The role of adenosine receptors in the central action of caffeine (2024)

Abstract

The behavioral effects of caffeine appear likely to be due in largemeasure to antagonism of the action of endogenous adenosine at A₁-and A_2a-receptors in the central nervous system. Other biochemicalmechanisms of action of caffeine, such as release of intracellular calcium,inhibition of phosphodiesterases and blockade of regulatory sites ofGABA_A-reccptors, would require much higher concentrations thanthe micromolar concentrations of caffeine associated with behavioralstimulation. However, micromolar concentrations of caffeine also would beexpected to cause only a modest blockade of adenosine receptors. Selectiveadenosine agonists and xanthine antagonists have provided some insights intocentral roles for adenosine receptor subtypes. Thus, behavioral stimulation byxanthines appears to require blockade of both A₁- andA_2a-receptors. Chronic blockade of adenosine receptors by caffeinewould be expected to result in alterations in the central receptors and pathwaysthat are regulated by adenosine through A₁- andA_2a-receptors. Indeed, chronic caffeine docs alter the density notonly of adenosine receptors, but also of adrenergic, cholinergic, GABAergic andserotonergic receptors. Behavioral responses to agents acting throughdopaminergic and cholinergic pathways arc altered. As yet, a coherentexplanation of the acute and chronic effects of caffeine in terms of blockade ofadenosine receptors has not emerged. Interactions between pathways subserved byA₁ - and A_2a-adcnosine receptors complicate attemptsto interpret caffeine pharmacology, as does the complex control by adenosinereceptors of dopaminergic, cholinergic and other central pathways.

Keywords: Adenosine receptors, Calcium storage, Phosphodiesterase, Dopamine, Cocaine, Amphetamine, Nicotine, Muscarinic antagonists

The widespread usc of caffeine-containing beverages has focused research on themechanisms underlying the central effects of caffeine (Nehlig et al., 1992; Daly, 1993).While the effects of moderate doses of caffeine on behavior arc complex, it appearslikely that blockade of A₁- and/or A₂-adenosine receptors, are theprimary molecular site of action for caffeine. There are at lea st four types ofadenosine receptors in brain (Jacobson et al.,1992; , andref. therein). The A₁-class can be inhibitory of adenylate cyclase,stimulatory to potassium channels, inhibitory to calcium channels, and stimulatory tophosphoinositide breakdown. Selective agonists and selective xanthine and nonxanthineantagonists are available for A₁-receptors. The A_2a- andA_2b-Subclasses are stimulatory to adenylate cyclase. The A_2a-and A_2b-receptors differ in affinity and in agonist selectivity. Selectiveagonists and antagonists for A_2a-rcceptors are available. Selective agentsfor A_2b-reccptors arc not available. Caffeine is not selective forA₁- or A₂-receptors. The A₃-rcceptor also occurs inbrain, is inhibitory to adenylate cyclase, and is remarkable in being insensitive toblock ade by xanthines, at least in rodents.

Direct effects of caffeine on receptors, other than adenosine receptors, have notbeen reported. Indirect effects of caffeine on systems served by receptors other thanthe adenosine receptor will occur due to the blockade by caffeine of the tonicinhibitory input by adenosine through A₁-receptors on release ofnorepinephrine, dopamine, serotonin, acetylcholine, GABA, glutamate, and perhaps evenncuropeptides. Since A₂-adcnosinc receptors can modulate responses of brainsecond messenger systems to norepinephrine, serotonin, histamine (Daly, 1977), both the cyclic AMP-generation and thephosphoinositide-breakdown mediated by receptors for those neurotransmitters would beexpected to be altered by caffeine. Activation of A_2a-adenosine receptors hasbeen shown to reduce affinity of dopaminergic agonists for D₂-receptors(Ferré et al., 1991) and caffeine,thus, could affect dopamine function through blockade of tonic adenosine input to suchA_2a-adenosine receptors. Thus, caffeine, through blockade of adenosinereceptors, would be expected to indirectly influence the function of most neuronalpathways in the brain. There is evidence for central effects of caffeine invivo and in vitro on noradrenergic, dopaminergic,serotonergic, cholinergic, GABAergic, and glutaminergic systems (Daly, 1993), but current research is focused on the interrelatedadenosine-dopamine-acetylcholine systems of the basal ganglia, in particular thestriatum (Ferré et al., 1992).

