H. Müller, J. Zierski, R. D. Penn (Editors) Local-spinal Therapy of Spasticity

With 140 Figures and 48 Tables

Springer-Verlag Berlin Heidelberg New York London Paris Tokyo Prof. Hermann Müller, MD Dept. Anesthesiology and Intensive Care Mediane Justus-Liebig-University Klinikstr. 29 UnWersitäts- D-6300 Gießen, FRG Bibliothek Priv.-Doz. Jan Zierski, MD München Dept. Neurosurgery Justus-Liebig-University Klinikstr. 29 D-6300 Gießen, FRG Richard D. Penn, MD Rush-Presbyterian-St. Luke's Medical Center Rush Medical College Dept. Neurosurgery 1753 West Congress Parkway Chicago, Illinois 60612, USA

ISBN 3-540-18295-0 Springer-Verlag Berlin Heidelberg New York ISBN 0-387-18295-0 Springer-Verlag New York Berlin Heidelberg

Library of Congress Cataloguing-in-Publication Data: Local-spinal therapy of spasticity / H. Müller, J. Zierski, R. D. Penn (editors), p. cm. Based on papers presented at an international meeting held in Giessen, FRG in Apr. 1986 as well as some additional contributions. ISBN 0-387-18295-0 I. Spasticity-Chemotherapy-Congresses. 2. Spinal anesthesia-Congresses. I. Müller, H. (Hermann) II. Zierski, J. (Jan), 1940- III. Penn, Richard D. [DNLM: 1. Muscle Spasticity-drug therapy-congresses. WE 550 L811 1986] RC935.S64L63 1988 616.8'3-dcl9 DNLM/DLC for Library of Congress 87-37653 CIP This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically those of translation, reprinting, reuse of illustrations, broadcasting, reproduc• tion by photocopying machine or similar means, and storage in data banks. Under § 54 of the German Copyright Law, where copies are made for other than private use, a fee is payable to Verwertungsgesellschaft Wort, Munich. © Springer-Verlag Berlin Heidelberg 1988 Printed in Germany Product Liability: The publisher can give no guarantee for information about drug dosage and appli• cation thereof contained in this book. In every individual case the respective user must check its accu• racy by consulting other pharmaceutical literature. The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typesetting, printing and binding: Graphischer Betrieb, Konrad Triltsch, Würzburg 2329/3321-543210 Table of Contents

Preface (H. Müller, J. Zierski, R. D. Penn) 1 A Short Historical Review of Spasticity and Its Therapy (D. Dralle, H. Müller, J. Zierski) 3 Basic Experiences Spinal Pharmacology of Agents which Alter Pain Transmission and Muscle Tone (T. L. Yaksh, P. A. C. Durant, M. S. Monasky, C. W. Stevens, R. R. Schick) ... 19 The Neuropharmacology of Baclofen (W. Zieglgänsberger, J. R. Howe, B. Sutor) . . 37 Animal Experiments on the Spinal Action of Midazolam (W. Gerlach, H. Müller, J. Boldt, G. Hempelmann) 51 Intrathecal Injection of Antispastic Drugs in Rats: Muscle Relaxant Action of Midazolam, Baclofen, 2-Aminophosphonoheptanoic Acid (AP7) and Tizanidine (M. Schwarz, T. Klockgether, L. Turski, K.-H. Son tag) 65 CSF Compatibility of Antispastic Agents (U. Börner, H. Müller, J. Zierski, G. Hempelmann, M. Reinacher) 81 Dural Permeability to Bupivacaine, Baclofen and Midazolam: In Vitro Determination with Human Dura Mater (J. Biscoping, G. Michaelis, D. Friess, D. Krauß, E. Mutschier, G. Hempelmann) 85 Clinical Studies Pharmacotherapy of Spasticity (J. Noth) 93 Pumps in Pharmacotherapy (H. Müller, J. Zierski) 97 Implantation of Ports and Pumps. Technique for Intrathecal Administration of Drugs (J. Zierski, H. Müller) 125 Control of Spasticity with Intrathecal Morphine Sulfate (D. L. Erickson, P. Moreno, J. Lo, J. Cameron, M. Michaelson) 137 Clinical Experience with Spinal Morphine, Midazolam and Tizanidine in Spasticity (H. Müller, J. Zierski) 143 Chronic Intrathecal Baclofen for Severe Rigidity and Spasms (R. D. Penn) .... 151 Intrathecal Baclofen in Spasticity (H. Müller, J. Zierski, D. Dralle, O. Hoffmann, G. Michaelis) . . . 155 Chronic Intrathecal Administration of Baclofen in Treatment of Severe Spasticity (Y. Lazorthes) 215 Pharmacokinetics of Intrathecal Baclofen (H. Müller, J. Zierski, D. Dralle, D. Krauß, E. Mutschler) 223 Intrathecal Baclofen in tetanus (H. Müller, J. Zierski, U. Börner, G. Hempelmann) . 227 The Value of Spinal Cord Stimulation (SCS) in Treatment of Disorders of the Motor System (J.-U. Krainick, H. Weisbrod, H. U. Gerbershagen) .... 245 Physostigmine Reversal of Baclofen-induced Sedation (G. Müller-Schwefe) .... 253 Surgical Treatment of Spasticity - A Review (J. Zierski, H. Müller) 255 Subject Index 265 The Neuropharmacology of Baclofen W. Zieglgänsberger , J. R. Howe, B. Sutor

Introduction Ganima-aminobutyric acid (GABA) is one Baclofen reduces muscle tone in patients of the major inhibitory neurotransmitters with spinal transections and reduces muscle in the mammalian central nervous system rigidity and tonic stretch reflexes in de• [see: 36, 69, 92, 98]. GABAergic neurons cerebrated animals [see: 13]. It is therefore have been identified throughout the central suggested that the therapeutically relevant nervous system with histochemical tech• effect of baclofen results from a direct ac• niques [see: 75]. GABA is, however, of no tion at the spinal level. In addition to its therapeutical value because the amino acid antispastic activity, however, baclofen can does not pass the blood-brain barrier in cause muscle weakness, ataxia, drowsiness, sufficient amounts to affect neuronal excit• insomnia, nausea, hypertension, a decrease ability [9, 64]. in growth hormone release and an increase The GABA analogue baclofen (ß-[4- in prolactin secretion. Also, antinociceptive chlorophenyl]-gamma-aminobutyric acid; actions of baclofen have been reported LioresalJi>) was designed to act as a GABA [107; 110] (see also Yaksh, this volume). mimetic that because of its lipophilicity Particularly in elderly patients, the sudden would distribute into the central nervous withdrawal of baclofen after chronic use is system after systemic application. Baclofen occasionally associated with the appear• effectively reduces exaggerated stretch re• ance of psychotic symptoms, including flexes and muscle tone after oral and intra• dysphoric episodes and even hallucinations venous administration and is widely used [see: 12]. These latter effects clearly indicate in the treatment of spasticity caused by that baclofen acts also on receptors remote traumatic spinal lesions, degenerative, neo• from the spinal cord. plastic or infectious diseases of the spinal This chapter examines the neuronal pro• cord, and multiple sclerosis. It is less ef• cesses affected by baclofen, with an em• fective in ameliorating the spasticity after phasis on electrophysiological findings. The stroke or cerebral palsy [see: 109] (see also enormous literature precludes an exhaus• various authors, this volume). tive documentation of all aspects.

