@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Day, Jamie Catherine"@en ; dcterms:issued "2009-04-14T19:12:21Z"@en, "1994"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The regulation of forebrain cholinergic systems, specifically those neurons in the cholinergic basal nuclear complex (CBC) which project to the hippocampus and cortex are of great interest, given the involvement of acetyicholine (ACh) in cognitive function. To assess the activity of CBC neurons, in vivo microdialysis has been used in the present experiments to measure ACh release in the hippocampus and cortex of freely moving rats. Dialysate concentrations of ACh in the hippocampus and cortex (and striatum) of freely moving rats were found to correlate positively and significantly with locomotor activity, a behavioural measure of arousal. Two arousing stimuli, injection of vehicle and onset of the rats’ dark phase, increased locomotor activity and ACh release in all three brain regions, as did injection of the muscarinic antagonist scopolamine. These data suggest that forebrain cholinergic neurons are responsive to arousing stimuli and that ACh release in the crtex, hippocampus and striatum generally correlates with arousal. The dopaminergic regulation of CBC neurons was examined by determining the extent to which dopamine (DA) receptor agonists and antagonists affect cortical and hippocampal ACh release. The indirect DA agonist d-amphetamine (AMPH) and the DA receptor agonist apomorphine increased ACh release in both the cortex and hippocampus as did the selective D1 receptor agonist CY 208-243. D2 receptor agonists (quinpirole and/or PHNO) had no effect on ACh release in the cortex and produced slight decreases in the hippocampus. In addition, the AMPH-induced increases in ACh release in both regions were attenuated more by the D1 receptor antagonist SCH 23390 than by the D2 antagonists haloperidol and/or raclopride, as was the apomorphine induced release of ACh in the cortex. That DA mediates AMPH-induced increases in cortical ACh release was supported by the finding that prior selective lesions of ascending dopaminergic but not noradrenergic systems attenuated this effect of AMPH. These results suggest that CBC neurons are regulated in an excitatory manner by DA acting primarily at D1 receptors. The extent to which ACh release in the cortex and hippocampus is related to the performance of a learned behavioural task was assessed in rats trained to anticipate and consume a palatable liquid diet. Hippocampal ACh release increased during the anticipatory and consummatory periods of the task, but the increase observed in rats trained with the liquid diet was not higher than the increases seen in rats trained with water or in naive rats. In contrast, cortical ACh release increased to a greater extent in rewarded rats than it did in the two control groups. This suggests that cholinergic activity in both the cortex and hippocampus is increased by a reward-independent aspect of the task, such as arousal or attention, while an additional reward-dependent component is seen with respect to cortical ACh release."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/7057?expand=metadata"@en ; dcterms:extent "2495489 bytes"@en ; dc:format "application/pdf"@en ; skos:note "BASAL FOREBRAIN CHOLINERGIC NEURONS: REGULATION BY DOPAMINEAND RESPONSES TO AROUSING STIMULIByJAMIE CATHERINE DAYB.Sc., University of Victoria, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESProgram in NeuroscienceWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAugust 1994©Jamie Catherine DayIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)__________________________________Department of .6l44ufl tt n AJ..Q1MSC €A’(CQThe University of British ColumbiaVancouver, CanadaDate /kLkL2JAJi 1 (qc1LtDE-6 (2)88)11ABSTRACTThe regulation of forebrain cholinergic systems, specifically those neurons in thecholinergic basal nuclear complex (CBC) which project to the hippocampus and cortexare of great interest, given the involvement of acetyicholine (ACh) in cognitive function.To assess the activity of CBC neurons, in vivo microdialysis has been used in the presentexperiments to measure ACh release in the hippocampus and cortex of freely movingrats.Dialysate concentrations of ACh in the hippocampus and cortex (and striatum) offreely moving rats were found to correlate positively and significantly with locomotoractivity, a behavioural measure of arousal. Two arousing stimuli, injection of vehicleand onset of the rats’ dark phase, increased locomotor activity and ACh release in allthree brain regions, as did injection of the muscarinic antagonist scopolamine. Thesedata suggest that forebrain cholinergic neurons are responsive to arousing stimuli andthat ACh release in the crtex, hippocampus and striatum generally correlates witharousal.The dopaminergic regulation of CBC neurons was examined by determining theextent to which dopamine (DA) receptor agonists and antagonists affect cortical andhippocampal ACh release. The indirect DA agonist d-amphetamine (AMPH) and theDA receptor agonist apomorphine increased ACh release in both the cortex andhippocampus as did the selective D1 receptor agonist CY 208-243. D2 receptor agonists(quinpirole and/or PHNO) had no effect on ACh release in the cortex and producedslight decreases in the hippocampus. In addition, the AMPH-induced increases in AChrelease in both regions were attenuated more by the D1 receptor antagonist SCH 23390than by the D2 antagonists haloperidol and/or raclopride, as was the apomorphineinduced release of ACh in the cortex. That DA mediates AMPH-induced increases incortical ACh release was supported by the finding that prior selective lesions ofUiascending dopaminergic but not noradrenergic systems attenuated this effect of AMPH.These results suggest that CBC neurons are regulated in an excitatory manner by DAacting primarily at D1 receptors.The extent to which ACh release in the cortex and hippocampus is related to theperformance of a learned behavioural task was assessed in rats trained to anticipate andconsume a palatable liquid diet. Hippocampal ACh release increased during theanticipatory and consummatory periods of the task, but the increase observed in ratstrained with the liquid diet was not higher than the increases seen in rats trained withwater or in naive rats. In contrast, cortical ACh release increased to a greater extent inrewarded rats than it did in the two control groups. This suggests that cholinergicactivity in both the cortex and hippocampus is increased by a reward-independent aspectof the task, such as arousal or attention, while an additional reward-dependentcomponent is seen with respect to cortical ACh release.ivTable of ContentsPageAbstract iiTable of Contents ivList of Tables viList of Figures viiAcknowledgments ixI. Introduction 1(A) The Role of Acetylcholine in the CentralNervous System 1(B) In Vivo Microdialysis and Assay of ACh 7II. Acetyicholine Release in the Rat Hippocampus, Cortex andStriatum Correlates with Behavioural Arousal 13(A) Introduction 13(B) Materials and Methods 14(C) Results 18(D) Discussion 26III. Dopaminergic Regulation of Cortical Acetyicholine Release 35(A) Introduction 35(B) Materials and Methods 38(C) Results 39(D) Discussion 48IV. Attenuation of Amphetamine-Induced Increases of CorticalAcetyicholine Release by Forebrain Depletions of Dopamine,but not Noradrenaline 57(A) Introduction 57(B) Materials and Methods 58(C) Results 63(D) Discussion 70V. Dopaminergic Regulation of Hippocampal Acetyicholine Release 75(A) Introduction 75(B) Materials and Methods 77(C) Results 78(D) Discussion 84VII. Selective Enhancement of Cortical, but not Hippocampal,Acetyicholine Release during the Anticipation and Consumptionof a Palatable Meal 89(A) Introduction 89(B) Materials and Methods 90(C) Results 93(D) Discussion 101VPageVII. General Discussion 107(A) Technical Considerations of CholinergicResearch 107(B) Regulation of Basal ForebrainCholinergic Neurons 110VIII. References 114viList of TablesPageTable I Probe characteristics and baseline output of ACh and Chin the striatum, hippocampus and frontal cortex of rats 19Table 2 Correlation of regional ACh outputs and activity countsduring experimental treatments 25Table 3 Effects of 6-OHDA lesions of the mesotelencephalicdopamine system on regional tissue concentrations ofdopamine, noradrenaline and serotonin 64Table 4 Effects of 6-OHDA lesions of the dorsal noradrenergicbundle on regional tissue concentrations of dopamine,noradrenaline and serotonin 65Table 5 Consumption and latency to drinking during dialysis 96Table 6 Average acetylcholine release during anticipatory andconsummatory periods 98vi.’List of FiguresPageFigure 1 Schematic of in vivo microdialysis with on-line assay for ACh,and steps in the enzymatic conversion and electrochemicaldetection of ACh 12Figure 2 Dialysis probe placements in the frontal cortex, dorsalstriatum and dorsal hippocampus 16Figure 3 Striatal dialysate values of ACh and Ch after injections ofvehicle and scopolamine and exposure to dark 20Figure 4 Hippocampal dialysate values of ACh and Ch after injections ofvehicle and scopolamine and exposure to dark 21Figure 5 Frontal dialysate values of ACh and Ch after injections ofvehicle and scopolamine and exposure to dark 22Figure 6 Dialysate ACh concentrations and activity counts from threeindividual rats after injections of vehicle and scopolamineand exposure to dark 24Figure 7 Frontal cortex dialysate values of ACh after injections ofd-amphetamine, apomorphine or vehicle, and after localadministration of d-amphetamine through the dialysis probe 40Figure 8 Frontal cortex dialysate values of ACh after injections ofquinpirole and PHNO 41Figure 9 Frontal cortex dialysate values of ACh after injections ofCY 208-243 or a combination of quinpirole and CY 208-243 42Figure 10 Frontal cortex dialysate values of ACh after injections ofSCH 23390, d-amphetamine and vehicle, alone or in combination 44Figure 11 Frontal cortex dialysate values of ACh after injections ofhaloperidol or raclopride, d-amphetamine and vehicle, aloneor in combination 45Figure 12 Frontal cortex dialysate values of ACh after injections ofSCH 23390 or raclopride and apomorphine 46Figure 13 Tetrodotoxin-sensitivity of cortical ACh recovered by atransverse dialysis probe made of acrylonitrile fibre 67Figure 14 Dialysate values of ACh in the frontal cortex of rats havingunilateral 6-OHDA lesions of the mesotelencephalic dopaminesystem or of control rats after injection of d-amphetamine 68Figure 15 Dialysate values of ACh in the frontal cortex of rats havingbilateral 6-OHDA lesions of the dorsal noradrenergic bundleor of control rats