Molecular sites of action other than adenosine receptors are known for caffeine.Historically, the first site of action of caffeine to be identified was stimulation ofrelease of calcium from intracellular storage sites. Caffeine binds to a site on acalcium-channel, which is associated with the intracellular, so-called calcium-sensitivepool of calcium, and thereby enhances calcium-dependent activation of the channel (McPherson et at., 1991). This calcium-channel isthe one blocked by ryanodine. Caffeine is now a widely-used tool for studies of the roleof this pool in nerve and muscle function, particularly with regard to oscillations inmembrane potentials and calcium levels. Caffeine, however, has a very low affinity forsuch sites with thresholds for effects on release of intracellular calcium at about 250μM, while 5 to 20 mM concentrations are required for robust effects. This is incontrast to the higher affinities of caffeine as an antagonist for adenosine receptors,where thresholds are less than 10 μM and Ki values arc 40 to 50 μM, wellwithin plasma and brain levels attained by htlmans and ani mals with behaviorallyeffective doses of caffeine. Certain xanthines that arc more potent than caffeine ascalcium releasing agents () may prove to be valuable in probing the role of the intracellularcalcium-sensitive calcium channels in the behavioral pharmacology of caffeine.

Historically, the second site of action of caffeine to be identified wasinhibition of phosphodiesterases. Other xanthines that are much more potent thancaffeine as phosphodiesterase inhibitors have been developed and most have proved rathernonspecific as inhibitors of various phosphodiesterase isozymes. Caffeine itself hasIC₅₀ values for phosphodiesterase isozymes ranging from 500 μM to1 mM, again well above the range at which caffeine blocks adenosine receptors. Xanthinesthat are potent phosphodiesterase inhibitors, in particular towards a braincalcium-independent cyclic AMP phosphodiesterase (rolipram-sensitive type IV isozyme)are behavioral “depressants”. in contrast to the behavioral stimulantactivity of caffeine and other xanthines that are weak inhibitors of that isozyme (Choi et al., 1988). The “depressant”part of the bell-shaped dose response curve of caffeine with respect to open-fieldlocomotor activity (Fig. 1) may be due toinhibition of the calcium-independent phosphodiesterase, which would become significantonly at the highest doses of caffeine.

Open in a separate window

Fig. 1

Typical Bell-shaped Dose-Response Curves for Effects of Xanthines on Open-FieldLocomotor Activity in Mice (Daly, 1993).Caffeine (●), theophylline (□), 3,7-dimethyl-1-propargylxamhine(○). Activity measured for 60 min in a circular arena afterintraperitoncal injection of xanthine to male NIH Swiss strain mice.

In the late 1970’s, caffeine was found to inhibit binding ofbenzodiazepines to sites on the GABAAreceptor channel (Marangos et al., 1979). Although exciting from the standpoint of theanxiogenic properties of caffeine, the affinity of caffeine (K_i 280μM) was several-fold higher than in vivo concentrations ofcaffeine that would be reached at non-toxic doses of caffeine. Interaction s atGABA_A-receptors may be relevant to the convulsant activity of caffeine.Xanthines more potent than caffeine at benzodiazepine sites have not been developed.

Thus, in spite of extensive studies on possible biochemical sites of action forcaffeine in vivo, only adenosine receptors have the requisite 10-50μM affinities for caffeine. Other sites, such as intracellular calcium-sensitivecalcium release channels, phosphodiesterases and GABA_A receptorsrequire> 200 μM concentrations of caffeine. At such concentrationscaffeine is a convulsant in vivo.

Chronic treatment of animals with caffeine, not surprisingly, results in anup-regulation of A₁-adenosine receptors as first reported by Fredholm (1982) for caffeine, and by Murray (1982)for theophylline. Almost all s ub sequent studies on chronic caffeine or theophyllinehave documented an increase in cortical A₁-adenosine receptors, except forene study with rats (Holtzman et al., 1991). TheA_2a-adenosine receptors do not appear to be up-regulated (Johansson et al., 1993; Shi et al., 1993), although there is one report of an increase inlevels of A_2a-aclenosine receptors in striatum after chronic ingestion ofcaffeine by mice (Hawkins et at., 1988). TheA_2b-receptor-mediated stimulation of cyclic AMP in rat brain slices doesnot appear altered after chronic caffeine (Fredholm,1982; ).