The GABAB Receptor The inhibitory actions of GABA are me• Picrotoxin [26; see: 36]. These two antago• diated in most neurons studied until now nists seem to block the actions of GABA via through an increase in chloride conduc• different mechanisms. Whereas bicuculline tance of the postsynaptic membrane [see: competes for the GABA binding site [74], 39, 69. 101]. These actions are mimicked by Picrotoxin seems to interact with the muscimol and THIP (4,5,6,7-tetrahy- chloride-channel in a more direct manner droisoxazolo-[5,4-c]pyridine-3[2H]-one) [see: 36]. Because GABA also evokes bicu- and can be antagonized by bicuculline and culline-insensitive responses which are 38 W. Zieglgänsberger et al. mimicked by baclofen (e.g. on transmitter binding was reduced [62]. This suggests that release) [15, 67], two classes of GABA re• these low affinity GABAB receptor binding ceptors, GABAA and GABAB, have been sites are located presynaptically on norad• proposed. The bicuculline-reversible ac• renergic terminals. Baclofen reduces the tions of GABA and the GABA mimetic evoked release of noradrenaline [15, 16, 40] muscimol are mediated through GABAA and several other neurotransmitters (see receptors, whereas baclofen is considered below). as a prototypical bicuculline-insensitive agonist at GABAB receptors [15, 54]. This Stereoselectivity of GABAB Actions novel GABAB receptor mediates ionic pro• cesses which are clearly distinct from those Therapeutically used baclofen (LIO- sensitive to bicuculline (see below). GABAA RESAL®) is a racemic mixture of the two receptor-mediated inhibition is potentiated isomers. In a number of behavioral studies by barbiturates and by the benzodiazepines and electrophysiological investigations in [see: 48, 49] and is blocked by some con- vivo, it has been shown that the (-)isomer vulsants like penicillin or pentetrazol [see: of baclofen is more potent than the 36, 68]. Unfortunately, a selective GABAB (H-)isomer [4, 52, 53, 80, 81, 107] (see also receptor antagonist is not yet available. Yaksh, this volume). Several in vitro elec• Data from binding studies also support trophysiologic studies have shown that the the existence of different types of GABA re• (-)isomer is at least 100-fold more potent ceptors. Various studies have corroborated than the (-H)isomer [5-7, 20, 47, 56, 59, 78]. the initial finding that baclofen does not Similar stereoselectivity was demonstrated displace 3(H)-GABA from its binding sites in experiments where baclofen reduced the on neuronal membranes [111]. Further• in vitro release of exogeneously loaded more, the binding of GABA is enhanced by 3(^-neurotransmitters [15, 61] and in stud• benzodiazepines [see: 48], whereas the ies of baclofen binding to bicuculline-in• binding of baclofen is unaffected by these sensitive receptors on synaptic membranes compounds [108]. [54]. Autoradiographic studies with 3(H)- It was reported by various authors that baclofen have shown that, with a few ex• the (-h)isomer can antagonize the actions ceptions, the distribution of GABAA and of the (-)isomer [94, 104, 105] (see also GABAB binding sites overlap in most re• Yaksh, this volume). In recent elec• gions of the brain [46]. There are indi• trophysiological experiments in which in• cations, however, that the neuronal distri• tracellular recording techniques were em• bution of these bindng sites are not identi• ployed however, no such antagonism could cal. For example, GABAA binding and high be demonstrated [4, 58]. Thus (-^-baclo• affinity GABAB binding were unchanged fen does not appear to be an antagonist at by interruption of forebrain noradrenergic all GABAB receptors. projections, whereas low affinity GABAB

Mode of Action of Baclofen Numerous electrophysiological studies the release of excitatory neurotransmitters. have shown that baclofen has profound in• It is now clear, however, that baclofen can hibitory effects on synaptic transmission in also reduce inhibitory synaptic transmis• the spinal cord and many other regions of sion and that baclofen directly increases the the mammalian central nervous system. postsynaptic potassium conductance of Until recently, the majority of the results many central neurons. In addition, there suggested that baclofen acted presynapti• are several reports that baclofen depresses cally to selectively reduce excitatory synap• the firing of central neurons induced by ex• tic transmission, a conclusion consistent citatory substances. The following sections with reports that baclofen directly reduces review the evidence for each these actions The Neuropharmacology of Baclofen 39 of baclofen and their relevance to baclo• synaptic transmission were inhibited [cf. fen's effects on central nervous function. 60]. The excitatory synaptic transmission from the lateral perforant path was insensi• tive to baclofen. A selective action of baclo• Actions of Baclofen and Synaptic fen on synaptic tranmission mediated by Transmission excitatory amino acids is also suggested by In the first study employing intracellular the findings of Ault and Evans [3]. These recording techniques, Pierau and Zim• authors described that baclofen reduced mermann [85] reported that baclofen de• dorsal root potentials recorded from the pressed excitatory postsynaptic potentials neonatal isolated spinal cord, potentials (EPSPs) evoked in cat motoneurons at which are also reduced by excitatory amino doses that did not affect inhibitory pro• acid antagonists [38], whereas the excit• cesses. Since neither the membrane po• atory responses of cervical ganglion tential (Em), input resistance (RN), nor the neurons to preganglionic stimulation is not direct excitability (action potentials evoked affected even by much higher concentra• by intracellular current injection) of these tions. Both excitatory afferent inputs from cells were affected by baclofen, these descending pathways employing still un• authors concluded that baclofen exerts its known transmitters and cholinergic inputs action through presynaptically located re• to motoneurons are not affected by baclo• ceptors. The occasionally observed slight fen [11, 63]. hyperpolarizations were interpreted as a In addition to these various extracellular disfacilitatory effect caused by inhibition of studies, there are now several reports from excitatory interneurons. investigations in which intracellular record• In several subsequent in vivo and in vitro ings were obtained that baclofen depresses electrophysiological studies employing ex• EPSPs in mammalian central neurons [14, tracellular recording techniques in various 42, 47, 55, 56, 60, 65, 73, 85, 102]. Although structures, further evidence was provided in some of these studies baclofen also hy- that baclofen preferentially reduces exci• perpolarized the cells (see below), it was tatory synaptic