UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Involvement of dopamine in feeding behaviours Blackburn, James Robert 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1985_A8 B53.pdf [ 5.62MB ]
JSON: 831-1.0096423.json
JSON-LD: 831-1.0096423-ld.json
RDF/XML (Pretty): 831-1.0096423-rdf.xml
RDF/JSON: 831-1.0096423-rdf.json
Turtle: 831-1.0096423-turtle.txt
N-Triples: 831-1.0096423-rdf-ntriples.txt
Original Record: 831-1.0096423-source.json
Full Text

Full Text

INVOLVEMENT OF DOPAMINE IN FEEDING BEHAVIOURS by JAMES ROBERT BLACKBURN B.Sc, McGill Universtity, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1985 • James Robert Blackburn, 1985 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. J . R. Blackburn Department of Psychology The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 D a t e October 11, 1985 DE-6(3/81) ABSTRACT This study investigated the involvement of the neurotransmitter dopamine in feeding behaviours. A conditioned feeding paradigm was used to study incentive responses. After conditioning rats responded to a conditional stimulus (CS+) by approaching a feeding site. Approach responses were attenuated by 0.4 or 0.6mg/kg of the dopamine antagonist pimozide. Neurochemical investigation revealed that exposure to the CS+ increased dopamine turnover in the forebrain. Thus, dopamine appears to be actively involved in the initiation of appetitive responses. In contrast, another experiment indicated that consumption of a liquid diet was not altered by up to 0.6mg/kg pimozide. These data were interpreted as supporting an "incentive-response hypothesis" of dopamine function, which states that "When an animal observes an incentive stimulus, the release of dopamine in the forebrain is increased, resulting in approach to the stimulus by the animal. Once the animal is in contact with a goal object, consummatory reactions occur which are not mediated by dopamine systems". A final experiment investigated the activity of dopamine systems following ingestion. After one hour during which food pellets or liquid diet were available to rats, dopamine turnover was increased in the n. accumbens and the striatum, relative to non-fed animals. No increase was observed in the brains of rats which had consumed similar quantities of saccharin solution. Thus, the increase observed following consumption of pellets or liquid diet could not be attributed to motor or "reward" effects. It was concluded that in addition to their involvement in incentive-responding, dopamine systems are also affected by the ingestion of nutrients. Dr. A.G. Phillips ' i i TABLE OF CONTENTS Abstract i i List of Tables v List of Figures vi Acknowledgement vii i Introduction 1 Reduction of feeding by dopamine lesions 5 Pharmacological blockade of dopamine receptors 8 Feeding induced by dopaminergic stimulation 10 Correlatons between feeding and dopamine release 15 Motivational interpretations of dopamine's role in feeding 28 Experiment 1 41 Methods 42 Results 45 Discussion 63 Experiment 2 65 Methods 66 Results 67 Discussion 73 Experiment 3 76 Methods 76 Results 77 Discussion 77 Experiment 4 81 Methods 82 i i i Results Discussion General Discussion, Conclusions References LIST OF TABLES Table I. Concentrations of dopamine, DOPAC and HVA in various brain regions of rats sacrificed before or after exposure to the CS+ 72 Table II. Concentrations of dopamine, DOPAC and HVA in various brain regions of rats exposed to various feeding conditions 89 v LIST OF FIGURES Figure 1: Cumulative time spent in niche during CS+ period by animals receiving 0.6% tartaric acid vehicle on test day 47 Figure 2: Cumulative time spent in niche during CS+ period by animals receiving 0.2mg/kg pimozide on test day 49 Figure 3: Cumulative time spent in niche during CS+ period by animals receiving 0.4mg/kg pimozide on test day 51 Figure 4: Cumulative time spent in niche during CS+ period by animals receiving 0.6mg/kg pimozide on test day 53 Figure 5: Mean latency to enter niche following CS+ onset 55 Figure 6: Mean number of nosepokes during CS+ period prior to onset of food delivery 57 Figure 7: Mean area under the cumulative response curve during CS+ prior to onset of food delivery 59 Figure 8: Mean number of entries in to feeding niche during CS+ period, following first entry but prior to onset of food delivery 61 vi Figure 9: DOPAC/DA ratios, in each of the three brain regions analyzed, after the rats had been exposed to the CS+ for four minutes 69 Figure 10: HVA/DA ratios, in each of the three brain regions analyzed, after the rats had been exposed to the CS+ for four minutes 71 Figure 11: Consumption of liquid diet in a twenty minute session four hours after the injection of pimozide or vehicle 79 Figure 12: DOPAC/DA ratios, in each of the three brain regions analyzed, one hour after the onset of various feeding treatments 86 Figure 13: HVA/DA ratios, in each of the three brain regions analyzed, one hour after the onset of various feeding treatments 88 vii ACKNOWLEDGEMENTS Many people have given their kind assistance to me in preparing this work. Fritz Le Paine devoted much effort to setting up the conditioned feeding apparatus and in writing the software for it's control. He also performed the dissections for the neurochemistry analyses. Lisa Wong was most helpful in her work on the feeding experiments. Neurochemical analyses were performed by Alex Jakubovic and Davina Lin. My advisor, Tony Phillips, has been very supportive throughout the entire study. I would like to extend special thanks to Jake and Linda for their inspiration and encouragement. vi i i INTRODUCTION An adequate neurological analysis of behaviour must begin with an adequate description of the behaviour to be analyzed. Although this is recognized in theory, in practice it is often forgotten or laid aside for the sake of experimental expediency. Thus, much of the literature on the physiological psychology of feeding suffers from a failure to recognize this fundamental principle. It is simply not possible to determine the neurological substrate of "feeding" if "feeding" is variously defined as operating the mouth and tongue in certain ways, ingesting a quantity of food, gaining weight, contacting a food-related cue or performing an operant response for food reward. Clearly such events are distinguishable at a behavioural level, and will often refer to different underlying neurological events. In order to relate these neurobiological events to behaviour it is neccessary to be sure which behaviours they are being related to. One strategy available in describing feeding is to fractionate feeding behaviour into a series of separate stages. Although some stages may eventually collapse or meld into each other, their tentative identification at least provides more managable units to which neurological enquiry can be directed. A first step in establishing a taxonomy of "feeding behaviours" is to recognize two broad categories: Preparatory (or "appetitive") behaviours, and consummatory behaviours (Sherrington, 1906; Craig, 1918). Woodworth defined these in his book Dynamic Psychology. A consummatory reaction is one of direct value to the animal - one directly bringing satisfaction - such as eating or escaping from danger. The objective mark of a consum-matory reaction is that it terminates a series of acts, and is followed by rest or perhaps a shift to some new series... The preparatory reactions are only mediately of benefit to the organism, and their value lies in the fact that they - 1 -lead to, and make possible, a consummatory reaction. Objectively, the mark of a preparatory reaction is that it occurs as a preliminary stage in a series of acts leading up to a consummatory reaction. (Woodworth, 1918, p. 40) In addition to these two categories, which result in the ingestion of food, it is also necessary to recognize an additional two types of "feeding behaviour". The first of these is the cessation of feeding, or the commencement of satiety. The second is the post-prandial period or intermeal interval, during which the animal is not feeding, and is not engaging in appetitive, feeding-related behaviours. These four catagories will now be examined in more detail. Appetitive reactions are performed in the absence of food. These can be divided into three classes: general activity, approach, and procurement. These distinctions serve a heuristic function, and are logically distinct even if they are not ethologically or neurologically valid. The first of these, general activity, is characterized by increased levels of locomotion and exploration. Because activity levels are higher with increasing time since the animal last fed, some relationship beween feeding and activity may be inferred, even if not all activity is actually involved in feeding. Approach behaviour can be defined either as taxis towards food or food related cues (Schneirla, 1959) or as increased time spent in their vicinity. For the rat, examples of approach reactions include directed sniffing of food odours, orienting to a light that signals imminent food reward (Holland 1977) or running down an alley toward food. The third class of appetitive behaviour, procurement, is defined as the performance of responses which result in the availability of food. An operant response, such as a lever press, which occurs more often when it is reliably followed by the presentation of a food pellet, will serve as an example of a procurement reaction. Obviously, running down an alley for food may fall into this category as well as into "approach". Rather than - 2 -discarding the distinction between these classes, it should simply be accepted that some responses may be considered as belonging to one, the other, or both classes, depending on the context in which they occur. Consummatory feeding reactions are defined as those that occur after the animal has made contact with the food and that result in it's ingestion. Because is does not immediately lead to ingestion, hoarding behaviour is tentatively identified as a naturally-occuring procurement response, even though the animal comes into direct contact with the food. Biting, lapping, chewing and swallowing are clear examples of consummatory responses. The handling of food as the animal eats it shall also be considered to be a member of this class. Typically, consummatory behaviours occur in "bouts": Brief bursts of constant contact with the food (handling or eating) punctuated by other behaviours not directly tied to feeding, such as grooming. This classification of feeding behaviours is still incomplete. As the consummatory reaction advances, the animal has consumed more and more food. At some point, the consumption must cease. Often neglected, the problem of satiety is as central to an analysis of feeding as are the problems of appetite and consumption (Smith and Gibbs, 1979). In addition to technical issues, such as the question of what physiological or psychological events bring feeding to a halt, satiety raises conceptual issues. For example, should satiety be considered simply as the end of feeding, or as the beginning of the post-prandial period? Both perspectives have some merit: At some point the food is no longer a sufficent stimulus to maintain feeding, while other stimuli beckon the animal to move on. Finally, the classification of feeding behaviour should include a description of the inter-meal period. After an animal has moved away from it's food, it is - 3 -generally unresponsive to food and food-related stimuli. This unresponsiveness is not complete: if the animal is presented with fresh food, particularly a different food, the animal may begin eating again. Still, the animal is not as responsive to food in the inter-meal period as after a day of food deprivation. Historically, such unresponsiveness was interpreted as reflecting an absence of hunger, but logically it may equally well reflect the presence of active inhibition. For example, the terminal sensillae of blowflies are inhibited by a diuretic hormone, released by the osmotic actions of food, for an hour or more following the ingestion of a large meal. Reducing sensory input appears to reduce the probability of further feeding during this period (Bernays and Simpson, 1982). This analysis provides only a crude outline of the intricacies of the behavioural regulation of feeding. It does, however, provide a somewhat more detailed account than is commonly employed as a basis for investigations of the neurobiological substrates of feeding. As such it may provide a framework for a modest advance in our understanding of these substrates. This thesis will begin by examining the experimental evidence accumulated to date concerning the involvement of the neurotransmitter dopamine in the various stages of feeding behaviour outlined above. Dopamine is believed to play a central role in feeding, yet at present there are few clearly articulated views regarding the nature of this role, and little consensus between these views. Consequentally, several theories of dopamine function shall then be critically examined in the light of the evidence reviewed. As well, additional experimental evidence relevant to the evaluation of these theories shall be presented and suggestions offered for their further development. - 4 -REDUCTION OF FEEDING BY DOPAMINE LESIONS Lesions of the lateral hypothalamus (LH) result in severely reduced food intake (Anand and Brobeck, 1951). Complete aphagia is often observed after such lesions: Animals may refuse food even when it is placed in their mouths. The animals are also completely adipsic, refusing to drink water. If left on their own, such animals usually die. However, if kept alive through gastric intubation of food they often gradually recover much of their normal behaviour (Teitlebaum and Stellar, 1954). However, even after "recovery", the animal does not display fully normal feeding. For example, i t will only drink water at meal times. "Recovered" animals are exceptionally "finicky" in their food selection: Highly palatable foods are consumed in considerably greater quantities than standard chow, while quinine-adulterated food is strongly avoided (Teitlebaum and Epstein 1962). Stellar (1954), in his classic dual-hypothalamic model of motivated behaviours, attributed these deficits to damage of the brain's "feeding centre", said to be located in the LH. However, later work indicated that much of the damage observed could be attributed to the destruction of fibres coursing through the LH (eg., Morgane, 1961a,b). Among the fibre tracts damaged by LH lesions are several ascending catecholamine (CA) systems. These include the dorsal and ventral noradrenaline (NA) bundles, plus the so-called "mesolimbic" and nigrostriatal dopamine (DA) bundles. The "mesolimbic" dopamine system has its cell bodies in the ventral tegmental area (VTA), and projects to many di- and telencephalic sites. The heaviest innervation is of the n. accumbens, while other projection sites include amygdala, olfactory tubercle, prefrontal cortex, septum and globus pallidus. The nigrostriatal bundle (NSB) originates in the pars compacta of the substantia nigra (SN) and projects to the neostriatum. Both systems have been extensively described by Fallon and Moore (1978). - 5 -A link between the LH syndrome and destruction of ascending catecholamine systems was first indicated by Heller and Harvey (1963) and Moore and Heller (1967), who noted that lesions of the LH resulting in aphagia reduce brain levels of noradrenaline. Oltmans and Harvey (1972) found that the extent of behavioual disruption produced by lesions of the NSB at the level of the LH was highly correlated with the extent of the striatal catecholamine depletion. Ungerstedt (1971b) employed the catecholamine-selective neurotoxin 6-hydroxydopamine (6-OHDA) to investigate the possibility that aspects of the syndrome associated with LH lesions could be attributed to loss of forebrain catecholamine innervation. Injections of 6-OHDA into the substantia nigra produced aphagia similar to that seen following electrolytic lesions of the LH. Such deficits were not observed following 6-OHDA lesions of ascending noradrenaline fibres. The case for attributing LH lesion-induced deficits was strengthened by Zigmond and Strieker (1972; Strieker and Zigmond, 1974). They found that pretreatment of animals with the monoamine oxidase inhibitor pargyline, or pretreatment with both pargyline and desmethylimipramine (DMI), which resulted in 95-99% dopamine depletions while lowering noradrenaline levels by less than 5%, were highly effective in producing prolonged aphagia. Recovery of function, in parallel with that seen following LH lesions, was observed by Fibiger, Zis and McGeer (1973), and was attributed to compensatory increases in the nigrostriatal system, such as post-synaptic receptor supersensitivity (Strieker and Zigmond, 1974). Similar lesions have been observed to attenuate not only consummatory, but also approach responding, which may be viewed as a form of appetitive feeding behaviour. Fibiger, Phillips and Zis (1974) found that injection of 6-OHDA into the substantia nigra completely prevented acquisition of a goal box approach response, even though the animals were no longer aphagic. - 6 -Lesions of the nigrostriatal dopamine projection consistently produce aphagia (eg., Oltmans and Harvey, 1972; Fibiger, Zis and McGeer, 1973; Marshall, Richardson and Teitlebaum, 1974; see Strieker and Zigmond, 1976 for review). On the other hand, lesions of the mesolimbic dopamine system do not result in severe disruption of food intake. Although Ungerstedt (1971b) found that injections of 6-OHDA into the VTA, the origin of this projection, did result in aphagia, the deficits induced by this treatment may have been caused by incidental NSB damage, as almost no dopamine terminals could later be observed in the neostriatum. Ungerstedt only observed transient deficits when he lesioned the pathway more anteriorly, despite the almost complete disappearance of dopamine from the n. accumbens and olfactory tubercle. 6-OHDA lesions of mesolimbic terminal sites, specifically the n. accumbens and olfactory tubercle, have actually been reported to increase feeding in 30 minute sessions (Koob, Riley, Smith and Robbins, 1978). In general, no significant decreases in ad lib feeding have been found following comparable lesions (eg., Le Moal, Stinus, Simon, Tassin, Thierry, Blanc, Glowinski and Cardo, 1977; Koob et al, 1978; Kelley and Stinus, 1985). Curiously, such lesions appear to result in sloppier feeding: The animals spill more food through the floors of their cages (Le Moal et al, 1977) and leave more partially eaten pellets (Kelley and Stinus, 1985). In contrast to the lack of effect 6-OHDA injections into the VTA or n. accumbens have on free feeding, they dramatically reduce the hoarding of food by hungry rats (Le Moal et al, 1977; Kelley and Stinus, 1985). Thus although nigrostriatal lesions disrupt both appetitive and consummatory feeding behaviours mesolimbic lesions appear to preferentially disrupt food procurement. - 7 -PHARMACOLOGICAL BLOCKADE OF DOPAMINE RECEPTORS If the nigrostriatal dopamine pathway is a necessary substrate of normal feeding behaviour, pharmacological blockade of dopamine transmission would also be expected to disrupt feeding. Compared to the consistent and profound effects of 6-OHDA lesions on feeding, the evidence on this point is suprisingly weak and inconsistent. However, some reports have noted attenuated feeding following administration of these drugs. For example, Heffner, Zigmond and Strieker (1977) found decreased food intake over l h when rats which had been deprived of food overnight were treated with the dopamine antagonist spiroperidol. Blundell and Latham (1978) provide interesting perspectives on the nature of the feeding deficit produced by pimozide, a relatively specific dopamine antagonist (Jansenn, Niemegeers, Schellekens, Breese, Lenaerts, Pinchard, Schgre, van Neuten and Verbrugger, 1968). Rats deprived of food for 16h were allowed access to food pellets 2h after the injection of 0.45mg/kg pimozide. Over the course of the next l h period food intake by these animals was 35% less than that of controls. Mean latency to begin feeding was not significantly reduced: It remained less than one minute. Feeding rate, while the animals were feeding, was reduced by half, although the number of feeding bouts remained constant. The duration of bouts was actually increased. Thus, the animals took longer to eat less. Tombaugh, Tombaugh and Anisman (1979) observed different effects following injection of l.Omg/kg pimozide. Animals were presented with five 45mg Noyes pellets in their home cages. Latency to begin feeding was increased seven-fold by pimozide treatment, but the time required to consume the pellets, once feeding began, was not increased. Tombaugh et al also found that the time required to finish consuming all the pellets after feeding had begun was - 8 -unaffected by pimozide. Again, the primary deficit in feeding was one involving food procurement, rather than one of food consumption. Note, however, that the consumption time referred to here applies to a restricted meal of five 45mg pellets. In contrast, Blundell and Latham (1978) observed slower feeding over several bouts in a one hour test. More recently, Wise and Colle (1984) offered rats five Noyes pellets every 36 seconds and recorded both latency to begin feeding and the total time to consume all pellets. Mean latencies and durations were increased by l.Omg/kg pimozide. A general deficit in ability to respond rapidly was not observed: More pimozide treated rats exhibited short latencies to begin feeding than did controls. It is difficult to integrate these studies due to their pronounced methodological differences. Their only point of agreement is the net disruption of feeding by dopamine blockade, through increased latency to begin feeding or a decreased feeding rate. However, other studies have failed to confirm even this. Indeed, Lawson, Byrd and Reed (1984) found that intermediate doses of pimozide, as well as two other neuroleptics (chlorpromazine and trifluperazine) all increased the intake of milk, although high doses of each decreased intake. K.B.J. Franklin (personal communication) has similarly found that 0.5 mg/kg pimozide can increase the intake of mash or pellets in animals which are 6 or 24 hours food deprived. In summary, it appears that profound destruction of dopaminergic systems, as produced by appropriate administration of 6-OHDA, results in severe disruption of feeding. However, milder interference with dopaminergic transmission, as produced by dopamine receptor blockade, has at most a mild suppresant effect on food intake. Although the effects of neuroleptics on food intake may be inconsistent, their effects on other feeding-related activities is clearer. Wise, Spindler, deWit - 9 -and Gerber (1978), Tombaugh et al (1979) and Tombaugh, Anisman and Tombaugh (1980) all observed marked reductions in lever pressing for food reward following injection of 0.5 or l.Omg/kg pimozide. Acquisition of lever