Most chronic studies related to levels or function of adenosine receptors havebeen caffeine or theophylline. In the 1980’s, chronic administration ofN⁶-R-phenylisopropyladenosine was shown to reduce the analgetic andlocomotor depressant effects of caffeine (). Levels of A₁-adenosine receptors wereunaltered. Recently, chronic injections of A₁-selective agents,8-cyclopentyl-1, 3-dipropylxanthine or N⁶-cyclopentyladenosine were shown tohave opposite effects on NMDA-induced seizures (VonLubitz et al., 1994). Chronic xanthine treatment greatly reduces theNMDA-effects, while chronic treatment with the adenosine analog enhances theNMDAelicited seizures. Levels of A₁-adenosine receptors were unaltered.

Adenosine receptors are not the only central receptors, whose levels are alteredafter chronic caffeine ingestion. This is not surprising, since removal of tonicadenosine inhibition of neurotransmitter release might be expected to increaseneurotransmitter release and lead to a downregulation 0f the relevant neurotransmitterreceptor. Hewever, in most cases an up-regulation rather than a down-regulation ofreceptors occurs. There has been only one broad study of effects of chronic caffeine onlevels of central receptors (Shi et al., 1993).Chmnic caffeine ingestion in male Swiss strain mice was found to affect the density ofreceptors subserving noradrenergic, serotonergic, cholinergic and GABAergic pathways(Table 1). Remarkably, since a variety ofevidence indicates that caffeine affects dopaminergic function (Ferré et al., 1992), the levels of dopaminergic receptorsappear unaffected. The levels of cortical and striatal At adenosine receptors areincreased by 15-20% by chronic caffeine, while the level of striatalA_2a-adenosine receptors is unaltered. The levels of corticalß₁- and cerebellar ß₂-adrenergic receptors arereduced by about 25%, while the levels of cortical α₁ andα₂-adrenergic receptors are not significanLly altered. The levelsof striatal D₁- and D₂-dopaminergic receptors are not altered.Levels of cortical muscarinic and nicotinic receptors are increased by 40-50%. Theapparent up-regulation of nicotinic receptors may actually represent conversion ofnicotinic receptors to a desensitized state. The level of corticalbenzodiazepine-binding sites associ-ated with GABA_A-receptors is increased by65% anci in this case the affinity for diazepam appears slightly decreased. The level ofcortical MK-801 binding sites associated with NMDA-glutaminergjc receptors appearunaltered. The level of cortical delta-opioid receptors is increased by 25%, while thelevels of cortical mu- and kappa-opioid receptors are unchanged. The level of corticalsigma receptors is unchanged. The density of cortical nitrendipine-binding sitesassociated with L-type calcium channels is increased by 18%. Thus, there is anincredible array of alterations in levels of central receptors elicited by chroniccaffeine ingestion in NIH Swiss strain mice. In addition, basal levels of striataladenylate cyclase are decreased after chronic caffeine, while stimulationsvia D₂-dopamine receptors or A_2a-adenosinereceptors are unaltered (Shi et al., 1994). Inrats, the up-regulation of adenosine receptors (see Daly,1993), the down-regulation of ß-adrenergic receptors (Goldberg et al., 1982; Fredholm et al., 1984; ) and an up-regulation of benzodiazepine sites, associated withGABA_A-receptors (; ) havebeen reported after chronic caffeine or theophylline. Effects on other receptors do notappear to have been examined systematically in rats. The levels of forskolin-bindingsites, associated with adenylate cyclase, have been reported to be increased in ratcerebral cortex after chronic caffehiC (Daval et al.,1989).

Clearly, the plethora of biochemical alterations in mice after chronic caffeinewill make difficult interpretations of behavioral alterations in the chronicallycaffeine-treated animal. The most studied behavioml alteration has been tolerance tocaffeine. An “insurmountable tolerance” has been reported in rats (Holtzman, 1983; Holtzman et al., 1988, 1991). Anexplanation as to how up-regulation of A₁-adenosine receptors could lead toan “insurmountable” tolerance to an agent, caffeine, that acts as anantagonist has not been forthcoming. The answer may lie in the biphasic dose-responsecurve to caffeine (see Fig. 1) where low doses ofcaffeine cause stimulation, while higher doses cause depression of locomotor activity.Thus, after chronic caffeine the depressant effects may predominate, leading to theappearance of an “insurmountable tolerance” with respect to behavioralstimulation. The effects of chronic caffeine on open-field locomotor activity have beenalso evaluated thoroughly, not in Sprague-Dawley rats, but recently in NIH Swiss strainmice. Tolerance does not occur in these mice and indeed the threshold for stimulatoryeffects of caffeine is significantly lowered (Fig.2, Nikodijevic et al., 1993a).Sensitization to behavioral effects of caffeine after chronic caffeine has also beenreported in rats (Meliska et al., 1990).Behavioral depression by high doses of caffeine is perhaps slightly enhanced afterchronic caffeine ingestion by mice (Nikodijevic et al.,1993a). The choreiform (dance-like) movements elicited in mice by high dosesof caffeine are significantly reduced after chronic caffeine ingestion (Nikodijevic et al., 1993c).