after injection of d-amphetamine and vehicle 69vi”PageFigure 16 Hippocampal dialysate values of ACh after injection of apomorphine 79Figure 17 Hippocampal dialysate values of ACh after injections ofCY 208-243, quinpirole or vehicle 80Figure 18 Hippocampal dialysate values of ACh after injections ofd-amphetamine, SCH 23390, raclopride or vehicle, alone orin combination 82Figure 19 Effect of d-amphetamine applied through the dialysis membraneon hippocampal ACh output 83Figure 20 Volume of Sustacal consumed and latency to drink by trainedrewarded rats 95Figure 21 Dialysate concentrations of ACh in the hippocampus of trainedrewarded rats, trained non-rewarded rats and naive rats duringthe anticipatory and consummatory periods of the task 99Figure 22 Dialysate concentrations of ACh in the cortex of trainedrewarded rats, trained non-rewarded rats and naive rats duringthe anticipatory and consummatory periods of the task 100ixAcknowledgmentsIn addition to being a daunting role-model, my supervisor Dr. H.C. Fibiger hasprovided an atmosphere of scientific excellence in the Department of Psychiatry,Division of Neurological Sciences at U.B.C. My experiences in this department haveincluded: training by Dr. Geert Damsma, one of the world’s experts in AChmicrodialysis; superb technical (and moral) support from Chui-Se Tham and CatrionaWilson; statistical assistance from Dr. Campbell Clark; and valuable discussions andcollaborations with fellow students and postdoctoral fellows including Drs. Elio Acquas,Erin Brown, Lilli Collu, Fiona Inglis, Mark Klitenick, George Nomikos, GeorgeRobertson, and Julie Williams. I sincerely thank each of these people.My research has been supported by a Medical Research Council of CanadaStudentship. Some of the drugs for the experiments included in this thesis were kindlydonated by Astra, McNeil, Merck Sharp and Dohme, Sandoz, and Schering Plough.For my mother, Barbara Day, for always expecting my bestFor my husband, Alan Matheson, for accepting me at my worstFor Geert, an inspiring scientist and friend who is and will be greatly missed1I. INTRODUCTION(A) THE ROLE OF ACETYLCHOLINE IN THE CENTRAL NERVOUS SYSTEMAs a result of Loewi’s classic experiments in the early 1920’s, acetylcholine (ACh)was the first chemical demonstrated to act as a neurotransmitter. These studies involvedstimulating the vagus nerve of a frog, causing a decrease in heart rate, while perfusingthe heart with a physiological solution. The solution recovered, when perfused througha denervated frog heart also decreased stimulation-induced heart rate, suggesting that ahumoral factor was responsible for the effect of the nerve stimulation. This factor,termed “vagusstoff”, was later identified as ACh. Eccles subsequently demonstrated inthe 1950’s that ACh could also act as a transmitter in the mammalian central nervoussystem (CNS). Dale had previously identified cholinergic transmission at theneuromuscular junction; based on this information and the famous Dale’s principle,Eccles reasoned that ACh must also be the. neurotransmitter responsible for thecommunication between the spinal motorneuron and the electrophysiologically identifiedRenshaw interneuron. These landmark studies were among the first of a great body ofresearch on cholinergic systems which has helped to establish many of the basicprinciples of synaptic function and neurotransmission (see Karczmar, 1993).Since these early studies, extensive research into the anatomy, chemistry,pharmacology and physiology of central cholinergic systems has yielded several wellaccepted principles. For example, cholinergic neurons are now known to innervate muchof the mammalian brain: nuclei in the brainstem send projections mainly to thethalamus and other diencephalic structures, neurons in the forebrain innervate the entirecortical mantle including the hippocampus and amygdala, and intrinsic cholinergicinterneurons are found within the basal ganglia (Semba and Fibiger, 1989). Cholinergicneurons can be defined by the presence of the enzyme choline acetyltransferase (ChAT),2which catalyzes the synthesis of ACh from choline and acetylcoenzyme A (Tuek, 1984).Early investigations of cholinergic actions led to the discovery of two classes of receptorsfor ACh, nicotinic and muscarinic, which have more recently been subdivided basedboth on pharmacology and molecular biology (Ladinsky, 1993). The neurophysiologicaleffects of ACh, mediated by nicotinic cation channel/receptor macromolecules and/or byG—protein coupled muscarinic receptors, are determined by the cellular location of thereceptors and the specific second messenger systems and effectors to which they arecoupled (Krnjeviã, 1993).Given the complex anatomy of central cholinergic systems, the diversity ofreceptors which bind ACh, and the variety of ACh’s electrophysiological effects, it isnot suprising that ACh has complex and diverse roles in behaviour. Indeed, AChneurotransmission has been reported to participate in a variety of behavioural functionsincluding circadian rhythms (Rusak and Bina, 1990), antinociceptiOn (Iwamoto, 1989;Klamt and Prado, 1991), locomotion (Brudzynski et at., 1991; Flicker and Geyer, 1982),regulation of the sleep/wake cyOle (Baghdoyan et al., 1993; Steriade et a!., 1991), stress(Gilad, 1987; Imperato et a!., 1991) and electrocortical arousal (Semba, 1991). Stimulatedby the discovery that a consistent correlate of Alzheimer’s dementia is a loss ofcholinergic neurons in the basal forebrain (Coyle ci al., 1983), an intensively studiedfunction of ACh has been its proposed role in learning and memory. Along with thiscorrelational evidence of cholinergic loss in Alzheimer’s dementia, studies have shownthat administration of anticholinergic drugs to animals or humans or lesions of forebrainregions containing cholinergic cell bodies in animals cause cognitive impairments, whichmay be ameliorated by cholinergic agonists (Collerton, 1986; Hagan and Morris, 1988).These findings have supported a cholinergic hypothesis of cognition’, and haveencouraged much research into possible cholinergic therapies for the treatment ofcognitive deficits.3The excitement generated by this hypothesis has been tempered by the overallineffectiveness of cholinergic therapies in treating dementia (Holttum and Gershon,1992) and by criticisms of the above-mentioned supporting experimental evidence. Forexample, one major confound of the axon-sparing neurotoxic lesions used in animals tomodel Alzheimer’s dementia is that non-cholinergic neurons in the vicinity of the targetcholinergic neurons are lesioned as well (Arbogast and Kozlowski, 1988; Dunnett et al.,1987; Everitt et a!., 1987). In fact, the extent to which excitotoxic lesions of the basalforebrain decrease cortical ChAT activity does not correlate with the severity of theresulting behavioural deficits in several learning and memory tasks (Dunnett et al.,1987). Specifically, quisqualic acid lesions of the basal forebrain produce largerdecreases in cortical ChAT, but less severe deficits in the acquisition and performance ofseveral behavioural tasks, than do ibotenic acid lesions (Markowska et al., 1990; Robbinset a!., l989b). This suggests that the observed lesion-induced effects may be largelynon-cholinergic in origin. Along with the lack of a selective cholinergic neurotoxin,several other shortcomings of the data supporting the cholinergic hypothesis of dementiahave been outlined (Fibiger, 1991), including the idea that regionally specific effects ofperipherally administered cholinergic drugs have not been considered.Defining a cholinergic role in cognition is also complicated by the inherentdifficulties in designing and interpreting the results of behavioural tasks. It is possible,for example, that the influence of ACh on cognition may be more accurately defined bythe participation of forebrain cholinergic neurons in attentional processes rather than inlearning and memory per Se. Attention has various operational definitions but isgenerally assessed by a subject’s ability to use sensory information to follow some rule orcontingency; whereas a learning and memory task might assess the acquisition, retentionor recall of this rule, an attention task assesses the subject’s ability to follow the rule insituations where the initially salient sensory cue is unexpectedly switched to a new cue(attention reversal) or where the salient cue is presented among distractions (selective4attention). In support of a cholinergic involvement in attention, cholinergic agonistsimprove such measures of attention in Alzheimer’s patients (Sahakian et at., 1989; 1993)and the cholinergic receptor antagonist, scopolamine, disrupts attention in normal humansubjects (Dunne and Hartley, 1985; Wesnes and Warburton, 1983). Additionally,excitotoxic lesions or pharmacological inhibition of the basal forebrain in animals causesimpairments in attentional tasks (Olton et at., 1988; Pang et at., 1993; Robbins et at.,1989a; Voytko et al., 1994) which, in some cases, have been ameliorated by cholinergicagonists or tissue grafts (Muir et at., 1992a; 1992b; 1994). It is important to note,however, that these experiments suffer from the same limitations as those describedabove (for example, lack of a selective cholinergic neurotoxin), and that contradictorydata have also been reported (Roberts et at., 1992). It should also be noted that theseresults do not exclude a role for basal forebrain cholinergic neurons in learning andmemory. It has recently been suggested, for example, that excitotoxic lesions of basalforebrain regions more rostral than those which cause attention deficits may impairmemory (Muir et at., 1993).Learning, memory and attention are complex cognitive constructs, operationallydefined using behavioural measures. In contrast, electroencephalographic (EEG) arousalis an electrophysiological measure, defined by low voltage fast activity (LVFA), ordesynchrony, in the cortex and rhythmic slow activity (RSA), or theta rhythym, in thehippocampus. The importance of ACh in the generation of these averaged electricalpatterns is well supported and has been reviewed (Buzsáki and Gage, 1989; Riekkinen etat., l991a; Semba, 1991; Smythe et at., 1992; Vanderwolf, 1988). Briefly, the activity ofpresumed cholinergic neurons in the basal forebrain is increased during EEG arousal(Détári and Vanderwolf, 1987; Sweeney et a!., 1992); stimulation of cholinergic cell bodyregions induces LVFA (Metherate et a!., 1992); RSA can be induced or blocked byadministration of cholinergic agonists or antagonists, respectively, into the hippocampus(Golebiewski et at., 1992; Rowntree and Bland, 1986) or the basal forebrain (Monmaur5et a!., 1993); cortical ACh release is increased during