pressing was slowed by injection of as little as 0.25mg/kg (Wise and Schwartz, 1981). Hoarding behaviour, known to be disrupted by lesions of the mesolimbic dopamine system, was greatly suppressed by 0.45mg/kg pimozide, although latency to begin was not affected (Blundell, Strupp and Latham, 1977). Again, food procuring behaviours appear especially sensitive to disruption of dopamine systems. This point shall be addressed at more length in a later section. FEEDING INDUCED BY DOPAMINERGIC STIMULATION If reduced dopaminergic activity generally produces decreases in feeding and feeding related behaviours, it might be expected that enhanced dopamine activity produces increases in these behaviours. Increases have been reported, but only in some situations. This section shall examine the effects of direct and indirect stimulation of dopaminergic systems by pharmacological, electrical and other means. Pharmacological Stimulation: Dopamine release is enhanced by systemic injection of d-amphetamine (Fuxe and Ugerstedt, 1970; McMillen, 1983), as is the release of noradrenaline (Glowinski and Axelrod, 1965). Post-synaptic dopamine receptors are directly stimulated by apomorphine (Ernst, 1967). Thus, both amphetamine and apomorphine result in increased activity at post-synaptic dopamine receptors. However, instead of resulting in increased food intake, both of these agents produce marked anorexia (Cole, 1978; Heffner et al, 1977). Amphetamine induced anorexia is partially attenuated by LH lesions, and Ahlskog and Hoebel (1973) have shown that at this site the the anorectic effect of - 10 -amphetamine is largely mediated by noradrenaline. Still, noradrenaline is not responsible for all of the anorectic effect: Fibiger et al (1973) reported that animals, largely recovered from the aphagia produced by 6-OHDA lesions of the NSB, showed relatively little amphetamine-induced anorexia. Further evidence for dopaminergic involvement in amphetamine and apomorphine-induced anorexia includes reports that such anorexias are attenuated by neuroleptics (Heffner et al, 1977; Burridge and Blundell, 1979). Although amphetamine and apomorphine are usually considered as effective anorectic agents, low doses of each have sometimes been reported to increase food intake. Holtzman (1977) found that 0.3mg/kg amphetamine increased food intake over 2h, while Winn, Williams and Herberg (1982) found that 0.25mg/kg increased intake over 3h. Blundell and Latham (1978) found that a single dose of 0.125mg/kg increased intake over the next 24h period. Interestingly, Eichler and Antelman (1977) found that some of the same doses which decreased food intake in hungry rats (from over 5.0 to less than l.Og) could increase food intake (up to 1.2 from 0.25g) in rats "pre-sated" by exposure to wet mash. Note that the intake reported by Eichler and Antelman after low doses of apomorphine was nearly equal to the intake observed after high doses, and the apparantly "opposed" effects are attributable to different baselines. Rather than making the animal eat, apomorphine may make the animal less concerned with his internal state, as opposed to external cues. We shall return to this possibility later. It also seems reasonable to suggest that amphetamine and apomorphine have opposed anorectic and food-intake promoting effect by acting at multiple sites in the brain. This hypothesis has been supported by studies in which dopaminergic drugs have been micro-injected into specific dopamine terminal sites. Winn et al - 11 -(1982) observed increased food intake following unilateral injection of 2.0ug amphetamine into the striatum, and Carr (1984) observed decreased food intake when amphetamine was injected bilaterally into the n. accumbens or the amygdala. Decreased feeding following injection into the n. accumbens may be attributable to the general increases in activity produced by the action of amphetamine at this site (Kelly, 1977; Makanjuola, Dow and Ashcroft, 1980; Carr, 1984): As with low doses of apomorphine, the animal may be relatively indifferent in what it responds to. Another treatment that induces feeding deserves mention here. This is the injection of cholinergic substances into the substantia nigra. Winn and Redgrave (1979) found that injection of acetylcholine into the substantia nigra of non-deprived rats increased food intake to more than four times that of control levels. Winn, Farrell, Maconick and Robbins (1983) repeated the experiment using the cholinergic agonist carbachol and found similar increases in feeding. This enhancement appears to be specific to feeding: Even in the absence of food, comparable injections did not increase gnawing, drinking, locomotion, grooming, sniffing or rearing. This provides strong evidence for a cholinergic mechanism in the substantia nigra that is selectively related to food intake. Although it is tempting to suppose that these behaviours are mediated by the dopaminergic nigrostriatal bundle, there is little direct evidence to support this claim at present. Taha and Redgrave (1980) administered 0.5mg/kg haloperidol to rats prior to injecting carbachol into the substantia nigra and found that food intake was reduced to zero - well below the baseline observed with no injection of either drug. Haloperidol was not injected in the absence of carbachol injections. Thus, although haloperidol did abolish the increase in feeding produced by 12 -carbachol, this may have been due to general debilitation rather than selective interference with carbachol-stimulated mechanisms. Electrical Stimulation: Lateral hypothalamic stimulation in the presence of food often results in feeding (Delgado and Anand, 1953; Coons, Levak and Miller, 1965). This effect was originally attributed to activation of a hypothalamic "feeding centre" in parallel to the analysis of LH lesion-induced aphagia. More recently, several investigators have indicated that much of stimulation-induced feeding may be attributable to the activation of ascending dopaminergic pathways. Feeding may be elicited by stimulating the origins of these systems in the VTA (Wyrwick and Doty, 1966) and the substantia nigra (Phillips and Fibiger, 1973b). Further, feeding elicted by LH stimulation is attenuated by intraventicular administration of 6-OHDA (Phillips and Fibiger, 1973a) and systemic injection of haloperidol (Philllips and Nikaido, 1975). Feeding elicited by electrical stimulation differs from natural feeding in several respects. The behaviour elicited by electrical stimulation is not at all specific to feeding and cannot be attributed, for example, to an elevation of "hunger drive". Stimulation of the LH can evoke a plethora of activities, including hoarding, drinking, grawing, attacking, tail preening and sexual behaviour. The different behaviours elicited in different rats cannot be attributed to minor differences in anatomical localization within the LH, rather they are related to subject and environmental variables (see review by Valenstein, Cox and Kakolewski, 1970). The behaviour elicited by stimulation at any one site does not appear to be related to any single motivational state. For example, if rats that preferentially eat a particular food in response to LH stimulation are stimulated in the absence of that food, another "stimulus bound" behaviour gradually emerges, and is as likely to be drinking as eating of a different food, however - 13 -familiar and palatable that other food may be, and will generally maintain that new response even if the initially-consumed food is returned (Valenstein, Cox and Kakolewski, 1968a, b; Valenstein and Phillips, 1970). Clearly, stimulation-induced feeding is different from that evoked by food deprivation: A hungry rat does not drink water in preference to eating a palatable food simply because the particular food he is eating is removed from the chamber. The true nature of stimulation-induced behaviours remains unclear. They have been described as coping responses to a state of high arousal (Valenstein, 1975) and products of the excitation of neural systems underlying well-established response patterns (Valenstein et al, 1970). Although these possibilities are of considerable interest in their own right, their relationship to natural feeding is neither clear nor established. Tail pinch-induced feeding: Feeding can be induced by indirect stimulation of dopaminergic systems as well as direct stimulation. One curious finding is that moderately intense tail pinch provokes feeding, and such feeding is blocked by neuroleptics or 6-OHDA lesions of the MFB (Antelman and Szechtman, 1975; Antelman Szechtman, Chin and Fisher, 1975). However, as with electrical stimulation, tail pinch can also induce aggression, copulation, drinking, maternal behaviour, licking and gnawing, depending on the stimuli in the environment. Again, there is no reason to believe that these are "normal" manifestations of motivated behaviours. In summary, stimulation of dopaminergic systems through administration of amphetamine or apomorphine, electrical stimulation, or tail pinch elicits a variety of behaviours. These include locomotion, drinking, licking, gnawing, maternal behaviour and copulation. In the presence of food, the animals will often eat at least small quantitites of food. Although these elicited behaviours may reflect the - 14 -activation of neurological substrates related to normal feeding, they may equally well reflect the non-specfic activation of aberrant patterns of neural firing unrelated to normal behaviour. For example, the animals appear to be influenced by environmental stimuli to an exceptional extent. The evoked behaviours may be related to those seen when an animal is highly stressed, and may have little to do with feeding in day-to-day life. On the other hand, it is worth observing that few stimulation studies have addressed the question of whether dopamine is involved in non-consummatory aspects of feeding behaviour. CORRELATIONS BETWEEN FEEDING AND DOPAMINE RELEASE The previous sections have indicated that central dopamine systems may play some role in feeding behaviour, to the extent that disruption of dopaminergic transmission generally leads to decreased food intake and food procurement, although enhancemant of dopamine activity will lead to enhanced food intake in only some circumstances. However, few of the studies described thus far have addressed the issue of ethological validity: Is the normal, day-to-day feeding of animals or humans correlated with alterations in central dopamine activity? Although it appears that a minimal level of dopamine function is required for feeding to occur, this may only indicate that dopamine plays a permissive role in feeding. Furthur, although elevation of dopamine activity may increase feeding, this usually appears to reflect a general increase in behavioural activity rather than a selective enhancement of feeding. Thus, lesion, neuroleptic and stimulation studies may not be adequate to elucidate the neurobiological substrates of natural feeding. More direct evidence linking dopamine and feeding behaviour is required. - 15 -In recent years evidence has accumulated that dopamine activity does in fact change in relation to some aspect or aspects of feeding or nutrient balance. Dopaminergic activity has been studied following food or glucoprivation, during or subsequent to feeding, and following the administration of nutrients and putative satiety agents. Each of these situations will be considered in turn. Dopaminergic activity and lowered nutrient status: Several authors have reported that dopamine activity is increased by food deprivation. Freidman, Starr and Gershon (1973) reported that dopamine levels were increased in the hypothalamus of rats which had been deprived of food for 22h, but not in the rest of their brains. Fuenmayor (1979) found that food deprivation increased striatal levels of the dopamine metabolite homovanillic acid (HVA) in mice. On the other hand, these tissues contained high levels of dopamine following the inhibition of dopamine synthesis by the tyrosine hydoxylase inhibitor alpha-methyl-para-tyrosine, suggesting a decrease in striatal dopamine turnover. Fuenmayor suggested that there may be dual, antagonistic, actions of fasting on various dopamine subpopulations. Enhanced levels of amphetamine-induced activity were observed following food deprivation (Campbell and Fibiger, 1971). The quantity of stabilimeter activity induced by l.Omg/kg d-amphetamine became progressively more intense over days, and was more than four times greater after four days of deprivation than after one. This appears to indicate increased availabilty of dopamine at synapses. Similarly, Glick, Waters and Milloy (1973) observed that amphetamine-stimulated dopamine release in the striatum was increased by depriving the animals of food for 24h. In contrast to these reports, other investigators have failed to observe increased dopamine activity with fasting. Heffner, Hartman and Seiden (1980) - 16 -measured levels of DOPAC and dopamine in rats which had continuous access to food and in rats which had been deprived of food for 20h. No significant between groups differences were found in the striatum, n.accumbens, hypothalamus, olfactory tubercle, or amygdala. An electrophysiological study also failed to indicate any change in the activity of single dopamine neurons in the substantia nigra of unrestrained cats as a result of food deprivation (Trulson, Crisp and Trulson, 1983). Over a 48h period of food deprivation plasma glucose levels dropped 21.4% in a group of cats maintained on a high carbohydrate diet, and increased slightly in cats fed a high protein, low carbohydrate diet. Nigral unit activity was unchanged from baseline levels in either group. Although of considerable interest, there are several reasons why the generality of this and similar studies may be limited. First, it exclusively involved neurons in the substantia nigra, pars comapacta. It is possible that mesolimbic dopamine neurons, originating in the VTA, could change their rates of firing while their nigral cousins do not. Second, it is possible that only a small proportion of dopamine neurons change their rate of firing, but do so in such a pronounced manner that net dopamine release is sufficiently altered to affect behaviour. Third, the subjects employed in the Trulson et al study may not provide an adequate model for feeding in humans or rodents. Cats are carnivorous and have somewhat unusual feeding habits (eg., they are among the only adult mammals that do not show a preference for sweet solutions, and as carnivores they are typicallly binge-feeders). As well, their dopamine systems have a somewhat different anatomical structure (eg., dopaminergic neurons throughout the feline pars compacta contain the peptide cholecystokinin: see below). For these reasons at least, these results may not readily generalize to onmivorous humans or rats. - 17 -Increased dopamine turnover as a result of insulin-induced hypoglycemia has been consistently reported in studies with rats. When insulin was administered to anesthetized rats levels of dopamine decreased in limbic, striatal and cortical regions increased as blood glucose levels fell (Agardh, Carlsson, Lindqvist and Siejso, 1979). A similar treatment increased dopamine release in the hypothalamus (Sauter, Ueta, Engel and Goldstein, 1981). Lower doses of insulin (5U/kg) which reduced blood gucose levels to 30% of control levels increased levels of dopamine metabolites in the striatum and hypothalamus of freely moving rats, but not in the n. accumbens (Cottett-Emard and Peyrin, 1982). Urinary levels of the metabolites were also increased, and this could not be attributed to increased release of dopamine from the adrenals. (Urinary levels of dopamine were also elevated in humans following insulin injection: Woolf, Akowuah, Lee, Kelly and Feibel, 1983). Similarly, Rowland, Bellush and Carlton (1985) recently reported that 5U/kg insulin increased dopamine turnover in the striatum but not the n. accumbens. Finally, McCaleb and Myers (1979) used a push-pull cannulation technique to investigate the release of dopamine in the striatum of freely moving rats. If insulin was added to the perfusate, recovery of ^ H-DA was increased. Levels of the dopamine metabolites 3,4-dihydoxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) were elevated in the cerebrospinal fluid of freely moving rats following injection of 6U/kg insulin (Danguir, Elghozi and Laude, 1984). However, if food was available glucose levels did not fal l and dopamine turnover did not increase. Glucose prevented similar increases when it was injected before blood glucose levels f e l l , and reversed increases if injected after they had occurred. Similarly, increased telencephalic dopamine turnover levels, induced by insulin injection, were returned to baseline following feeding or spontaneous glucorecovery (Beliin and Ritter, 1981). - 18 -Taken together, these studies suggest that dopamine activity will be reliably increased during pronounced hypoglycemia or severe food deprivation. However, in cases of less severe deprivation (eg., 24h without food for rats), increases are less consistently reported and are generally of lower magnitude. This weakens the case for the existence of a role for increased dopamine activity due to privation as a trigger for normal feeding, although increased dopamine activity may accompany near-starvation when nutrient reserves reach a critical point. On the other hand, blood glucose levels fa l l in the few minutes prior to meal onset in rats with continuous access to food (Louis-Sylvestre and Le Magnen, 1980; Campfiled, Brandon and Smith, 1985). This fall could produce an increase in dopaminergic activity related to meal onset. Dopaminergic activity and feeding: Relatively few studies have examined the effect of feeding pjer se on dopamine activity. Technically, it is difficult to separate the effects of having engaged in feeding behaviours from those induced by the physiological consequences of ingestion. One approach to this problem has been to monitor the activity of single dopamine neurons electrophysiologically while the animal feeds. However, it is necessary to reiterate that there are possible limitations to the generality of these electrophysiological studies, both with respect to the cell populations they investigate and the species they employ. Nonetheless, Trulson et al (1983) and Strecker, Steinfels and Jacobs (1983) both found no change in nigral unit discharge rate, discharge pattern or waveform when food deprived cats were allowed to eat. Few neurons in the Strecker et al study changed firing rate by more than + 10%. Martin and Myers (1976) examined the release of ^C-dopamine during feeding and lever pressing for food on an FR6 schedule using the technique of repeated push-pull perfusion. During free feeding, dopamine activity did not - 19 -increase at any site in the anterior hypothalamus, increased at one of five sites adjacent to the third ventrical (in the vicinity of the n. oriens and the paraventricular n.) and at the most medial of five sites in the substantia nigra. There were also increases at two other sites in the vicinity of the third ventrical and at one site in the substantia nigra during lever pressing for food. These results are not conclusive because of limitations inherent in the technique, the limited number of sites investigated, and the fact that no sites were examined for both free-feeding and lever pressing at the same site. Still , the study suggests that dopaminergic activity increases at some limited sites in the brain as a result of certain aspects of feeding behaviours. Other studies that have examined the effect of feeding on dopamine activity have assayed chemically the levels of dopamine and its metabolites after sacrificing animals at some time following consumption of a meal. For example, Biggio, Porceddu, Fratta and Gessa (1977) reported that brain levels of HVA and DOPAC levels were increased by feeding in rats which had previously been deprived of food for 21h. The increase was significant l h after food presentation and reached a peak after 3h. In animals that received food for 3h, levels remained elevated for at least 4h after food was removed. The most thorough study of the impact of feeding on dopamine turnover to date was conducted by Heffner, Hartman and Seiden (1980). They allowed rats that had been deprived of food for 20h to eat for l h . At the end of the hour the animals were decapitated and their brains immediately dissected. Amounts of dopamine and DOPAC in several brain regions were determined by radioenzymatic assay. DOPAC/DA ratios were increased by 25% in n. accumbens, by 99% in amygdala, and by 22% in hypothalamus, compared to non-fed controls. Ratios were unchanged in striatum, olfactory tubercle, septum or frontal cortex. - 20 -Further experiments showed that tube feeding produced increases in the amygdala that were similar to those seen following feeding, but similar increases were not observed in other regions. On this basis, Heffner et al (1980) argued that increases observed in n. accumbens and hypothalamus do not reflect post-ingestive effects but rather an effect of having performed feeding behaviours. This interpretation was challanged by later data from the same laboratory. Heffner, Vosmer and Seiden (1984) reported that if rats were sacrificed l h after feeding had begun, when they had consumed an average of 10.3g of food, hypothalamic DOPAC levels were not significantly elevated. On the other hand, if the rats were sacrificed two hours after feeding began, when mean consumption was 11.3g (a mere l.Og more than the l h group) hypothalamic DOPAC levels were nearly twice that of controls. Apparantly a large increase in dopamine metabolism occurred in the second hour, when almost no food was consumed. It seems probable that some post-ingestional factor must be responsible for this increase. Although data were only reported for the hypothalamus in the Heffner et al (1984) report, one might speculate that such an increase also occured in the n. accumbens and other structures. Although Heffner et al (1980) only found increases in mesolimbic, as opposed to nigrostriatal, dopamine turnover, Chance, Foley-Nelson, Nelson, Kim and Fischer (1985) have reported recently that elevated DOPAC and HVA levels were found in both n. accumbens and striatum following one hour of free access to food. This is consistent with the long-hypothesized role for the nigrostriatal dopamine system in feeding. On the other hand, both studies have reported increases in the n. accumbens, a structure typically not implicated in feeding by lesion studies. Apparantly, a full analysis of feeding and dopaminergic systems - 21 -will require consideration of both mesolimbic and nigrostriatal dopamine projections. It is important to note an apparantly paradoxical aspect of the results of the Martin and Myers (1976), Heffner et al (1980, 1984) and Chance et at (1985) studies: Each report indicated that there are increases in dopamine turnover as a result of feeding, while