transmission by a pre• shown that the depressions of EPSPs were synaptic mechanism [5-7, 29, 30, 42, 43, 52, not a direct consequence of these hy• 53, 63, 70, 82-84, 86, 88]. Thus baclofen perpolarizations [14, 60]. In neocortical was shown to produce profound reductions neurons, baclofen's reductions of EPSP am• of responses to orthodromic stimulation of plitudes were independent of membrane excitatory afferent pathways without sig• potential over the range of values ± 30 mV nificantly affecting responses to antidromic from resting Em [56]. In our studies, baclo• responses or presnyaptic fiber volleys [5, 6, fen applications that produced 70 to 100% 29, 42, 70, 82]. Synaptically evoked re• depressions of EPSPs typically produced sponses were shown to be reduced at doses only 20 to 30% decreases in RN; baclofen's or concentrations of baclofen that had little depressions of EPSPs typically outlasted its or no effect on chemically evoked excita• effects on Em and RN for several minutes. tion or spontaneous firing [30, 42, 52, 53, Thus in neocortical neurons, baclofen's re• 82]. ductions of EPSP amplitudes do not appear The selectivity of baclofen's depressions to be solely the consequence of its action to of synaptically evoked responses in some increase postsynaptic conductance. Anoth• structures was also interpreted as being in• er possible interpretation of such findings, consistent with a postsynaptic depressant however, is that baclofen preferentially in• action of baclofen. Electrophysiological ex• creases dendritic conductance, and that on• periments in slice preparations of the hip• ly a portion of this conductance increase is pocampus suggest that baclofen selectively detected from intracellular recordings at inhibits transmission at putative glu- the neuronal soma. Such an interpretation tamatergic synapses (Lanthorn and Cot- has been made of results obtained in the man 1981). In this study, only the CA3-pro- hippocampus [47]. jections to pyramidal cells in the CA1 re• gion (Schaffer collaterals) and mossy fiber 40 W. Zieglgänsberger et al. The most detailed analyses of baclofen's diated [100, 101], and baclofen's de• effects on excitatory synaptic transmission pressions of these IPSPs were independent have been performed on nonmammalian of membrane potential between values of preparations. Shapovalov and Shiriaev [97] -50 and - 110 mV [56]. Importantly, baclo• studied the effect of baclofen on mono• fen's reductions of short-latency GABAerg• synaptic EPSPs in the frog motoneuron. ic IPSPs in olfactory cortical and neocorti• These EPSPs are composed of both a cal neurons were shown to be accompanied chemically mediated and an electrically by reductions in the conductance increases mediated component. Baclofen produced measured during these IPSPs [56, 96]. Be• marked reductions of the chemically me• cause baclofen does not reduce responses diated component, but had no effect on the to exogenously applied GABA or muscimol electrically mediated component of these in these same neurons (see below), a post• EPSPs. From statistical analyses of single- synaptic blockade of the IPSP conductance fiber EPSPs, Shapovalov and Shiriaev con• can be excluded and these results argue cluded that baclofen acted presynaptically strongly that baclofen reduces GABAA re• to reduce the release of the transmitter ceptor-mediated IPSPs by a presynaptic ac• which generates chemically mediated tion. EPSPs at these synapses. At the crayfish In addition to depressing chloride-de• neuromuscular junction, Barry [8] found pendent GABAA receptor-mediated IPSPs, that baclofen depressed excitatory trans• baclofen has also been reported to decrease mission without affecting the input resis• the amplitude of potassium-dependent tance of the muscle. She was further able to long-latency IPSPs (slow IPSPs) in hip• show that baclofen reduced the frequency pocampal neurons [14, 60]. Similar potas• of spontaneous excitatory junction po• sium-dependent long-latency IPSPs (times tentials without affecting their size, results to peak of 150 to 250 ms) are also evoked in which clearly indicate a presynaptic site of neocortical neurons [57], and baclofen con• action. At present, the complexity of mam• sistently and markedly reduces the ampli• malian preparations has prohibited these tude of these IPSPs [56, 57]. As for re• sort of detailed analyses. ductions of GABAA receptor-mediated Fox et al. [42] noted that, in addition to IPSPs in these neurons, baclofen's re• baclofen's consistent reductions of EPSPs, ductions of long-latency IPSPs are in• baclofen also reduced the amplitude of dependent of membrane potential and are IPSPs recorded in some cells. That baclofen accompanied by a reduction of the conduc• reduces stimulation-evoked GABAergic tance increases associated with these IPSPs inhibition in the olfactory cortex and the [56, 57]. hippocampus was suggested from ex• Despite baclofen's action to reduce tracellular recordings of field potentials [6, IPSPs, in most neurons studied baclofen's 24]. Intracellular studies in vitro have action to depress EPSPs appears to pre• shown that baclofen reduces the amplitude dominate and baclofen causes an increase of short-latency GABAA receptor-mediated in the stimulation intensity required to pro• IPSPs evoked in various hippocampal duce a synaptically evoked action potential neurons [14, 60, 73], neurons in the ol• [42, 47, 57, 85]. These baclofen-induced in• factory cortex [96], and neurons in the fron• creases in action potential stimulation tal neocortex [55, 56, 102]. These de• thresholds are consistent with the many ex• pressions of IPSP amplitudes are not due to tracellular studies cited above in which it the concomitant action of baclofen to hy- was found that baclofen decreased synaptic perpolarize these neurons. Blaxter and Car• excitability. There are reports, however, len [14] reported that baclofen's de• that baclofen can either increase or de• pressions of short-latency GABAergic crease action potential stimulation IPSPs in hippocampal neurons persisted thresholds, depending on the cell popula• when the membrane potential was returned tion investigated and the concentration of to its resting value by direct current in• baclofen applied [60, 73]. Although baclo• jection. Short-latency IPSPs in rat neocorti• fen increased the stimulation threshold of cal neurons are also GABAA receptor-me• synaptically evoked action potentials in The Neuropharmacology of Baclofen 41 virtually every neocortical neuron tested, lar junction. In most studies, it was con• baclofen often caused an increase in the cluded that the action was presynaptic number of action potentials produced by either because no evidence of a post• suprathreshold stimulation intensities due synaptic action was found or the post• to baclofen's