LVFA (Casamenti et a!., 1986;Celesia and Jasper, 1966; Kanai and Szerb, 1965); and cholinergic antagonists or lesionsof cholinergic cell body regions disrupt EEG arousal (Buzsâki et a!., 1988; Ray andJackson, 1991; Riekkinen et a!., 1992) which in some cases can be ameliorated bycholinergic agonists (Riekkinen et a!., 1991a Vanderwolf et a!., 1993).Although the critical significance of ACh in EEG arousal is well established, themechanisms by which ACh and other neurotransmitters interact to create theelectrophysiological phenomena underlying LVFA and RSA are complex and still beingelucidated (Foote and Morrison, 1987; Smythe et a!., 1992, Steriade et a!., 1993;Vanderwoif, 1988). In addition, the behavioural significance of EEG arousal has notbeen definitively determined. It should be noted, for example, that LVFA and RSAoccur during rapid eye movement (REM) sleep. In the waking state, however, EEGarousal occurs during voluntary movement in laboratory animals (Vanderwolf, 1988) andduring periods of facilitated sensory processing (Livingstone and Hubel, 1981; Simon andEmmons, 1956). It is thus often speculated, but not universally accepted (Vanderwolf,1988), that EEG arousal may be a physiological representation of attentiveness, and maybe required for cognitive functioning. This theory is supported by correlationalevidence including reports that the degree of disruption of EEG patterns correlates withthe severity of cognitive deficits in Alzheimer patients (Penttilä et at., 1985) and thatcognition enhancers activate the EEG (Delacour et at., 1990a). The fact that LVFA andRSA also occur during REM sleep may not dispute the theory that aroused EEG patternsare required for cognitive functioning, given that REM sleep may be involved ininformation processing (Dujardin et a!., 1990).Consistent with the speculation that increased ACh release occuring during EEGarousal contributes to increased attentiveness, cholinergic mechanisms are involved infacilitating the activity of hippocampal (Markram and Segal, 1990) and cortical neuronsresponding to afferent neural transmission. Single or multiple unit responses of cortical6neurons to visual, auditory or somatic stimulation, or to stimulation of the auditorythalamus, are enhanced by iontophoretic application of ACh and/or by stimulation ofthe basal forebrain (Donoghue and Carroll, 1987; Hars et at., 1993; Metherate and Ashe,1993; Metherate et at., 1987; Sillito and Kemp, 1983; Tremblay et a!., 1990; Webster etat., 1991). In some cases, the facilitory effects of basal forebrain stimulation have beenblocked by intracortical iontophoretic or systemic administration of the muscarinicreceptor antagonist atropine (Hars et al., 1993; Metherate and Ashe, 1993; Tremblay eta!., 1990). Electrical stimulation of the basal forebrain may approximate an increase inthe firing of basal forebrain neurons, such as that which occurs in monkeys presentedwith stimuli that precede, or have been previously paired with reinforcement(Richardson and DeLong, 1990; Wilson and Rolls, 1990) or in rabbits presented with aconditioned stimulus which predicts a mildly aversive stimulus (Whalen et al., 1994).Accordingly, the increased cortical ACh release presumably accompanying this increasedfiring may facilitate cortical neuronal responses to other, co—incident afferent signals,thereby providing a mechanism by which the relevance of sensory stimuli could beneurally encoded. Cholinergic facilitation of neuronal responsiveness may alsoparticipate in sensory cortex plasticity (Delacour et at., 1990b), cellular associativeconditioning (Pirch et al., 1992; Rigdon and Pirch, 1986), long-term potentiation in boththe cortex and hippocampus (Blitzer et al., 1990; Hunter et at., 1994; Lin and Phillis,1991) and afferent stimulation-induced protein synthesis in the hippocampus (Feig andLipton, 1993), all of which have been proposed as cellular models or correlates oflearning and memory.Considered together, the research outlined above points to the importance offorebrain cholinergic systems in determining an animal’s perception of its environmentas to which aspects are deemed significant, attended to, and thus perhaps remembered.Although terms such as learning, memory, attention and arousal may be insufficientlyprecise to define the roles of ACh in the CNS, they do succeed in suggesting a role for7this neurotransmitter in cognitive function. To complement research aimed atdetermining the functional significance of forebrain cholinergic systems, basiccharacterizations of these neurons are required, including their response to environmentalstimuli and their regulation by endogenous neurotransmitters. Given the anatomy andproposed functions of cholinergic systems, such research should ideally include measuresof regionally specific cholinergic activity in awake, behaving animals.(B) IN VIVO MICRODIALYSIS AND ASSAY OF ACHIn vivo microdialysis allows the measurement of interstitial concentrations ofneurotransmitters in discrete brain regions of awake, behaving animals (Benveniste andHüttemeier, 1990; Ungerstedt, 1984; Westerink et at., 1987) and is presently the bestavailable technique for estimating ACh release. Microdialysis exploits the ability ofsmall molecules to diffuse down a concentration gradient from the brain into a solutionapproximating brain interstitial fluid perfused through a semi-permeable, tubular dialysismembrane. A neurochemical which is permeable to the dialysis membrane, and forwhich there is a sensitive assay, can then be measured in the collected dialysate.Similarly, using reverse dialysis, drugs included in the perfusion solution can be appliedby diffusion into the discrete sampling area of the dialysis probe, and this can beaccomplished concurrently with neurochemical measurement. An obvious advantage ofmicrodialysis over ex vivo techniques is the ability to monitor neuochemica1s frequently(several times an hour) and for many hours in each animal. Compared to earlier in vivosampling methods, such as push-pull perfusion (Philippu, 1984) and the cortical-cuptechnique (Moroni and Pepeu, 1984), microdialysis produces less tissue damage andperturbation, can be performed in most brain regions, and is amenable to use in awake,freely-moving animals. Microdialysis also has an advantage over another current in vivo8neurochemical sampling technique, in vivo electrochemistry (Kawagoe et at., 1993),which can only be used to detect electroactive neurochemicals.Despite its many advantages, in vivo microdialysis has several limitations whichmust be considered. First, the concentrations of neurotransmitters measured usingmicrodialysis do not represent absolute synaptic concentrations. Factoring into thecalculations the in vitro recovery of the dialysis probe succeeds in estimating theconcentration of molecules reaching the dialysis probe, but neglects the influence ofimportant in vivo factors (Amberg and Lindefors, 1989; Benveniste et at., 1989; Justice,1993) such as metabolism, active uptake and tortuosity, which is a descriptor of theimpedance of the interstitial pathway through which the transmitter diffuses between itsrelease site and the probe. Because the absolute synaptic concentration of transmittermolecules can not yet be accurately determined, most microdialysis data are interpretedwith regard to relative changes in interstitial concentrations of transmitter arising,presumably, from synaptic overflow.The possibile origin of the recovered transmitter from activity-independent releasefrom neurons or from non-neuronal sources is the second limitation of microdialysis thatmust be considered. To assess the degree to which diaysate concentrations of aneurochemical are dependent on activity-dependent release, the sodium channel blockertetrodoxin (TTX) can be added to the perfusion solution, effectively blocking theconduction of action potentials in the vicinity of the dialysis probe. In optimalmicrodialysis conditions, TTX treatment reduces the concentrations of severalneurotransmitters by at least 80%, and often to below detectable levels (Damsma et at.,1987b;1987c; Santiago and Westerink, 1990; van Veldhuizen et a!., 1990). Removal ofcalcium ions from the perfusion solution, or their replacement with competitive bivalentcations, can also be used to estimate the proportion of dialysate transmitters originatingfrom calcium-dependent vesicular release (Benveniste and HUttemeier, 1990), although9non-vesicular, calcium-independent release may still be neuronal in origin and activityrelated (Levi and Raiteri, 1993).It has been demonstrated that the tissue trauma resulting from probe implantationaffects interstitial concentrations of neurochemicals (de Boer et a!., 1992; Reiriz et a!.,1989); the amount of time after implantation required to optimize recovery ofneurochemicals related to activity-dependent release, usually one to two days, depends inpart upon the type of dialysis probe used (Santiago and Westerink, 1990). In the case ofneurochemicals such as glutamate and adenosine, which have constitutive roles in cellularfunctioning as well as proposed roles as neurotransmitters, those molecules originatingfrom neurotransmission-related release may only represent a fraction of the total amountrecovered. Thus, for example, basal microdialysate concentrations of glutamate areunaffected or only partially reduced by TTX infusion (Girault et a!., 1986; Moghaddam,1993; Westerink et a!., 1987).A third important consideration to be made concerning microdialysis is whetherbasal, unstimulated neurotransmitter release is being measured. Dialysate concentrationsof neurotransmitters have been shown to be dependent on the ionic composition of theperfusate (Moghaddam and Bunney, 1989; Osborne et aL, 1991). Perfusate solutionscontaining higher concentrations of calcium or potassium than those found in theinterstitial space surrounding the probe would be expected to artificially stimulatetransmitter release by affecting the respective ion gradients and currents across theneuronal membrane. Indeed, it has been shown that perfusion solutions differing incalcium concentration can yield qualitatively different results in ACh microdialysisexperiments (de Boer et a!., 1990a). To ensure that basal release is being assessed, theionic composition of the perfusate should closely approximate that in the interstitialspace (Hansen, 1985).A final characteristic of microdialysis that must be considered is the basicfunctional principal of dialysis itself: that any molecule or ion which is permeable to10the dialysis membrane and which is not in equal concentrations in the brain and in theperfusion solution will cross the membrane, flowing down its concentration gradient.Although this allows