there have been no reports of decreases. It was previously noted that increases in dopamine turnover are observed folowing severe deprivation, and that increases in dopaminergic system activity makes animals feed. Taken together, i t would seem that the effect of feeding on dopamine systems is to put them into a state of elevated activity, where feeding is more likely to occur. Suppression of dopaminergic activity by satiety factors: Inhibition or disruption of dopamine systems results in attenuated feeding. Conversely, it may be suspected that decreased dopaminergic activity characterizes the brain of a satiated animal, when it is phasically unresponsive to food. In order to investigate this hypothesis investigators have studied dopamine activity following administration of putative satiety factors. No consensus has ever been reached on the question of what cues lead to meal termination. St i l l , considerable evidence exists that both blood glucose and cholecystokinin play some role in satiety. The post-prandial period is characterized by relatively elevated levels of both glucose (Steffans, 1969) and CGK (Walsh, Lamers and Valenzuela, 1982; Smith, Greenberg, Falasco, Gibbs, Liddle and Williams, 1985). Adminstration of both glucose (Mayer, 1953) and CCK (Gibbs, Young and Smith, 1973; Antin, Gibbs, Holt, Young and Smith, 1975; Mueller and Hsiao, 1978; Smith and Gibbs, 1981; Collins, Forsyth and Weingarten, - 22 -1983) suppress food intake. Finally, as the following paragraphs document, both substances appear capable of suppressing central dopamine activity. Dopamine and glucose: In the push-pull cannualation experiment described above, McCaleb and Myers (1979) also found that, when glucose was added to the perfusate, release of H-dopamine was decreased. Similarly, in an in vitro study of striatal slices, Dorris (1978) found that potassium- or amphetamine-stimulated release of a false dopaminergic neurotransmitter q ( H-alpha-methyl-m-tyramine) was inhibited by glucose infusion. Sailer and Chiodo (1980) injected glucose intravenously into anesthetized rats while measuring the activity of dopaminergic neurons in the substantia nigra. Injection of 250mg/kg of glucose completely inhibited these neurons for at least 30 minutes, and a small injection of 15mg/kg decreased firing rate for 10 - 12 minutes. To round out their litany of negative findings, Strecker et al (1983) and Trulson et al (1983) found no such effect when glucose was injected into freely moving cats. Further, Westerink and Spaan (1981) found that levels of dopamine and its metabolites in rat striatum were unaffected by injection of 500mg/kg glucose. Several reports indicate that the behavioural effects of dopaminergic drugs are modulated by blood glucose levels. White and Blackburn (submitted) found that i.p. injection of l.Og/kg d-glucose shifted the dose response curve for the stereotypy-inducing effect of amphetamine to the right, indicating suppression of dopaminergic release. However, they found that similar injections did not affect amphetamine-induced motor activity (as measured by photocell interruptions) or circling induced by amphetamine in rats with unilateral 6-OHDA lesions of the substantia nigra. Amphetamine induced stereotypy, locomotion and circling are mediated in part by dopamine systems (Kelly, 1977; Ungerstedt, 1971a). The - 23 -preferential attenuation of stereotypy may reflect differential effects on separate subpopulations of dopamine neurons (Kelly, 1977) or receptors (for review see Joyce, 1983). The previous studies examined the effects of exogenously applied glucose on dopamine activity. Other work has examined dopamine activity in animals with altered glucometabolism. Amphetamine-induced stereotypy, as well as locomotor activity and anorexia, are reduced in diabetic rats (Marshall, Friedman and Heffner, 1976, 1978; Marshall, 1978). These effects are restored by insulin injections. The data support the suggestion that the decrease in sensitivity to amphetamine observed in these animals is due to elevated blood glucose levels, rather than to general i l l health. The possibility that diabetic rats have chronically depressed dopamine systems due to elevated blood glucose levels is furthur supported by the finding that the number of striatal dopamine receptors is increased in rats with experimentally induced diabetes, as indicated by increased H-spiperone binding (Lozovsky, Sailer and Kopin, 1981). Such increased binding was not found in other diabetic rats that received insulin therapy for 12 days prior to sacrifice. Although the effects of amphetamine, which enhance dopamine transmission, are attenuated in rats with elevated blood glucose, those of haloperidol, which blocks dopamine effects post-synaptically, are enhanced. Sailer and Kopin (1981) found that haloperidol-induced catalepsy was enhanced by injection of 1.25g/kg d-glucose. In summary, it appears that many of the behavioural effects of amphetamine and haloperidol are altered in a manner which is consistent with Sailer and Chiodo's suggestion that elevated blood glucose levels inhibit dopaminergic acitivity. However, several reports indicate that this effect is not universal: It is - 24 -quite possible that only selected subpopulations of dopaminergic neurons or receptors are affected. Dopamine and cholecystokinin: Several lines of evidence have indicated that the putative satiety hormone cholecystokinin (CCK) also affects the activity of dopamine systems. Interestingly, this peptide co-exists with dopamine in many neurons in the VTA and some, mostly medial, neurons of the substantia nigra (Hokfelt, Skirboll, Rehfeld, Goldstein, Markey and Dann, 1980). These neurons appear to belong to the mesolimbic projection, rather than the nigrostriatal system, with the heaviest projection to the posterior-medial n. accumbens (Williams, Gayton, Zhu and Dockray, 1981, Studler, Simon, Casselin, Legrand, Glowinski and Tassin, 1981; Marley, Emson and Rehfeld, 1982; Gilles, Lostra and Vanderhaegen, 1983). Although these is a large CCK presence in the striatum, it does not appear to be of nigral origin (Meyer, Beinfeld, Oertel and Brownstein, 1982). Electrophysiological studies have provided evidence of CCK effects on dopamine activity. Skirboll, Grace, Hommer, Rehfeld, Goldstein, Hokfelt and Bunney (1981) examined the effects of intravenously (i.v.) or iontophoretically applied CCK on cells of the substantia nigra and the ventral tegemental area (VTA). Units in areas of the substantia nigra in which dopamine co-existed with CCK showed transient increases in firing rate following i.v. application of CCK, but units in areas without CCK/DA cells were unresponsive. In the VTA, of 31 cells examined, 7 were unaffected by i.v. CCK, 15 were fleetingly suppressed and 9 exhibited increases like those seen in the substantia nigra. In three cases the increases were apparently of sufficient magnitude to send the cell into depolarization inactivation. Iontophoretically applied CCK increased firing rates in VTA and those areas of substantia nigra with CCK/DA co-existence. - 25 -Hommer, Paklovitis, Crawley, Paul and Skirboll (1985) again showed that nigral activity was increased some 50-80% in the 30 second period beginning 20 seconds after i.v. injection of CCK. Chiodo and Bunney (1983) blocked a similar increase in activity in cells of the VTA with the CCK antagonist proglumide. White and Wang (1984) determined that CCK could also increase activity in cells of the n. accumbens in the regions where the axons of many CCK/DA neurons appear to terminate. Although CCK by itself produces transient increases in dopamine unit activity, it potentiated the inhibition of dopamine activity by apomorphine (Hommer and Skirboll, 1983). This sugests that CCK also plays an inhibitory role on dopamine systems, possibly potentiating dopamine autoreceptor sensitivity. Inhibitory, rather than excitatory effects of CCK on dopaminergic systems have also been indicated by several studies of dopamine turnover. Intraventricular injection of CCK has been reported to decrease dopamine turnover in several regions including the n. accumbens, striatum, hypothalamus, mesencephalon and septum (Fuxe, Andersson, Locatelli, Agnati, Hokfelt, Skirboll and Mutt, 1980; Fekete, Kadar and Telegdy, 1981; Mashal, Owen, Deakin and Poulter, 1983). Release of dopamine in the n. accumbens, as determined by push-pull canulation, was reduced when CCK was added to the buffer (Voigt and Wang, 1984). In a recent study employing in vivo voltametry, Lane, Blaha and Phillips (personal communication) have found that dopamine release was decreased beginning about ten minutes after CCK injection, reaching a minimum half an hour later, and remaining greatly suppressed for over an hour. Similar suppressant effects are observed in vitro: Low concentrations of CCK inhibited release of both basal and electrically evoked outflow of 3 H-dopamine from slices of cat striatum (Markstein and Hokfelt, 1984). - 26 -Several reports have indicated suppression of dopaminergic behaviours and neuroendocrine effects by CCK. Central injection of CCK in rats reduced methamphetamine-induced activity (Katsuura and Itoh, 1982), reduced spontaneous rearing but not locomotion (Schneider, Alpert and Iversen, 1983; Widerlov, Kalivas, Lewis, Prange and Breese, 1983) and interfered with both acquisition and maintainance of active avoidance behaviour (Fekete, Szabo, Balasz, Penke and Telegdy, 1981). In humans, i.v. injection of CCK suppressed the dopamine mediated growth hormone-release response to apomorphine (Lai, Nair, Eugenio, Thavundayil, Lizondo, Wood, Etienne and Guyda, 1983). Two other behavioural effects of CCK are worth mentioning, although their reationship to dopamine is less well established. First, Crawley has demonstrated in experiments with mice (Crawley, Hays, Paul and Goodwin, 1981) and rats (Crawley, Hays and Paul, 1981) that exploration of an open field which contains several objects of interest to rodents is reduced by CCK. Second, even modest doses of CCK can lead to sedation or behavioural quiesence (Itho and Katsuura, 1981; Katsuura and Itoh, 1981; Crawely, Rojas-Ramirez and Mendelson, 1982; Rojas-Ramirez, Crawley and Mendelson, 1982). In conclusion, it is clear that CCK and dopamine interact physiologically, but the precise nature and extent of this interaction is uncertain. Electrophysiological work suggests that only those neurons in which CCK and dopamine co-exist are excited by CCK (Skirboll et al, 1981). These neurons appear to project primarily to the medial posterior n. accumbens (Williams et al, 1981, Studler et al , 1981; Marley et al, 1982). On the other hand, injection of CCK seems to decrease dopamine turnover in other brain regions as well, including the striatum (Fuxe et al, 1980; Mashal et al, 1983; Markstein and Hokfelt, 1984; Lane, Blaha and Phillips, personal communication). Possibly the - 27 -short term excitatory effect of CCK on CCK/DA cells is an artifact of the administration procedure (rapid bolus injections of moderately large amounts of CCK). The longer term inhibitory effect on a larger population of dopamine cells is likely more representative of what occurs following a meal, which would result in the slow release of a modest quantity of CCK resulting in plasma levels approximately equal to that seen after an injection of 2ug/kg (Smith et al, 1985). The plausiblity of multiple CCK/DA interaction is increased by the recent report of Hommer et al (1985) who found that excitatory effect of CCK on nigral cells was attenuated, but not abolished, by lesions of afferents to or efferents from the nucleus of the solitary tract (NTS). The NTS is a brainstem nucleus responsible for integrating many sensory and gustatory inputs. Along with its vagal afferents, it has been implicated as playing a role in the exploration-inhibiting (Crawley, Hays and Paul, 1981, Crawley and Schwaber, 1984) and satiety-promoting effects of CCK (Smith, Jerome, Cushin, Eterno and Simansky, 1981; Lorenz and Goldman, 1982; Morley, Levine, Kreip and Grace, 1982). MOTIVATIONAL INTERPRETATIONS OF DOPAMINE'S ROLE IN FEEDING The previous sections have reviewed the experimental data concerning relationships between dopamine and feeding. This section shall examine several theoretical interpretations of dopaminergic contributions to behaviour and see how they may be applied to an analysis of feeding. Of course, many studies have examined the role of dopamine in other behaviours, and although no attempt will be made here to review these studies comprehensively, those that aid in understanding the possible role or roles played by dopamine in feeding will be described. - 28 -The Sensorimotor Hypothesis: Following Ungerstedt's (1971a,b) reports that 6-OHDA lesions of the nigrostriatal bundle produced severe aphagia and motor deficits, the first general theoretical interpretation of dopamine function to receive serious consideration was based on the suggestion that the syndrome observed following NSB or lateral hypothalamic lesions could be attributed to sensorimotor deficits (Marshall, Turner and Teitlebaum, 1972; Marshall, Richardson and Teitlebaum, 1974; Ungerstedt, 1974). Following unilateral 6-OHDA lesion of the NSB rats show a chronic tendency to turn towards the lesioned side (Ungerstedt, 1971a) and display difficulty in using the limbs contralateral to the lesion for righting, climbing and resisting gravitational pull (Marshall et al, 1974). This postural asymmetry does not merely reflect a motor deficit, i t appears to be related to a state of "sensory neglect": Stimuli do not evoke normal orienting responses when presented on the side of the body contralateral to the lesion. Such deficits are not complete, rather, the orienting response appears variable and quickly disappears. For example, if the whiskers of a rat are touched it may turn toward the stimulus, but if they are brushed again soon afterward the rat does not orient. Thus, it appears that deficits observed following NSB damage are neither wholly sensory nor wholly motoric, but rather they appear to reflect disruption of a higher level of sensorimotor integration: Apparently the animal perceives, but does not respond. Bilateral lesions have a much more severe effect on the animal. The animals are initially akinetic, but do not seem somnolent. When presented with a novel stimuls, such as fresh food, they may orient to it, and may even approach it. However, because all responses rapidly wane following the presentation of any stimulus, the animal does not sustain or even commence feeding. Although - 29 -Marshall et al (1974) felt that sensorimotor deficits could not account for all of the observed feeding and drinking deficits observed (eg., the active rejection of food in the initial period after lesioning) such deficits may contribute strongly to the NSB and LH syndromes The sensorimotor hypothesis accounts satisfactorily for several other experimental obsevations. If dopamine is involved in responding to stimuli, then increases in dopamine activity would be expected to make an animal hyper-responsive to environmental stimuli (as is observed), and suppression of dopaminergic activity by neuroleptics would be expected to make animals less responsive to the environment. The preferential disruption of consummatory vs preparatory aspects of feeding can be attributed within the theory to the relative lack of salience of food related cues compared to the salience of food itself. The anhedonia hypothesis: As described in the section concerning the effects of neuroleptics on feeding behaviours, pimozide treated animals decrease the rate of operant responding through the course of a test session. On the next test these animals begin at a lower rate of responding and rapidly fall to sti l l lower rates of responding. Similar response decrements are observed when responding is reinforced by brain stimulation (Fouriezos and Wise, 1976; Fouriezos, Hannson and Wise, 1978; Franklin, 1978), water (Gerber, Sing and Wise, 1981), thermal stimulation (Ettenberg and Carlisle, 1985) and intravenous administration of amphetamine (Yokel and Wise, 1975) or cocaine (deWit and Wise, 1977). The "anhedonia hypothesis" attributes this continuing decline in operant rates to the proposition that "neuroleptics attenuate the hedonic impact of a variety of positive reinforcers" (Wise, 1985, p. 184). Thus, once the reinforcer is - 30 -no longer rewarding, the response extinguishes. This analysis seems to apply to all rewards, not just feeding. Immediately after the anhedonia hypothesis was proposed by Wise et al (1978) several studies were conducted demonstrating that operant responding in animals, trained on intermittent schedules of reinforcement, are more seriously disrupted by neuroleptics than by nonreinforcement. For example, although animals trained on intermittent schedules typically show considerable resistence to extinction (the so-called partial reinforcement extinction effect, or PREE), Phillips and Fibiger (1979) and Tombaugh et al (1980) observed rapid deterioration in performance following administration of neuroleptics. Further, a decrease in responding was observed before the presentation of the first reinforcer on a partial reinforcement schedule, that is before the first experience the animal had with the reinforcer while in it's "anhedonic" state (Mason, Beninger, Fibiger and Phillips, 1980; Gray and Wise, 1980). In addition, the effects of neuroleptics sum with those of extinction (Ettenberg, Cinsavich and White, 1979; Gray and Wise, 1980; Mason et al, 1980; Tombaugh et al, 1980). That is, animals respond at sti l l lower rates during extinction if they are treated with neuroleptics. Finally, when animals are transferred from pimozide to extinction responding recovers completely. These effects should not be observed if the effects of pimozide and extinction are functionally equivalent. To explain the partial reinforcement and summation effects Gray and Wise (1980) proposed that neuroleptics not only attenuate the hedonic impact of posiitve reinforcers, but also blunt the secondary reinforcing effects of stimuli associated with reward. That is, during operant conditioning the animal comes to associate various cues (such as the lever or the response itself) with reward. Normally, these "secondary reinforcers" act to elicit approach and manipualtion of - 31 -the bar for some time, even in the absence of reward (see Franklin and McCoy, 1979, for a demonstration of how these effects apply to neuroleptic-induced deficits). Other data is consistent with the idea that neuroleptics attenuate the conditioned reinforcing properties of environmental cues paired with food (Spyraki, Fibiger and Phillips, 1982). In fact, it is tempting to speculate that neuroleptics only disrupt the salience of food-related cues without attenuating the reward value of food itself: Animals often eat as much as usual when drugged with moderate doses of pimozide; the same doses produce clear deficits in both operant responding for food and hoarding. The anhedonia hypothesis of neuroleptic action is similar to the sensorimotor deficit hypothesis of NSB lesion-induced deficits in that both predict deficits in responding to environmental stimuli. However, they differ in the nature of the stimuli to which they refer, and they differ in the severity of the deficits to which they are addressed. This may reflect differences in the precise substrate to which they refer. The sensorimotor deficits, observed by Marshall et al in response to all environmental stimuli are typically seen only after nearly total destruction of the NSB. In contrast, while Wise reports deficits in responses for hedonically-positive rewards, or signals of these rewards, after relatively mild pharmacological disruption of dopamine transmission. One possibility is that NSB lesions disrupt a non-dopaminergic system. Another is that NSB lesions destroy part of a dopaminergic system that is relatively insensitive to the effects of neuroleptics. On the other hand, the mesolimbic dopamine may be an especially sensitive substrate, given the similarities between the deficits seen in hoarding following neuroleptics and VTA lesions. Although it is possible that NSB lesions and neuroleptics produce qualitatively different deficits it is also possible that the difference between the - 32 -effects of the two manipulations is purely quantitative. By extension of the anhedonia hypothesis, sensorimotor deficits may represent a loss of attractiveness of all stimuli, including reinforcers. On the other hand, apparent "reward" deficits could reflect a general decrement in the salience of stimuli, most obvious in situations with clearly defined stimulus-response contingencies (such as in operant conditioning). Yet again the greater sensitivity of operant and appetitive responses to the disruptive effects of neuroleptics could reflect the fact that these are typically learned responses, while consummatory and orienting responses are more reflexive and species-typical (see Beninger, 1983). The sensorimotor hypothesis and the anhedonia hypothesis differ in another way. According to the sensorimotor hypothesis dopamine is actively involved in responding. Although Wise (eg., 1985) takes an explicitly agnostic position on the issue of whether dopamine is involved in response production, it is clear that the most straightforward form of the anhedonia hypothesis has dopamine playing a passive, evaluative role. That is, dopamine release is proportional to how "rewarding" a reward is; if it was sufficiently rewarding in the past the animal will respond again in the future. Thus, the sensorimotor hypothesis advocates an active role for dopamine, while the anhedonia hypothesis views dopamine's role as passive. The response-initiation deficit hypothesis: Another model of dopamine function requires the dopamine system to play an active role in responding. This is the response initiation deficit hypothesis (Posluns, 1962; Fibiger, Zis and Phillips, 1975). This hypothesis suggests that an organism under the influence of a neuroleptic is unable to initiate a motor reaction in response to an environmental cue, even though the animal knows - and cares - what the cue signals. The idea was originally advanced to explain deficits seen following - 33 -neuroleptic administration or 6-OHDA lesions of the NSB in the avoidance responding paradigm. During avoidance learning a stimulus is presented for some standard period of time (typically 10 seconds to a minute), while an animal is confined in a compartment equipped with a grid floor. At the end of this period the grid is electrified. Shock is terminated by the performance of an arbitrarily chosen response. By performing the response, the animal has "escaped" the shock. If the animal responds prior to shock onset, no shock is administered, the animal has successfully "avoided" the shock. In cases where the response required is movement out of the shock compartment into a "safe" place, most normal animals are reliably avoiding shock within a few trials. In this