic agonist are needed to produce inhibition in the presence of substance B. This is entirely consistent with observed effects on neurotransmitter release occurring only at greatly elevated presynaptic agonist concentrations in vitro. Results of experiments employing the guinea pig ileum myenteric plexus synaptosomes would, at first glance, suggest that substance B is not present in these nerve endings. However, comparison of the concentrations of presynaptic receptor agonist required to inhibit the release of AcCho from the intact longitudinal muscle strip preparation and the synaptosomes derived from the myenteric plexus indicates dramatic differences. Although effects of 2-chloroadenosine on muscle contractions are observed in the nanomolar range, micromolar concentrations are required to see effects on AcCho release from synaptosomes. We have observed that the intact muscle strip preparation requires several hours to show consistent and maximal responses to inhibitory agonists and, if a fresh muscle strip is simply added to this tissue bath, reversal of inhibition by 200 nM 2-chloroadenosine is observed. These observations indicate that the myenteric plexus contains substance B but at a level not sufficient to completely block the effect of inhibitory presynaptic agonists. (In direct preliminary experiments we have found substance B in a small concentration in the ileum, a large amount in the heart, and none in kidney or liver.) The disinhibitory action of substance B cannot simply be attributed to a general stimulation of neurotransmitter release since addition of brain extracts to assays in which the nicotinic-evoked release of AcCho was measured did not result in an increase of basal or evoked release. In fact, on

occasion, the crude factor produced a slight decrease in basal and evoked release of AcCho. Further, it is clear from the results presented here that the noninhibitory action of the factor is not due to the action of substance B as a presynaptic receptor antagonist. This explanation would require that this endogenous neuromodulator be an antagonist at purinergic, muscarinic, and adrenergic presynaptic receptors and this is highly unlikely. Reversal of inhibitory presynaptic receptor action may involve effects on a second messenger system and/or ionic conductance that is common to each of the receptor systems studied. A possible mechanism, consistent with the observed biochemical correlates (30, 31) of these presynaptic receptor agonists, is one in which substance B antagonizes the ability of inhibitory presynaptic receptors to decrease intraneuronal cAMP levels. This could come about in several ways: (i) inhibition of phosphodiesterase, (ii) activation of adenylate cyclase, and (iii) direct antagonism of the actions of the guanine nucleotide regulatory protein that mediates the activity of adenylate cyclase. With regard to the first two possibilities, agents that stimulate adenylate cyclase, such as forskolin, have been shown to stimulate the release of neurotransmitters, but the evoked release of AcCho cannot be modulated by oxotremorine, (-)-PIA, or clonidine (15). This finding also implies that release and modulation of release involve two different mechanisms. However, the ability of substance B to prevent inhibition of neurotransmitter release without directly stimulating release would be predicted if this factor antagonized the ability of the inhibitory G protein to affect adenylate cyclase or possibly phosphatidylinositol phosphate hydrolysis. The importance of substance B as a tool for the study of the biochemical basis for presynaptic inhibition is obvious. In summary, this report presents evidence for the existence of a factor in brain that possesses the ability to reverse presynaptic modulation observed in synaptosomes and longitudinal muscle strips isolated from guinea pig ileum. Despite the presence of amino acids in hydrolysates of partially purified preparations, attempts to inactivate this suspected

Neurobiology: Pearce et al.

Proc. Natl. Acad. Sci. USA 83 (1986)

7983

small peptide did not provide direct evidence that substance B is a peptide. The ability of this factor to reverse the inhibition of neuronal activity involving a number of receptor systems suggests that this endogenous neuromodulator regulates the response of cholinergic neurons to inhibitory neuromodulators. Conceivably, the discovery of this regulatory factor in brain may have relevance to diseases involving cholinergic dysfunction. Still to be determined is the effect of this factor on systems that utilize transmitters other than AcCho. Preliminary observations (T. Dunwiddie, personal communication) suggest that substance B is active in slices of hippocampus.