the recovery and analysis of neurochemicals, it also disturbs thelocal chemical microenvironment by removing not only the molecular species of interestbut also many others. The brain tissue surrounding the probe for up to 1mm is drainedby this neurochemical “sink” (Benveniste, 1989; Blaha, 1991) and may be influenced bysuch factors as rate of perfusion, perfusate composition, and probe geometry. It musttherefore be recognized that microdialysis does not assess neurotransmitter release inentirely normal physiological conditions, but rather within a zone of artificialequilibrium between the tissue and the dialysis probe, where many dynamic neuronalprocesses are presumably affected. The recently developed technique of quantitativedialysis (Justice, 1993) attempts to deal with this issue, at least with respect to thespecies of interest, by including this neurochemical in the perfusate at a concentrationwhere no net flow occurs across the membrane. While this modification is prohibitivelylabour-intensive for most dialysis applications, it is informative about the dialysistechnique itself.A special consideration must be made in the case of microdialysis for recovery ofACh. Because the degradation of released ACh by acetylcholinesterase (AChE) activityis extremely efficient, an AChE inhibitor is usually included in the perfusion solution.The reversible, non-lipophilic, quaternary AChE inhibitor neostigmine has been arguedto be best suited for this purpose (Damsma and Westerink, 1991). Depending on thesensitivity of the specific assay, this adaptation is necessary in most cases to allow therecovery of reliably detectable amounts of ACh. High-maintenance assay procedures orlong sampling intervals have been used to carry out dialysis experiments in the absenceof an AChE inhibitor (Damsma et at., l987c; de Boer et at., 1990b; Messamore et at.,1993), although it is usually reported that release is not detectable in all animals tested.To increase the number of successful experiments, inclusion of low concentrations of an11AChE inhibitor is generally considered an acceptable compromise of the dialysistechnique. Indeed, presently it is not known to what extent the presence of AChEinhibition might actually ameliorate the above—mentioned dialysis “sink” effect, at leastwith respect to the local depletion of ACh surrounding the membrane.Although the above-mentioned issues must be considered in assessing microdialysisas an in vivo neurochemical sampling technique, the question of quantitative, assays forACh need only address two points: sensitivity and specificity. Several assays for AChhave been coupled with in vivo monitoring techniques, including bioassays (Celesia andJasper, 1966; Kanai and Szerb, 1965), gas chromatography/mass spectrometry (Marienand Richard, 1990), a radioenzymatic assay (Consolo et a!., 1987) and aradioimmunoassay (Kawashima et at., 1991). The highest sensitivities routinely reported,however, have been from chromatographic methods coupled to electrochemical detection(ECD; Damsma et at., 1987a’). ECD of the electrically inactive ACh molecule isachieved by post-chromatographic enzymatic conversion of ACh, via anacetylcholinesterase, to choline and acetic acid, and the subsequent conversion ofcholine, via choline oxidase, to betaine and hydrogen peroxide (Fig. 1). The hydrogenperoxide, stoichiometrically produced from ACh, is detected by its oxidation at aplatinum electrode. The chromatographic separation and subsequent enzymaticprocessing of the ACh fulfills the requirement of specificity in the assay.Given the above-mentioned considerations, the experiments reported here use invivo microdialysis coupled to on-line assay of ACh by high performance liquidchromatography with ECD (HPLC-ECD; Damsma and Westerink, 1991) to assessneuronal overflow of ACh in forebrain regions of awake, freely moving rats (Fig. 1).Experiments were conducted on the second day after implantation of a transversemicrodialysis probe and 100 nM neostigmine was included in the perfusate.Microdialysate concentrations of ACh using this methodology have previously beendemonstrated to be calcium-dependent and TTX-sensitive (Damsma et al., 1987c; 1988).12CH3\\+ /,2CHN_CHrcOH 2 12°2CH3Fig. 1 Upper panel: Schematic of in vivo microdialysis with on-line assay foracetyicholine (ACh). a) Mobile phase reservoir, b) HPLC pump, c) pulse dampener, d)guard column, e) standard ioop and valve, f) sample loop and valve, g) timer, h)analytical column, i) immobilized enzyme reactor, j) electrochemical detector includingworking and reference electrodes and potentiostat, k) chart recorder and 1) perfusatesyringe driver. An average time delay of 20 mm occurs between the neurochemicalevent in the rat’s brain and its representation on the chart recorder. See Chapter II,Materials and Methods section for details.Lower panel: Steps in the enzymatic conversion and electrochemical detection of ACh,occuring in the enzyme reactor and on the working electrode (i and j above).bC dgaCH3,CH3CHN-CCOCCH3Acetlycholine CH3Esterase + ,CH3+ H20 -CHN-CH2CHO +HO—CCH3CH3 Choline\\ + OxidaseCHi—N—CIk-CHPH + H20 + 202CH3PlatinumElectrode14202+ 500 mV+ 2W + 2e13II. ACETYLCHOLINE RELEASE IN THE RAT HIPPOCAMPUS, CORTEX ANDSTRIATUM CORRELATES WITH BEHAVIOURAL AROUSAL(A) INTRODUCTIONACh has been implicated in a variety of behavioural functions including learningand memory (Collerton, 1986; Hagan and Morris, 1988). Although the strength of thedata supporting the cholinergic hypothesis of dementia has been questioned (Fibiger,1991), forebrain ACh may participate in cognitive functions other than learning andmemory per Se, such as attention (Muir et al., 1994; Sahakian et al., 1993; Voytko et al.,1994) or arousal (Semba, 1991). Arousal can be operationally defined in laboratoryanimals either electroencephalographically or behaviourally. Central cholinergicneurotransrnission has been associated with both EEG and behavioural measures ofarousal. Using cortical cup and push/pull perfusion techniques, increased concentrationsof ACh have been recovered during spontaneous or stimulated EEG desynchronization incats (Celesia and Jasper, 1966; Kanai and Szerb, 1965; Phillis and Chong, 1965; Szerb,1967) and rats (Casamenti et al., 1986), during increased behavioural activity in rabbits(Collier and Mitchell, 1967), and during sensory stimulation or treadmill-running in rats(Dudar et al., 1979). More recently, significant correlations have been reported betweencortical, hippocampal and striatal dialysate ACh concentrations (Or whole brain tissueconcentrations) and motor activity counts after anticholinergic treatment (Frances andJacob, 1971; Toide, 1989; Watanabe and Shimizu, 1989).Microdialysate concentrations of ACh from awake, freely-moving rats in drug-freeconditions have been demonstrated to vary considerably (Damsma et al., 1987b), andpreliminary studies in this laboratory suggested that these variations may be due tochanges in animals’ activity levels. The experiments reported here were undertaken tofurther examine the possible relationship between ACh release and arousal, as measured14by locomotor activity. Photocell beam interruptions and in vivo microdialysateconcentrations of ACh and choline (Ch) in the dorsal hippocampus, frontal cortex ordorsal striatum of rats were measured simultaneously under three conditions: 1) after aninjection of vehicle; 2) after administration of the muscarinic receptor antagonist,scopolamine; and 3) before and after the beginning of the rats’ night cycle. The twonon-pharmacological stimuli are associated with increased locomotion, while muscarinicreceptor antagonists also increase locomotion and are known to increase ACh release inthe cortex, hippocampus, and striatum via the blockade of an inhibitory feedbackmechanism (Aquilonius et al., 1972; Dolezal and Wecker, 1990; Lefresne et al., 1978;Nordstrom and Bartfai, 1980; Rospars et a!., 1977; Szerb et a!., 1977).(B) MATERIALS AND METHODSExperimental protocol and drugsExperiments were performed on male Wistar rats (250-330g) two days after theimplantation of a microdialysis probe. Following surgery, rats were housed individuallyin Plexiglas cages (35x35x25 cm) and maintained on a 12:12 h light:dark schedule withfood and water available ad libitum. During the light phase of the rats’ daily cycle,subjects were first injected with water (1 mi/kg, s.c.), and then 3 h later with themuscarinic antagonist, scopolamine hydrobromide (0.4 mg/kg, Sigma). Experimentscontinued into the rats’ usual dark phase (when the room lights were turned off)approximately 4 h later. During each of these conditions, locomotor activity andhippocampal, cortical or striatal interstitial concentrations of ACh and Ch were measuredsimultaneously.15Surgery and microdialysisTrans-cerebral microdialysis sampling of ACh was performed according to themethodology of Damsma et a!. (1987b) which has been reviewed (Damsma andWesterink, 1991). Rats were stereotaxically implanted with a transverse dialysis probe(Fig. 2; Damsma et a!., 1987b; Imperato and Di Chiara, 1984) under pentobarbitalanesthesia (50-60 mg/kg, i.p.). The probe was placed in one of three sites according tothe atlas of Paxinos and Watson (1986), measured from bregma in mm: striatum A:+l.7,V:-4.75; hippocampus A:-4.3, V:-3.3; frontal cortex A:+2.7, V:-2.5. The probes weremade of saponified cellulose ester dialysis fibre (ID = 0.22 mm, OD = 0.27 mm,molecular weight cut off > 10 000 Dalton; Cordis Dow Medical), having an activesurface length of 6.4 mm, 6.8 mm, or 10.9 mm for the striatum, hippocampus, or frontalcortex, respectively. On completion of each experiment, the probe location was verifiedusing standard histological procedures.During microdialysis experiments the dialysis fibre was perfused at 5 uI/mm,controlled by a syringe pump (Carnegie Medicin). The syringe was connected to theprobe inlet by polyethylene tubing (800x0.28 mm); the probe outlet was connected to thesample loop (100 UI) of the analytical system by fused silica tubing (800x0.1 mm). Thesample valve was controlled by an adjustable timer (Valco), and samples (50 ul) werecollected and injected at ten minute intervals.The composition of the perfusion solution was selected to approximate the ioniccomposition of the interstitial fluid in the brain (Hansen, 1985) and contained NaCI (125mM), KC1 (3 mM), CaCl2 (1.3 mM), MgCI2 (1.0 mM), NaHCO3 (23 mM) in aqueousphosphate buffer (pH 7.3). To recover detectable dialysate concentrations of ACh, areversible acetylcholinesterase (AChE) inhibitor (neostigmine bromide, 0.1 uM; Sigma)was included in the perfusion solution. Thirty minutes of perfusion preceded samplecollection to allow perfusate concentrations of ACh and Ch