paradigm, neuroleptic-treated animals do not avoid, but will escape with a short latency once shock begins (Hunt, 1956). The quick escape response indicates that the animal is physically capable of responding. What is more, prior to shock onset the animals squeal and defaecate. In addition, their subsequent response to the cue in a test for classically-conditioned fear elicitation indicates that they have learned the significance of the signal (Beninger, Mason, Phillips and Fibiger, 1980). The response deficit is strongest during acquisition of the response, and is not seen when well-trained animals are tested under the effects of neuroleptics (Fibiger et al, 1975; Beninger, Phillips and Fibiger, 1983). In order to interpret this deficit it is interesting to note that well-trained animals will often perform an avoidance response without showing any sign of fear (Solomon and Wynne, 1953; Mineka, 1979; Mineka and Gino, 1980) and with too short a latency for autonomic signs of fear to have been generated (Champion, 1964). Rather than acting out of fear, these animals have been described as acting "cognitively" (Solomon and Wynne, 1953, p. 17). Although no study has ever simultaneously examined the decline of fear during avoidance - 34 -training and the development to immunity from the disruptive effects of neuroleptic, i t is possible to view this coincidence as indicating that dopamine release is required to make a response motivated by emotion (fear), but not for a response under "cognitive" control, or, alternatively, when the avoidance response habit is well established. The anhedonia hypothesis cannot readily explain the avoidance response deficits seen following neuroleptic administration. Avoidance can be interpreted as having a rewarding component (eg. Mowrer, 1947; Masterson and Crawford, 1982). If one assumes that termination of escape is also rewarding, then a simple reward deficit has trouble specifying why the escape from shock is sufficiently rewarding to produce reliable, low latency escape, if the avoidance of shock is not rewarding enough to reinforce responding prior to shock onset. An alternative way to interpret the avoidance deficit data is to consider that aviodance and escape are the result of two different processes (Ehrman and Overmier, 1976; Jacobs and Harris, in press). By this interpretation, avoidance can be seen as an appetitive response, while escape can be viewed as a consummatory reaction. Thus in avoidance, as in feeding behaviour, dopamine receptor blockade appears to produce particularly severe deficits in the appetitive responding. The Incentive-Response Hypothesis: It is possible that neuroleptics may interfere with operant responding for positive reinforcement and with avoidance responding through separate neural mechanisms. However, i t is preferable to devise a theory which can account for both simultaneously. Accordingly, an alternative hypothesis is provided here, one that combines elements of several other hypotheses. It is an extension of ideas presented previously by Mogenson and Phillips (1976), Clody and Carlton (1980), Phillips, McDonald and Wilkie - 35 -(1981), Beninger (1983). Carr and White (1984) and Fibiger and Phillips (1985). It is also similar, in some respects, to a proposal made by Panksepp (1982). The alternative hypothesis borrows the assertion that dopamine systems are actively involved in reponse production from the response initiation hypothesis. In agreement with Wise, the current hypothesis recognizes a connection between dopamine systems and motivationally relevant stimuli. However, the hypothesis assigns dopamine an active role in responses to these stimuli, rather than merely in the post hoc evaluation of their consequences: It is concerned with incentive, not reward. Accordingly, the hypothesis shall be referred to as the "incentive-response hypothesis" of dopaminergic function. It's core idea may be stated in the following way: "When an animal observes an incentive stimulus, the release of dopamine in the forebrain is increased, resulting in approach to the stimulus by the animal. Once the animal is in contact with a goal object, consummatory reactions occur which are not mediated by dopamine systems". Before elaborating the hypothesis and applying it to data i t is necessary to indicate that i t is not tied to any specific theory of incentive motivation, such as those of Spence (1956), Bolles (1967, 1975) or Bindra (1976, 1978). Accordingly, it is not based on any fully articulated statement concerning the relationship between stimuli, central representations, expectations and responses, or concerning the associations between these constructs. Rather, the incentive-response hypothesis simply rests on the assumption that stimuli that are associated with primary rewards attain incentive value. No formal statement shall be presented defining the incentive strength of a stimulus, but i t is assumed to be determined by monotonically increasing functions depending on the degree of association between the stimulus and a biologically relevant stimulus, and on the magnitude of the biological significance of that relevant stimulus. - 36 -Application of the incentive-response hypothesis to feeding: The hypothesis predicts that when the animal observes a food-related stimulus, the level of dopamine release in the forebrain is increased. This increase in dopamine release results in approach to the incentive which may be a discrete cue or an element in a stimulus compound (see Schneirla, 1958 and White, Messier and Carr, 1984 for discussions of the complex nature of approach). In the context of feeding, the animal would be expected to approach food or the place where food is usually available. Thus, the hypothesis makes clear predictions with respect to appetitive behaviours. Stimulation of dopamine activity should be functionally equivalent to an increase in the incentive value of food and lead to approach of food related stimuli. This is observed with food itself and with conditioned-reinforcing signals that have previously been paired with food, but only with some levels of stimulation. It may be suggested that higher levels of dopamine stimulation indiscriminantly enhance the incentive value of all stimuli, and beyond a certain level of stimulation the pre-potency of food-related stimuli is overwhelmed by any and all other objects in the environment. Alternatively, activation of dopaminergic mechanisms involved in appetitive behaviours may overwhelm other mechanisms involved in consummatory behaviour. The animal may repeatedly approach food but immediately move on before feeding. The status of dopamine systems during consummatory behaviour is less clear. On one hand, food is of biological significance and should have a naturally high incentive value. On the other hand, there is some evidence that different mechanisms subserve appetitive and consumatory behaviours (Konorski, 1967; Jacobs and Harris, 1985). It is possible that dopamine, and the forebrain regions that dopamine neurons project to, are exclusively involved in appetitive - 37 -behaviour. Functional dopamine disruption does not disrupt escape, which may be viewed as a consummatory response, and even cecerebrate animals will ingest food if it is placed directly in contact with their mouths or on their bodies (Woods, 1964). Although high levels of dopamine activity are apparantly not necessary for the performance of a consummatory response, high levels could normally accompany such behaviours. High dopamine release while feeding could keep an animal in contact with food, but it could also result in the animal approaching extraneous environmental stimuli. Interpretation of the role of dopamine in the onset of. satiety is again problematic. Satiety is characterized by the cessation of contact with food that should, according to the hypothesis, be associated with a decrease in dopamine activity. However, when an animal stops eating it initiates other activities and may be quite active, a state which may well be associated with increases in dopamine release. In the post-prandial period an animal is less attracted by food-related incentives (eg., Cabanac, 1971). The hypothesis predicts that during this period incentives should evoke less response from dopaminergic systems, and dopamine neurons may even be actively suppressed. This is consistent with the observed suppression of dopamine activity by glucose and CCK. However, this prediction is incompatible with the elevation in dopamine turnover in the post-prandial period observed by Heffner et al (1980, 1984) and Chance et al (1985). This point shall be discussed more extensively below. In summary, the incentive-response hypothesis handles most of the data available concerning dopamine and feeding quite well, with the exception of the observed increases in dopamine activity following ingestion of a meal. However, there is little evidence that is directly related to the fundamental assertion of - 38 -the hypothesis, namely that incentive, food-related stimuli elicit an increase in dopamine activity, which in turn results in behavioural responses to these stimuli. One way of testing this is to observe respones to conditional stimuli that have previously signalled the delivery of food, and to examine the role of dopamine in these responses. Migler (1975) conducted such a study with results that are in accord with the incentive-response hypothesis. In his "conditioned approach" paradigm, four monkeys were presented with a compound stimulus consisting of the illumination of a yellow panel for 15 sec, and the presentation of a tone in ther first second of panel illumination. A food pellet was delivered behind a flap coincident with the onset of the compound stimulus. Thus, the panel-tone combination served as an excitatory conditional stimulus and acquired incentive value. Administration of chlorpromazine increased the proportion of long-latency responses for each animal. Each monkey typically took the pellet in the first few seconds of panel illumination while undrugged, but when drugged they often did not respond until the panel was no longer illuminated. Interestingly, responses were often initiated as soon as the panel light went off. Unfortunately, there are several problems with this study. First, only four monkeys were used, too few for statistical analysis. Second, the use of chlorpromazine as the neuroleptic in this study was unfortunate, because chlorpromazine lacks the specifity to dopamine receptors that other drugs have. It also has pronounced effects in other neurotransmitter systems, notably on noradrenergic receptors. Finally, no controls were run to ensure that the deficit in responding was not due to some general performance deficit. The incentive-response hypothesis requires a more definitive test, using a sufficiently large number of animals to provide reasonable statistical power, and employing a - 39 -specific dopamine receptor blocker. The present thesis presents the details of a related experiment incorporating these features and a second experiment directly investigating changes in dopamine turnover as a result of exposing animals to a food-related conditional stimulus. The results of both experiments confirm Migler's study and provide support for the incentive-response hypothesis. These experiments indicate a role for dopamine systems in appetitive feeding behaviours. However, a comprehensive analysis of the role of dopamine in feeding must extend to consummatory behaviour and the post-prandial period. Accordingly, two other experiments investigated the role of dopamine in these phases of feeding behaviour. First, the effect of dopamine receptor blockade on consummatory behaviour was investigated. Injections of equivalent doses of the same neuroleptic used in the experiment on appetitive behaviour were administered to rats prior to the onset of a free-feeding session. Finally, a fourth experiment was conducted to clarify the nature of forebrain dopamine activity during the post-prandial period. To this end dopamine turnover following the consumption of food was analyzed using a more reliable technique than the radioenzymatic assay employed by Heffner et al (1980), namely high pressure liquid chromatography coupled with electrochemical detection. In addition, the analysis was extended to the ingestion of other types of substances, to investigate whether different patterns of dopaminergic activity would be produced by feeding with different moter, incentive and post-ingestional consequences. - 40 -EXPERIMENT 1 Weingarten (1983, 1984) has developed an elegant technique to investigate responses to food-related stimuli. Rats are housed in individual sound-attenuating chambers, with water available at all times. During a conditioning phase the only food the animals receive is delivered as six discrete meals per day, each consisting of an 8ml portion of liquid diet, delivered to a feeding niche at one end of the cage. These meals provide the rats with as much food as they would normally consume during free feeding. Presentation of the meal coincides with the last minute of an extended conditional stimulus (CS+) consisting of a buzzer-light combination. By the end of an 11 day training phase the animals reliably enter the feeding niche with a short latency once meal begins. More interestingly, the animals spend much of the CS+ period in the niche, and as soon as the meal is delivered they consume it. After training, a test phase commences in which the animals have unlimited free access to food at a site distinct from the feeding niche. The free food is identical to that used in training. One CS+ trial is given per day. As in the training phase the animals approach the niche and consume the food when it is delivered. The presence of free food ensures that the animals are not eating in response to a nutritional deficit. Instead, the feeding appears to be a conditional response (CR) evoked by the CS+. In terms of the incentive-response hypothesis described above the CS+ should, by acting as a food-related incentive stimulus, cause an increase in dopamine release, resulting in approach of the feeding niche. Weingarten's (1984) conditioned feeding paradigm was adopted with only minor modifications to investigate this possibility . Animals received all their daily food rations as six - 41 -discrete meals which were signalled, prior to delivery, by a conditional stimulus (CS+). Once the animals were responding reliably to the CS+ by approaching the food source they were given one of three doses of the dopaminergic antagonist, pimozide. Attenuation of responding was attributed to interference with dopaminergic transmission. METHOD Subjects: A l l rats used in this study were hooded male rats obatained from Charles River Laboratories of Canada, and weighed 350-480g at the start of the experiment. Apparatus: Conditioning was conducted in 8 individual sound attenuating chambers (modified Coleman "Sno-Lite Low Boy" coolers). Each chamber was fitted with a dim houselight which was illuminated from 6am to 6pm, plus a fan for air circulation and to provide background white noise. Contained within each outer chamber was a smaller plexiglas compartment (29 x 23 x 18cm) in which a rat was housed. Below this compartment was a bed of Sani-cel, seperated from the rat by a wire grid floor. Two holes were drilled in one end wall of the compartment and a Richter tube containing tap water was always available to the rat in one of these. A feeding niche was recessed into the opposite wall. The niche was 5 X 7cm and had a 1cm high lip. Entry into the niche was detected by the interruption of a photocell. For conditioning, a buzzer, a tone generator and a bright cue light were mounted on the ceiling of the chamber, outside the plexiglas compartment. The cue light and tone were located above and to the left side of the feeding niche, while the buzzer and houselight were located at the opposite end. A peristaltic pump (Cole-Parmer) delivered liquid diet (Sustacal, Mead Johnson) into the rear of the feeding niche via silastic tubing. - 42 -A l l cues, the pump and the photocells were connected to an interface under the control of a Nova 3 computer with Manx software. Procedure: Prior to conditioning all rats were pre-exposed to the liquid diet so that neophobia would not interfere with conditioning, and to ensure that each rat found the diet palatable. Standard lab chow and water were removed overnight and each rat was given 40ml of liquid diet. Any rat which did not consume the diet overnight received a similar 40ml portion the next day, plus water. Any rat which did not consume the diet by the second day (about 5-10% of rats) was discarded from the experiment. Free food and water were returned to the rats following pre-exposure, but food was removed the day prior to the onset of conditioning. During conditioning each rat was housed continuously in an individual test compartment. The chambers remained closed and undisturbed except for a brief daily mantainance period. CS+ trials consisted of a 210sec buzzer-light compound. An 8ml portion of liquid diet was delivered into the feeding niche coincident with the final 60sec of the CS+ period. Thus, the CS+ alone period lasted 150sec. CS-trials consisted of a 210sec tone which did not signal any other event. CS+ and CS- trails alternated, and were presented on a quasi-random schedule with a mean inter-trial interval of 2h. Thus, each rat received apporoximately 6 CS+ trials per day, for a total volume of 48ml liquid diet. This was enough food for some rats to maintain a constant body weight, though many lost up to 10% of their free feeding weight over the conditioning period. The conditioning phase lasted 9-15 days. Some variability was caused by occasional equipment failures. Preliminary experiments indicated that all rats would reliably eat in rsponse to the CS+ after this period, even if food was made freely available in a Richter tube. However, in order to study the effect of - 43 -pimozide on conditioned appetitive responding, such responding was maximized by testing the rats in the absence of free food. Even so, some animals did not reliably spend appreciable amounts of time in the feeding niche, the operational definition of appetitive responding. Videotaped observation of some rats indicated that they would spend much of their time in other, apparently appetitive behaviours, such as orienting to the light, grooming, and running around the compartment. As the only response which could be identified during drug tests was time spent in the niche it was necessary to exclude any animal not spending at least 5sec in the niche during the CS+ period, prior to the delivery of food, on the final baseline day of testing. This resulted in the exclusion of 10 of 28 animals subjected to conditioning. Most animals rarely entered the niche at all during CS- trials. However, some responding did occur, and was occasionally quite pronounced. No attempt was made to investigate the effect of pimozide on this behaviour. Drug testing was always conducted during the first CS+ trial after the onset of the dark phase of the cycle ie., between 1800 and 2200h. (Some, but not all, of the rats displayed diurnal variance in their responses, and this time appeared to be near the maximal peak for most of these.) Animals were randomly assigned to receive 0.2, 0.4 and 0.6mg/kg pimozide or it's vehicle (0.6% tartaric acid). A l l injections were administered 4h prior to the test CS+, at least 40min after the previous trial. No CS- occurred in the interim. Following pimozide tests rats were sacrificed in a carbon dioxide chamber. Rats receiving vehicle injections were given 1-2 days of additional baseline training and were then tested again at one of the doses of pimozide. A total of 6 rats were tested at each level of drug dose and vehicle. - 44 -Statistical Analysis: Because of large differences in the quantity of baseline responding observed between different rats, the response of each rat while drugged was compared to his response to the first CS+ of the previous evening. The latency of the response was determined as an index of response onset. Rats not entering the feeding niche prior to the onset of food delivery were assigned latency scores of 150sec. Number of entries into the feeding niche and the total area under the cumulative response curve were used as global measures of response intensity, while frequency of entry following the first entry was determined as an index of response intensity following response initiation. A l l measures were analyzed using two-way analysis of variance (ANOVA) with day (baseline day vs test day) and level of pimozide dose as the two factors. When significant effects were observed post-hoc analyses were conducted using the Newman-Kuel's test, with the alpha level arbitrarily set to 0.05. RESULTS The effect of different doses of pimozide and its vehicle on responding during the CS+ period can is can be seen by examining the cumulative response diagrams of Figures 1 through 4. In each figure, the response curve for the animals during the drug test are shown in the lower diagram (B). The responses of the same rats on the control trial, that is the first of the previous evening, are shown in the upper half (A). It is evident from the figures that the animals receiving 0.4 or 0.6mg/kg pimozide were much less responsive to the CS+ in the period prior to food delivery. This was confirmed statistically. The latencies of the first entry into the feeding niche are presented in Figure 5. The ANOVA indicated that there were significant effects of baseline vs - 45 -Figure 1: Cumulative time spent in niche during CS+ period by animals receiving 0.6% tartaric acid vehicle on test day. A: Baseline trial of evening prior to test trial. B: Test trial, four hours after injection of vehicle. - 46 -Figure 2: Cumulative time spent in niche during CS+ period by animals receiving 0.2mg/kg pimozide on test day. A: Baseline trial of evening prior to test trial. B: Test trial, four hours after injection of pimozide. - 48 -A. TRIAL TIME / s e c . - 49 -Figure 3: Cumulative time spent in niche during CS+ period by animals receiving 0.4mg/kg pimozide on test day. A: Baseline trial of evening prior to test trial. B: Test trial, four hours after injection of pimozide. - 50 -A 140 H TRIAL TIME /sec . - 51 -Figure 4: Cumulative time spent in niche during CS+ period by animals receiving 0.6mg/kg pimozide on test day. A: Baseline trial of evening prior to test trial. B: Test trial, four hours after injection of pimozide. - 52 -A. ONSET OF FOOD DELIVERY TRIAL TIME / s e c . - 53 -Figure 5: Mean latency to enter niche following CS+ onset. Dark bars represent baseline trial of evening prior to test trial. Light bars represent test trial, four hours after injection of pimozide or vehicle. Vertical lines represent standard error of the mean for six rats. Figures along lower axis represent pimozide doses in mg/kg. - 54 -Latency / sec CM > 0> CD O cn o Cn o cn O cn _l l l I I I I Figure 6: Mean number of nosepokes during CS+ period prior to onset of