Direct Observation of the Effect of Autoreceptors on Stimulated Release of Catecholamines from Adrenal Cells

The Journal of Neuroscience, April 1994, 14(4): 2402-2407

Rong Zhou, Guoan Luo, and Andrew G. Ewing Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania 16802

The direct effect of a,-autoreceptors was studied by mea- cellular catecholaminesact on presynaptic oc,-autoreceptorsto suring the effects of piperoxan, an a,-autoreceptor antag- inhibit further catecholamine release(Starke, 1977; Westfall, onist, and clonidine, an agonist on catecholamine exocyto- 1977) and oc,-autoreceptoragonists such as clonidine or guan- sis, from single bovine chromaffin cells in culture. facine causea suppressionof locus cell firing and decreasethe Catecholamine release was elicited by stimulation with 100 releaseof catecholamines(Pelayo et al., 1980; Cooper et al., PM nicotine and was monitored electrochemically with a car- 199 1; Jackisch et al., 1992) whereasol,-autoreceptor antagonists bon-fiber microelectrode placed adjacent to the cell. These such as piperoxan, yohimbine, and idazoxan reverse these in- electrodes allowed the number of exocytotic release events hibitory effects and increase release(Dennis et al., 1987; Ab- to be monitored and reported as total charge for release ercrombie et al., 1988; Cooper et al., 1991). following a specific stimulus. Repeated stimulation with 100 Although some experiments have provided evidence about PM nicotine showed that total release caused by the second presynaptic mechanisms,it is not always possibleto determine exposure to nicotine was 32% of the first, and release caused if the drugs affect the presynaptic terminals directly, or whether by the third exposure to nicotine was 80% of the second. they act primarily on nearby cells that then emit a secondsignal Total release of catecholamine increased significantly after to the presynaptic terminals. This is particularly difficult to de- application of 20 PM piperoxan relative to a control appli- cide in tissue studies (Starke, 1981). Basically, there are two cation of balanced salt solution. Application of 20 @I pipe- methods to study the interaction of neurotransmitters or drugs roxan alone did not cause release. After the cells were in- with cells. The first procedure is to determine the biological cubated in culture medium containing 20 MM clonidine, a responseofan intact isolatedorgan, suchas the guinea pig ileum, significant decrease in nicotine-stimulated catecholamine to applied agonists or antagonists. The disadvantage to this release was observed. These results confirm that there are procedure lies in the difficulty in discriminating between a large autoreceptors on chromaffin cells and, when relatively high number of processestaking place before the drug interacts with levels of catecholamine are released, the catecholamine the receptor. The second approach to studying receptors is to stimulates the a,-autoreceptors, which inhibits subsequent measureligand binding to a homogenateor slice preparation. release through a negative feedback mechanism. In addition This technique becamefeasible with the development of ligands to piperoxan, the sympathomimetic drug amphetamine also of a high specificradioactivity and a high affinity for the receptor increases quanta1 release after application of nicotine. Am- (Cooper et al., 199 1). Neither of these techniques eliminate phetamine increases the extracellular concentration of cat- postsynaptic effects. The use of amperometric monitoring of echolamine, and these data appear to indicate that at least catecholamine exocytosis (Leszczyszyn et al., 1990) combined part of the pharmacology of amphetamine might involve with the application of receptor agonistsor antagonists,is a new blocking catecholamine autoreceptors. method that can functionally identify autoreceptors directly, [Key words: a,-autoreceptors, catecholamines, adrenal since there are no postsynaptic cells. The electrochemical mea- medullaty chromaffin cells, microelectrodes, piperoxan, clo- surements described here are spatially resolved to the single- nidine, amphetamine] cell level becauseof the size of the microelectrode sensoras well as the small volume of the applied chemical stimulus. The releaseof catecholaminesto the extracellular synaptic cleft Adrenal medullary chromaffin cells were selectedin the pres- can be regulated by several factors, which include presynaptic ent study becausethey contain secretory granulesthat store the cY,-autoreceptoractivation (Starke, 1981). The releasedextra- catecholamine hormonesepinephrine and norepinephrine and they are similar to sympathetic neurons in neuroectodermal origin and biochemical function (Douglas, 1968; Viveros, 1975; Wightman et al., 1991). The cultured adrenal medullary cells Received duly 16, 1993; revised Sept. 22, 1993; accepted Oct. 7, 1993. have most of the properties of the cells in the intact gland (Wil- This work was supported by grants from the Office of Naval Research and the son and Viveros, 1981; Livett, 1984). In the present work we National Science Foundation. A.G.E. is a Camille and Henry Dreyfus Teacher useda new electrochemistry method (Leszczyszyn et al., 1990), Scholar. We would like to acknowledge Ta Kung Chen for helpful discussions about electrochemistry and exocytosis and Dr. David Sulzer for advice on the combined with the application of an cu,-autoreceptorantagonist, Dharmacoloev of catecholamine autoreceotors. piperoxan, and an agonist, clonidine (Feldman and Quenzer, Correspondence should be addressed to G. Ewing, Department of Chemistry, The Pennsylvania State University, 152 Davey Laboratory, University Park, PA 1984; Cooper et al., 199I), to investigate the direct effect of +- 16802. autoreceptors on nicotine-stimulated releaseof catecholamines Copyright 0 1994 Society for Neuroscience 0270-6474/94/142402-06$05.00/0 from single adrenal medullary chromaffin cells. The Journal of Neuroscience, April 1994, 14(4) 2403