to equilibrate with those inthe brain.16-‘-2.7Figure 2. Dialysis probe placements in the frontal cortex, dorsal striatum and dorsalhippocampus. Measurements are from Bregma, in mm. Thinner areas of the proberepresent those areas of the fibre open to dialysis, whereas the thicker areas are blockedby glue.6 /\\÷1.7-4.317Assay of AChACh was assayed by HPLC-ECD in conjunction with an enzyme reactor (Damsmaet a!., 1987a; Fig. 1). ACh and Ch were separated on a reverse phase column (75x2.1mm) pretreated with lauryl sulphate. The eluent from this analytical column then passedthrough an enzyme reactor (10x2.1 mm) containing acetylcholinesterase (EC 3.1.1.7;Sigma, type VI-S) and choline oxidase (1.1.3.17; Sigma), covalently bound toglutaraldehyde-activated Lichrosorb NH2 (10 urn; Merck). The separated ACh and Chreacted to give a stoichiometric yield of hydrogen peroxide, which was electrochemicallydetected at a platinum electrode at a potential of +500 mV versus an Ag/AgC1 referenceelectrode (BAS-LC4B). The mobile phase, 0.2 M aqueous potassium phosphate bufferpH 8.0, was delivered by a pump (LKB-2150) at 0.4 ml/min. The detection limit of theassay was approximately 50 fmol/injection. The time required to complete achromatogram was 4-5 mm. Standards of ACh/Ch were injected hourly.Motor activityDuring the microdialysis experiments, a Digiscan Animal Activity Monitor (modelRXYZCM(l6); Omnitech Electronics, Inc.) was used to measure locomotor activity in 10mm blocks corresponding to the 10 mm dialysate samples. Successive interruptions ofthe same photocell beam were categorized by the monitor as “stereotypy”, and thesestereotypy counts were subtracted from the total horizontal counts to yield a measure ofambulation.Statistical analysesBiochemical data were calculated as a percent of baseline concentrations, 100%baseline being defined as the average of the last three pre-scopolamine values. Aunivariate analysis of variance (ANOVA) with repeated measures was used to comparethe effects of scopolamine and vehicle on ACh and Ch output. ANOVAs were also18conducted to evaluate the effects of the vehicle injection and exposure to dark on AChand Ch. These analyses included the three samples prior to and the three samples afterthe initiation of these two treatments. Comparisons of the biochemical responses to thethree experimental variables between brain regions were also made using ANOVAs.Greenhouse-Geisser adjustments of the degrees of freedom were made to account forthe use of time as a repeated measure. All reported values refer to the interaction effectof time with experimental treatment. Pearson’s correlation coefficients between AChconcentrations (in fmol/min) and motor counts were determined for the three treatmentconditions in each individual rat. Group correlations were carried out as well, in whichthe data of all animals having probes in the same brain region were combined foranalysis. In this case, the biochemical measures were expressed as percentage ofbaseline. Because motor counts were not normally distributed, for all the correlationanalyses these values were normalized using logarithmic transformation.(C) RESULTSThe average baseline outputs of ACh and Ch in the three brain regions are shownin Table I and have been corrected for the differences in probe surface length.The neurochemical and motor effects of vehicle and scopolamine injections, andexposure to dark, are shown in Figs. 3, 4 and 5 for the striatum, hippocampus andcortex, respectively. Vehicle injections transiently increased extracellular ACh by 54%in the hippocampus (pczO.05) and by 161% in the frontal cortex (n.s.), while no changewas evident in the striatum. No significant effect of this treatment on dialysateconcentrations of Ch was observed in any brain region. The responses of ACh and Chin the three brain regions to vehicle treatment were not significantly different. Injectionof vehicle caused short lasting increases in motor activity.19Table 1. Probe characteristics and baseline output of ACh and Ch in the striatum,hippocampus and frontal cortex of ratsBaseline Dialysate Output Probe In vitroBrain (fmol/min/mm) Surface ProbeRegion ACh Ch Lenath (mml Recovery (%‘lStriatum 9.04 ± 0.67 588 ± 130 6.4 21.4 ± 6.2Hippocampus 2.10 ± 0.48 368 ± 82 6.8 22.5 ± 5.4Frontal cortex 2.30 ± 0.56 294 ± 82 10.9 33.3 ± 7.6Dialysate outputs are expressed as the mean (± S.E.M.) of the last 3 pre-scopolamine valuesfrom 4-6 rats. Outputs are corrected for the active surface length of the probes, as shown.Neostigmine (100 nM) was included in the perfusion solution.20500 -a,U,150 (Uw -400-JW 100CI) ID0 300 50 HD00.. 200F— 00o 100-.3000 >I-0- 2000 >I—1000 00TIME (hr)Figure 3. Striatal dialysate values of ACh (closed circles) and Ch (open circles) afterinjections of vehicle and scopolamine and exposure to dark. Activity counts (open bars)were measured concurrently. Vehicle (1 mI/kg) and scopolamine (0.4 mg/kg) wereinjected subcutaneously, and “lights off” occured at the appropriate time in the rats’daily cycle. Data points represent group means (n=4-6) ± S.E.M.%ooo999STRIATUMScopolamine Lights Off0 1 2 3 4 5 6 7 8 9 10 11211600-jI 1400-z --JW 1200-C,)fXl 1000-F— 800-600-0400-• 2000-0-120(U• --100I--80 D0-I—000HIPPOCAMPUS0 1 2 3 4 6 7 8 9TIME (hr)3000 >I-2000 >F-1000 00Figure 4. Hippocampal dialysate values of ACh (closed circles) and Ch (open circles)after injections of vehicle and scopolamine and exposure to dark. Activity counts (openbars) were measured concurrently. Vehicle (1 ml/kg) and scopolamine (0.4 mg/kg) wereinjected subcutaneously, and “Lights off’ occured at the appropriate time in the rats’daily cycle. Data points represent group means (n=4) ± S.E.M.222000 -1750-W 180z =j1500w (U(0-140aiI-1000I— 100 ft.I—I— 750 Qo 60- 500 0o 0• 2503000 >0 !z2000 >I—1000 00TIME (hr)Figure 5. Frontal dialysate values of ACh (closed circles) and Ch (open circles) afterinjections of vehicle and scopolamine and exposure to dark. Activity counts (open bars)were measured concurrently. Vehicle (1 mi/kg) and scopolamine (0.4 mg/kg) wereinjected subcutaneously, and lights off” occured at the appropriate time in the rats’daily cycle. Data points represent group means (n=4-5) ± S.E.M.FRONTAL CORTEX0 1 2 3 4 5 6 7 8 9 1023Scopolamine (0.4 mg/kg, s.c.) caused large increases in interstitial ACh and inmotor activity. Peak ACh increases of 400, 1200 and 1400% in the striatum,hippocampus and cortex, respectively, occurred within 20 mm and persisted forapproximately 1 h. Compared to vehicle values, scopolamine had a significant effect onACh output (striatum p<0.0O1, hippocampus and cortex p<0.05). The drug significantlyaffected Ch only in the striatum (p.cO.0l) where there was an increase in the secondpost-injection sample (159% of baseline). The effect of scopolamine on ACh wassignificantly smaller in the striatum than in either the cortex or hippocampus (p<0.05).The drug responses in the hippocampus and cortex did not differ significantly from eachother. Motor activity was greatly increased by scopolamine with approximately the sametime course as was observed for the increase in ACh.At the onset of the dark phase of the rats’ day-night cycle, motor activityincreases coincided with ACh increases of 58% in the striatum, 169% in thehippocampus, and 77% in the cortex. In the striatum, ACh showed a significantillumination by time interaction effect (p<0.05); the increases in the other areas failed toreach significance. The three brain regions did not differ significantly with respect tothe ACh nor Ch response to dark.To illustrate the nature of the relationship between behavioural activity and AChmeasured in either the striatum, hippocampus or cortex, Fig. 6 shows the motor anddialysate profiles of three individual animals. Most of the animals tested exhibitedsignificant correlations between locomotor activity and ACh release under all threeexperimental conditions. After vehicle injection, significant correlations (p<0.05) werefound in 2 of 4 animals with striatal probes (r=0.572-0.643), 3 of 4 animals withhippocampal probes (r=0.6ll-0.808) and in 3 of 5 animals with frontal cortex probes(r=0.665-0.704). Scopolamine treatment yielded significant values in all 6 animals withstriatal probes (r=0.532-0.867), in all 4 with hippocampal probes (r=0.517-0.736) and in 4of 5 animals with frontal cortex probes (r=0.741-0.877). During lights-off, motor counts24E4oEE300E20.010..STRIATUM40302010- 7505000-50 -2500Scopr=O.330[IIPPOCAMPUSLightsr=O.603*I>-F-1000>I—05000ScopFRONTALLights CORTEXI r=O.6650-403020ID0Figure 6. Dialysate ACh concentrations (closed circles) and activity counts (open bars)from three individual rats after injections of vehicle (1 mI/kg) and scopolamine (0.4mg/kg), and exposure to dark. Insets show Pearson’s correlation coefficients of the twomeasures for the time points within each experimental treatment. * p<0.05.1500100050003 4 5 6TIME (hr)7 8 925Table 2. Correlation of regional ACh outputs and activity counts during experimentaltreatmentsBrain TreatmentRegion Vehicle Scopolamine Lights Off* *Striatum 0.256 0.599 0.300* * *Hippocampus 0.472 0.630 0.595* *Frontal cortex 0.151 0.468 0.515Pearson’s correlation coefficients of ACh outputs (as % baseline) in the striatum, hippocanipusor cortex (4-6 animals/group) and locomotor activity are shown. Analyses included the twomeasures for all time points (10-26) within each experimental treatment. * p<0.05.26correlated significantly with dialysate ACh concentrations in the striatum (4 of 5animals, r=O.677-O.844), in the hippocampus (all 4 animals, r=O.493-O.926) and in thecortex (3 of 4 animals, r=O.696-O.899).As evident in Table 2, when the data were grouped across animals, the correlationvalues became less robust. This grouping facilitates comparisons between the brainregions across the experimental conditions. The highest correlation between motoractivity and ACh output occurred in the hippocampus under the scopolamine condition.The highest value for the striatum was also after scopolamine, while in the cortex AChconcentrations correlated best with motor activity during the dark phase. The lowestcorrelations for each group were seen after vehicle treatment.