14. Reese, J. H. & Cooper, J. R. (1984) Biochem. Pharmacol. 33, 1145-1147. 15. Reese, J. H. & Cooper, J. R. (1984) Biochem. Pharmacol. 33, 3007-3011. 16. Snyder, S. H. (1985) Annu. Rev. Neurosci. 8, 123-321. 17. Vizi, E. S., Somogyi, G. T. & Magyar, K. (1983) in Physiology and Pharmacology of Adenosine Derivatives, eds. Daly, J. W., Kuroda, Y., Phillis, J. W., Shimizu, H. & Ui, M.

This work was supported by Grant BNS83-11393 from the National Science Foundation. 1. Frank, K. & Fuortes, M. G. F. (1957) Fed. Proc. Fed. Am. Soc. Exp. Biol. 16, 39-40. 2. Dudel, J. & Kuffler, S. W. (1%1) J. Physiol. (London) 155, 543-562. 3. Eccles, J. C., Eccles, R. M. & Magni, F. (1961) J. Physiol. (London) 159, 147-166. & Schmidt, R. F. (1962) J. 4. Eccles, J. C., Kastyuk, P. Physiol. (London) 161, 258-281. 5. Dudel, J. (1965) Pflugers Arch. 284, 81-94. 6. Starke, K. (1977) Rev. Physiol. Biochem. Pharmacol. 7, 1-124. 7. Langer, S. Z. (1981) Pharmacol. Rev. 32, 337-362. 8. Starke, J. (1981) Annu. Rev. Pharmacol. Toxicol. 21, 7-30. 9. Chesselet, M.-F. (1984) Neuroscience 12, 347-375. 10. Dubocovich, M. L. & Langer, S. Z. (1974) J. Physiol. (London) 237, 505-519. 11. Briggs, C. A. & Cooper, J. R. (1981) J. Neurochem. 36, 1097-1108. 12. Briggs, C. A. & Cooper, J. R. (1982) J. Neurochem. 38, 501-508. 13. Reese, J. H. & Cooper, J. R. (1982) J. Pharmacol. Exp. Ther. 223, 612-616.

20. Reddington, M., Lee, K. S. & Schubert, P. (1982) Neurosci. Lett. 28, 275-279. 21. Reddington, M. & Schubert, P. (1979) Neurosci. Lett. 14, 37-42. 22. Phillis, J. W. & Wu, P. H. (1981) Prog. Neurobiol. 16, 187-239. 23. Daly, J. W. (1983) in Physiology and Pharmacology of Aden-

0.

(Raven, New York), pp. 209-217. 18. Phillis, J. W., Edstrom, J. P., Kostopoulos, G. K. & Kirkpatrick, J. R. (1979) Can. J. Physiol. Pharmacol. 57,

1289-1312.

19. Phillis, J. W., Kostopoulos, G. K. & Limacher, J. J. (1975) Eur. J. Pharmacol. 30, 125-129.

osine Derivatives, eds. D4ly, J. W., Kuroda, Y., Phillis, J. W., Shimizu, H. & Ui, M. (Raven, New York), pp. 59-69. 24. Corrier, A. G., Barberis, C. & Gayet, J. (1981) Biochem.

Pharmacol. 30, 2732-2735. 25. Cotman, C. W. (1974) Methods Enzymol. 21, 445-452. 26. Paton, W. D. M. & Zar, M. A. (1968) J. Physiol. (London) 144, 13-33. 27. Benishin, C. G., Pearce, L. B. & Cooper, J. R. (1986) J. Pharmacol. Exp. Ther., in press.

28. Vizi, E. S. (1979) Prog. Neurobiol. 12, 181-290.

29. Marchi, M., Paudice, P., Caviglia, A. & Raiteri, M. (1983) Eur.

J. Pharmacol. 91, 63-68. 30. Miller, R. J. (1985) Trends Neurosci. 8, 463-465. 31. Cooper, J. R., Bloom, F. E. & Roth, R. H. (1986) The Biochemical Basis of Neuropharmacology (Oxford Univ. Press, New York), 5th Ed., pp. 106-121.