food delivery. Dark bars represent baseline trial of evening prior to test trial. Cross-hatched bars represent test trial, four hours after injection of pimozide or vehicle. Vertical lines represent standard error of the mean for six rats. Figures along lower axis represent pimozide doses in mg/kg. - 56 -40 35-30-- 57 -Figure 7: Mean area under the cumulative response curve during the CS+ period prior to onset of food delivery. Square root of area is shown. Dark bars represent baseline trial of evening prior to test trial. Cross-hatched bars represent test trial, four hours after injection of pimozide or vehicle. Vertical lines represent standard error of the mean for six rats. Figures along lower axis represent pimozide doses in mg/kg. - 58 -Square Root of Area Under Curve / sec cn _L_ ho o cn (j4 O _ l _ cn _ l _ o [ Figure 8: Mean frequency of niche entry during the CS+ period, after first entry, but prior to onset of food delivery. Dark bars represent baseline trial of evening prior to test trial. Cross-hatched bars represent test trial, four hours after injection of pimozide or vehicle. Vertical lines represent standard error of the mean for six rats. Figures along lower axis represent pimozide doses in mg/kg. - 60 -0.32 - i 0.28-- 61 -test day (F(l,20)=8.94, £<.01) and level of drug (F(3,20)=4.19, E<05), plus a significant drug x day interaction (F(3,20)=3.51, p_<.05). Post-hoc tests revealed that latencies were significantly longer following injection of 0.4mg/kg pimozide than in the vehicle and 0.2mg/kg conditions. The 0.6mg/kg dose did not cause a further increase in latency relative to baseline, over the effect observed with a dose of 0.4mg/kg. Rats were not only slower to initiate entery into the feeding niche following pimozide treatment, they also entered it less frequently. The number of entries into the niche prior to onset of food delivery is shown in Figure 6. The ANOVA again indicated a significant day effect (F(l,20)=19.46, E < n 0 1 ) and a significant dose x day interaction (F(3,20)=5.79, p_<.01). Post-hoc tests revealed that 0.4 and 0.6mg/kg significantly decreased the number of niche entries, relative to baseline controls. The area under the cumulative response curve prior to food delivery indicates the total amount of appetitive responding. The square rooot of the area was determined to restore linearity to the function, and is illustrated in Figure 7. The ANOVA indicated a strong effect of day (F(l,20)=24.96, £<.0001) and a significant day x dose interaction (F(3,20)=4.40, p_<.05). Post-hoc test revealed that 0.4 and 0.6mg/kg pimozide decreased responding relative to baseline. Measures of total responses are influenced by the latency to begin responding. Thus, the decreased responding observed could simply reflect slower response initiation, without any changes in responding after initiation occured. However, Figure 8 reveals that the frequency of entries into the feeeding niche, in the period between the first entry and the onset of food delivery, was also decreased by the high doses of pimozide. The ANOVA conducted on these data excluded two rats from the 0.4mg/kg test and one from the 0.6mg/kg test, as - 62 -these animals did not respond at all prior to food delivery. A significant effect effect of day was observed (F(l,17)=77.04, p_<.0001), as well as a significant dose x day interaction (F(3,17)=84.25, E<.0001). Both 0.4 and 0.6mg/kg pimozide decreased response frequency by rats in this period, relative to their performance on the baseline day. DISCUSSION This experiment indicated that dopaminergic transmission is necessary for the production of appetitive responses. Entry into the feeding niche where a signalled meal was about to be delivered was attenuatd by 0.4 and 0.6mg/kg pimozide. Under these drug conditions rats took longer to initiate responding. Once they began, the magnitude of these responses was diminished. Despite the clarity of these effects, some caution must be exercised in the interpretation of these data. In particular, i t is necessary to question the validity of niche-entering as a measure of appetitive responding. Decreases similar to those observed could occur if, for example, the animals spent increased amounts of time grooming, orienting to the CS+ itself, or began drinking in response to the CS+. Therefore, the effect of pimozide on conditioned appetitive behaviour was directly examined by videotaping the responses of three other rats, not included in the present study, both while they were undrugged and again after injections of pimozide. This analysis indicated that pimozide made the animals extremely unresponsive. Their undrugged response to the CS+ was very intense and had a very short latency. They would explore the niche, the compartment and the cue light. They groomed and ran around. In contrast, under the influence of pimozide they lay motionless for most of the CS+ period. Only slowly did they rouse themselves to stand, to groom, and to make their way 63 -slowly across the chamber to the feeding niche. Once there, they would remain in the niche and eat the signalled meal in an apparently normal manner. These observations lend credibilty to the use of niche-entry as an index of appetitive responding. Future studies employing video analyses more extensively will be conducted to examine the nature of the appetitive deficit more closely. The deficits observed in responses to the CS+ were greatly reduced by the onset of food delivery. As may be seen clearly in Figure 3, following 0.4mg/kg pimozide 4 of 6 rats spent virtually all (>58sec) of the 60sec food delivery (UCS) period in the feeding niche, apparently consuming food. These four rats include one which had only entered the niche once in the previous 150sec. Two rats never entered the niche prior to food delivery, yet when they did enter it (5.43 and 48.22sec later), they remained there for the remainder of the food delivery period without interuption. Figure 6 illustrates that even though all rats exhibited attenuated reactions to the CS+ when they were adminstered 0.6mg/kg pimozide, three subjects spent >56sec of the food delivery period in the niche. As with 0.4mg/kg, all rats remained in the niche almost constantly once they entered it during the food delivery period. Thus, although appetitive reactions were markedly reduced by 0.4 and 0.6mg/kg pimozide, consummatory reactions were apparently not disrupted. This suggests that dopamine systems may be exclusively involved in the appetitive aspects of feeding. That is, dopamine may only be involved in feeding behaviour when an animal is distant from the food, or must perform some response to get it, and may not be required for the consumption of food once the animal is in direct contact with it. This issue will be addressed further in Experiment 3, but first a more direct examination of dopaminergic activity during conditioned appetitive responding will be presented. - 64 -EXPERIMENT 2 The first experiment demonstrated that a dopaminergic substrate is necessary for the production of the appetitive reactions observed when a rat is presented with a CS+ in the conditioned feeding paradigm. However, the experiment did not directly address the question of whether the substrate was actively involved in the production of the appetitive reactions. That is, although the results of the experiment indicated that dopamine was required for appetitive responses to occur, it is not clear whether the response was actually a result of increased dopaminergic activity or whether it was simply necessary that dopaminergic systems be functional for the animal to be capable of performing such responses. Further, the use of a peripherally adminstrated pharmacological agent such as pimozide, which disrupts dopaminergic transmission at all dopamine receptor sites in the brain, did not permit the identification of which dopaminergic systems are involved in these responses. The second experiment was designed to shed light on these two issues through direct neurochemical investigation of dopaminergic activity. The level of metabolites of dopamine in brain tissue relative to the level of the parent molecule' provides an index of the release of dopamine at the time of sacrifice. (La Vielle, Tassin, Thierry, Blanc, Hevre, Barthelemy and Glowinski, 1978). Thus, it is possible to test a major prediction of the incentive-response hypothesis, namely that presentation of an incentive stimulus should result in increased forebrain dopamine turnover. An increase in turnover would be consistent with the incentive-response hypothesis, while a negative finding would be consistent with a suggestion that only basal levels of dopamine release are necessary for appetitive responding to occur. - 65 -Measurement of dopamine and its metabolites also provides an indication of which dopaminergic system or systems are involved in appetitive responding. Only those regions in which dopamine turnover is increased are implicated as playing an active role in such responding. Accordingly, tissue from three different dopamine projection sites was analyzed (refer to Fallon and Moore, 1978, for a detailed description of the anatomy of the dopaminergic innervation of the forebrain). The first of these was the striatum, the brain region with the heaviest dopaminergic innervation. This innervation originates in the substantia nigra. Together, the nigra and the striatum are major components of the extrapyramidal motor system. In addition, analyses were performed on tissue from two limbic sites receiving dopaminergic projections from the ventral tegmental area, namely the n. accumbens and the olfactory tubercle. METHOD Materials and Apparatus: Conditioning chambers, diet and type of rats used in Experiment 2 were the same as those in Experiment 1. Procedure: The conditioning procedure was the same as in Experiment 1 except that the CS+ duration was increased to 630sec, with liquid diet again being delivered in the final 60sec. This longer CS+ period was used in an attempt to maximize any induced neurochemical changes. Conditioning lasted 8-9 days. 16 rats were conditioned, and of these the 14 which spent the greatest amount of time in the feeding niche during CS+ delivery were assigned to either the "CS+" group or the "control" group. Each CS+ rat was exposed to the CS+ for 4min before being removed from the chamber and sacrificed by cervical fracture. The brain was removed, mounted on a microtome and frozen. Slices containing the neostriatum, n. accumbens and olfactory tubercle were placed on - 66 -ice. These structures were then bilaterally dissected from the slices and prepared for analysis of dopamine and its metabolites 3,4dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) by high pressure liquid chromatography with electrochemical detection (HPLC-ED), as described by Jakubovic, Lin and Fibiger (in press). Control rats were treated identically to those receiving exposure to the CS+ except that they were removed from the chamber just prior to CS+ onset. Statistical Analysis: Levels of dopamine and the two metabolites were measured in terms of ug/g wet tissue, and the ratios of the levels of the metabolites to dopamine were computed. These five measures were each subjected to a two way ANOVA. In addition to the group factor (CS+ ys control), hemisphere was treated as a factor, as both left and right samples were taken from each of the brain regions. RESULTS The ratio of DOPAC to dopamine in the three brain regions is shown in Figure 9. The DOPAC/DA ratio was significantly increased in the n. accumbens, indicative of increased dopamine turnover (F(l,7)=6.678, £<.05). The ratio was increased from .325 to .389, a 20% increase. An apparantly similar 20% increase in the striatum was not statistically significant (F(l,7)=3.036, n.s.). No significant changes in DOPAC/DA ratio were observed in the olfactory tubercle. No significant hemisphere effects were observed in any region. Ratios of HVA to dopamine are displayed in Figure 10. Comparison of this figure with Figure 9 suggests that increases HVA/DA ratio occurred which were - 67 -Figure 9: DOPAC/DA ratios in the three brain regions analyszed, after rats had been exposed to the CS+ for four minutes. Cross-hatched bars represent left hemisphere tissue. Dark bars represent right. Vertical lines represent standard error of the mean for seven rats. - 68 -STRIATUM ACCUMBENS O. TUBERCLE Figure 10: HVA/DA ratios in the three brain regions analyszed, after rats had been exposed to the CS+ for four minutes. Cross-hatched bars represent left hemisphere tissue. Dark bars represent right. Vertical lines represent standard error of the mean for seven rats. - 70 -STRIATUM ACCUMBENS O. TUBERCLE 0.12n - i - i Table I. Concentrations of dopamine, DOPAC and HVA in various brain regions of rats sacrificed before or after exposure to the CS+ DOPAMINE DOPAC HVA Left Right Left Right Left Right STRIATUM Controls 11.343 11.671 * 2.839 2.894 0.837 0.858 0.412 0.356 .183 .156 .053 .073 CS+ 10.643 t 10.971 *+ 3.077 3.213 0.963 0.936 0.302 0.249 .224 .204 .096 .090 ACCUMBENS Control 0.825 0.814 2.702 2.602 0.686 0.652 .022 .033 .094 .132 .032 .035 CS+ 0.810 0.708 * t 3.120 * 2.799t 0.798 t 0.677= .012 .016 .197 .170 .077 .049 OLF. TUBERCLE Controls 4.457 4.567 2.055 2.124 0.376 0.356 .330 .367 .241 .209 .033 0.026 CS+ 4.256 4.324 1.842 2.042 0.352 0.386 .396 .393 .126 .183 .025 .031 Values are expressed as mean ug/g wet tissue in each hemisphere. Lower figure represents standard error of the mean of seven rats. * - differs from contralateral hemisphere, p < .05. t -- differs from ipsilateral control, p < .05. - 72 similar to the significant increase observed in the DOPAC/DA ratio in the n. accumbens, however these were not statistically significant (striatum: F(l , 7 ) = 2 . 3 3 1 , n.s.; n. accumbens: F(l , 7 )=3 .732 , £<.10). No increase is apparant in the olfactory tubercle (F<1.0). Again, no significant hemisphere effects were observed. Absolute levels of dopamine, DOPAC and HVA are indicated in Table 1. No significant main effects of treatment were observed for any chemical in any brain region. There were, surprisingly, strong hemisphere effects, particularly in the n. accumbens. There, higher levels of dopamine were found in the left hemisphere (F(l , 12)=17 .059 , E< 005), while higher levels of HVA (F(l , 12)=6 .369 , E<-05) and DOPAC (F(l , 12)=11 .551 , £<.01) were observed in the right hemisphere. As well, dopamine levels were elevated in the right striatum (F(l , 12)=17 .058 , E<.005), while DOPAC levels were higher in the right olfactory tubercle (F(l , 12 )=5 .703 , E<-05). Examination of Table 1 suggests that the hemisphere differences are more pronounced in the treated animals. However, the only significant hemisphere x treatment interaction was for dopamine levels in the n. accumbens (F(l , 12)=11 .083 , E ^ O D . which reflects a lower dopamine level in the right hemisphere of the CS+ animals. DISCUSSION When rats were exposed to a CS+ which signalled the imminent onset of food delivery for four minutes there was a 20% increase in dopamine turnover in some forebrain regions. This increase in release is in agreement with the prediction of the incentive-response hypothesis thatexposure to an incentive stimulus should result in an increase in forebrain dopamine turnover. Such release - 73 -may occur to naturally attractive food related stimuli (eg., odours) and, as the current results show, are released as a conditional response (CR) to an arbitrary simulus, such as a buzzer or a light. Together with the results of Experiment 1, that indicated that dopamine receptor blockade interfered with appetitive responding, the result of this experiment is consistent with the suggestion that increased dopamine release in reponse to incentive stimuli dopamine release is a critical component of appetitive responding. Although the effect was not statistically robust, in the case of the n. accumbens it is reasonably reliable, because there was a significant increase in DOPAC to dopamine ratio. An increase of apparently equal magnitude occurred in the striatum, certainly there is no evidence that the increase in release in the n. acumbens was greater than that in the striatum. Thus, the results do not help in clarifying the relative roles of the mesolimbic and nigrostriatal dopamine systems in appetitive responding. It appears that any increase in dopamine turnover in the olfactory tubercle is not as large as that observed in the n. accumbens. An increase of twenty percent in dopamine turnover is quite remarkable, given the relatively brief period for which the CS+ was presented. Other studies examining changes in dopamine turnover as a result of behavioural activity often employ a much longer period. For example, in their studies of dopamine release accompanying reinforced circling, Yamato and Freed (1982, 1984) had their animals perform for twently minutes prior to sacrifice. The increase after only four minutes appears quite substantial in this context. The differences in absolute levels of dopamine and its metabolites in the two hemispheres was a complete surprise. Hemispheric differences in dopamine or metabolite levels are often reported in individual rats, but these are not consistent across individuals (jerussi and Glick, 1976; Jerussi and Taylor, 1982). - 74 -That is, although some asymmetry is to be expected in a given rat, levels should not be consistently higher in right vs left accumbens across animals. These effects could be a spurious result of experimental procedure. Because hemispheric differences were not expected, no procedures were employed to control for them. Specifically, the left hemisphere tissues were always dissected out and homogenized before their right hemisphere counterparts, and were analyzed first in the HPLC columns. Thus, differing amounts of degredation could have occured in the samples prior to analysis. This explanation is not very satisfactory, particularly in light of the finding that dopamine levels were higher in the right stiatum and the left accumbens, while DOPAC levels were higher in the left accumbens and right olfactory tubercle. One possible interpretation of the data is that there is a bilateral increase n dopamine turnover but that in one hemisphere only (ie. the left n. accumbens) there is a compensatory increase in dopamine synthesis. - 75 -EXPERIMENT 3 The results of the first two experiments are consistent with the assertion of the incentive-response hypothesis that appetitive responding to an incentive stimulus is dependent on dopaminergic activity. However, other intrepretations of the results of these experiments are possible. For example, in Experiment 1 the decrease in appetitive responding observed following dopaminergic receptor blockade may not be specific to appetitive responding at all, but may merely reflect a general disruption in feeding behaviour, global motor deficits, or an attenuation of "hunger". The apparent consumption of food in that experiment argues against this interpretation, but in the absence of quantitative data it cannot be ruled out. Consequently, Experiment 3 examined the effects of pimozide, at the same doses used in Experiment 1, on the consumption of the same liquid diet which was used in the earlier experiment. METHOD Subjects and apparatus: 18 male rats of the same strain used in Experiments 1 and 2, and of similar size, were used in this experiment. Liquid diet (Sustacal) was presented to the subjects in 50ml Nalgene tubes fitted with drinking spouts. Procedure: Rats were familiarized with a feeding regime consisting of a 20min period of access to liquid diet at about the same time each day, followed by a one hour break, and then a one hour period of ad lib. access to food pellets. Water was available at all times except during the liquid diet period. After the animals were reliably eating a constant volume of liquid diet - 76 -(17.6+0.6ml) in the 20min period, that is after nine days, they were assigned to three groups of six rats each, approximately equated for consumption. Drug testing was conducted on two days, separated by one baseline day. On the first day half of the rats in each group received pimozide, while the other half received 0.6% tartaric acid vehicle. On the second test day injection conditions were reversed and the rats which had received pimozide were given vehicle, and vice versa. Rats from the three different groups received 0.2, 0.4 or 0.6mg/kg pimozide subcutaneously, 4h prior to liquid diet delivery, on their test day. Statisitical analysis: The results were analyzed using a two-way ANOVA. Drug vs vehicle administration constituted one factor, while dose of drug (0.2, 0.4 and 0.6mg/kg) constituted the second. RESULTS Examination of Figure 11 indicates that no dose of pimozide had any noticable effect on home cage consumption of liquid diet. This impression is confirmed by the ANOVA, which showed no significant effect of drug or drug dose, and no significant interaction (F<1 in all cases). DISCUSSION In this experiment no attenuation of feeding was seen even with 0.6mg/kg pimozide. This is in marked contrast to the strong suppressant effects on conditioned appetitive responding seen in Experiment 1, even with 0.4mg/kg. This result is contrary to those of Heffner et al (1977), Blundell and Latham (1978), - 77 -Figure 11: Consumption of liquid diet in twenty minute session four hours after injection of pimozide or vehicle. Cross-hatched bars represent consumption following vehicle injection. Light bars represent right consumption by the same rats following pimozide injection. Vertical lines represent standard error of the mean for six rats. Figures along lower axis represent dose of pimozide in mg/kg. - 78 -28 Tombaugh et al (1979) and Wise and Colle (1984), all of whom observed suppression of food intake by comparable doses of neuroleptics. Nor were the results of this study similar to those of Lawson et al (1984) or Franklin (personal communication), who observed increased food consumption following administration of pimozide. These discrepancies may be due to differences in drugs, injection procedures, food, deprivation levels, familiarity of feeding regime or other factors. For present purposes the critical fact is that doses of pimozide which strongly suppressed conditioned appetitive responses in Experiment 1 had no discernable effect on food consumption in the present experiment. This was true even though the two experiments employed the same food, the same drug vehicle, the same injection procedure, and rats of the same stock. In each case the rats had been on restricted feeding regimes for about a week prior to the pimozide test. Clearly, conditioned appetitive responses are much more vulnerable than are consummatory responses to the disruptive effects of dopamine receptor blockade. - 80 -EXPERIMENT 4 The previous experiment demonstrated that the doses of pimozide that seriously disrupt appetitive responding for cues related to food delivery do not diminish consummatory responding for the same food. Thus, dopamine release does not play an active role in the