Materials and Methods A Cultured chromajin cells. Primary cultures of bovine adrenal medullary chromaffincells were prepared as previouslydescribed by Wilson and 20PAL Viveros (1981) and Leszczyszynet al. (1991) from freshadrenal glands 10 s obtained from a local slaughterhouse. Experiments were performed at 37°C between days 4 and 6 ofculture. The culture medium was removed and the cells were washed three times and then maintained in a balanced salt solution containing 150 mM NaCl, 4.2 mM KCl, 1.O mM NaH,PO,, 11.2 mM glucose, 0.7 mM MgCl,, 2 mM CaCl,, and 10 mM HEPES, adjusted to pH 7.4. MgCl, was substituted for Caz+-free solutions (Wightman et al., 1991). Chemical stimulations were performed with the desired agent dissolved in this solution. Electrodesand voltammetricprocedures. Working electrodes were prepared from 5-pm-diameter carbon fibers sealed in glass capillaries (Kelly and Wightman, 1986). The sensing tip of the electrode was polished at a 45” angle on a micropipette beveler (World Precision Instruments, New Haven, CT). The reference electrode was a sodium-saturated cal- omel electrode (SSCE). Constant-potential amperometry was performed with a commercial potentiostat (EI-400, Ensman Instrumentation, Bloomington, IN). The applied potential was 650 mV. The output was connected to an A/D converter (Labmaster, Scientific Solutions, Solon, OH) interfaced with a IBM personal computer. Singlechromajincell release experiments. The culture dish was placed on the stage of a microscope (IM35, Zeiss, Germany) resting upon a B vibration isolation table (Kinetic Systems, Boston, MA). A working electrode was positioned over the top of a cell with a three-dimensional micromanipulator (Narishige, Tokyo, Japan). The electrode was ma- nipulated so that it lightly touched a cell and then was retracted to the desired position (Wightman et al., 199 1). Release of catecholamine from a single chromaffin cell was induced through short-duration (3 set) local ~OPAL applications of 100 PM nicotine. Both the cu,-autoreceptor antagonist 10 s piperoxan and agonist clonidine were used to show the effect of autoreceptors. The antagonist was examined following short-duration (3 set) local application of 100 PM nicotine followed by local application of 20 PM piperoxan or balanced salt solution from two micropipettes attached to a two-channel pressure-application device (Picospritzer, General Valve Corporation, Fairfield, NJ). For the clonidine experiments, the cells were incubated for 30-90 min in the culture medium containing 20 PM clonidine and secretion of catecholamine was elicited through local application of 100 PM nicotine with 20 PM clonidine. Typical tip dimensions of micropipettes were 10 pm. They were calibrated to deliver volumes of approximately 3 nl at 1 sec. The micropipettes were positioned with a micropositioner (Zeiss) 30 pm from the cell. Data treatment.Transient current responses were evaluated by a peak- a b detection-integration algorithm. This program evaluated the current Figure I. Current versus time at carbon-fiber electrodes placed outside response by comparing the difference between consecutive data-point single chromaffin cells. A, A 3 set application of 100 I.IM nicotine was values, which is the first derivative of the original current data. When administered at the arrow. Cells were maintained in 7 ml of balanced the difference between consecutive data points is greater than the peak- salt solution. B, The cell was exposed to 100 PM nicotine in the absence detection threshold (also called the noise threshold and is dependent on and presence of Ca*+. Cells were maintained in a Ca2+-free balanced the experiment), then spikes were considered significant. Transient cur- salt solution. Arrows:a, 3 set application of 100 PM nicotine without rent responses measured in the amperometric mode were integrated Ca*+;b, 3 set applicationof 100PM nicotine with 2.0 mM Ca2+.De- with respect to time to determine the amount of charge in units of tection was performed in the amperometric mode (constant potential) picocoulombs (PC). The charge in each spike is related to the number at 650 mV versus SSCE. of moles electrolyzed by Faraday’s law. Significance testing employed a two-side Student’s t test. Reported values are given with the standard error of the mean (SEM). in the injection solution evoked a seriesof rapid current re- Results sponses. Nicotine-stimulated catecholamine releasefrom single cells The response of a single cell to a 3 set application of 100 PM Repeated nicotine-stimulated catecholamine releasefrom nicotine monitored by amperometry at a microelectrode is shown single cells in Figure 1A. Current responsesowing to the releaseof cate- To demonstrate the effect of autoreceptors on quanta1 release cholamine from single granules are rapid and transient, as has in the adrenal cell system, it is preferable to usethe responseto been shown previously (Leszczyszyn et al., 1990, 1991; Wight- repeated nicotine stimulations of the samecell. Individual cells man et al., 199 1). The average spike area in our experiments were repeatedly stimulated with 3 set exposuresto nicotine (100 was 0.7 + 0.02 pC (n = 2 18). Oxidation current for catechol- PM) at intervals of 100400 sec. The current responsesfor neu- amines at a carbon-fiber microelectrode placed adjacent to a rohormone secretion were integrated to obtain the total charge single cell in Ca2+-free buffer is shown in Figure IB. In this for the responseover a 200 set time window. Representative experiment, an injection of 100 PM nicotine without Ca2+ in data are shown in Figure 2. For repetitive 3 set exposuresto either the cell medium or injection pipette resultedin no current nicotine, the ratio of total secretion of the second to the first spikes.However, a secondinjection of nicotine with 2 mM Ca*+ exposure (S,:S,) is 0.32 * 0.05 (n = 7) and the ratio of total 2404 Zhou et al. l Autoreceptors and Stimulated Release from Adrenal Cells