Open in a separate window

Fig. 2

Dose-Response Relationships for Effects of Caffeine on Locomotor Activity in Mice(Nikodijevic et al., 1993a). Activitywas measured for 30 min in a circular arena after intraperitoneal injection ofcaffeine in control and chronic caffeine male NIH Swiss strain mice. Caffeineingestion (100 mg/kg/day) was for 7 days. followed by 2-4 hr withdrawal forcaffeine clearance.

More consonant with the up-regulation of A₁-receptors is theobservation that the behavioral depressant effects of an A₁-selectiveadenosine analog, N⁶-cyclohexyladenosine (CHA) are slightly enhanced afterchronic caffeine ingestion in mice (Nikodijevic et al.,1993a,b). However, the behavioraldepressant effects of an A_2a-selective adenosine analog, APEC, arc alsoslightly enhanced, as are those of a potent mixed A₁/A₂-adenosineanalog, NECA. A simple interpretation of these results is complicated by the fact thatthere appea rs to be a synergism between the behavioral depressant effects of activationof A1-receptors and A_2a-receptors by selective adenosine analogs in mice(Nikodijevic et al., 1991). This may explainthe high potency of the mixed A₁/A₂ agonist NECA as a behavioraldepressant, and might explain the enhanced depressant effects of all adenosine analogsafter the chronic caffeine-elicited up-regulation of A₁-adenosinereceptors.

The converse to synergism s for agonists appears to apply with respect to thebehavioral stimulation elicited by xanthines. Thus, 8-cyclopentyltheophylline (CPT), anA₁-selective antagonist, is a weak behavioral stimulant, 8-cyclopentyl-1,3-dipropylxanthine (CPX), an even more A1-selective antagonist, is not a behavioralstimulant (Nikodijevic et al., 1991, 1993b) and 8-(3-chlorostyryl) caffeine (CSC), anA_2a-seleclive antagonist, is a very weak behavioral stimulant (Jacobson et al., 1993). However, a combination ofCPX and CSC results in a synergistic stimulation of open-field locomotor activity (Fig. 3). Thus, caffeine and other xanthines that arerelatively non-selective as adenosine receptor afltagonists may owe their effectivenessas behavioral stimulants to blockade of both A₁- and A_2a-adenosinereceptors.

Open in a separate window

Fig. 3

Effects of an A₁-Selective Antagonist (CPX) and anA_2a-Selective Antagonist (CSC) on Locomotor Activity in Mice (Jacobson et al., 1993). Activity wasmeasured for 30 min after intraperitoneal injection of 8-cyclopentyl-1,3-dipropyll(anthine (CPX, 0.25 mg/kg), and/or 8-(3-chlorostyryl) xanthine (CSC,1.0 mg/kg) in male NIH Swiss strain mice.

Geminal studies in the early 1980’s proposed a correlation ofA₁-receptor affinity and behavioral stimulation for xanthines (Snyder et al., 1981; Katims et al., 1983). This no longer appears true and indeedA_2a-receptors have been proposed to have a more dominant role inregulation of behavioral activity (). Recent studies suggest that stimulation of locomotor activity bycaffeine and development of tolerance to caffeine arc more closely related to blockadeof A₁-adenosine receptors (Kaplan et al.,1992, 1993). The demonstration ofsynergistic interactions of A₁- and A_2a-adenosine receptors incontrol of locomotor activity (Nikodijevic et al.,1991, Jacobson et al., 1993), suggeststhat questions as to relative importance of blockade of A₁- versusA₂-adenosine receptors to the effects of caffeine will be difficult toanswer. It is noteworthy that the effectiveness with which caffeine and other xanthinesreverse the depressant effects of adenosine analogs is usually greater than theirability to cause behavioral stimulation alone (Katims etal., 1983; Coffin et al., 1984; Holtzman et al., 1991).