blockade of IPSPs [56]. Thus synaptic changes that were observed were there are circumstances in which baclofen considered to be insufficient to account for can in fact produce increases rather than the effects on postsynaptic potentials. As decreases in synaptic excitability. The ac• we already mentioned, there is, however, tion of baclofen is clearly distinguishable direct evidence for effects of baclofen on from GABAA receptor antagonists such as neurotransmitter release and also recent bicuculline, however, which commonly evidence that baclofen has a direct effect on produces significant reductions in action postsynaptic membrane conductance. In potential stimulation thresholds and pro• addition, there are several electrophysio• motes the generation of epileptiform bursts logical studies which have directly ad• of action potentials in mammalian central dressed the effects of baclofen on pre• neurons. In contrast, baclofen has been synaptic afferent terminals or experimental shown to block bicuculline-induced models thereof. These results are presented epileptiform activity in in vitro preparations in the following sections. of the hippocampus [6, 7, 20; but see: 73] and the frontal neocortex [56]. These results also indicate that baclofen's reductions of Presynaptic Actions of Baclofen excitatory synaptic transmission are resis• The most direct evidence for a presynaptic tant to blockade by bicuculline and indeed action of baclofen is its demonstrated inhi• baclofen's reductions of EPSP amplitudes bition of neurotransmitter release. Baclofen are not antagonized by this GABAA re• has been shown to decrease the evoked re• ceptor antagonist [56]. lease of several putative neurotransmitters, The advent of in vitro slice preparations including monoamines [15, 40, 94] and ex• of the mammalian CNS has made it pos• citatory amino acids [24, 61, 82, 89, 90]. sible to evaluate the action of known con• Baclofen's action to reduce the release of centrations of drugs on neurons in these excitatory amino acids is consistent with preparations under steady state conditions. baclofen's selective inhibition of synaptic The EC50 for baclofen's depressions of ex• transmission that is thought to be mediated citatory synaptic transmission is approxi• by these excitatory amino acids. mately 1 uM and significant depressions Although baclofen depresses GABAA re• are observed at concentrations between 10 ceptor-mediated IPSPs by an action which and 100 nM [5-7, 24, 47, 56, 70, 82]. These is not postsynaptic [56, 96] and dendriti- latter concentrations are approximately cally located GABAB receptors have been equal to those obtained in the cerebro• demonstrated on central GABAergic spinal fluid after systemic administration of neurons [19, 106], baclofen does not reduce therapeutic doses in man [66, 103]. In our the directly evoked release of GABA from study of neocortical neurons, there was no brain slices [24, 6i, 89]. Collins et al. [24] apparent difference in the concentration demonstrated, however, that baclofen sig• dependence of baclofen's reductions of nificantly reduces GABA release that is EPSPs and its reductions of either type of evoked by electrical stimulation of excita• IPSP [56]. tory afferents. They proposed that baclofen In summary, the majority of the studies reduces stimulation-evoked GABA release on the action of baclofen on synaptic trans• and GABAergic inhibition via its direct ac• mission seem to favor the conclusion that tion to reduce the release of excitatory baclofen reduces synaptic transmission by a amino acid neurotransmitters and conse• presynaptic action. It should be noted, quently, the excitatory drive of GABA re• however, that none of these studies provide leasing interneurons [cf. 56, 96]. direct evidence for such a mechanism, with Several careful electrophysiological stud• the exceptions of the studies on the frog ies have addressed the mechanism of baclo• motoneuron and the crayfish neuromuscu- fen's putative presynaptic inhibition of syn- 42 W. Zieglgänsberger et al. aptic transmission. Unfortunately these Baclofen had no effect, however, on in• studies have not provided any consistent ward calcium currents recorded from cul• and positive evidence in this regard. The tured hippocampal neurons under voltage- first suggestion that baclofen may activate clamp conditions [45], nor on the duration presynaptic inhibitory mechanisms in the of the calcium component of action po• spinal cord by enhancing primary afferent tentials recorded from rat neocortical depolarization as described for ben• neurons [56]. Barry [8] concluded that zodiazepines [see: 49, 95] was abandoned baclofen's presynaptic inhibition of synap• after more experimental data were ob• tic transmission at the crayfish neuro• tained. It was shown that baclofen de• muscular junction was unlikely to be the presses rather than enhances the excit• result of an effect on presynaptic calcium ability of primary afferents [22, 29, 42] and influx. In an in vitro preparation of the hip• presynaptic inhibition was unchanged or pocampus, baclofen reduced the extracellu• reduced by baclofen [1,71, 72]. lar calcium concentration measured with Davidoff and Sears proposed that baclo• ion-sensitive microelectrodes ([50] de• fen's depressions of afferent excitability creases in extracellular calcium concentra• were secondary to its action of hy- tion reflect the movement of calcium ions perpolarizing presynaptic terminals and into pre- and postsynaptic elements as a suggested that this action might account for consequence of neuronal activity). In these baclofen's depressions of synaptic trans• experiments, however, consistent re• mission [29]. The magnitude of baclofen's ductions in the presynaptic component of reductions of terminal excitability were stimulation-evoked calcium entry were on• considered to be insufficient, however, to ly observed when baclofen was applied at a account for its depressions of synaptically concentration of 50 u.M. This is approxi• evoked responses [42]. The action potential mately 50-fold greater than the EC50 for invasion of the terminal region of afferent baclofen's depressions of synaptic trans• fibers does not seem to be impaired by mission (see above). Thus the findings that baclofen, because its depressions of spinal baclofen reduces somatic calcium currents monosynaptic reflex responses can be tem• in dorsal root ganglion or myenteric plexus porarily overcome by post-tetanic potenti• neurons do not necessarily extrapolate to ation [22]. Shapovalov and Shiriaev also CNS neurons, and at present there is little concluded that baclofen did not impair evidence to support the claim that baclofen presynaptic terminal invasion [97]. inhibits transmitter release and, thereby, The terminal region of primary afferent synaptic transmission