20pAL A :

Figure 2. Amperometric response of one chromaffin cell to three suc- cessive 3 set applications of 100 PM nicotine at the arrows.Electrode potential was 650 mV versus SSCE.

d e f !4 secretion of the third to the second exposure (S,:S,) is 0.80 & 0.02 (n = 4). . . Effects of piperoxan on nicotine-stimulated catecholamine relea’sefrom single cells B The effects of autoreceptors on single adrenal medullary chromaffin cells were characterized by 5 set applications of pipe- .: roxan (20 PM) approximately 80 set after a 3 set injection of t nicotine. A representative experiment is shown in Figure 3A. . . : 20~~4 L After a 3 set application of nicotine, the characteristic current 25s responseshowing a large number of oxidation spikes is observed. At approximately 80 set after the stimulation, the oc- currence of oxidation spikes has significantly decreased.At this . i point, application of 20 PM piperoxan results in a significant 8’ i increase in the oxidation spikes observed. A second 5 set ap- . . ! plication of 20 PM piperoxan after another 80 set interval again results in oxidation spikes, apparently indicating a resurgence of catecholamine releaseby exocytosis. Three additional 5 set applications of 20 FM piperoxan resultedin very little increase in oxidation spikes. However, a final application of nicotine resulted a sharp increasein releaseevents as observed by am- a b C perometry. A control experiment involving application of nic- Figure 3. Effect of piperoxan (20 PM on nicotine-induced exocytosis otine followed by only a balancedsalt solution is shown in Figure of catecholamines. A:a and g, 3 set application of 100 PM nicotine; & 3B. The data in Figure 3B are very similar to that obtained k 5 set application of 20 PM piperoxan. &a andc, 3 set application of following only nicotine stimulation (Fig. 1A). 100 PM nicotine. b, 5 set application of balanced salt solution. Electrode To evaluate the effect of piperoxan on catecholamine exo- potential was 650 mV versus SSCE. cytosis, the integrated current for releaseevents following nicotine and piperoxan together has been compared to the total charge for releaseevents after only nicotine. This comparison 30-90 min and then stimulated with a 3 set injection of nicotine is shown as a ratio of the total chargefor all the current spikes containing 20 FM clonidine. The action of clonidine is slow and after the sequenceof nicotine and piperoxan stimulations to the has little or no effect on stimulated releaseafter transient ap- total charge after only nicotine in Figure 4. A 15% (n = 3) plication. The comparison between the total charge for cate- increasein catecholamine releaseis observed after the nicotine cholamine releaseevents after incubation with clonidine to that and piperoxan compared to a 46% (n = 6) decreasein release of a control group without clonidine is shown in Figure 5. After for the control experiment (identical protocol without appli- incubation with clonidine for 30-60 min, secretion decreases cation of piperoxan). 