(D) DISCUSSIONSource of microdialysate ACh and ChThe source of ACh recovered in the microdialysates from the three studied brainregions is anatomically distinct. The striatum contains intrinsic cholinergic interneurons(Phelps et al., 1985; Semba et a!., 1987) while the hippocampus and cortex receivecholinergic projections from the basal forebrain. Cholinergic perikarya in the basalforebrain are located among other non-cholinergic neurons in nuclei including themedial septal nucleus, the nucleus of the diagonal band of Broca, the magnocellularpreoptic nucleus, substantia innominata and the globus pallidus. An alternativesubdivision of these cholinergic neurons, independent of these nuclear groups and basedinstead on their projection patterns, has also been suggested (Mesulam et a!., 1983).Although theoretical subdivisions of basal forebrain cholinergic neurons are tenable, ithas also been demonstrated that these neurons form an anatomical continuum (Schwaber27et a!., 1987), with hippocampal and cortical cholinergic projections originating mainly inthe anterior and posterior regions, respectively, of this cholinergic basal nuclear complex.The possibility exists that extrinsic cholinergic projections from the basal forebrainare not the sole source of ACh in the rat cortex and hippocampus.Immunohistochemical studies have suggested the presence of intrinsic cholinergicneurons in these regions (see Semba and Fibiger, 1989), a finding unique to rodents.However, a sensitive in situ hybridization technique failed to detect mRNA for cholineacetyltransferase in the rat hippocampus or cortex (Butcher et at., 1992). The possibilitythat cholinergic interneurons are located in these regions thus remains controversial.With regard to this controversy, attempts have been made to estimate the percentage ofACh in cortical and hippocampal microdialysates derived from basal forebrain projectionneurons. Mechanical deafferentations or excitotoxic lesions of cholinergic projectionneurons have decreased ACh in hippocampal microdialysates by 90% (Leanza et at.,1993) and in cortical microdialysates by up to 60% (Ammassari-Teule et a!., 1993;Herrera-Marschitz et at., 1990; Rosenblad and Nilsson, 1993). Although thecompleteness of the lesions may be questioned, these results may also be interpreted tosupport the possibility that cholinergic interneurons exist in the cortex and hippocampusand contribute to microdialysate concentrations of ACh.The source and significance of microdialysate concentrations of Ch are not wellunderstood. Ch is both a precursor of ACh and a product of its breakdown. Furthercomplicating any interpretation of changes in extracellular Ch is the fact that at least80% of the Ch turnover in the brain is thought to be involved in phospholipidmetabolism (Choi et al., 1975). Additional investigations of the source of extracellularCh are required before the meaning of the experimentally manipulated concentrations ofdialysate Ch can be determined.28Basal concentrations of microdialysate ACh and ChPrevious microdialysis studies comparing baseline outputs of ACh in the frontalcortex, hippocampus and striatum have reported values that are an order of magnitudehigher than those reported here (Ajima and Kato, 1988; Toide and Arima, 1989; Wu eta!., 1988). These discrepancies are likely due to differences in probe type and brainplacement, post-operative recovery times, type and concentration of the cholinesteraseinhibitor, and composition and pH of the perfusate. The lower concentration ofcholinesterase inhibitor used in this study, and the more physiological pH and calciumconcentration of the perfusion solution are the most likely sources of theabovementioned differences in basal outputs. Indeed, the values reported here formicrodialysate ACh in the individual brain regions are comparable to those reported inexperiments using more similar methodologies (Damsma et al., 1987b; Durkinet at.,1992; Mark et a!., 1992).The rank order of baseline concentrations measured in the three brain regionsgenerally compares well with those found previously in dialysates and tissuehomogenates: for ACh, striatum >> frontal cortex hippocampus; for Ch, striatum >hippocampus > frontal cortex (Ajima and Kato, 1988; Jope, 1982; Toide and Arima,1989; Wecker and Dettbarn, 1979; Wu et at., 1988). Incontrast, a very different rankorder for basal ACh (hippocampus >> frontal cortex striatum) was found in anexperiment using a perfusate lacking an AChE inhibitor (Xu et al., 1991). However,given that this result does not agree with earlier reports from these researchers, againusing AChE inhibitor-free perfusion (Ajima and Kato, 1988),further experiments willbe necessary to determine the possibly differential effects of AChE inhibition on basaldialysate concentrations of ACh in these three brain regions.29Effects of vehicle injection on ACh releaseSubcutaneous injection of vehicle tended to increase locomotion and dialysate AChconcentrations in the hippocampus and frontal cortex, although the effect in the latterstructure failed to reach significance due to high inter-subject variability. This findingis reliably replicated, for both cortical and hippocampal ACh release, in later chapters(Chapters III, IV, V). It has also been demonstrated that ACh release is increased in thehippocampus of rats being handled (Nilsson et a!., 1990). Arousing stimuli such ashandling and injecting the animals may involve stress, which has also been shown toincrease microdialysate concentrations of ACh in the hippocampüs (Imperato et a!., 1992;Nilsson and Björklund, 1992). The finding that a control injection increases ACh releasein the hippocampus and cortex emphasizes the necessity of including control vehicleinjections when studying the effects of drugs on cholinergic activity in these regions;this finding also suggests that it may be possible to study cholinergic correlates ofbehaviour in one or both of these structures.Effects of the muscarinic receptor antagonist on ACh releaseIt has been previously demonstrated, using in vitro techniques, that muscarinicreceptor antagonists including scopolamine increase ACh release in the striatum (Dolezaland Wecker, 1990; Lefresne et al., 1978), hippocampus (Nordstrom and Bartfai, 1980;Szerb et al., 1977), and cortex (Aquilonius et al., 1972; Rospars et a!., 1977). Suchantagonists are thought to block an inhibitory feedback mechanism. This phenomenonhas also been confirmed using microdialysis (de Boer et a!., 1990c; Marien and Richard,1990; Toide, 1989; Toide and Arima, 1989; Watanabe and Shimizu, 1989). However, itmust be noted that these dialysis experiments and those reported here have been carriedout in the presence of AChE inhibition. Decreasing the degradation of ACh in thesynapse with an AChE inhibitor may be expected to increase cholinergic autoreceptoroccupancy and thereby alter the effects of autoreceptor ligands on ACh release. Indeed,30when an AChE inhibitor is not included in the perfusate, the muscarinic antagonistatropine does not increase microdialysate concentrations of ACh in the cortex or striatum(de Boer et at., 1990b; Messamore et at., 1993). In addition, the muscarinic receptoragonist oxotremorine decreases striatal ACh release in the absence, but not in thepresence, of cholinesterase inhibition (de Boer et at., 1990b) leading the authors of thisstudy to conclude that under physiological conditions, muscarinic autoreceptors are notfully occupied. However, given that microdialysis depletes neurochemicals such as AChfrom the sampled tissue (Benveniste, 1989), it is unclear whether perfusion without anAChE inhibitor more accurately reflects physiological conditions than does perfusionwith AChE inhibitors.The differences in magnitude of scopolamine’s effect on ACh release (cortexhippocampus >> striatum) may reflect regional differences in muscarinic autoreceptorlocation and/or function. In the cortex and hippocampus, the effect of the antagonist onACh release is thought to be mediated by muscarinic autoreceptors located oncholinergic terminals (Marchi et a!., 1983; Molenaar and Polak, 1980; Sethy and Hyslop,1990). In contrast, evoked ACh release in the striatum is apparently not controlled bymuscarinic terminal autoreceptors (James and Cubeddu, 1987; Marchi et at., 1983;Raiteri et at., 1984). The necessary components for the effects of muscarinic antagonistson striatal ACh are known to be intrinsic to this region because intrastriatal applicationof antimuscarinics through the dialysis probe increases dialysate ACh concentrations (deBoer et at., 1990c; Marien and Richard, 1990). The increased release of ACh measuredin the striatum after scopolamine administration (this study, Toide and Arima, 1989;Watanabe and Shimizu, 1989) may therefore be mediated by multi-neuron circuits withinthe striatum, or possibly by dendritic or somal receptors located on the cholinergicinterneurons.31Effects of the onset of dark phase on ACh releaseExposure to dark at the appropriate time in the rats’ day-night cycle tended totransiently increase both locomotion and microdialysate concentrations of ACh in allthree brain regions. Only the increase in the striatum was statistically significant due tothe high variability of the locomotor responses and the neurochemical responses in theother two brain areas. The effect in the cortex has recently been confirmed in studieswhich used the onset of darkness as a stimulus to reliably increase cortical ACh release(Moore et a!., 1992; 1993). It has also been reported that sensory stimuli other thandarkness, both tactile and non-tactile, increase hippocampal but not cortical ACh releaseas assessed using a superfusion technique (Dudar et a!., 1979). This is in contrast to thelability of cortical ACh release found here and may be due to methodologicaldifferences.In addition to a transient increase in ACh release at the onset of the dark phase,cholinergic activity is reportedly higher on average over the course of the dark phasewhen compared to the light phase, as discussed below. Wholebrain ACh concentrationsexhibit diurnal oscillations; the lowest concentrations of ACh in tissue extracts, thoughtto reflect high cholinergic activity, have been found six hours after the onset of the darkperiod (Hanin et a!., 1970). Microdialysate concentrations of ACh in the cortex(Jimenez-Capdeville and Dykes, 1993; Kametani and Kawamura, 1991) and hippocampus(Mizuno et at., 1991) have also been demonstrated to be higher during the dark phase ofthe day/night cycle. Together, these data and the results of the experiments reportedhere suggest that environmental stimuli can influence cholinergic neurotransmission inthe hippocampus and frontal cortex.32Correlation between ACh release and behavioural arousal after non-pharmacologicalstimuliThe present results indicate that changes in dialysate concentrations of ACh in thestriatum, hippocampus and frontal cortex induced by the non-pharmocological stimulicorrelate with locomotor activity, a measure of behavioural arousal. It should be notedthat such a correlation has not been reported previously for other neurotransmitters.These findings corroborate previous suggestions of a relationship between ACh releaseand arousal, defined both electroencephalographically (Casamenti et at., 1986; Celesiaand Jasper, 1966; Kanai and Szerb, 1965; Phillis and Chong, 1965; Szerb, 1967) andbehaviourally (Collier and Mitchell, 1967). ACh release-has also been reported toincrease in the hippocampus of swimming rats (Nilsson and Björklund, 1992) and in thehippocampus (Dudar et al., 1979) and cortex (Kurosawa et al., 1993) of rats walking orrunning on a treadmill. It should be noted, however, that it is uncertain to what degreespontaneous locomotion, which is thought to reflect an animal’s level of arousal, issimilar to forced locomotion, which may involve stress.Although the accumulated evidence suggests that a correlation exists betweenbehavioural arousal and cholinergic activity in the brain regions examined, the source ofthis correlation is unknown. The locomotor measure of arousal defined here can bedissociated pharmacologically from ACh release in the cortex and hippocampus: -increased ACh release can occur in the absence of increased locomotion and the reverseis also true (see Chapters III, IV, V). This suggests that neither measure is absolutelynecessary for the occurence of the other, although they do coincide in the non-drugsituations described here. Discovering whether a cause-effect relationship is responsiblefor the observed correlation between behavioural arousal and ACh release, or on whatcommon element both measures are dependent, is a challenge for future research. Toaccomplish this goal, more detailed behavioural descriptions, coupled with an in vivoACh measurement with improved temporal resolution will be required.33Correlation between AC/i release and behavioural arousal after scopolamine administrationThe positive correlations between locomotor activity counts and ACh release in thethree brain regions after scopolamine administration extend previous findings thatmuscarinic receptor antagonist-induced increases in motor activity are associated withincreases in frontal and hippocampal dialysate ACh (Toide, 1989), and with decreases inwholebrain tissue levels of ACh (Frances and Jacob, 1971).The correlation evident after application of scopolamine may be qualitativelydifferent than that discussed above, for the non-pharmacological stimuli. In drug-freeconditions, increased ACh release is associated with increased chblinergicneurotransmission; the correlation between ACh release and behavioural arousal wouldtherefore suggest a correlation between cholinergic transmission and behavioural arousal.After injection of scopolamine, however, the increased ACh release occurs as a result of,and during, muscarinic receptor blockade. The behavioural hyperactivity caused by thisanticholinergic agent thus occurs in the presence of decreased muscarinic cholinergictransmission. Therefore, while behavioural arousal correlates with cholinergictransmission after the non—pharmacological stimuli, it does not appear to be associatedwith muscarinic cholinergic transmission in scopolamine-treated rats. This apparentinconsistency may be accounted for in several ways: 1) Pre- and postsynapticmuscarinic receptors may differ such that the dose of scopolamine used here mightpreferentially bind the presynaptic autoreceptors, resulting in increased ACh release,while the postsynaptic receptors remain unblocked. This would serve to increasecholinergic transmission. This possibility can be rejected, however, as Szerb et al. (1977)have reported that scopolamine has a ten fold lower efficacy at presynaptic than atpostsynaptic muscarinic receptors. 2) While scopolamine blocks muscarinic transmission,it would at the same time increase nicotinic cholinergic transmission by increasing AChrelease. The results obtained here could be explained if the action of ACh on nicotinicreceptors was responsible for the correlation between ACh release and behavioural34arousal. 3) As discussed above, the source of the correlation after the non-pharmacological stimuli is not known, and it is possible to pharmacologically dissociatethis correlation. This additional example of a pharmacological dissociation suggests thatthe correlation does not involve a direct cause-effect relationship between ACh releaseand behavioural arousal. Thus, pharmacological agents such as scopolamine, whichincrease locomotion while decreasing muscarinic tone, may affect these two measuresindependently, perhaps by acting at different loci within the central and peripheralnervous systems.Significance of the correlation between behavioural arousal and ACh release in the cortex,hippocampus and striatumGiven the previously suggested involvement of basal forebrain cholinergic neuronsin arousal (see Semba, 1991), the responsiveness of ACh release in the frontal cortex andhippocampus to arousing stimuli and the correlation of This release with a behaviouralmeasure of arousal is not surprising. The correlation of ACh release in the s.triatum withbehavioural arousal is perhaps more unexpected. In addition to the anatomicaldifferences between cholinergic interneurons and basal forebrain cholinergic neurons, thefunctions usually ascribed to the striatum are quite different from those of thehippocampus and cortex. The striatum is part of the basal ganglia, a group of nucleitraditionally thought to be involved in initiating and modulating motor output.However, the striatum is considered the most cognitive of the basal ganglia nuclei, beingclosely connected with the cortex. Indeed, mnemonic functions of the striatum havebeen suggested (Divac, 1972; Mishkin and Petri, 1984; Packard and McGaugh, 1992).In summary, these experiments illustrate the feasibility of carrying out AChmicrodialysis experiments of long duration, coupled to behavioural monitoring, indifferent brain regions. The data reported here indicate that arousing stimuli canincrease cholinergic transmission in a regionally selective manner.35III. DOPAMINERGIC REGULATION OF CORTICAL ACETYLCHOLINE RELEASE(A) INTRODUCTIONThe extrinsic cholinergic innervation of the cortex and hippocampus emanatesfrom posterior and anterior portions, respectively, of the cholinergic basal nuclearcomplex (CBC). The CBC is an anatomical continuum of cholinergic neurons which areinterspersed among non-cholinergic neurons within several classically defined nucleiincluding the medial septal nucleus, the nucleus of the diagonal band of Broca, themagnocellular preoptic area, the substantia innominata and the globus pallidus (Schwaberet a!., 1987; Semba and Fibiger, 1989). Direct synaptic contacts made with these ChAT-containing neurons are relatively sparse and generally localized to distal dendrites(Ingham et a!., 1985). However, afferents to the CBC arise from many brain regions, asoutlined below, suggesting that hippocampally- and cortically-projecting cholinergicneurons may be weakly innervated by multiple sources.The CBC receives many inputs among which are reciprocal projections from thecortex and hippocampus, as well as ascending projections from the hypothalamus andbrainstem nuclei including the ventral tegmental area, substantia nigra, dorsal andmedian raphe, locus coeruleus, pedunculopontine and laterdorsal tegmental nuclei (Sembaand Fibiger, 1989; Woolf, 1991). The neurotransmitter contents of some of theseafferents have been identified immunohistochemically. Catecholaminergic neuronsdefined by the presence of tyrosine hydroxylase (TH) have been retrogradely labelled,from the CBC, in regions known to contain dopaminergic and noradrenergic perikaryaincluding the ventral tegmental area, substantia nigra and locus coeruleus (Jones andCuello, 1989; Semba et al., 1988). Serotonin-containing neurons localized in severalnuclei including the dorsal raphe, and ChAT-positive neurons in the mesopontinetegmentum were also retrogradely labelled in these experiments.36Ultrastructural characterizations using electron microscopy have provided evidenceof direct synaptic contacts with basal forebrain neurons. Terminals immunoreactive forTH and ChAT synapse on ChAT-labeled neurons in the CBC (Milner, 1991). Serotonincontaining terminals synapse on hippocampally-projecting basal forebrain neurons, someof which may be cholinergic (Milner and Veznedaroglu, 1993). Synaptic contacts havealso been demonstrated between GABAergic terminals and ChAT containing neurons(TOth et at., 1993; Zaborszky et al., 1986b), or presumably cholinergic cortically-projecting neurons (Ingham et a!., 1988) in the basal forebrain. GABAergic synapsesmay arise from projections of the nucleus accumbens (Walaas and Fonnum, 1979) orhippocampus (TOth et al., 1993) or from intrinsic GABAergic neurons (KOhler and -Chan-Palay, 1983; Zaborszky et al., l986a). Recent electrophysiological studies ofidentified cholinergic neurons in the basal forebrain of guinea pig brain slices alsoprovide evidence of GABAergic, noradrenergic, as well as histami-nergic effects on thefiring rates of CBC neurons (Fort et at., 1993; Khateb et a!., 1993; Pegna et a!., 1993).Presumptive glutamate/aspartate projections to the basal forebrain from manybrain regions including the amygdala, thalamus and cortex have also been described(Carnes et at., 1990), although neurons using excitatory amino acid transmitters aredifficult to characterize anatomically. Evidence supporting a glutamatergic innervationof the CBC includes the finding that 1-glutamate binding sites represent a large portionof the total binding sites autoradiographically identified in the basal forebrain of the rat(Zilles et at., 1991) and that the AMPA subtype of glutamate receptor has beenimmunocytochemically localized to cholinergic basal forebrain neurons in the monkey(Martin et al., 1993). In addition, Rasmusson et at. (1994) have recently reported thatincreases in cortical ACh release caused by brainstem stimulation in anaesthetized ratscan be blocked by application of a glutamate antagonist into the basal forebrain. Withregard to the abovementioned proposition that the presence of appropriate receptorsfurther supports the regulatory role of a transmitter within a certain brain region, it37should be noted that receptors for ACh, serotonin, GABA, dopamine (DA) andnoradrenaline have also been autoradiographically identified in the basal forebrain (Zilleset at., 1991). Evidence of whether some of these receptors may be situated oncholinergic neurons awaits further ultrastructural studies.In addition to the classical neurotransmitters, a variety of neuropeptide transmittersand growth factors may regulate CBC neurons. There is evidence for interactionsbetween basal forebrain neurons, in some cases identified as cholinergic, and terminalscontaining substance P. enkephalin, somatostatin, neuropepetide Y, neurotensin orvasopressin (reviewed by Záborszky et at., 