by reducing pre• fibers carry GABAB binding sites [91] and synaptic calcium influx. GABAA and GABAB receptors coexist on the perikarya of small caliber primary afferent fibers [33]. However, due to their small size, primary afferent terminals have Postsynaptic Actions of Baclofen resisted analysis with intracellular record• Baclofen has a dose- and concentration-de• ing techniques. An often used substitute for pendent hyperpolarizing action on some the analysis of ionic mechanisms that are mammalian central neurons [65, 73, 79] supposed to occur in the terminal region which is associated with an increase of are recordings from dorsal root ganglion postsynaptic conductance [20, 44, 47, 56, cells in vivo and in vitro. In dorsal root 59, 76, 78, 87, 99]. This action is unaffected ganglion neurons in culture, baclofen re• by blockade of synaptic transmission and is duces the duration of the calcium com• therefore indeed a direct postsynaptic ac• ponent of action potentials [35]. Similar re• tion and not a disfacilitation secondary to sults were obtained in neurons of the removal of tonic excitatory input [20, 76, myenteric plexus [23]. These results provide 78, 87, 99]. This action of baclofen results suggestive evidence that baclofen reduces in decreases in the direct excitability of neurotransmitter release by blocking in• central neurons [56, 60]. The baclofen ward calcium currents in presynaptic termi• concentrations at which these effects nals [see also: 97]. become apparent are approximately equal The Neuropharmacology of Baclofen 43 to the concentrations at which baclofen The extent to which baclofen's action to produces effects on synaptic transmission increase postsynaptic potassium conduc• (EC50 < 2 uM) [56, 59, 78,99]. tance contributes to its depressions of syn• The mean reversal potential of baclofen- aptic transmission in various structures or induced changes in Em and the dependence to the in vivo effects of baclofen is debat• of this reversal potential on the extracellu• able. As noted, there are many studies in lar potassium concentration indicate that which baclofen produced profound effects these changes are secondary to an increase on synaptic transmission and yet no evi• in the conductance of the postsynaptic dence for a postsynaptic action was found. membrane to potassium ions [56, 59, 78, That an increase in postsynaptic potassium 99]. In contrast to GABAA receptor- conductance is inhibitory is obvious, how• mediated responses, the amplitude and re• ever, and in some neurons the hyperpolari- versal potential of baclofen-induced chang• zations observed are as great as 20 mV. It is es in Em are unaffected by reductions of the unfortunate that in many of these studies extracellular chloride concentration [59, 78] the effects on synaptic responses were not or by intracellular injections of chloride investigated. Gähwiler and Brown [45] pro• ions [14, 76, 78]. posed that if baclofen increased the potassi• It was suggested on the basis of current um conductance of presynaptic terminals, clamp recordings from hippocampal this action might indirectly lead to a de• neurons that baclofen-induced conduc• crease in presynaptic calcium influx by de• tance changes were voltage-dependent [59, creasing the duration of the action po• 78]. This was verified directly in voltage tential. In neocortical neurons, however, clamp experiments on cultured hip• baclofen's depressions of postsynaptic po• pocampal neurons in which baclofen was tentials and its action to increase somatic shown to activate a potassium conductance potassium conductance could be tempo• which is voltage-dependent and inward rally dissociated. rectifying [45]. Baclofen-induced currents were also recorded under voltage clamp conditions from rat neocortical neurons Effect of Baclofen on Exogeneously Applied [56]. The conductance activated by baclo• Neurotransmitters fen is sensitive to blockade by the potas• Saito et al. [93] reported that, in low con• sium channel blockers 4-aminopyridine centrations, baclofen selectively reduced and barium ions; however, the present re• the depolarizing actions of substance P in sults indicate that it is different from any of spinal motoneurons and suggested that it the previously identified potassium con• was a substance P antagonist. This sugges• ductances [45, 59, 78, 99; but see: 14]. tion was not supported by subsequent elec• Baclofen-induced increases in post• trophysiological studies however [31, 37, synaptic potassium conductance are in• 41, 42, 51, 84], and baclofen does not in- sensitive to blockade by concentrations of terfer with the binding of 3H substance P GABAA receptor antagonists that com• [see: 17]. pletely antagonize GABA-mediated in• It was shown in several extracellularly creases in chloride conductance [20, 44, 45, conducted studies that baclofen can inhibit 56, 59, 76, 78, 87; but see: 60]. Pentobarbi• responses to exogenously applied excita• tone, which increases the effect of GABA tory neurotransmitters [27, 32, 41, 80, 83, on chloride-conductance, does not alter 93]. In further such investigations, however, the action of baclofen [78]. GABA can, it was demonstrated that inhibitions of re• however, mimic the action of baclofen to sponses to these chemical excitants re• increase postsynaptic potassium conduc• quired doses or concentrations of baclofen tance when GABAA receptors are blocked significantly greater than those at which with appropriate antagonists [44, 45, 78]. baclofen inhibited synaptically evoked re• These results are consistent with the con• sponses [30, 42, 52, 82]. In our intracellular clusion that the postsynaptic effects of study of neocortical neurons, we found that baclofen are mediated by GABAB re• applications of baclofen that produced ceptors. virtually complete depressions of stimu- 44 W. Zieglgänsberger et al. lation-evoked EPSPs did not significantly neurons by iontophoretically applied reduce depolarizations produced by GABA [56]. L-glutamate, L-aspartate, or N-methyl- D-aspartate [56]. The observed occasional and modest reductions of depolarizations GABAB Receptor Mediated Synaptic produced by these excitatory amino acids Processes were similar in magnitude and duration to It was proposed by Newberry and Nicoll baclofen-induced decreases in direct excit• [77, 78] that slow (long-latency) IPSPs ability, thus suggesting that they were the evoked in hippocampal CA1 neurons and result of baclofen's action to increase the hyperpolarizations produced by baclofen postsynaptic potassium conductance of rat may each be secondary to activation of neocortical neurons. That baclofen does not GABAB receptors. According to this as• postsynaptically block conductance in• sumption, baclofen and the endogenous creases produced by these substances is di• transmitter responsible for the long-latency rectly supported by our findings that baclo• IPSPs should act on the same population of fen had no effect on L-glutamate-evoked postsynaptic receptors to increase the same inward currents recorded in neocortical postsynaptic potassium conductance. Due neurons under voltage clamp conditions. to the lack of established antagonists of Baclofen applications that reduce the either baclofen-induced hyperpolarizations conductance increases associated with or long-latency IPSPs, this possibility can• stimulation-evoked GABAergic IPSPs have not be tested directly at present. Interest• no effect on conductance increases pro• ingly, however, baclofen reduces slow duced by the direct application of GABA IPSPs in hippocampal neurons [14, 60] and or the GABAA agonist muscimol [56, 96]. similar IPSPs evoked in neocortical Even at high concentrations, baclofen had neurons [56, 57]. The mechanism of this ef• no effect on currents evoked in neocortical fect of baclofen is unresolved.