production of consummatory feeding behaviour as measured by amount of food consumed, allthough a certain minimal amount of dopamine activity (ie., levels above those seen following NSB lesions) may be neccessary to permit consumption to occur, the quantity of dopamine release in unlesioned animals is unlikely to ever fluctuate low enough to interfere with consummatory behaviour. In contrast to this analysis, Heffner et al (1980) interpreted the increased dopamine turnover they observed one hour after food delivery as evidence that dopamine was released during feeding. The lack of effect of blocking post-synaptic dopamine receptors during a feeding session suggests that such release has no immediate impact on consummatory behaviour. Therefore, it is hard to understand why such release would occur. Heffner et al (1980) claim that the observed increase in dopamine turnover must have been due to the act of feeding itself, because increases were not observed following intubation of food. However, the tube-fed and self-fed treatments differ in several important ways. In addition to not going through the motor patterns of feeding, the tube-fed animals do not taste the food and may not experience the post-ingestive effects of feeding in the same way as self-fed animals. Experiment 4 was designed to extend the analysis of the impact of feeding on dopamine turnover by examining the impact of manipulating motoric and nutritive aspects of this feeding. Thus, while some animals ate solid food 81 pellets, similar to those used by Heffner et al, others consumed the liquid diet used in Experiments 1 through 3. Such consumption does not involve gnawing or chewing, but is dependent on a somewhat different set of tongue and mouth movements. Another group of rats consumed comparable quantities of a saccharin solution. Rats in the saccharin group will have performed motor actions similar to those displayed by rats consuming liquid diet. Those rats that consume saccharin will also have experienced strong, if different, taste sensations. Consumption of saccharin solution may also be confidently described as "rewarding", because rats avidly consume this solution in much much greater quantities than water. However, although saccharin is sweet it is a metabolically inert substance, so its' consumption produces no direct post-ingestive effects. This fact makes consumption of saccharin solution a useful technique for separating the post-ingestive consequences of feeding from the motoric and reward properties of feeding itself. METHOD Subjects and Apparatus: 30 male rats were used which were similar to those used in the other experiments. Liquid diet (Susatcal) and 0.4% saccharin (Fisher) solution were delivered in Nalgene tubes, as in Experiment 3. Standard Purina lab pellets were available to the rats in food hoppers mounted on the outside of the cage. Procedure: The procedure employed was derived from that of Heffner et §1 (1980). Rats were randomly assigned to five groups (n=6 for each group). Four groups were adapted for a week to a feeding schedule in which they received food for only 4h/day in their home cages. The feeding period occurred near the middle of the 12h lights-on period in the colony. Times were slightly varied to - 82 -prevent the formation of strong associations between feeding and temporal cues. A fifth group of rats (CA) had continuous free access to food. Water was freely available to all rats at all times. Two groups, the food deprived group and the pellet group (FD and Pe) received standard pellets in each 4h feeding period. One group (LD) received liquid diet during the 4h feeding period on the fourth day, and in the first hour on the eighth and twelfth days, followed by 3h of pellets. On all other days they received 4h of pellets. Rats in the Sa group were treated exactly like those in group LD, except that 0.4% saccharin solution was substituted for liquid diet. This procedure ensured that rats in the four restricted feeding groups were maintained at nearly equivalent levls of deprivation, while it allowed the rats in groups Sa and LD to become familiar with these other solutions. On the test day, the fifteenth overall, all the rats were sacrificed and their brains were removed. Rats in group FD were sacrificed when they had been deprived of food for about 20h, before food was delivered on the test day. Care was taken to avoid exposing these rats to odours or sounds signalling food delivery- Rats in groups Pe, LD and Sa were sacrificed l h after they had been given pellets, liquid diet and saccharin solution, respectively. Rats in group CA, which had continuous access to food, were sacrificed at similar times. Note that these rats were unlikely to have consumed much food immediately prior to sacrifice, because this occured in the first half of the lights on period. Once the the brain was removed, it was dissected, homogenized and subsequently analyzed as described in Experiment 2. Data Analysis: Two-way ANOVAs (treatment by hemisphere) were conducted for dopamine, both metabolites and each ratio in each of the three brain regions. - 83 -As in Experiment 2 Newman-Keuls post-hoc tests, with alpha arbitraily set to 0.05, were conducted when the ANOVA indicated a significant effect. RESULTS In the hour in which food was available the LD rats consumed 24.25 + 0.75ml (mean + s.e.m.) of liquid diet, while the Sa rats consumed 27.60 + 1.86ml of saccharin solution. Pellet intake was not measured for group Pe, but as with Heffner et al (1980) it was apparant that most food intake by this group occurrred in the first hour food was available. The ratios of HVA to dopamine in the three brain region are shown in Figure 12. ANOVA indicated that in the striatum there was a main effect of treatment (F(4,25)=3.538, £<.05) but no significant effect of hemisphere (F(l,25)=1.167, n.s.) and no significant hemisphere by treatment interaction (F<1). Post-hoc Newman-Kuels tests only indicated that the HVA/DA ratio was higher in the Pe group than in the FD group. In the n. accumbens there was also a significant treatment effect (F(4,25)=6.209, £<.005) on HVA/DA ratio, but no significant hemisphere effect (F<1). There was also a significant treatment by hemisphere interaction (F(4,25)=2.715, £=.05). Post-hoc tests indicate that the Pe and LD groups had significantly higher HVA/DA ratios than any other groups. As well, the tests indiated that the levels were higher in the right n. accumbens than in any hemisphere of any other group, and were higher than in the contralateral hemisphere of the the same animals. No main effects on HVA/DA ratio were observed in the olfactory tubercle (Fs<l), and there was no significant treatment by hemisphere interaction (F(4,25)=2.623, n.s.). - 84 -Figure 12: HVA/DA ratios in the three brain regions analysed, one hour after various feeding treatments. CA: Continuous access to food. FD: Food deprived (no food for 20h). Pe: Had access to standard food pellets for l h prior to sacrifice. LD: Had access to liquid diet for l h prior to sacrifice. Sa: Had access to 0.4% saccharin solution for l h prior to sacrifice. Cross-hatched bars represent left hemisphere tissue. Dark bars represent right. Vertical lines represent standard error of the mean for six rats. - 85 -STRIATUM 0.18-1 0.16-0.14-0V ^ # ACCUMBENS 0. TUBERCLE 0v 45 § 0v <P <c* # Figure 13: DOPAC/DA ratios in the three brain regions analysed, one hour after various feeding treatments. CA: Continuous access to food. FD: Food deprived (no food for 20h). Pe: Had access to standard food pellets for l h prior to sacrifice. LD: Had access to liquid diet for l h prior to sacrifice. Sa: Had access to 0.4% saccharin solution for l h prior to sacrifice. Cross-hatched bars represent left hemisphere tissue. Dark bars represent right. Vertical lines represent standard error of the mean for six rats. - 87 -ACCUMBENS 0. TUBERCLE 0.8 n cJf <? $ # cj <? $ e? Table II. Concentrations of dopamine, DOPAC and HVA in various brain regions of rats exposed to various feeding conditions.  DOPAMINE HVA DOPAC Left Right Left Right Left Right STRIATUM Continuous Access 11.425 11.833 a 1.111 1.197 a 2.281 2.455 a 0.560 0.470 .072 .068 .099 .068 Food Deprived 11.408 12.750 1.081 1.206 2.194 2.429 * 0.651 0.530 .090 .061 .120 .104 Pellets 11.383 11.550 1.393 1.438 2.164 2.372 0.370 0.370 .056 .096 .047 .064 Liquid Diet 10.958 10.539 1.263 1.343 b 2.385 2.714 *' 0.449 0.436 .073 .055 .115 .072 Saccharin 10.539 11.000 1.033 1.100 2.064 2.274 0.435 0.253 .038 .053 .094 .051 ACCUMBENS Continuous Access 9.174 8.539 a 0.989 0.898 2.575 2.507 .521 .438 .068 .057 .225 .188 Food Deprived 9.113 8.858 1.026 0.915 2.513 2.503 .428 .267 .048 .060 .122 .114 Pellets 9.154 8.677 1.231 1.159 b 2.704 2.564 .304 .354 .074 .074 .137 .102 Liquid Diet 9.248 8.677 1.218 1.311 b 2.888 3.100 b .309 .314 .063 .120 .163 .071 Saccharin 8.634 8.192 0.895 0.845 2.475 2.465 .179 .315 .055 .062 .071 .051 OLF. TUBERCLE Continuous Access 4.152 3.724 a 0.529 0.538 2.356 2.230 a 0.225 0.290 .069 .042 .326 .294 Food Deprived 3.641 4.186 0.506 .562 1.853 2.206 * 0.346 0.322 .053 .047 .250 .226 Pellets 3.515 4.058 0.496 0.541 1.663 1.934 0.310 0.436 .059 .054 .323 .297 Liquid Diet 4.570 + 4.927 + 0.570 0.606 2.201 + 2.454 0.448 0.475 .023 .033 .276 .258 Saccharin 3.281 3.557 0.464 0.450 1.801 1.974 0.401 0.362 .050 .069 .314 .347 Values are expressed as mean ug/g wet tissue in each hemisphere. Lower figure represents standard error of the mean of six rats. * - differs from contralateral hemisphere, p < .05. t -- differs from food deprived rats for ipsilateral hemisphere, p < .05. a - main effect of hemisphere in brain region, p < .05. b -- treatment differs bilaterally from food deprived rats, p < .05. - 89 -DOPAC/DA ratios are shown in Figure 13. Treatment did not cause a significant change in DOPAC/DA ratios in the striatum, although there was a trend in this direction <F(4,25)=2.185, £<.l), apparantly due to slightly higher levels in the LD group. There was a significant hemisphere effect (F(l,25)=17.52, p_<.0005) with a higher DOPAC/DA ratio in the right striatum, but there was no significant hemisphere by treatment interaction (F<1). In the n. accumbens there was no significant main effect of treatment on DOPAC/DA ratios (F(4,25)=1.369, n.s.), however there was a significant effect of hemisphere (F(l,25)=15.597, £<.005) and a significant treatment by hemisphere interaction (F(4,25)=4.113, £<.05). The interaction reflects the elevated DOPAC/DA ratio in the right hemisphere of the LD animals which is higher than any observed in any hemisphere of any other group, and higher than in the contralateral hemisphere of the animals in the same group. The left hemisphere of the LD animals had higher ratios than those observed in the same hemisphere of the FD animals. Thus, liquid diet consumption increased DOPAC/DA ratios in this brain region, but preferentially in the right hemisphere. In the olfactory tubercle there was no main effect of treatment (F<1) and no treatment by hemisphere interaction (F<1), but the DOPAC/DA ratio was higher in the right hemisphere (F(l,25)=4.605, £<.05). Examination of Table 2 indicates that observed increases in HVA/DA and DOPAC/DA ratios are primarily attributable to increases in metabolite levels, as opposed to decreases in dopamine levels. In the n. accumbens and striatum dopamine levels were not altered by feeding conditions (Fs<l), but treatment did alter HVA levels in the n. accumbens (F(4,25)=8.885, £<.0001) and in the striatum (F(4,25)=5.707, £<.005). In the n. accumbens HVA levels were higher in the Pe - 90 -and LD groups than in rats of any other feeding treatment, while in the striatum HVA levels were higher in the Pe and LD groups than in the FD and Sa rats. There were also significant effects on DOPAC levels in both the n. accumbens and the striatum. In the n. accumbens (F(4,25)=3.050, p_<.05) levels were higher in the LD group than in any other, while in the striatum (F(4,25)=3.189, £<.05) DOPAC levels were higher in the LD group than in the Sa group. Although there were no significant treatment x hemisphere interactions in the n. accumbens or the striatum for dopamine, HVA or DOPAC levels, several main effects of hemisphere were observed. Dopamine levels were higher in the left n. accumbens (F(l,25)=8.986, £<.01) and in the right striatum (F(4,25)=13.531, £<.005). HVA levels not significantly different in the two hemispheres of the n. accumbens (F(l,25)=1.557, n.s.), but were higher in the left than the right striatum (F(l,25)=6.932, £<.0001). DOPAC levels did not differ between left and right n. accumbens (F<1), but were 10% higher in right than in left striatum (F(l,25)=51.319, £<.0001). One noteworthy effect was observed in the olfactory tubercle. Although there was no main effect of treatment on dopamine levels (F(4,25)=1.888, n.s.), there was a significant main effect of hemisphere (F(l,25)=7.603, £<.05) and a significant treatment by hemisphere interaction (F(4,25)=3.665, £<.05). Post hoc tests indicated that dopamine levels were elevated in the right hemisphere of LD animals relative to the same hemisphere of any other group. With the exception of group CA dopamine levels were higher in the left hemisphere of rats of the LD group than in the sinister hemisphere of rats in any other group. Thus, dopamine levels were elevated in the olfactory tubercle by consumption of the liquid diet. - 91 -Overall HVA and DOPAC levels were not altered by treatment in the olfactory tubercle (Fs<l). However, DOPAC levels were higher in the right hemisphere (F(l,25)=14.230, p_<.001) except in the CA group (interaction: F(4,25)=*2.843, £<.05). HVA levels were not significantly different between hemispheres, but there was a trend towards higher levels in the right (F(l,25)=3.287, E<1). DISCUSSION In agreement with Heffner et al (1980) the present study found increases in forebrain levels of dopamine metabolites relative to the level of dopamine one hour after rats were presented with food. This increase does not appear to be related specifically to the motor actions involved in the consumption of food pellets because such increases were even more pronounced in the brains of animals which had consumed a liquid diet. The changes observed in HVA/DA and DOPAC/DA ratios are even more pronounced in light of the lack of decrease in dopamine levels. Possibly dopamine synthesis was increased in the hour during which food was available. This suggestion is supported by the observed increases in dopamine levels in the olfactory tubercle of LD rats. Some differences with the results of the Heffner et al study should be noted. In particular, in agreement with the results of Chance et al (1985), increases generally larger than those occurring in the striatum were observed in the n. accumbens, whereas Heffner et al only observed increased dopamine turnover in the accumbens. These differences may have been due to the use of somewhat different portions of these regions. Heffner et al used much larger tissue samples. For example, the striatal samples were dissected out of slices - 92 -4.5mm thick, and weighed 68.0 +1.5mg (Heffner and Seiden, 1980). The samples used in the current experiment were taken out of thinner slices approximately 1.2mm thick, taken midway through the rostral-caudal extent of the structure, and weighed 12.2 +0.2 mg. An increase in dopamine turnover in the mid portion of the caudate in the Heffner et al (1980) study could have been obscured by the inclusion of other tissue. In marked contrast to the increases observed following consumption of liquid diet, dopamine turnover was not increased following consumption of similar volumes of saccharin solution. In the n. accumbens DOPAC/DA ratios were lower for Sa than in LD and Pe rats. Absolute DOPAC and HVA levels were also higher in LD than in Sa rats. In the striatum differences between LD and Sa animals were less pronounced, but HVA and DOPAC levels were both higher in the LD rats, and there was a trend towards higher DOPAC/DA ratios in the LD rats. The pronounced difference between dopamine turnover in rats which had consumed liquid diet and in those that had consumed saccharin cannot be attributed to differences in motor responding because the Sa rats consumed at least as much fluid from identical dispensers. Nor can the differences be readily accounted for by appealling to differences in the "reward value" between the two fluids. Although no attempt was made to equate the attractiveness of the foods through two bottle preference tests, other experiments conducted in this lab have demonstrated that 0.4% saccharin solution is consumed in considerably larger quantities than is water (Blackburn, Jacobs and Phillips, 1984). In addition, the fact that the rats consumed similar quantities of liquid diet and saccharin solution in the present experiment argues against the suggestion that the liquid diet is more attractive to the rats than is the saccharin solution. - 93 -These considerations lead to the conclusion that the different effects of saccharin and liquid diet on dopamine turnover must be the result of differing post-ingestive consequences of consuming these solutions. The data of Heffner et al (1984), which indicated that major increases in the level of dopamine turnover occurred in the second hour in which food was available, even though only a minimal amount of food was ingested, is consistent with the assertion that the increase in dopamine turnover was due to post-ingestive factors. The absence of any indication of increased dopamine turnover following ingestion of a non-nutritive substance firmly establishes the point. It is st i l l possible that some dopamine release occurs during feeding, but i f i t is, levels have returned to baseline within an hour. The results of the current experiment, along with those of Experiment 3, do not support the hypothesis that dopamine release is necessary for the performance of consummatory feeding behaviours. The post-ingestive increase in dopamine turnover elicited by pellets and liquid diet must reflect some post-ingestive physiological event. The brief increase in dopamine activity seen following the i.v. injection of CCK may, for some reason, be more pronounced when CCK is released from the gut following food consumption by freely moving animals. Alternatively, if CCK, glucose or some other post-ingestive factor decreases dopamine activity in the immediate post-prandial period, the increase observed one hour after food presentation could be a rebound effect, with dopamine activity increasing to above baseline levels following a period of inhibition. A more precise analysis of the time course of changes in dopamine turnover following feeding is necessary to settle this point. As in Experiment 2, the observed inter-hemispheric differences are difficult to interpret. Again, no steps were taken to control for different sacrifice to homogenization or homogenization to analysis intervals, but as in the earlier - 94 -experiment such explanations are insufficient to explain why dopamine levels are higher in the left striatum but are higher in the right n. accumbens and olfactory tubercle. In some cases the hemisphere differencs are more pronounced in the fed animals. For example, the difference between HVA/DA and DOPAC/DA levels which were observed between right and left n. accumbens were particularly pronounced in the LD animals. Conceivably this difference could be attributed to the fact that the Nalgene tubes which contained the liquid diet were mounted at the right side of the front of the cage (from the rat's perspective). Possibly the rats had a consistent postural asymmetry while feeding. Postural asymmetry has been related to asymmetrical dopamine release (Freed and Yamato, 1985), but posture would have to interact with nutrient content to explain why similar asymmetries are not observed in the DOPAC/DA and HVA/DA ratios of Sa rats. - 95 -GENERAL DISCUSSION The four experiments described above demonstrate that dopamine is differentially involved in different aspects of feeding behaviour. Appetitive feeding behaviour, indexed as approach to a feeding site in response to an incentive stimulus, appears to be dependent on the activation of a dopaminergic substrate. Disruption of dopaminergic transmission by post-synaptic receptor blockade severly attenuated these appetitive behaviours, while exposure to an incentive stimulus for four minutes was sufficient to cause increased dopamine turnover in the forebrain. In contrast, consummatory behaviour does not appear to be dopaminergically mediated: Food intake was not altered by the same doses of pimozide that disrupted appetitive behaviour. Dopamine release was not examined during the consummatory phase itself. However, subsequent to feeding, during the post-prandial period, increased dopamine turnover was observed in the n. accumbens and the striatum. This increase does not appear to have been a direct consequnce of having performed the consummatory behaviours, but rather appears to have been the result of some post-ingestive effect: Increased turnover was only obvserved in groups of rats which had consumed nutritive substances, not in those which had consumed a non-nutritive saccharin solution. The results of these experiments provide additional evidence that can be used to evaluate theoretical interpretations of dopaminergic invovement in behaviour. Experiments 1, 2 and 3 are particularly relevant to a comparison of the anhedonia hypothesis (Wise, 1982; 1985) and the incentive-response hypothesis outlined in the introduction. First, the two hypotheses are in agreement that the deficits observed following dopamine receptor blockade are motivational in nature, and are not purely motoric. The unattenuated feeding observed in Experiment 3, - 96 -plus the responding that occurred in Experiment 1 following delivery of food, rule out the possibility that the animals were seriously debilitated by the adminsitration of pimozide. The two hypotheses are at variance in predicting deficits in various types of feeding behaviour. For example, the anhedonia hypothesis predicts that the animals of Experiment 3 would have found the taste of the food less emjoyable when they were drugged with the higher doses of pimozide, and so should have consumed less of the liquid diet. However, no decrease in consumption was observed. Intake of the diet may have been attenuated if the rats had repeated experience with consummatory sessions while drugged (Wise and Colle, 1984), but this was not investigated in the current series of experiments. The anhedonia hypothesis offers no ready explanation for the fact that dopamine receptor blockade decreases the attractiveness of the food less severely on the first occasion than on the second or third occasions on which the animal consumes food while drugged. In contrast to the anhedonia hypothesis, the incentive-response hypothesis does not predict any deficit at all in feeding behaviour once the animal is in close proximity to the food. In a home cage feeding experiment, this is virtually all of the time. Once the animal makes