75% over the control group (P < 0.05). Secretion decreases82% over the control group (P < 0.05) when a 90 min incubation Eflect of clonidine on nicotine-stimulated catecholamine with clonidine is used. releasefrom single cells To demonstrate further the effect of autoreceptors on single Eflect of amphetamine on nicotine-stimulated catecholamine adrenal medullary chromaffin cells, the responseof stimulated releasefrom single cells releaseto an agonist was evaluated. Here, adrenal cells were Amphetamine is thought to act via a mechanismthat increases incubated in culture medium containing 20 WM clonidine for the extracellular concentration ofcatecholamines. Here, we have The Journal of Neuroscience, April 1994, 74(4) 2405

l *

Figure 4. Effect of piperoxan(20 PM) and amphetamine (20 PM) on nicotine- stimulated release. In each experiment, a 3 set application of 100 PM nicotine was followed by a 5 set application of balanced salt solution (&KS’) (n = 6), 20 PM piperoxan (n = 3) or 20 PM amphetamine (n = 3). Catecholamine release is expressed as the ratio of total charge for all release events (integrated current under all the spikes monitored) following nicotine and balanced salt solution, nicotine and piperoxan, or nicotine and amphetamine, relative to the total charge for all release events following nicotine only. Both piperoxan and amphetamine increase nicotine- stimulated release significantly: *, P < BSSlnieotine pipemxanlniwtinc amphetamindnicntine 0.05; **, P < 0.01). examined the effect of amphetamine on release of catechol- et al., 1991). In the previous work, voltammetric and phar- amines from cultured adrenal cells. A 5 set application of am- macological evidence indicated that the amperometric signals phetamine (20 KM) at approximately 20 set after stimulation of observed were due to exocytosis of catecholamines. The am- the cell with nicotine resulted in a dramatic increase in cate- perometric responseto nicotine stimulation is Ca2+dependent cholamine release events (Fig. 6). A comparison of the ratio of and consistsof a seriesof rapid current spikeswhen the electrode total release following nicotine and amphetamine relative to is held at a sufficient oxidation potential (Fig. 1). This is con- nicotine only is included in Figure 4. The increase in catechol- sistent with the requirement of Ca2+ for exocytosis (Douglas, amine exocytosis (only the current under the current spikes is 1968; Viveros, 1975; Schweizer et al., 1989). Assuming a two- integrated) over the control experiment is 870%. electron oxidation process,the quantity of chargein the average spike correspondsto the detection of 4 x lo-l8 mol of cate- Discussion cholamine. This value is in the range calculated for the average Voltammetry at microelectrodes can be used to monitor single catecholamine vesicle in adrenal cells (Winkler and Westhead, exocytotic events at catecholamine containing adrenal medul- 1980; Wightman et al., 1991). lary chromaffin cells (Leszczyszyn et al., 1990, 199 1; Wightman It is apparent from the data in Figure 2 that the total cate-