Conclusions The presently available data indicates that At present, the mechanism by which baclofen has both presynaptic and post• baclofen reduces presynaptic transmitter synaptic effects in the mammalian CNS. release is unclear. Presynaptic inhibition Baclofen's action to increase postsynaptic secondary to depolarization of afferent ter• potassium conductance directly depresses minals can be excluded, however, and the neuronal excitability, however, several lines data indicate that baclofen does not reduce of evidence indicate that this effect is not synaptic transmission by hyperpolarizing alone responsible for baclofen's marked presynaptic fibers sufficiently to impair ac• depression of postsynaptic potentials. tion potential invasion of the terminal re• Although most of the recordings probably gion. There is some evidence which sug• have been obtained by somatic impale• gests that baclofen may reduce transmitter ments, the data suggest that the primary release by reducing inward calcium cur• mechanism by which baclofen depresses rents in presynaptic terminals. There is no synaptic transmission is a presynaptic re• direct evidence for such a mechanism, how• duction of transmitter release. This con• ever, and at least some calcium currents do clusion is supported directly by the es• not appear to be affected by baclofen. tablished effect of baclofen to reduce the evoked release of several putative neurotransmitters. The Neuropharmacology of Baclofen 45 References 1. Ashby P, White DG (1973) "Presynaptic" 15. Bowery NG, Hill DR, Hudson AL, Doble inhibition in spasticity and the effect of ß(4- AL, Middlemiss A, Shaw J, Turnbull M chlorophenyl)-GABA. J Neurol Sei 20: (1980) (-)Baclofen decreases neurotrans• 329-338 mitter release in the mammalian CNS by 2. Auden N, Wachtel H (1977) Biochemical an action at a novel GABA receptor. Na• effects of baclofen (ß-parachlorophenyl- ture 283:92-94 GABA) on the dopamine and the norad• 16. Bowery NG, Doble A, Hill DR, Hudson renaline in the rat brain. Acta Pharmacol AL, Shaw JS, Turnbull MJ, Warrington R Toxicol (Copenh) 40:310-320 (1981) Bicuculline-insensitive GABA re• 3. Ault B, Evans RH (1978) Central depres• ceptors on peripheral autonomic nerve ter• sant action of baclofen. J Physiol (Lond) minals. Eur J Pharmacol 71:53-70 284:13 IP 17. Bowery NG (1982) Baclofen: 10 years on. 4. Ault B, Evans RH (1981) The depressant Trends in Pharmacological Science 3:400 action of baclofen on the isolated spinal -403 cord of the neonatal rat. Eur J Pharmacol 18. Bowery NG, Hill DR, Hudson AL (1983) 71:357-364 Characteristics of GABAB receptor binding 5. Ault B, Nadler JV (1982) Baclofen selective• sites on rat whole brain synaptic mem• ly inhibits transmission at synapses made branes. Br J Pharmacol 73:191-206 by axons of CA3 pyramidal cells in the hip• 19. Bowery NG, Prive GW, Turnbull MJ, Wil• pocampal slice. J Pharmacol Exp Ther kin GP (1983) Evidence for the presence of 223:291-297 GABAB receptors on cerebellar Purkinje 6. Ault B, Nadler JV (1983) Anticonvulsant- dendrites. Br J Pharmacol 79:9P like actions of baclofen in the rat hip• 20. Brady RJ, Swann JW (1984) Postsynaptic pocampal slice. Br J Pharmacol 78:701 actions of baclofen associated with its an• -708 tagonism of bicuculline-induced epilepto- 7. Ault B, Nadler JV (1983) Effect of baclofen genesis in hippocampus. Cell Mol Neuro- on synaptically-induced cell firing in the rat biol 4:403-408 hippocampal slice. Br J Pharmacol 80:211 21. Cain CR, Simmonds MA (1982) Effects of -219 baclofen on the olfactory cortex slice prep• 8. Barry SR (1984) Baclofen has a presynaptic aration. Neuropharmacology 21:371-373 action at the crayfish neuromuscular junc• 22. Capek R, Esplin B (1982) Baclofen-induced tion. Brain Res 311:152-156 decrease of excitability of primary afferents 9. Bein HJ (1972) Pharmakologische Differen• and depression of monosynaptic transmis• zierung von Muskelrelaxantien. In: Birk• sion in cat spinal cord. Can J Physiol mayer W (ed) Aspekte der Muskelspastik. Pharmacol 60:160-166 Huber, Wien, pp 76-82 23. Cherubini E, North RA (1984) Inhibition of 10. Ben-Ari V, Krnjevic K, Reiflfenstein RJ, Re• calcium spikes and transmitter release by inhardt W (1981) Inhibition conductance gamma-aminobutyric acid in the guinea- changes and action of y-aminobutyrate in pig myenteric plexus. Br J Pharmacol rat hippocampus. Neuroscience 6:2445 82:101-105 -2463 24. Collins GGS, Anson J, Kelly EP (1982) 11. Benecke R, Meyer-Lohmann J (1974) Ef• Baclofen: Effects on evoked field potentials fects of an antispastic drug ß-(4- and amino acid neurotransmitter release in chlorophenyl)-gamma-aminobutyric acid the rat olfactory cortex slice. Brain Res on Renshaw cell activity. Neuropharma• 238:371-383 cology 13:1067-1075 25. Connors BW, Gutnick MJ, Prince DA 12. Bianchine JR (1985) Drugs for Parkinson's (1982) Electrophysiological properties of disease, spasticity, and acute muscle neocortical neurons in vitro. J Neuro- spasms. In: Goodman, Gilman (eds) The physiol 48:1302-1320 Pharmacological Basis of Therapeutics. 