1991). In addition, receptors forneurotrophins and estrogen have been located in CBC neurons (Toran-Allerand et at.,1992).Considered together, the above anatomical data suggest that the activity of CBCneurons may be regulated by a great variety of neurotransmitters. The regulation ofthese cholinergic neurons, specifically those projecting to the cortex, by theneurotransmitter DA has been investigated in the experiments discussed in this chapter.To this end, in viva dialysate concentrations of ACh from the frontal cortex ofconscious, freely moving rats were measured after systemic administration of thepsychomotor stimulant d-amphetamine (AMPH) or the DA receptor agonistapomorphine, and after local application of AMPH through the dialysis probe. Todetermine which DA receptor subtype(s) may mediate the dopaminergic regulation ofcortically-projecting cholinergic neurons, the effects of selective D1-like and D2-likereceptor agonists on cortical ACh release were examined. In addition, the extent towhich selective D1-like and D2-like receptor antagonists attenuate AMPH- andapomorphine-induced effects on cortical ACh release was determined. Furtherreferences to “D1” and “D2 should be interpreted as “D1—like” and “D2-like” respectively(see Discussion).38(B) MATERIALS AND METHODSExperimental protocol and drugsThe size and strain of subject rats were as in Chapter II. Dialysis experimentswere conducted as described in Chapter II and involved the subcutaneous (s.c.) injectionof the following drugs, alone or in combination: d-amphetamine sulfate (AMPH, 2.0mg/kg; BDH), apomorphine hydrochloride (1.0 mg/kg; Sigma), CY 208-243 (1.0 mg/kgcontaining 0.4% acetic acid; Sandoz), quinpirole hydrochloride (0.2 or 0.5 mg/kg; RB!),(+)-4-propyl-9-hydroxynaphthoxazine (PHNO, 0.05 mg/kg; Merök, Sharp and Dohme),haloperidol base (0.15 mg/kg; McNeil Pharmaceutical), raclopride (1.0 mg/kg; Astra),and SCH 23390 (0.3 mg/kg; Schering Plough). The drugs were dissolved in distilledwater, and injected in a volume of 1 mI/kg. AMPH (10 uM) was also administereddirectly through the dialysis probe by its addition to- the perfusion solution. The dosesused in this study have significant effects on behaviour (Christensen et a!., 1984;Markstein et a!., 1988; Martin et a!., 1984; Ogren et a!., 1986) and/or have reliableneurochemical effects on dopaminergic systems (Imperato and DiChiara, 1985; Kuczenskiand Segal, 1989; Sharp et at., 1987) or on striatal cholinergic neurons (Bertorelli andConsolo, 1990; Damsma et at., 1990; 1991; Robertson et al., 1993; Westerink et at., 1990).Surgery and microdialysisTransverse dialysis probes were stereotaxically implanted into the frontal cortex ofrats as described in Chapter II. Microdialysis was carried out as previously describedwith the following exception: the probe outlet was connected to the sample valve by alength of polyethylene tubing, rather than fused silica tubing, containing 50 ul.Assay of AChACh was assayed by HPLC-ECD as described in Chapter II.39Statistical analysesBiochemical data were calculated as a percent of baseline concentrations, 100%baseline being defined as the average of the last three pre-drug values. UnivariateANOVAs with repeated measures were used to compare the effects of the differentdrugs on dialysate ACh concentrations. Reported values refer to the main group effectof the experimental treatment, unless otherwise noted as referring to the interactioneffect of time with experimental treatment. In the latter case, Huynh-Feldt adjustmentsof degrees of freedom were made to account for the use of time as a repeated measure.For the local application of AMPH, the last baseline sample and the six samples duringAMPH perfusion were used to test for an effect of the drug over time.(C) RESULTSThe average baseline output of ACh ( S.E.M.) from the frontal cortex in all ofthe animals was 46.66 ± 6.69 fmol/min (n=88). Systemically administered AMPH (2.0mg/kg) increased ACh output to 280% of baseline values within I h and the increaselasted for approximately 3 h. This was statistically significant [F(l,lO)=46.98, p 60 000 Dalton; FiltralAN69, Hospal) and had an active surface length of 10.9 mm in the DNB lesioned andcontrol groups or 5.45 mm in the MDS lesioned and control groups. In the latter groups,only the half of the probe situated in the cortical hemisphere ipsilateral to the lesion wasleft unblocked and thus active for dialysis.To prevent twisting of inlet and outlet tubing due to excessive turning in the MDSlesioned rats treated with AMPH, the tubings of these rats and their controls wereattached to a liquid swivel which was anchored to a “vest” worn by the rat.61In vitro recovery of probesUnused bilateral cortical microdialysis probes were perfused with the neostigminecontaining perfusion solution while immersed in a standard solution of ACh/Chmaintained at 37°C. The recovery of ACh across the membrane was calculated as apercentage of the standard solution.Assay of AChACh was assayed by HPLC-ECD as described in Chapter II, with the followingadditions to the technical equipment: an HPLC pump (Shimadzu) and a platinumelectrode/reference electrode assembly (Antec).Tissue level analysisTo determine the efficacy of the neurotoxic lesions, representative dopaminergicterminal regions of the MDS (striatum and nucleus accumbens (NAc)) and representativenoradrenergic terminal regions of the DNB (frontal cortex (FCTX), parietal cortex(PCTX) and hippocampus) were assayed for biogenic amine content. Rats were killedby cervical dislocation and the brains were rapidly removed. To assess DA depletion inthe MDS lesioned rats, 2 coronal sections of forebrain were cut with a freezingmicrotome, spanning the following anterior coordinates (in mm) measured from bregma(from the atlas of Paxinos and Watson, 1986): 1.2-2.2 for NAc and 0.2-1.2 for striatum.To assess NA depletion in the DNB lesioned rats, 2 coronal sections were cut using aMcllwain tissue chopper, spanning 1.2-5.2 for the FCTX and (-l.8)-(+0.2) for thePCTX. From the remaining caudal portion of brain, the hippocampi were handdissected. The brain sections were placed on ice-cold, dampened filter paper where DAand NA terminal regions were hand-dissected from both hemispheres with the aid ofbinocular magnifiers. This procedure yielded unilateral tissue samples having meanweights of 9.6 mg for NAc, 13.2 mg for striatum, 26.4 mg for FCTX, 18.6 mg for62PCTX and 17.0 mg for hippocampus. Tissue samples were sonicated in ice-cold 0.2 MHC1O4 containing 0.15% Na2SO3and 0.05% Na2EDTA (NAc, FCTX and PCTX in 300ul, striatum in 400 ul, hippocampus in 500 ul). After centrifugation, samples were storedat -80°C until the time of thçir assay. Thawed samples were injected from a 100 ulloop on a valve (Valco) onto a reverse phase analytical column (nucleosil 5 urn C18;Phenomenex). The mobile phase consisted of 0.1 M sodium acetate buffer, 0.01 mMNa2EDTA, 0.75-0.85 mM octane sulfonic acid, 10-12% methanol, pH 3.6 and wasdelivered at 0.35-0.8 mi/mm by a pump (PM-48, BAS or Shimadzu LC-600). Soluteswere quantified by electrochemical detection using an LC-4B amperometric detector(BAS) with a glassy carbon working electrode at +0.7 V vs. a Ag/AgCl referenceelectrode. The detection limit for all compounds under these conditions was between 10and 100 fmol/injection.Retrograde labellingUnilateral cortical probes, in MDS lesioned and control rats which could not beused for dialysis experiments due to technical problems (n=3), were filled with a 4%solution of Fluoro-Gold (Fluorochrome, Inc.) in 0.9% saline, and flushed with saline 12h later. One week later these animals were sacrificed and coronal forebrain slices werecollected, processed and examined under ultraviolet light as described by Robertson eta!. (1991).Statistical analysesMicrodialysate outputs in each animal were calculated as a percent of baseline,100% baseline being defined as the average of the last six pre-drug concentrations(fmol/min). As an alternate method of correcting for inter-animal variability in baselineoutput, data from individual animals were also transformed into “delta fmol/min” bysubtracting the average of the last six pre-drug concentrations (fmol/min) from the63absolute concentration (fmol/min) of each dialysate sample. Univariate ANOVAs wereused to compare basal microdialysate outputs (fmol/min) of cortical ACh in lesion vs.control groups for the two lesion types. ANOVAs with repeated measures were used tocompare treatment-induced changes in cortical microdialysate outputs of ACh (in both %baseline and delta fmol/min scores) in surgical control vs. 6-OHDA lesioned rats, andthe main treatment effect is reported. Treatment x time interaction effects are alsoreported which include Huynh-Feldt adjustments of the degrees of freedom to accountfor the use of time as a repeated measure. One-way analyses of variance with Tukey“highly significant differences” post-hoc comparisons were used to assess differences inlesion-induced tissue concentrations of monoamines.(C) RESULTSEfficacy of lesionsTable 3 shows the effects of 6-OHDA lesions of the MDS on tissue concentrationsof DA, NA and 5-HT. DA was significantly depleted in the striatum and NAcipsilateral to the lesioned MDS, and decreased to a lesser extent in the contralateral NAc.NA concentrations were also significantly reduced ipsilaterally by these lesions. Tissueconcentrations of 5-HT were not affected in any region.The results of 6-OHDA DNB lesions on regional tissue concentrations of DA, NAand 5-HT are summarized in Table 4. NA was significantly depleted in all 3 regions,leaving tissue concentrations of DA and 5-HT unaffected.Site of neurons projecting to unilateral cortical probeApplication of Fluoro-Gold through the unilateral cortical probe retrogradelylabeled cell bodies in, among other areas, the basal forebrain (data not shown). While64Table 3. Effects of 6-OHDA lesions of the mesotelencephalic dopamine system on regionaltissue concentrations of dopamine, noradrenaline, and serotoninTissue concentrationpmol/mg tissue ± S.E.M. (% surgical control)Region Surgical Control Lesion Controla LesionDOPAMINEN. accumbens 39.1±2.8 27.9±1.5 (71.4%)# 0.19±0.06 (0.49%)*Striatum 57.6±3.3 57.6±2.5 (100.0%) 0.06±0.02 (0.10%)*NORADRENALINEN. accumbens 0.96±0.09 0.71±0.08 (74.0%) 0.36±0.07(37•5%)**Striatum 0.48±0.05 0.62±0.07 (129.2%) 0.11±0.05 (22.9%)SEROTONINN. accumbens 2.68±0.26 2.24±0.15 (83.7%) 2.59±0.26 (96.8%)Striatum 1.23±0.08 1.08+0.06 (87.6%) 1.04+0.13 (84.3%)Tissue was assayed from 6 surgical control rats and 10 6—OHDA lesioned rats.Lesion control samples are from the hemisphere contralateral to the lesion.Values are significantly different from both the surgical control and the lesion control, p