100

Figure 5. Effect of clonidine (20 PM) 40 on nicotine-stimulated catecholamine release. The cells were incubated with 30 clonidine for 30-60 min (n = 4), or 90 min (n = 4) and then stimulated with a 3 set injection of nicotine containing 20 20 PM clonidine. The current responses for catecholamine release after the first 10 nicotine stimulation were integrated to obtain the total charge. The total release decreased significantly after incubation 0 with clonidine (P < 0.05) as compared control 30-60 min !lOmin to the control stimulations (n = 13). 2406 Zhou et al. l Autoreceptors and Stimulated Release from Adrenal Cells

than the decreaseafter incubation of cells for 30-60 min (75%). This suggeststhat the effect of clonidine is slow and agreeswith the hypothesis that clonidine stimulation of autoreceptors re- 2OpAL sults in the inhibition of catecholaminesynthesis (Feldman and 10 s Quenzer, 1984). Perhaps the most widely known sympathomimetic drug is amphetamine. The actions of amphetamine appear to be man- ifested as an increasedlevel of extracellular catecholamine in regions of the CNS (Zetterstrom et al., 1983; Hernandez et al., 1987; Sharp et al., 1987). A great deal of work has shown that amphetamine enhancesrelease of catecholamines(Kuczenski, 1983), inhibits the catecholamine transporter (Wise and Boz- arth, 1987) and acceleratesexchange diffusion acrossthe plasma membrane (Fischer and Cho, 1979; Kuczenski, 1983). In ad- t t dition, recent evidence suggeststhat amphetamineacts as a weak a b base to disrupt catecholamine vesicularization, and therefore enhance the cytoplasmic level of neurotransmitter for subse- Figure 6. Amperometric response of a carbon-fiber electrode placed quent releasevia reversetransport (Sulzer and Rayport, 1990). at a single chromaffin cell following a 3 set application of 100 PM nicotine All these actions appear to play a role in the responseto am- (a), and a 5 set application of 20 PM amphetamine (b). phetamine. The data presentedin this article point to a specific mechanismwhereby amphetamine can also increaseextracel- cholamine release(number of releaseevents) from exocytosis lular catecholamine levels via an increasein exocytotic release following nicotine stimulation diminishesby 68% from the first by acting as an antagonist on a,-autoreceptors. This appearsto to the secondconsecutive stimulation. From the second to the be completely consistentwith data suggestingthat amphetamine third stimulation only a 20% decreasein total releaseis ob- does not bind strongly to the catecholamine transporter (An- served. As shown below, this cannot be explained simply by dersen, 1987; Ritz et al., 1987). However, the data are not depletion of catecholamine vesicles, but rather, appearsto be consistent with studiesshowing an effect of amphetamine due the result of cell inhibition by autoreceptors. to binding at dopamine receptors (Shalaby et al., 1983; Ku- To support the hypothesis that autoreceptorsare activated by czenski et al., 1990). This might reflect a specific difference in a threshold concentration of releasedneurotransmitter, the re- action between D, and a,-autoreceptors. Blockade of cu,-auto- leaseof catecholamineshas been estimated from overflow into receptorsleading to enhancedrelease increases the extracellular an incubation or perfusion fluid (Langer, 1977; Langer et al., level of catecholamine, which is the clear end result of am- 1977; Starke, 1977, 1987; Langer and Dubocovich, 1981). The phetamine administration. present work provides direct evidence for