26. Curtis DR, Duggan AW, Felix D, Johnston Macmillan, London Basingstoke, pp 473 GAR, McLennan H (1971) Antagonism be• -490 tween bicuculline and GABA in the cat 13. Birkmayer RW (1972) Spasticity - a topical brain. Brain Res 33:57-73 survey. Huber, Wien 27. Curtis DR, Game CJA, Johnston GAR, 14. Blaxter TJ, Carlen PL (1985) Pre- and post• McCulloch RM (1974) Central effects of synaptic effects of baclofen in the rat hip• /3-(p-chlorophenyl)-gamma-aminobutyric pocampal slice. Brain Res 341:195-199 acid. Brain Res 70:158-170 46 W. Zieglgänsberger et al. 28. Curtis DR, Lodge D, Bornstein JC, Peet MJ 44. Gähwiler BH, Maurer R, Wüthrich HJ (1981) Selective effects of (-)-baclofen on (1984) Pitrazepin, a novel GABA A antago• spinal synaptic transmission in the cat. Exp nist. Neurosci Lett 45:311-316 Brain Res 42:158-170 45. Gähwiler BH, Brown DA (1985) GABAB 29. Davidoff RA, Sears ES (1974) The efTects of receptor-activated K+ current in voltage- Lioresal on synaptic activity in the isolated clamped CA3 pyramidal cells in hip• spinal cord. Neurology 24:957-963 pocampal cultures. Proc Natl Acad Sei USA 30. Davies J (1981) Selective depression of syn• 82:1558-1562 aptic excitation in cat spinal neurons by 46. Gehlert DR, Yamamura HI, Wamsley JK baclofen: an iontophoretic study. Br J Phar• (1985) Gamma-aminobutyric acid-B-re- macol 72:373-384 ceptors in the rat brain: Quantitative 31. Davies J, Dray A (1976) Substance P in the autoradiographi3 c localization using substantia nigra. Brain Res 107:623-627 (A H°(-)-baclofen. Neurosci Lett 56: 32. Davies J, Watkins JC (1974) The action of 183-188 ß-phenyl-GABA derivatives on neurones of 47. Haas HL, Greene RW, Olpe HR (1985) the cerebral cortex. Brain Res 70:501-505 Stereoselectivity of L-baclofen in hip• 33. Desarmenien M, Feltz P, Occhiopinti G, pocampal slices of the rat. Neurosci Lett Santangelo F, Schlichter R (1984) Coexis• 55:1-4 tence of GABAA and GABAB receptors on 48. Haefely W (1983) Tranquillizers. In: Gra- A delta and C primary afferents. Br J Phar• hame-Smith DG, Cowen PJ (eds) Psycho- macol 81:327-333 pharmacology I, part 1: Preclinical psycho- 34. Dreifuss J J, Kelly JS, Krnjevic K (1969) pharmacology. Excerpta Medica, Amster• Cortical inhibition and gamma-aminobu- dam Oxford Princeton, Elsevier, Amster• tyric acid. Exp Brain Res 9:137-154 dam, pp 107-151 35. Dunlap K (1981) Two types of gamma- 49. Haefely W, Pole P (1983) Electrophysiologi• aminobutyric acid receptor on embryonic cal studies on the interaction of anxiolytic sensory neurons. Br J Pharmacol 74:579 drugs with GABAergic mechanisms. In: -585 Malick JB, Enna SJ, Yamamura HI (eds) 36. Enna SJ, Gallagher JP (1983) Biochemical Anxiolytics: Neurochemical, Behavioral and electrophysiological characteristics of and Clinical Perspecitives. Raven Press, mammalian GABA receptors. Int Rev New York, pp 113-145 Neurobiol 2:181-212 50. Heinemann U, Hamon B, Konnerth A 37. Evans RH, Francis AA, Watkins JC (1976) (1984) GABA and baclofen reduce changes The effects of substance P like peptides on in extracellular free calcium in area CA1 of spinal motoneurones in vitro and antago• rat hippocampal slices. Neurosci Lett nism by Lioresal. Proc Br Pharmacol Soc 47:295-300 C18 51. Henry JL, Ben-Ari Y (1976) Actions of the 38. Evans RH, Watkins JC (1978) Specific an• p-chlorophenyl derivative of GABA, li• tagonism of excitant amino acids in the oresal, on nociceptive and non-nociceptive isolated spinal cord of the neonatal rat. Eur units in the spinal cord of the cat. Brain Res J Pharmacol 50:123-129 117:540-544 39. Fagg GE, Foster AC (1983) Amino acid 52. Henry JL (1982) Pharmacological studies neurotransmitters and their pathways in the on the prolonged depressant effects of mammalian central nervous system. Neuro- baclofen on lumbar dorsal horn units in the science 9:701-719 cat. Neuropharmacology 21:1085-1093 40. Fillenz M, Fung SC (1983) Effect of GABA 53. Henry JL (1982) EfTects of intravenously on 3(H)noradrenaline release from rat hip• administered enantiomers of baclofen on pocampal synaptosomes. J Physiol (Lond) functionally identified units in lumbar dor• 339:39-40P sal horn of the spinal cat. Neuropharma• 41. Fotherby KJ, Morrish NJ, Ryall RW (1976) cology 21:1073-1083 Is lioresal (baclofen) an antagonist of sub• 54. Hill DR, Bowery NG (1981) 3H-baclofen stance P? Brain Res 113:210-213 and 3H-GABA bind to bicuculline-insensi- 42. Fox S, Krnjevic K, Morris ME, Puil E, Wer- tive GABAB sites in rat brain. Nature man R (1978) Action of baclofen on mam• 290:149-152 malian synaptic transmission. Neuroscience 55. Howe JR, Sutor B, Zieglgänsberger W 3:495-515 (1985) The hyperpolarization by baclofen 43. Fukuda H. Kudo Y, Ono H (1977) Effects of rat neocortical neurons is probably un• of ß-(p-chlorophenyl)-GABA (baclofen) on related to its depression of postsynaptic po• spinal synaptic activity. Eur J Pharmacol tentials. Neurosci Lett [Suppl] 22:S386 44:17-24 56. Howe JR, Sutor B, Zieglgänsberger W The Neuropharmacology of Baclofen 47 (1986) Baclofen reduces postsynaptic po• NG (ed) Actions and Interactions of GABA tentials of rat neocortical neurons by an ac• and Benzodiazepines. Raven Press, New tion other than its hyperpolarizing action. J York, pp 109-132 Physiol (submitted for publication) 69. Krnjevic K (1974) Chemical nature of syn• 57. Howe JR, Sutor B, Zieglgänsberger W aptic transmission in vertebrates. Physiol (1986) Characteristics of a long-duration in• Rev 54:418-540 hibitory postsynaptic potential in rat frontal 70. Lanthorn TH, Cotman CW (1981) Baclofen neocortical neurons in vitro. Cell Mol selectively inhibits excitatory synaptic Neurobiol (submitted for publication) transmission in the hippocampus. Brain 58. Howe JR, Zieglgänsberger W (1986) Res 225:171-178 D-Baclofen does not antagonize the actions 71. Laskey W (1974) Effects of a GABA-deriva- of L-baclofen on rat neocortical neurons. tive on excitability of la afferent terminals. Neurosci Lett (submitted for publication) Proc Can Fed Biol Soc 17:55 59. Inoue M, Matsuo T, Ogata N (1985) Baclo• 72. Laskey W, Esplin B, Capek R (1975) Effects fen activates voltage-dependent and of the antispastic agent, ß-(4-chlorophenyl)- 4-aminopyridine sensitive K+ conductance gamma-aminobutyric acid (CPG), on spi• in guinea-pig hippocampal pyramidal cells nal reflexes. Proc Can Fed Biol Soc 18:30 maintained in vitro. Br J Pharmacol 73. Misgeld U, Klee MR, Zeise ML (1982) Dif• 84:833-841 ferences in burst characteristics and drug 60. Inoue M, Matsuo T, Ogata N (1985) sensitivity between CA3 neurons and gran• Characterization of pre- and postsynaptic ule cells. In: Klee MR, Lux HD, Speck• actions of (-)-baclofen in the guinea-pig mann EJ (eds), Physiology and Phar• hippocampus in vitro. Br J Pharmacol macology of Epileptogenic Phenomena. 