contact with the food, consummatory behaviours are executed under the command of non-dopaminergic neural systems. The anhedonia hypothesis is unable to explain the severe disruption of appetitive behaviour produced by pimozide in Experiment 1. Although the anhedonia hypothesis, Mark II (Gray and Wise, 1980) predicts deficits in responding to secondary reinforcing stimuli, i t unable to explain why responses to secondary reinforcers are actually more vulnerable to the effects of pimozide than are primary reinforcers, such as food. The anhedonia hypothesis could be - 97 -extended further to specifically incorporate a statement that the rewarding properties of seconndary reinforcers are relatively more susceptable to the effects of neuroleptics than are primary reinforcers. However, secondary rewards are semantically equivalent to incentives in most situations. In effect, the resulting version of the anhedonia hypothesis would be more similar to the incentive- response hypothesis than to the current version of the anhedonia hypothesis. The incentive-response hypothesis predicts the deficit in appetitive responding seen after the administration of pimozide. By the hypothesis, when the animal is presented with an incentive stimulus - the CS+ - dopamine release is increased, but due to post-synaptic dopamine receptor blockade the released transmitter has no effect. Because dopamine systems are involved in the production of approach and other appetitive behaviours, the animal does not advance toward the incentive stimulus or towards the feeding niche. This analysis is supported by the results of Experiment 2, which indicated that presentation of the CS+ for four minutes resulted in a twenty percent increase in forebrain dopamine turnover. Increased dopaminergic activity in the n. accumbens and striatum results in increased activity, locomotion and approach. It was indicated in the introduction that elevated dopaminergic activity often results in feeding, but that this feeding does not appear to be normal in all respects. For example, stimulation of the LH may induce consumption of food pellets, as long as pellets are present. However, if the pellets are removed many animals begin responding to the stimulation by drinking water or engaging in other behaviours, rather than by eating another palatable food which is available (Valenstein et al, 1970). A similar situation may occur in conditioned feeding: After conditioning, even if food is available ad lib an animal will not eat the - 98 -free food during the CS+ period. Rather it will wait until the signalled meal is delivered into the feeding niche, and then it will consume that (Weingarten, 1984). Although apparantly maladaptive, such specific responding may not be totally aberrant. Unlike direct stimulation studies, where increased dopaminergic activity is not related to any identifiable environmental stimulus, in the conditioned feeding situation the eliciting stimulus has also been a good predictor of food availability in the past. Similarly, natural incentives may serve as both eliciters of dopamine release and cues for goal availability. A sufficiently salient stimulus (which the CS+ has come to be through overtraining) may naturally, and usefully, evoke a "use-it-or-lose-it" reaction from the animal. The data at hand do not settle the question of whether pimozide-treated animals perceive the incentive as an attractive stimulus, to which they are unable to respond, or whether they perceive the incentive as less attractive. In avoidance responding, neuroleptic-treated animals react to a warning signal as if they were frightened by it, squealing and defaecating in advance of shock onset. The same warning signal can later evoke a conditional emotional response (Beninger et al, 1980). In the conditioned feeding paradigm, autonomic responses to the CS+ could also indicate that the stimulus is perceived as attractive. In fact, the video analysis of one rat treated with 0.4mg/kg pimozide indicated that he salivated profusely following CS+ onset, even though he did not approach the feeding niche. Further analysis of similarity treated rats will be necessary to determine if this is a typical response by a rat with an incentive-response deficit. The current experiments have also failed to shed any additional light on the relative functions of the mesolimbic and nigrostriatal dopamine projections. Following exposure to the CS+, dopamine turnover was only significantly - 99 -increased in the n. accumbens, but an apparantly similar, though non-significant, twenty percent increase was observed in the striatum. Similarly, food consumption significantly increased DOPAC/DA ratios in the n. accumbens, while there was a trend toward significance in the striatum. However, HVA/DA ratios were significantly elevated in both the striatum and the n. accumbens following consumption of nutritive substances. Thus, in each situation examined, the n. accumbens and striatum appear to have acted in parallel. The incentive-response hypothesis is a theoretical device that is useful for interpreting the role of dopaminergic systems in appetitive and consummatory aspects of feeding. However, as it is currently formulated it provides little insight into the cause or the significance of the increased dopamine turnover observed following the ingestion of nutrients. In light of the results of Experiment 4 it appears that no matter how important a role dopamine plays in appetitive responding, incentives are not the only causes of increased dopamine release. What could produce the increased dopamine turnover observed following ingestion of liquid diet or food pellets? In addition to the purely physiological factors discussed earlier, the consumption of nutrients may have other, more "psychological" effects as well. Specifically, the consumption of a meal after twenty hours of food deprivation may provide "satisfaction", to borrow Thorndike's (1899) term. This "satisfaction" has an obvious conceptual relationship to "reward". However, it is important to note the limited sense of reward implied by the release of dopamine following feeding, given that only the ingestion of nutrients is rewarding by this criterion, while the ingestion of saccharin solution is not. Both the saccharin and the liquid diet appear to taste good to the animals, given - 100 -the avidity with which they are consumed. As well, if the performance of a consummatory response is in itself rewarding (Sheffield, Roby and Campbell, 1954) then consumption of the two solutions should be equally rewarding, because equal volumes of each were consumed. However, the consumption of nutrients has other consquences which do not follow the ingestion of saccharin. These consequences include the elevation of blood levels of glucose, pyruvate and CCK, increases in blood temperature and the relief of "hunger". Any of these may ultimately prove to be sufficient or necessary factors for the increase in dopamine turnover observed one hour after presentation of a nutritive food source to a rat. The suggestion that central dopamine systems may be involved only in this limited sense with "satisfaction" or "reward" may surprise some readers. Perhaps dopamine is actually involved in a more global set of rewards and rewarding events. Ingestion of saccharin may in fact elevate dopamine turnover, but only during the actual period of ingestion: If most consumption of saccharin solution occurred in the first few minutes it was available then dopamine turnover could return to baseline levels by the time the hour is over and the brain extracted. More detailed analysis of the time courses of saccharin ingestion and of dopamine release will be necessary to evaluate these suggestions. No matter what causes the elevation of dopamine turnover following the consumption of nutrients an additional question remains: Does this release have any functional consequences? It does not appear cause the animal to be more responsive to food-related incentive stimuli. Perhaps an hour after the beginning of food availability the animal has recovered from the effects of twenty hour's food deprivation and is exceptionally responsive to other sorts of incentive stimuli (social, sexual, cognitve puzzles). A more thorough analysis of the - 101 -behavioural responsiveness of similar animals would be necessary to evaluate this hypothsis. Can the elevation of dopamine turnover following the ingestion of nutrients have anything to do with feeding itself? At least two potential functions can be suggested. First, whatever the impact of elevated dopamine turnover during the post-prandial phase, the elevated release may function as an unconditional response (UCR) to the ingestion of food. In this context, food is an unconditional stimulus (UCS) and incentive cues, such as the buzzer-light compound in the conditioned feeding paradigm, function as conditional stimuli (CS+). Thus, with repeated conditioning trials the CS+ may gradually come to evoke increased dopamine release as a conditional reflex. In this way appetitive behaviour would be conditioned through a conditioned dopamine release mechanism. A second possible function for increased dopamine turnover in the post-prandial period is to maintain the memory of the association between the cues which had signalled food delivery and the presentation of the food itself. White and his colleagues have previously demonstrated that increased dopamine release in the striatum (the presumed consequence of microinjection of amphetamine into this site) is sufficient to enhance a classically conditioned association between a tone and a shock (Carr and White, 1984). If consumption of food increases dopamine turnover in a similar manner then association of a tone and shock should be stengthened by allowing animals to consume a nutritive solution following tone-shock pairings. Messier and White (1984) have demonstrated just such an effect: Conditional suppression was enhanced when rats were allowed to consume a glucose solution following contingent tone-shock pairings. It is intriguing, in the context of the current experimental results, to - 102 -note that no such increase in suppression was observed when rats consumed a saccharin solution following tone-shock pairings. The potential biological relevance of such a mechanism is obvious: If an animal investigates a place and finds food there, he will be more likely to retain the relationship between the food and the place i f consuming the food releases dopamine in the forebrain and consequently strengthens the association between them. Further support for this hypothesis is found in a recent report by White and Carr (1985) which indicated that consumption of a saccharin solution in a chamber does not produce a conditioned place preference for the chamber, although consumption of an equally preferred sucrose solution in the same chamber does produce a place preference. These authors agrue that although the animal may associate the place with a positive hedonic state induced by consuming both sweet solutions, the strength of the association between saccharin and the place is much weaker due to a minimal effect of saccharin on dopamine release during the post-prandial phase. This hypothesis was developed in the absence of solid neurobiological evidence of differing post-ingestive effects of consumption of sugar and saccharin solutions on dopamine release. The current study provides an independent validation of the hypothesis. CONCLUSIONS This thesis has developed a hypothesis describing the role of dopamine in appetitive and consummatory responding, with particular reference to feeding. The formulation, referred to as the incentive-response hypothesis, asserts that increased forebrain release of dopamine is elicited by food-related incentive stimuli, and that this increase in release plays a crucial role in the production of approach and appetitive behaviours. The hypothesis does not ascribe a role to dopamine in consummatory responding. - 103 -The experimental evidence presented here provides support for the incentive-response hypothesis. Presentation of an incentive stimulus was shown to increase dopamine turnover in the forebrain, and dopamine receptor blockade was shown to seriously attenuate responses to an incentive stimulus. On the other hand, dopamine receptor blockade had no effect on consummatory behaviour. A further experiment indicated that dopamine is releasd in the post-prandial period, after an animal has consumed food pellets or a liquid diet, but not after the consumption of similar quantities of a saccharin solution. The incentive-response hypothesis, as currently formulated, cannot explain the relationship between this release and incentive-based behaviours. The use of a slightly more detailed taxonomy of feeding behaviours than is commonly employed in neurobiological research permitted insights into the nature of dopaminergic involvement in feeding that would not otherwise have been possible. Specifically, the fractionation of feeding into appetitive, consummatory and post-prandial stages permitted separate analyses of dopamine function in each stage. This strategy permitted the identification of both weak and strong points in the incentive-response hypothesis. It is hoped that this method will be adopted and refined for further analysis of relationships between neural systems and behaviour. - 104 -REFERENCES Agardh, C.-D., Carlsson, A., Lindqvist, M. & Seisjo, B.K. (1979). The effect of pronounced hypoglycemia on monoamine metabolism in rat brain. Diabetes. 28, 804-809. Ahlskog, J.E. & Hoebel, B.G. (1973). Overeating and obesity from damage to a noradrenergic system in the brain. Science, 182. 166-169. Anand, B.K. and Brobeck, J.R. (1951) Hypothalamic control over food intake in rats and cats. Yale Journal of Biology and Medicine, 24. 123-140. Antin, J., Gibbs, J, Holt, J., Young, R.C., and Smith, G.P. (1975) Cholecystokinin elicits the complete sequence of satiety in rats. Journal of Comparative and Physiological Psychology. 89, 784-790. Antelman, S.M. & Szechtman, H. (1975). Tail pinch induces eating in sated rats which appears to depend on nigrostriatal dopamine. Science, 189, 731-733. Antelman, S.M., Szechtman, H., Chin, P. & Fisher, A.E. (1975). Tail pinch-induced eating, gnawing and licking behavior in rats: Dependence on the nigrostriatal dopamine system. Brain Research. 99, 319-337. Bellin, S.I., & Ritter, S. (1981). Insulin induced elevation of hypothalamic norepinephrine turnover persists after glucose restoration unless feeding occurs. Brain Research. 217, 327-337. Beninger, R.J. (1983). The role of dopamine in locomotor activity and learning. Brain Research Review. 6, 173-196. Beninger, R.J., Mason, S.T., Phillips, A.G. & Fibiger, H.C. (1980). The use of conditioned suppression to evaluate the nature of neuroleptic-induced avoidance deficits. Journal of Pharmacology and Experimental Therapeutics. 213. 623-627. Beninger, R.J., Phillips, A.G. & Fibiger, H.C. (1983). Prior training and intermittent retraining attenuate pimozide-induced avoidance defecits. Pharmacology, Biochemistry & Behavior, 18, 619-624. Bernays, E.A. and Simpson, S.J. (1982). Control of food intake. Advances In Insect Physiology, 16, 59-118. Biggio, G., Porceddu, M.L., Fratta, W. and Gessa, G.L. (1977). Changes in dopamine metabolism associated with fasting and satiation. In: E. Costa and G.L. Gessa (Eds.), Advances in Biochemical Psychopharmacology, 16, 377-380. Bindra, D. (1976). A Theory of Intelligent Behavior. New York: Wiley and Sons. Bindra, D. (1978). How adaptive behavior is produced: A perceptual motivational alternative to response-reinforcement. The Behavioral and Brain Sciences. 1, 41-91. (includes commentaries). - 105 -Blackburn, J.R., Jacobs, W.J. and Phillips, A.G. (1984). Paiatability and the suppression of food intake by cholecystokinin-octapeptide. Society for  Neuroscience: Abstracts. 10. 174.12. Blundell, J.E. & Latham, C.J. (1978). Pharmacological manipulation of feeding behavior: possible influences of serotonin and dopamine on food intake. In S. Garanttini and R. Samanian (Eds.), Central Mechanisms of Anorectic Drugs (pp 83-99). New York: Raven Press. Blundell, J.E., Strupp, B.J. & Latham, C.J. (1977) Pharmacological manipulation of hoarding: Furthur analysis of amphetamine isomers and pimozide. Physiological  Psychology. 5, 462-468. Bolles, R.C. (1967). Theory of Motivation. New York: Harper and Row. Bolles, R.C. (1975). Theory of Motivation (Second Edition). New York: Harper and Row. Burridge, S.L. & Blundell, J.E. (1979) Amphetamine anorexia: antagonism by typical but not atypical neuroleptics. Neuropharmacology. 18, 453-457. Cabanac, M. (1971). Physiological role of pleasure. Science. 73, 1103-1107. Campbell, B.A. & Fibiger, H.C. (1971). Potentiation of amphetamine-induced arousal by starvation. Nature. 233, 424-425. Campfield, L.A., Brandon, P and Smith, F.J. (1985). On-line continuous measurement of blood glucose and meal pattern in free-feeding rats: The role of glucose in meal initiation. Brain Research Bulletin, 14. 605-616. Carr, G.D. (1984). The neuroanatomical basis of the behavioral effects of amphetamine: an intracranial microinjection study- Unpublished PhD Thesis, McGill University. Carr, G.D. & White, N.M. (1984). The relationship between stereoptypy and memory improvement produced by amphetamine. Psychoparmacology. 82, 203-209. Champion, R.A. (1964). The latency of the conditioned fear-response. American  Journal of Psychology. 77, 75-83. Chance, W.T., Foley-Nelson, T., Nelson, J.L., Kim, M.W. & Fischer, J.E. (1985). Changes in neurotransmitter levels associated with feeding. Society for  Neuroscience: Abstracts. 11, 61. Chiodo, L.A. & Bunney, B.S. (1983). Proglumide: selective antagonism of excitatory effects of cholecystokinin in CNS. Science. 219, 1449-1451. Clody, D.E. & Carlton, P.E. (1980). Stimulus efficacy, chlorpromazine, and schizophrenia. Psychopharmacology. 69. 127-131. - 106 Cole, S.O. (1978). Brain mechanisms of amphetamine-induced anorexia, locomotion, and stereotypy: a review. Neuroscience and Biobehavioral Reviews. 2, 89-100. Collins, S.H., Forsyth, P. & Weingarten, H. (1983). Endogenous cholecystokinin reduces food intake in the rat. Gastroenterology. 84. 1128. Coons, E.E., Levak, M. and Miller, N.E. (1965). Lateral hypothalamus: Learning of food-seeking response motivated by electrical stimulation. Science. 150. 1320-1321. Cottett-Emard, J.M. & Peyrin, L. (1982). Conjugated HVA increase in rat urine after insulin-induced hypoglycemia: involvement of central dopaminergic structures but not of adrenal medulla. Journal of Neural Transmission. 55. 121-138. Craig, W. (1918). Appetites and aversions as constituents of instincts. Biological  Bulletin of Woods Hole. 34. 91-107. Crawley, J.N., Hays, S.E. & Paul, S.M. (1981). Vagotomy abolishes the inhibitory effects of cholecystokinin on rat exploraory behaviors. European Journal of Pharmacology. 55, 121-138. Crawley, J.N., Hays, S.E., Paul, S.M. & Goodwin, F.K. (1981). Cholecystokinin reduces exploratory behaviour in mice. Physiology and Behavior. 27, 407-411. Crawley, J.N., Rojas-Ramirez, J.A. & Mendelson, W.B. (1982). The role of central and peripheral cholecystokinin in mediating appetitive behaviors. Peptides, 3, 535-538. Crawley, J.N. & Schwaber, J.S. (1984). Abolition of the behavioral effects of cholecystokinin following bialteral radiofrequency lesions of the parvocellular subdivision of the nucleus tractus solitarius. Brain Research. 295. 289-299. Danguir, J., Elghozi, J.-L. & Laude, D. (1984). Increased dopamine and serotonin metabolites in CSF during severe insulin-induced hypoglycemia in freely moving rats. Neurochemistry International. 6, 71-75. Delgado, J.M.R. and Anand, B.K. (1953). Increase of food intake induced by electrical stimulation of the lateral hypothalamus. American Journal of Physiology. 172. 162-168. de Wit, H. & Wise, R.A. (1977). Blockade of cocaine reinforcement in rats with the dopamine receptor blocker pimozide but not with the noradrenergic blockers phentolamine or phenoxybenzamine. Canadian Journal of Psychology, 31, 195-203. Dorris, R.L. (1978). Influence of glucose on amine release from rat corpus striatum. Neuropharmacology, 17, 157-158. Ehrman, R.N. and Overmier, J.B. (1976). Dissimilarity of mechanisms for evocation of escape and avoidance responding in dogs. Animal Learning and Behavior. 4, 347-351. - 107 -Eichler, A.J. & Antelman, S.M. (1977). Apomorphine: Feeding or anorexia depending on internal state. Communications in Psychopharmacology. 1, 533-540. Ernst, A.M. (1967). Mode of action of apomorphine and dexamphetamine on gnawing compulsion in rats. Psychopharmacologia. 10, 316-323. Ettenberg, A. and Carlisle, H.J. (1985). Neuroleptic-induced deficits in operant responding for temperature reinforcement. Pharmacology, Biochemistry & Behavior, 00, 000-000. Ettenberg, A., Cinsavich, S.A. and White, N. (1979). Performance effects with repeated-response measures during pimozide-produced dopamine receptor blockade. Pharmacology, Biochemistry & Behavior, 11. 557-561. Fallon, J.H. & Moore, R.Y. (1978). Catecholamine innervation of the basal forebrain IV: Topography of the dopamine projection to the basal forebrain and neostriatum. Journal of Comparative Neurology. 180, 545-580. Fekete, M., Kadar, M. & Telegdy, G. (1981). Effect of cholecystokinin antiserum on the brain monoamine content in rats. Acta Physiologica Acadamiae  Scientarum Hungaricae. 57, 177-183. Fekete, M., Szabo, A., Balzas, B., Penke, B. & Telegdy, G. (1981). Effects of intraventricular administration of cholecystokinin octapeptide sulfate ester and unsulfated cholecystokinin octapeptide on active avoidance and conditioned feeding behaviour of rats. Acta Physiologica Acadamiae Scientarum  Hungaricae. 58, 39-45. Fibiger, H.C. & Phillips, A.G. (1985). Behavioral pharmacology of neuroleptic drugs: possible mechanisms of action. In: L.S. Seiden and R.L. Balster (Eds.), Behavioral Pharmacology: The Current Status (pp 243-259). New York: Alan Liss. Fibiger, H.C, Phillips, A.G. and Zis, A.P. (1973). Deficits in instrumental responding after 6-hydroxydopamine lesions of the nigro-neostriatal dopaminergic projection. Pharmacology. Biochemistry & Behavior. 2, 87-96. Fibiger, H.C, Zis, A.P. & McGeer, E.G. (1973). Feeding and drinking deficits after 6-hydroxydopamine adminstration in the rat: Similarities to the lateral hypothalamic syndrome. Brain Research, 55, 135-148. Fibiger, H.C, Zis, A.P. & Phillips, A.G. (1975). Haloperidol-induced disruption of conditioned avoidance responding: Attenuation by prior training or by anticholinergic drugs. European Journal of Pharmacology. 30. 309-314. Fouriezos, G., Hannson, P. & Wise, R.A. (1976). Neuroleptic-induced attenuation of brain stimulation reward. Journal of Comparative and Physiological Psychology. 92, 659-669. - 108 -Fouriezos, G. & Wise, R.A. (1976). Pimozide-induced of extiction of intracranial self-stimulation: Response patterns rule out motor performance deficits. Brain  Research. 103. 