autoreceptors at the single-celllevel and with millisecond time resolution. Although References it has been reported that functional adrenoreceptors are not Abercrombie ED, Keller RM, Zigmond MJ (1988) Characterization present on adrenal chromaffin cells (Powis and Baker, 1986; of hippocampal norepinephrine release as measured by microdialysis Orts et al., 1987), the results of repeated stimulation (Fig. 2) perfusion: pharmacological and behavioral studies. Neuroscience 27: suggestthat cY,-autoreceptorsdo exist on the adrenal medullary 897-904. chromaffin cells and act to decreasecatecholamine release.The Andersen PH (1987) Biochemical and pharmacological characteriza- data in Figure 3 strongly suggestthat a negative feedback mech- tion of [3H]GBR 12935 binding in vitro to rat striatal membranes; labeling of the dopamine uptake complex. J Neurochem 48: 1887- anism exists. The first and second injections of piperoxan lead 1896. to the promotion of releasespikes, but subsequentinjections do Cooper JR, Bloom FE, Roth RH (199 1) The biochemical basis of not result in additional release.Apparently, the effects of the neuropharmacology. New York: Oxford UP. initial nicotine stimulation on the nicotinic receptors are di- Dennis T, L’Heureux R, Carter C, Scatton B (1987) Presynaptic 02- adrenoreceptors play a major role in the effects of idazoxan on cortical minished at theselater times. Exposure to piperoxan enhances noradrenaline release (as measured by in vivo dialysis) in the rat. J the releaseof the catecholaminesfrom nicotine-stimulated cells; Pharmacol Exp Ther 241:642-649. however, piperoxan alone has no effect upon on releasefrom Douglas WW (1968) Stimulus-secretion coupling: the concept and cells not exposedto nicotine. The final exposure to nicotine (Fig. clues from chromaffin and other cells. Br J Pharmacol 34:45 l-474. 3) showsthat both the cell and the microelectrodestill responded Feldman RS, Quenzer LF (1984) Fundamentals of neuropsychophar- macology, p 189. Sunderland, MA: Sinauer. to nicotine stimulation and the presenceof catecholamines,re- Fischer JF, Cho AK (1979) Chemical release ofdopamine from striatal spectively. Thus, the lack of releasefor piperoxan exposures homogenates: evidence for an exchange diffusion model. J Pharmacol labeledd-f in Figure 3 can be attributed to a lack of stimulation Exp Ther 208:203-209. of the nicotinic receptors. Hernandez L, Lee F, Hoebel BG (1987) Simultaneous microdialysis and amphetamine infusion in the nucleus accumbens and striatum The effect of clonidine on stimulated release(Fig. 5) further of freely moving rats: increase in extracellular dopamine and sero- suggeststhat there are autoreceptors on adrenal medullary chro- tonin. Brain Res Bull 19:623-628. maffin cells. After the cell is incubated with clonidine, the total Jackisch R, Huang HY, Rensing H, Lauth D, Allgaier C, Hertting G catecholamine releasedeclines significantly after the first 3 set (1992) a,-Adrenoceptor mediated inhibition of exocytotic noradren- application of nicotine. Apparently, the releaseofcatecholamine aline release in the absence of extracellular calcium. Eur J Pharmacol 81245-252. is inhibited by autoreceptors activated by clonidine. The de- Kelly RS, Wightman RM (1986) Bevelled carbon-fiber ultramicroelec- creaseof catecholamine releaseafter incubation of cells with trodes. Anal Chim Acta 187:79-87. clonidine for 90 min compared to control group is greater (82%) Kuczenski R (1983) Biochemical actions of amphetamine and other The Journal of Neuroscience, April 1994, 14(4) 2407