One other aspect of behavioral effects of adenosine analogs and xanthinesfurther iiluslrates the complexity of interactions of adenosinc/xanthines. In the early1980’s it was noted that the combination of caffeine withN⁶-Rphcnylisopropyladenosinc not only reversed the behavioral depressanteffects of the adenosine analog, bllt actually caused a behavioral stimulation greaterthan that elicited by the xanthines alone (Snyder etal., 1981; Katims et al., 1983, seealso Phillis et al., 1986). This occurs withcaffeine and theophylline and even with a non-stimulatory xanthine,isobutylmethylxanthine (Fig. 4). An explanation wasnot apparent in 1980, nor has one been forthcoming. However, on examining dose-responseeffects on open-field locomotor activity for the adenosine analogs CHA, NECA and APEC incombination with caffeine, it appears that the synergistic stimulatory effect ofxanthine-adenosine analog combinations manifests itself as a stimulatory“bump” or at least a plateau in the behavioral dose-response curve asillustrated for NECA in Fig. 5A. It is noteworthythat the stimulatory “bump” is diminished after chronic caffeine ingestion(Fig. 5A and Nikodijevic et al., 1993b) as are the synergistic depressant effects ofA₁- and A_2a-agonists (Nikodijevic et al., 1993b). Such “bumps“ or plateaus in thedose-response curves for NECA have now been observed when NECA is administered incombination with central stimulants other than ca ffeine (see below).

Open in a separate window

Fig. 4

Effects of Xanthines and the Adenosine AnalogN⁶-R-phenylisopropyladenosine (R-PIA) on locomotor activity in mice(Daly, 1993). Open-field activity isfor the second thirty minute period after intraperitoneal injections to Male ICRmice of xanthines alone or in combination with R-PIA (0.2 μmol/kg). Notethe depressant effect of lowest dose of caffeine and stimulatory effect of R-PIAin combination with that dose of caffeine.

Open in a separate window

Open in a separate window

Fig. 5

Dose-Dependent Effects of NECA on Open-Field Locomotor Activity of Mice in thePresence of a Central Stimulant (Nikodijevic etal., 1993b; Shi et al., 1994).A. Caffeine (5 mg/kg). B. Amphetamine (1mg/kg). C. Cocaine (20 mg/kg}. D. Scopolamine(1 mg/kg}. Male NIH Swiss strain mice were injected intrapcritoneally with acentral stimulant (caffeine, amphetamine, cocaine, scopolamine) and varyingdoses of the adenosine analog NECA and open-field locomotor activity wasmeasured for a 30 minute period. Caffeine ingestion (100 mg/kg/day) was for 4days, followed by 2-4 hr withdrawal for caffeine clearance.

Since dopaminergic systems are strongly linked to caffeine pharmacology, it wasreasonable to examine alterations in behavior subserved by dopaminergic pathways afterchronic caffeine. Behaviorally, the stimulation of open-field locomotor activity byamphetamine, which releases dopamine, and cocaine which blocks reuptake of dopamine, areliule affected by chronic caffeine ingestion in mice (Nikodijevic et al., 1993a). This is at least consonant with the lack ofchange in density of dopamine receptors. However, it should be noted that a 1 mg/kg doseof amphetamine has significantly less effect after chronic caffeine ing estion {Nikodijevic et al., 1993a, sec Fig. 5B). Further studies are needed, since a large body of evidencesuggests that chronic caffeine ought to affect, via blockade of striatal adenosinereceptors, the function of dopaminergic receptors and/or dopaminergic sensitivity (Ferré et al., 1992 and ref. therein). Recentstudies, demonstrating that antagonists for Dt and Indopamine receptors blockcaffeine-induced stimulation of locomotor activity in rats () provide further evidence for theimportance of dopamine systems to the behavioral pharmacology of caffeine. Dopaminesystems do undergo hom eostatic changes as a result of denervation, chronic receptoractivation, or chronic receptor-blockade. Chronic treatment with adenosine analogs hasbeen shown to result in an attenuation of both A_2a-adenosine and Dt-dopaminereceptor-mediated stimulation of striatal adenylate cyclase (Porter et al., 1988). Dopamine denervation was reported to enhanceA_2a-adenosine and Indopamine receptor interactions in rat striatum (). Thus, it is clearthat dopamine and A_2a-adenosine systems are subject to homeostaticregulation, and the lack of major changes after chronic caffeine remains puzzling. It isof mechanistic interest that the dose-response curves for effects of NECA on open-fieldlocomotor activity can exhibit a stimulatory “bump” not only when assessedin the presence of a stimulatory dose of caffeine (Fig.5A), but also with a stimulatory dose of amphetamine or cocaine (Fig. 5BC). The “bump” for amphetamineis diminished after chronic caffeine ingestion.