84:843-851 Raven Press, New York, pp 131-139 61. Johnston GAR, Hailstone MH, Freeman 74. Möhler H, Okada T (1977) GABA receptor CG (1980) Baclofen: stereoselective inhi• binding with 3(H) + bicuculline-methiodide bition of excitant amino acid release. J in the rat CNS. Nature 267:65-67 Pharm Pharmacol 32:230 75. Mugnaini E, Oertel W (1985) An atlas of 62. Karbon EW, Duman R, Enna SJ (1983) the distribution of gabaergic neurons and Biochemical identification of multiple terminals as revealed by GAD immunohis- GABAB binding sites: association with tochemistry. In: Björklund A, Hökfelt T noradrenergic terminals in rat forebrain. (eds) Handbook of Chemical Neuroanato• Brain Res 274:393-396 my. GABA and Neuropeptides in the CNS, 63. Kato M, Waldmann U, Murakami S (1978) part I. Elsevier, Amsterdam, pp 436-608 Effects of baclofen on spinal neurones of 76. Newberry NR, Nicoll RA (1984) Direct hy• cats. Neuropharmacology 17:827-833 perpolarizing action of baclofen on hip• 64. Keberle H, Faigle JW (1972) Synthese pocampal pyramidal cells. Nature 308:450 sowie Beziehung zwischen Struktur und -452 Wirkung der Gamma-aminobuttersäurede- 77. Newberry NR, Nicoll RA (1984) A bicu- rivate. In: Birkmayer W (ed) Aspekte der culline-resistant inhibitory post-synaptic Muskelspastik. Huber, Wien, pp 90-93 potential in rat hippocampal pyramidal 65. Klee MR, Misgeld U, Zeise ML (1981) cells in vitro. J Physiol (Lond) 348:239-254 Pharmacological differences between CA3 78. Newberry NR, Nicoll RA (1985) Compari• and dentate granule cells in hippocampal son of the action of baclofen with gamma- slices. In: Feher O, Joo F (eds) Advances in aminobutyric acid on rat hippocampal Physiological Science, vol 36. Akademiai pyramidal cells in vitro. J Physiol (Lond) Kiado, Budapest, pp 155-164 360:161-185 66. Knutsson E, Lindblom U, Martensson A 79. Ogata N, Abe H (1982) Neuropharmacolo• (1974) Plasma and cerebrospinal fluid gy in the brain slice: Effects of substance P levels of baclofen (Lioresal) at optimal on neurons in the guinea pig hypothala• therapeutic responses in spastic paresis. J mus. Comp Biochem Physiol 72:171-178 Neurol Sei 23:473-484 80. Olpe HR, Koella WP, Wolf P, Haas HL 67. Koketsu K, Shoji T, Yamamoto K (1974) (1977) The action of baclofen on neurones Effects of GABA on presynaptic nerve ter• of the substantia nigra and of the ventral minals in bullfrog sympathetic ganglia. Ex- tegmental area. Brain Res 134:577-580 perientia 30:382-383 81. Olpe HR, Demieville H, Baltzer V, Bencze 68. Krogsgaard-Larsen P, Falch E, Jacobsen P WL, Koella WP, Wolf P, Haas HL (1978) (1984) GABA agonists: Structural require• The biological activity of D- and L-baclofen ments for interaction with the GABA-ben- (LioresaP). Eur J Pharmacol 52:133-136 zodiazepine receptor complex. In: Bowery 82. Olpe HR, Baudry M, Fagni L, Lynch G 48 W. Zieglgänsberger et al. (1982) The blocking action of baclofen on dual-action synapse (with special reference excitatory transmission in the rat hip• to baclofen). General Physiology and Bio• pocampal slice. J Neurosci 2:698-703 physics 1:423-433 83. Ono H, Fukuda H, Kudo Y (1979) Mecha• 98. Simmonds MA (1984) Physiological and nisms of depressant action of baclofen on pharmacological characterization of the ac• the spinal reflex in the rat. Neuro• tions of GABA. In: Bowery NG (ed) Ac• pharmacology 18:647-653 tions and interactions of GABA and ben• 84. Phillis JW (1976) Is ^-(4-chlorophenyl)- zodiazepines. Raven Press, New York, GABA a specific antagonist of substance P pp 27-41 on cerebral cortical neurons? Experientia 99. Stevens D, Gallagher JP, Shinnick-Gal- 32:593-594 lagher P (1985) Further studies on the ac• 85. Pierau FK, Zimmermann P (1973) Action tion of baclofen on neurons of the dorsolat• of a GABA-derivative on postsynaptic po• eral septal nucleus of the rat, in vitro. Brain tentials and membrane properties of cats' Res 358:360-363 spinal motoneurons. Brain Res 54:376-380 100. Sutor B, Zieglgänsberger W (1984) A 86. Pierau FK, Matheson GK, Wurster RE> GABA-mediated, chloride-dependent de• (1975) Presynaptic action of ß-(4-chloro- polarizing IPSP in neocortical neurons of phenyl)-GABA. Exp Neurol 48:343-351 the rat in vitro. Pflügers Arch [Suppl] 87. Pinnock RD (1984) Hyperpolarizing action 400: R37 of baclofen on neurons in the rat substantia 101. Sutor B (1985) Nachweis eines GABA-ver- nigra slice. Brain Res 322:337-340 mittelten, inhibitorischen postsynaptischen 88. Pole P, Haefely W (1976) EfTects of two Potentials in Neuronen des Neokortex der benzodiazepines, phenobarbitone and Ratte. Ph. D. Thesis, University of Er- baclofen on synaptic transmission in the cat langen-Nürnberg cuneate nucleus. Naunyn Schmiedebergs 102. Sutor B, Howe J, Zieglgänsberger W(1985) Arch Pharmacol 294:121-131 Baclofen depresses stimulation-evoked 89. Potashner SJ (1979) Baclofen: EfTects on postsynaptic potentials of rat neocortical amino acid release and metabolism in slices neurons in vitro. Naunyn Schmiedebergs of guinea pig cerebral cortex. J Neurochem Arch Pharmacol [Suppl] 329:381 32:103-109 103. Swahn CG, Beving H, Sedvall G (1979) 90. Potashner SJ, Gerard D (1983) Kainate-en- Mass fragmentographic determination of hanced release of D-3(H)aspartate from ce• 4-a m i n o- 3- p-ch loroph e ny lb u tyric aci d rebral cortex and striatum: reversal by (baclofen) in cerebrospinal fluid and serum. baclofen and pentobarbital. J Neurochem JChromatogr 162:433-438 40:1548-1557 104. Swaynok J, Dickson C (1984) D-Baclofen is 91. Price GW, Wilkin GP, Turnbull MJ, an agonist/antagonist at baclofen receptors Bowery NG (1984) Are baclofen-sensitive mediating antinociception in the spinal GABAB receptors present on primary af• cord. Soc Neurosci Abstr 10:32.13 ferent terminals of the spinal cord? Nature 105. Terrence CF, Sax M, Fromm GH, Chang 301:71-73 CH, Yoo CS (1983) Effect of baclofen enan- 92. Roberts E, Chase TN, Tower DB (1976) tiomorphs on the spinal trigeminal nucleus GABA in Nervous System Function. Raven and steric similarities with carbamazepine. Press, New York Pharmacology 27:85-94 93. Saito KS, Konishi S, Otsuka M (1975) An• 106. Wilkin GP, Hudson AL, Hill DR, Bowery tagonism between lioresal and substance P NG (1981) Autoradiographic localization of in rat spinal cord. Brain Res 97:177-180 GABAB-receptors in rat cerebellum. Nature 94. Schlicker E, Classen K, Göthen M (1984) 294:584-587 GABAB receptor-mediated inhibition of 107. Wilson PR, Yaksh TL (1978) Baclofen is serotonin release in the rat brain. Naunyn antinociceptive in the spinal intrathecal Schmiedebergs Arch Pharmacol 326:99 space of animals. Eur J Pharmacol 51:323 -105 -330 95. Schmidt RF (1971) Presynaptic inhibition 108. Wojcik WJ, Neff NH (1984) y-Amino- in the vertebrate central nervous system. butyric acid B receptors are negatively Erg Physiol Biol Exp Pharmacol 63:20-101 coupled to adenylate cyclase in brain and 96. Scholfield CN (1983) Baclofen blocks post• in the cerebellum these receptors may be synaptic inhibition but not the effect of associated with granule cells. Mol Phar• muscimol in the olfactory cortex. Br J Phar• macol 25:24-28 macol 78:79-84 109. Young RR, Delwaide PJ (1981) Drug ther• 97. Shapovalov AI, Shiriaev BI (1982) Selective apy: spasticity. N Engl J Med 304:28-33 modulation of chemical transmission at a 110. Zieglgänsberger W (1986) Central control The Neuropharmacology of Baclofen 49 of nociception. In: Handbook of Physiolo• Gamma-aminobutyric acid binding to re• gy. The Nervous System IV. pp 100-210 ceptor sites in the rat central nervous sys• 111. Zukin SR, Young AB, Snyder SH (1974) tem. Proc Natl Acad Sei USA 71:4802-4807