stimulants. In: Stimulants: neurochemical, behavioral, and clinical R, Thoenen H, Bookman RJ, Burger MM (1989) Inhibition of perspectives (Creese I, ed), pp 3 1-61. New York: Raven. exocytosis by intracellularly applied antibodies against a chromaffin Kuczenski R, Segal DS, Manley LD (1990) Apomorphine does not granule-binding protein. Nature 339:709-7 12. alter amphetamine-induced dopamine release measured in striatal Shalaby IA, Kotake C, Hoffmann PC, Heller A (1983) Release of dialysates. J Neurochem 54:1492-1499. dopamine from coaggregate cultures of mesencephalic tegmentum Langer SZ (1977) Presynaptic receptors and their role in the regulation and corpus striatum. J Neurosci 3: 1565-l 57 1. of transmitter release. Br J Pharmacol 60:48 l-497. Sharp T, Zetterstriim T, Ljungbers T, Ungerstedt U (1987) A direct Langer SZ, Dubocovich ML (1981) Cocaine and amphetamine an- comparison of amphetamine-induced behaviours and regional brain tagonize the decrease of noradrenergic neurotransmission elicited by dopamine release in the rat using intracerebral dialysis. Brain Res oiymetazoline but potentiate the inhibition by Lu-methylnorepine- 4Oi:322-330. phrine in the perfused cat spleen. J Pharmacol Exp Ther 216: 162- Starke K (1977) Regulation of noradrenaline release by presynaptic 171. receptor systems. Rev Physiol Biochem Pharmacol 77: 1-124. Langer SZ, Adler-Graschinsky E, Giorgi 0 (1977) Physiological sig- Starke K (198 1) Presynaptic receptors. Annu Rev Pharmacol Toxic01 nificance of cY-adrenoceptor-mediated negative feedback mechanism 21:7-30. regulating noradrenaline release during nerve stimulation. Nature 265: Starke K (1987) Presvnaptic ol-autoreceptors. Rev Physiol Biochem 648-650: Pharmacol 107:73-146. - Leszczyszyn DJ, Jankowski JA, Vieros OH, Diliberto EJ Jr, Near JA, Sulzer D, Rayport S (1990) Amphetamine and other psychostimulants Wightman RM (1990) Nicotine receptor-mediated catecholamine reduce pH gradients in midbrain dopaminergic neurons and chro- secretion from individual chromaffin cell: chemical evidence for exo- maffin granules: a mechanism of action. Neuron 5:797-808. cytosis. J Biol Chem 265:14736-14737. Viveros GH (1975) Handbook of physiology, section on endocrinol- Leszczvszvn DJ. Jankowski JA, Viveros OH, Diliberto EJ Jr, Near JA, oav. Vol 6 (Blaschko A. Smith AD, eds). DD 389-426. Washington.- Wig&man RM (199 1) Secretion of catecholamines from individual DC: American Physiological Society. ” - _ adrenal medullary chromaffin cells. J Neurochem 56: 1855-l 863. Westfall TC (1977) Local regulation of adrenergic neurotransmission. Livett BG (1984) Adrenal medullary chromaffin cells in vitro. Physiol Physiol Rev 57:659-728. Rev 64:1103-l 160. Wightman RM, Jankowski JA, Kennedy RT, Kawagoe KT, Schroeder Orts A, Orellana C, Canto T, Cena V, Gonzalez-Garcia C, Garcia AG TJ, Leszczyszyn DJ, Near JA, Diliberto EJ Jr, Viveros OH (1991) (1987) Inhibition of adrenomedullary catecholamine release by pro- Temporally resolved catecholamine spikes correspond to single ves- pranolol isomers and clonidine involving mechanisms unrelated to icle release from individual chromaffin cells. Neurobiology 88: 10754- adrenoreceptors. Br J Pharmacol 92:7951801. 10758. Pelavo F. Dubocovich ML. Laneer SZ (1980) Inhibition of neuronal Wilson SP, Viveros OH (198 1) Primary culture of adrenal medullary uptake reduces the presynaptyc effects of dlonidine but not a-me- chromaffin cells in a chemically defined medium. Exp Cell Res 133: thylnoradrenaline on the stimulation-evoked release of 3H-noradren- 159-169. aline from rat occipital cortex slices. Eur J Pharmacol 64: 143-155. Winkler M, Westhead E (1980) The molecular organization of adrenal Powis DA, Baker PF (1986) ol,-Adrenoceptors do not regulate cate- chromaffin granules. Neuroscience 5: 1803-l 823. cholamine secretion by bovine adrenal medullary cells: a study with Wise RA, Bozarth MA (1987) Brain mechanisms of drug reward and clonidine. Mol Pharmacol 29: 134-l 4 1. euphoria. Psycho1 Med 3:445460. Ritz MC, Lamb RJ, Goldberg SR, Kuhar MJ (1987) Cocaine receptors Zetterstrijm T, Sharp T, Marsden CA, Ungerstedt U (1983) In vivo on dopamine transporters are related to self-administration of co- measurement of dopamine and its metabolites by intracerebral di- caine. Science 237:1219-1223. alysis: changes after d-amphetamine. J Neurochem 4 1: 1769-l 773. Schweizer FE, Schafer T, Tapparelli C, Crob M, Karli UO, Heumann