Cholinergic systems also appear intimately linked to function of striataldopaminergic systems and to caffeine-sensitive adenosine systems. Chronic caffeineingestion does cause alterations in levels of both muscarinic and nicotinic receptors(Shi et al., 1993). Nicotine/caffeineinteracti ons have been extensively studied {White,1988; Cohen et al., 1991 and ref.therein). Behaviorally, the stimulation of open-field locomotor activity by themuscarinic agonist scopolamine in mice is significantly changed after chronic ca ffeineingestion (Nikodijevic et al., 1993a). Higherdoses of scopolamine are required for the same degree of locomotor stimulationsuggesting an increased tonic input of ace tylcholine to muscarinic receptors. The depressant effects of a muscarinic agonist, oxotremorine, appear only somewhat diminishedafter chronic caffeine ingestion. The dose-response curves for effects of NECA onlocomotor activity in the presence of a sti mulatory dose of scopolamine arc only marginally biphasic (Fig. 5D). Behavioral depressanteffects of a nicotinic agonist, nicotine, arc absent after chronic caffeine ingestion(Nikodijevic et al., 1993a, Shi et al., 1994) in spite of an apparentup-regulation of nicotine receptors. Probably, as is the case for tolerance to nicotineelicited by chronic nicotine, these “up-regulated” receptors are actuallydesensitized and non-functional (Marks et al.,1993). Nicotine, in combination with caffeine, has little effect in controlmice, but can cause behavioral stimulation after chronic caffeine (Shi et at., 1994). It should be noted that the open-field locomotoractivity of the NIH Swiss strain mice has been reduced by nearly 40% as a result ofchronic ingestion of caffeine, and it was postulated that this might be due in part toenhanced cholinergic function (Nikodijevic et al.,1993a). Mecamylamine, a nicotinic antagonist causes somewhat greaterdepression in mice after caffeine ingestion (Shi et at,1994).

Behavioral studies on noradrenergic, serotoninergic, GABA_A, andcalcium channel faneLion after chronic caffeine ingestion also are neeelecl, sincelevels of receptors subserving Stich systems are altered after chronic treatment of maleNIH Swiss strain mice (Shi et at, 1993).

In summary, a large body of evidence suggests that A₁- andA_2a-adcnosine receptors arc the most likely targets for pharmacologicalactions of caffeine in the central system. The A₁- andA₂-adenosinc receptor-regulated pathways are not independent and seem tointeract synergistically to cause betiavioral depression, while blockade byreceptor-selective xanthines interacts synergistically to cause behavioral stimulation.Thus, caffeine may owe many central-mediated behavioral effects to its nom-selectiveability to block A₁- and A_2a-adenosine receptors. Chronicingestion of caffeine by NIH Swiss strain mice results in a wide range of biochemicalalterations in the central nervous system (Table1). A depression in basal open-field locomotor activity occurs, accompaniedby changes in respoasi-vcness to caffeine and other xanthines, to adenosine analogs, andto cholinergic agents. Changes in behavioral responsiveness to dopaminergic agents areminimal. Combinations of caffeine with an adenosine analog reveal interestingmultiphasic effects of the adenosine analog on locomotor activity. Such multiphasiceffects also pertain for combinations of the adenosine analog with doparninergic(cocaine, amphetamine) agents, but not with muscarinic (scopolamine) agents. Chroniccaffeine ingestion reduces the depressant effects of ethanol, but chronic ethanolingestion has no effect on the locomotor stimulation evoked by caffeine (Daly et at, 1994).

Acknowledgements

The support of the International Life Science Institute for ottr program on“Mechanism of Action of Caffeine and Theophytline“ is gratefullyacknowledged.

References