377-380. Franklin, K.B.J. (1978). Catecholamines and self-stimulation: Reward and performance deficits dissociated. Pharmacology. Biochemistry & Behavior. 9, 813-820. Franklin, K.B.J. & McCoy, S.N. (1979). Pimozide-induced extinction in rats: Stimulus control of responding rules out motor deficit. Pharmacology. Biochemistry & Behavior, 11. 71-76. Freed, CR. and Yamato, B.K. (1985). Regional brain dopamine metabolism: A marker for the speed, direction, and posture of moving animals. Science. 229. 62-65. Friedman, E., Starr, N., & Gershon, S. (1973). Catecholamine synthesis and the regulation of food intake in the rat. Life Sciences, 12, 317-328. Fuenmayor, L.D. (1979). The effect of fasting on the metabolism of 5-hydroxytryptamine and dopamine in the brain of the mouse. Journal of Neurochemistry. 33, 481-485. Fuxe, K., Andersson, K., Locatelli, V., Agnati, L.F., Hokfelt, T., Skirboll, L. & Mutt, V. (1980). Cholecystokinin peptides produce marked reduction in dopamine turnover in discrete areas in the rat brain following intraventricular injection. European Journal of Pharmacology. 67, 329-331. Fuxe, K. & Ungerstedt, U. (1970). Histochemical, biochemical and funtional studies on central monoamine neurons after acute and chronic amphetamine administration. In E. Costa & S. Garattini (Eds.) Amphetamine and Related  Compounds. New York: Plenum Press. Gerber, G.J., Sing, J. & Wise, R.A. (1981). Pimozide attenuates lever pressing for water reinforcement in rats. Pharmacology. Biochemistry & Behavior. 14, 201-205. Gibbs, J., Young, R.C. & Smith, G.P.(1975). Cholecystokinin decreases food intake in rats. Journal of Comparative and Physiological Psychology. 84, 488-495. Gilles, C, Lotstra, F. & Vanderhaegen, J.-J. (1983). CCK terminals in the rat striatal and limbic areas originate partly in the brain stem and partly in telencephalic structures. Life Sciences. 32, 1683-1690. Glick, S.D., Waters, D.H. & Milloy, S. (1973). Depletion of hypothalamic norepinephrine by food deprivation and interaction with amphetamine. Research  Communications in Chemical Pathology and Pharmacology. 6, 775-778. Glowinski, J. & Axelrod, J. (1965). Effects of drugs on the uptake, release, and metabolism of 3H-norepinephrine in the rat brain. Journal of Pharmacology and  Experimental Therapeutics. 149. 43-49. - 109 Gray, T. & Wise, R.A. (1980). Effects of pimozide on lever pressing behavior maintained on an intermittent reinforcement schedule. Pharmacology. Biochemistry & Behavior, 12, 931-935. Heffner, T.G., Hartman, J., & Seiden, L. (1980). Feeding increases dopamine metabolism in the rat brain. Science. 208. 1168-1170. Heffner, T.G. & Seiden, L. (1980). Synthesis of catecholamines from 3H-tyrosine in brain during the performance of operant behavior. Brain Research. 183, 403-419. Heffner, T.G., Zigmond, M.S. & Strieker, E.M. (1977). Effects of dopaminergic agonists and antagonists on feeding in intact and 6-hyroxydopamine-treated rats. Journal of Pharmacology and Experimental Therapeutics. 201, 386-399. Heffner, T.G., Vosmer, G. & Seiden, L.S. (1984). Time-dependent changes in hypothalamic dopamine metabolism during feeding in the rat. Pharmacology, Biochemistry & Behavior 20, 947-949. Heller, A. & Harvey, J.A. (1963). Effects of CNS lesions on brain norepinephrine. Pharmacologist, 5, 264. Hokfelt, T., Skirboll, L., Rehfeld, J.F., Goldstein, M. & Markey, K. (1980). A subpopulation of mesencephalic dopamine neurons projecting to limbic areas contains a cholecystokinin peptide: evidence from immunohistochemistry combined with retrograde tracing. Neuroscience. 5, 2093-2124. Holland, P.C. (1977). Conditioned stimulus as a determinant of the form of the Pavlovian response. Journal of Experimental Psychology: Animal Behaviour  Processes, 3, 77-104. Holtzman, S.G. (1974). Behavioral effects of separate and combined administration of naloxone and d-amphetamine. Jouranal of Pharmacology and Experimental  Therapeutics. 189. 51-60. Hommer, D.W., Palkovits, M., Crawley, J.N., Paul, S.M. & Skirboll, L.R. (1985). Cholecystokinin-induced excitation in the substantia nigra: Evidence for peripheral and central components. Journal of Neuroscience. 5, 1387-1392. Hommer, D.W. & Skirboll, L.R. (1983). Cholecystokinin-like peptides potentiate apomorphine-induced inhibition of dopamine neurons. European Journal of Pharmacology, 91, 151-152. Hunt, H.F. (1956). Some effects of drugs on classical (type S) conditioning. Annals of the New York Academy of Science, 65, 258-267. Itoh, S. & Katsuura, G. (1981). Suppressive effects of cholecystokinin octapeptide on the behavioural effects of 1-dopa in the rat. European Journal of Pharmacology. 75, 313-316. - 110 -Jacobs, W.J. & Harris, C. (in press). Escape responding in a shuttlebox. Learning  and Motivation. Jakubovic, A., Lin, D. & Fibiger, H.C. (Submitted). Simultaneous measurement of dopamine, serotoinin and their metabolites by HPLC with electrochemical detection: factors affecting their seperation, retention time and stability. Janssen, P., Neimegeers, C, Schellekens, K., Breese, A., Lenaerts, F., Pinchard, A., Schgre, W., Van Neuten, J. & Verbrugger, J. (1968). Pimozide: A chemically novel highly potent and orally long acting neuroleptic drug. Jahrgang. Janssen Pharmaceutica. 21, 1245-1251. (1968). Jerussi, T.P. & Glick, S.D. (1976). Drug induced rotation in rats without lesions: Behavioral and neurochemical indications of a normal asymmetry in nigro-striatal function. Psychopharmacology. 47, 249-260. Jerussi, T.P. & Taylor, CA. (1982). Bilateral asymmetry in striatal dopamine metabolism: Implications for pharmacotherpy of schizophrenia. Brain Research. 246. 71-75 Joyce, J.N. (1983). Multiple dopamine receptors and behavior. Neuroscience and  Biobehavioral Reviews. 7, 227-256. Katsuura, G. & Itoh, S. (1982) Sedative effect of cholecystokinin octapeptide on behavioral excitiation by thyrotropin releasing hormone and methamphetamine in the rat. Japanese Journal of Physiology. 32, 83-91. Kelly A.E. & Stinus, L. (1985). Disappearance of hoarding behavior after 6-hydroxydopamine lesions of the mesolimbic dopamine neurons and its reinstatement with 1-dopa. Behavioral Neuroscience, 99, 531-545. Kelly, P.H. (1977). Drug-induced motor behavior. In L.L. Iversen, S.D. Iversen & S.H. Snyder (Eds.) Handbook of Psychopharmacology. 8, New York: Plenum Press. Konorski, J. (1967). The Integrative Activity of the Brain. Chicago: University of Chicago Press. Koob, G.F., Riley, S.J., Smith, S.C. & Robbins, T.W. (1978). Effects of 6-hydroxydopamine lesions of the nucleus accumbens »septi and olfactory tubercle on feeding, locomotor activity, and amphetamine anorexia in the rat. Journal of Comparative and Physiological Psychology. 92, 917-927. Lai, S., Nair, N.P.V., Eugenio, H., Thavundayil, J., Lizondo, E., Wood, P.L., Etienne, P. & Guyda, H. (1983). Neuroendocrine evaluation of CCK-peptides on dopaminergic function in man. Progress in Neuropsychopharmacology and  Biological Psychiatry. 7, 537-544. La Vielle, S., Tassin, J.-P., Thierry, A.-M., Blanc, G., Hevre, D., Barthelemy, C. & Glowinski, J. (1978). Blockade by benzodiazeptines of the selective high increase in dopamine turnover induced by stress in mesocortical dopaminerigic neurons of the rat. Brain Research. 168. 581-594. - I l l -Lawson W.B., Byrd, J. & Reed, D. (1984). Effects of neuroleptics on food intake. Society for Neuroscience: Abstracts. 10, 92.10. Le Moal, M., Stinus, L., Simon, H., Tassin, J.P., Thierry, A.M., Blanc, G., Glowinski, J. & Cardo, B. (1977). Behavioral effects of a lesion in the ventral mesencephalic tegmentum: Evidence for involvement of A10 dopaminergic neurons. In E. Costa & G.L. Gessa (Eds.), Advances in Biochemical  Psychopharmacology, 16 (pp 237-245). New York: Raven Press. Lorenz, D.M. & Goldman, S.A. (1982). Vagal mediation of the cholecystokinin satiety effect in rats. Physiology and Behavior. 29, 599-604. Louis-Sylvestre, J. & Lemagnen, J. (1980). A fal l in blood glucose level precedes meal onset in free feeding rats. Neuroscience and Biobehavioral Reviews. 4, SUPPI. 1, 227-256. Lozovsky, D., Sailer, C.F. & Kopin, I.J. (1981). Dopamine receptor binding is increased in diabetic rats. Science. 214, 1031-1033. Makanjuola, R.O.A., Dow, R.C. & Ashcroft, G.W. (1980). Behavioural responses to stereotactically controlled injections of monoamine neurotransmitters into the acumbens and caudate-putamen nuclei. Psychopharmacology. 71, 227-235. Markstein, R. & Hokfelt, T. (1984). Effect of cholecystokinin-octapeptide on dopamine release from slices of cat caudate nucleus. Journal of Neuroscience. 4, 570-575. Marley, P.D., Emson, P.C. & Rehfeld, J.F. (1982). Effect of 6-hydroxydopamine lesions of the medial forebrain bundle on the distribution of cholecystokinin in rat forebrain. Brain Research. 252. 382-385. Marshall, J.F. (1978) Further analysis of the resistance of the diabetic rat to d-amphetamine. Pharmacology. Biochemistry & Behavior. 8, 281-286. Marshall, J.F, Friedman, M.I. & Heffner, T.G. (1976) Reduced anorexic and locomotor-stimulant action of d-amphetamine in alloxan-diabetic rats. Brain  Research. 111. 428-432. Marshall, J.F, Friedman, M.I. & Heffner, T.G. (1978) Reduced anorexic and locomotor stimulant action of amphetamine in experimental diabetes mellitus: relation to brain catecholamines. In S. Garanttini and R. Samanian (Eds.), Central Mechanisms of Anorectic Drugs (pp 111-125). New York: Raven Press. Marshall, J.F., Richardson, J.S. & Teitlebaum, P. (1974). Nigrostriatal bundle damage and the lateral hypothalamic syndrome. Journal of Comparative and  Physiological Psychology. 87, 808-830. Marshall, J.F., Turner, B.H. & Teitlebaum, P. (1972). Sensory neglect produced by lateral hypothalamic damage. Science, 174, 523-525. - 112 -Martin, G.E. & Myers, R.D. (1976) Dopamine efflux from the brain stem of the rat during feeding, drinking and lever-pressing for food. Pharmacology. Biochemistry & Behavior. 4, 551-560. Mashal, R.D., Owen, F., Deakin, J.F.W. & Poulter, M. (1983). The effects of cholecystokinin on dopaminergic mechanisms in the neostriatum. Brain  Research. 277. 375-376. Mason, S.T., Beninger, R.J., Fibiger, H.C. & Phillips, A.G. (1980). Pimozide-induced suppression of responding: evidence against a block of food reward. Pharmacology. Biochemistry & Behavior. 12, 917-923. Masterson, M.A. & Crawford, M. (1982). The defence motivational system: A theory of avoidance behavior. The Behavioral and Brain Sciences, 5, 661-696. (includes commentaries). Mayer, J. (1953). Glucostatic mechanism of regulation of food intake. New  England Journal of Medicine. 249, 13-16. McCaleb, M.L. & Myers, R.D. (1979). Striatal dopamine release is altered by glucose and insulin during push-pull perfusion of the rat's caudate nucleus. Brain Research Bulletin. 4, 651-656. McMillen, B.A. (1983). CNS stimulants: Two distinct mechanisms of action for amphetamine-like drugs. Trends in Pharmacological Sciences. 4, 429-432. Messier, C. & White, N.M. (1984). Contingent and non-contingent actions of sucrose and saccharin reinfrocers: Effects on taste preference and memory. Physiology and Behavior. 23, 195-203. Meyer, D.K., Beinfeld, M.C., Oertel, W.H. & Brownstein, M.J. (1982). Origin of the cholecystokinin-containing neurons in the rat caudate putamen. Science. 215. 187-188. Migler, B. (1975) Conditioned approach: An analogue of conditioned avoidance; Effects of chlorpromazine and diazepam. Pharmacology, Biochemistry & Behavior. 3, 961-965. Mineka, S. (1979). The role of fear in theories of avoidance learning, flooding and extinction. Psychological Bulletin. 86. 985-1010. Mineka, S. & Gino, A. (1980). Dissociation between conditioned emotional response and extended avoidance performance. Learning and Motivation. 11, 476-502. Mogenson, G.J. & Phillips, A.G. (1976). Motivation: A psychological contruct in search of a physiological substrate. In: J.M. Sprague & A.N. Epstein (eds.), Progress in Psychobiology and Physiological Psychology, vol. 6, (pp 189-243). New York: Academic Press. - 113 -Moore, R.Y. & Heller, A. (1967). Monoamine levels and neuronal degeneration in rat brain following lateral hypothalamic lesions. Journal of Pharmacology and  Experimental Therapeutics. 156, 12-22. Morgane, P.J. (1961a). Alterations in feeding and drinking of rats with lesions in globi pallidi. American Journal of Physiology. 201, 420-428. Morgane P.J. (1961b). Medial forebrain bundle and "feeding centers" of the hypothalamus. Journal of Comparative Neurology. 117, 1-25. Morley, J.E., Levine, A.S. & Kneip, J. & Grace, M. (1982). The effect of vagotomy on the satiety effects of neuropeptides and naloxone. Life Sciences. 30. 1943-1947. Mowrer, O.H. (1947). On the dual nature of learning: A reinterpretation of "conditioning" and "problem-solving". Harvard Educational Review. 17, 102-148. Mueller, K. & Hsaio, S. Current status of cholecystokinin as a short-term satiety hormone. Neuroscience and Biobehavioral Reviews. 2, 79-87. Oltmans, G.A. & Harvey, J.A. (1972). LH syndrome and brain catecholamine levels after lesions of the nigrostriatal bundle. Physiology and Behavior, 8, 69-78. Panksapp, J. (1982). Toward a general psychobiological theory of emotions. The  Behavioral and Brain Sciences. 5, 407-467. (includes commentaries). Phillips, A.G. & Fibiger, H.C. (1973a). Deficits in stimulation-induced feeding after intraventicular administration of 6-hydroxydopamine in rats. Behavioral Biology. 9, 749-754. Phillips, A.G. & Fibiger, H.C. (1979). Decreased resistence to extinction after haloperidol: Implications for the role of dopamine in reinforcement. Pharmacology. Biochemistry & Behavior. 10. 751-760. Phillips, A.G., McDonald, A.C. and Wilkie, D.M. (1981). Disruption of autoshaped responding to a signal of brain-stimulation reward by neuroleptic drugs. Pharmacology. Biochemistry & Behavior. 10, 751-760. Phillips, A.G. & Nikaido, R.S. (1973a). Disruption of brain stimulation-induced feeding by dopamine receptor blockade. Nature. 258. 750-751. Posluns, D. (1962). An analysis of chlorpromazine-induced suppression of the avoidance response. Psychopharmacologia. 3, 361-373. Rojas-Ramirez, J.A., Crawley, J.N. & Mendelson, W.B. (1982). Electroencephalo-graphic analysis of the sleep-inducing actions of cholecystokinin. Neuropeptides, 3, 129-138. Rowland, N.E., Bellush, L.L. and Carlton, J. (1985). Metabolic and neurochemical correlates of glucoprivic feeding. Brain Research Bulletin. 14, 617-624. - 114 -Sailer, C.F. & Chiodo, L.A. (1980). Glucose suppresses the basal firing and haloperidol-induced increases in the firing rate of central dopaminergic neurons. Science. 210. 1269-1271. Sailer, C.F. & Kopin, I.J. (1980). Glucose potentiates haloperidol-induced catalepsy. Life Sciences. 29 2337-2341. Sauter, A., Ueta, K., Engel, J. & Goldstein, M. Effects of insulin on the release and turnover of dopamine (DA), noradrenaline (NA) and adrenaline (A) in rat brain. Experimentia, 37. 631. Schneider, B.S., Alpert, J.E. & Iverson, S.D. (1983) CCK-8 modulation of mesolimbic dopamine: Antagonism of amphetamine-stimulated behaviors. Peptides. 4, 749-753. Schneirla, T.C. (1959). An evolutionary and developmental theory of biphasic processes underlying approach and withdrawal. In M.R. Jones (Ed.) Nebraska  Symposium on Motivation. VII (pp 1-43). Lincoln, NE: University of Nebraska Press. Sherrington, C.S. (1906). The Integrative Action of the Nervous System. New Haven, CO: Yale University Press. Sheffield, F.D., Roby, T.B. and Campbell, B.A. (1954).Journal of Comparative and Physiological Psychology. 47, 349-354. Skirboll, L.R., Grace, A.A., Hommer, D.W., Rehfeld, J., Goldstein, M., Hokfelt, T. & Bunney, B.S. (1981). Peptide monoamine coexistence: studies of the actions of the actions of cholecystokinin-like peptide on the electrical activity of midbrain dopamine neurons. Neuroscience. 6, 2111-2124. Smith, G.P. & Gibbs, J. (1979). Postprandial satiety. In: J.M. Sprague & A.N. Epstein (eds.), Progress in Psychobiology and Physiological Psychology, vol. 9, (pp 179-242). New York: Academic Press. Smith, G.P. & Gibbs, J. (1981). Brain-gut peptides and the control of food intake. In: J.B. Martin, S. Reichlin & K.I. Bick (Eds)., Neurosecretion and Brain  Peptides. New York: Raven Press. Smith, G.P, Jerome, C, Cushin, B.J., Eterno, R. & Simansky, K.J. (1981). Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat. Science. 213, 1036-1037. Smith, G.P., Greenberg, D.,Falasco, J.D., Gibbs, J. Liddle, R.A. & Williams, J.A. (1985). Plasma levels of cholecystokinin produced by satiating doses of exogenous CCK-8. Society for Neuroscience: Abstracts, 11, 167.13. Solomon, R.L. & Wynne, L.C. (1953). Traumatic avoidnace learning. Psychological  Monographs. 67, whole #354. Spence, K.W. (1956). Behavior Theory and Conditioning. New Haven, CO: Yale University Press. - 115 -Spyraki, C, Fibiger, H.C. & Phillips, A.G. (1982). Attenuation by haloperidol of place preference conditioning using food reinforcement. Psychopharmacology. 7, 379-382. Steffans, A.B. (1969). Rapid ingestion of glucose in the intestinal tract of the rat after ingestion of a meal. Physiology and Behavior. 4, 829-832. Stellar, E. (1954). The physiology of motivation. Psychological Review. 61, 5-22. Strecker, R.E., Steinfels, G.F. & Jacobs, B.L. (1983). Dopaminergic unit activity in freely moving cats: Lack of relationship to feeding, satiety and glucose injections. Brain Research. 260, 317-321. Strieker, E.M. & Zigmond, M.J. (1974). Effects on homeostasis of intraventricular injection of 6-hydroxydopamine in rats. Journal of Comparative and  Physiological Psychology. 86, 973-994. Strieker, E.M. & Zigmond, M.J. (1976). Recovery of function following damage to central catecholamine-containing neurons: a neurochemical model for the lateral hypothalamic syndrome. In: J.M. Sprague & A.N. Epstein (Eds.), Progress in Psychobiology and Physiological Psychology, vol. 6, (pp 121-188). New York: Academic Press. Studler, J.M., Simon, H., Casselin, Legrand, J.C., Glowinski, J & Tassin, J.P. (1981). Biochemical investigation of the localization of the cholecystokinin octapeptide in dopaminergic neurons originating from the ventral tegmental area of the rat. Neuropeptides, 2, 131-139. Taha, E.B. & Redgrave, P. (1980). Neuroleptic suppression of feeding and oral stereotypy following microinjections of carbachol into substantia nigra. Neuroscience Letters. 20. 357-361. Teitlebaum, P. & Epstein, A.N. (1962). The lateral hypothalamic sydrome: recovery of feeding and drinking after lateral hypothalamic lesions. Psychological  Review. 69, 74-90. Teitlebaum, P. & Stellar, E. (1954). Recovery from the failure to eat produced by hypothalamic lesions. Science, 120. 894-895. Tombaugh, T.N., Anisman, H. & Tombaugh, J. (1980). Extinction and dopamine receptor blockade after intermittent reinforcement training: Failure to observe functional equivalence. Psychopharmacology. 70, 19-28. Tombaugh, T.N., Tombaugh, J. & Anisman, H. (1979). Effects of dopamine receptor blockade on alimentary behaviors: Home cage food consumption, operant acquisition, and performance. Psyhcopharmacology. 66, 219-225. Trulson, M.E., Crisp, T. & Trulson, V.M. (1983). Dopamine-containing substantia nigra units are unresponsive to changes in plasma glucose levels induced by dietary factors, glucose infusions or insulin adminstration in freely moving cats. Life Sciences, 32, 2555-2564. - 116 -Ungerstedt, U. (1971a). Striatal dopamine release after amphetamine or nerve degeneration revealed by rotational behaviour. Acta Physiologica Scandinavica. Supplementum 367, 49-68. Ungerstedt, U. (1971b). Adipsia and aphagia after 6-hydroxydopamine induced degeneration of the nigro-striatal dopamine system. Acta Physiologica  Scandinavica. Supplementum 367. 95-122. Ungerstedt, U. (1974). Brain dopamine neurons and behavior. In F.O. Schmitt & F.G. Worden (Eds.), The Neurosciences. Third Study Program (pp. 695-703). Cambridge: MIT Press. Valenstein, E.S. (1975). Stereotyped behavior and stress. In G. Serban (Ed.), Psychopathology of Human Adaptation (pp. 113-124). New York: Plenum Press. Valenstein, E.S., Cox, V.C. & Kakolewski, J.W. (1968a). The motivation underlying eating elicited by lateral hypothalamic stimulation. Physiology and Behavior. 3, 969-971. Valenstein, E.S., Cox, V.C. & Kakolewski, J.W. (1968b). Modification of behavior elicited by electrical stimulation of the hypothalamus. Science, 159, 1119-1121. Valenstein, E.S., Cox, V.C. & Kakolewski, J.W. (1970). Reexamination of the role of the hypothalamus in motivation. Psychological Review. 77, 16-31. Valenstein, E.S. & Phillips, A.G. (1970). Stimulus-bound eating and deprivation from prior contact with food pellets. Physiology and Behavior, 5, 279-282. Voigt, M.M. & Wang, R.Y. (1984). In vivo release of dopamine in the nucleus accumbens of the rat: modulation by cholecystokinin. Brain Research. 296, 189-1393. Walsh, J.N., Lamers, C.B. & Valenzuela, J.E. (1982). Cholecystokinin-octapeptide like immunoreactivity in human plasma. Gastroenterology. 82, 438-444. Weingarten, H.P. (1983). Conditioned cues elicit feeding in sated rats: a role for learning in meal initiation. Science. 220, 431-433. Weingarten, H.P. (1984). Meal initiation controlled by learned cues: behavioural properties. Appetite. 5, 147-158. Westerink, B.H.C. and Spaan, S.J. (1981). Effect of glucose on dopamine metabolism in the rat striatum. Journal of Pharmacology and Pharmacy, 33. 601-602. White, F.J. & Wang, R.Y. (1984). Interactions of choloecystokinin and dopamine on nucleus accumbens neurons. Brain Research, 300, 161-166. White, N.M. & Blackburn, J. (submitted). Effect of glucose on amphetamine-induced motor behavior. - 117 -White, N.M. and Carr, G.D. (1985). The conditioned place preference is affected by two independent reinforcement properties. Pharmacology. Biochemistry & Behavior. 23, 737-42. White, N.M., Messier, C. & Carr, G.D. (1984) Operationalizing and measuring the organizing influence of drugs on behavior. In M.A. Bozarth (Ed.), Methods of Assessing the Reinforcing Properties of Abused Drugs. Brunswick, ME: Haer Institute. Widerlov, E., Kalivas, P.W., Lewis, M.H., Prange, A.J. Jr. & Breese, G.R. (1983). Influence of cholecystokinin on monoamineregic pathways. Regulatory Peptides, 6, 99-109. Williams, R.G., Zhu, R.J. & Dockray, G.J. (1981). Changes in brain cholecystokinin octapeptide following lesions of the medial forebrain bundle. Brain Research. 213. 227-230. Winn, P., Farrell, A., Maconick, A. & Robbins, T.W. (1983). Behavioral and pharmacological specificy of feeding elicited by cholinergic stimulation of the substantia nigra in the rat. Behavioral Neuroscience. 97, 794-809. Winn, P. & Redgrave, P. (1979). Feeding following microinjection of cholinergic substances into substantia nigra. Life Sciences. 25. 333-338. Winn, P., Williams, S.F. & Herberg, L.J. (1982). Feeding stimulated by very low doses of d-amphetamine administered systemically or by microinjection into the striatum. Psychopharmacology. 78, 336-341. Wise, R.A. (1982). Neuroleptics and operant behavior: The anhedonia hypothesis. The Behavioral and Brain Sciences. 5, 39-88. (includes commentaries). Wise, R.A. (1985). The anhedonia hypothesis: Mark III. The Behavioral and Brain  Sciences. 8, 178-186. Wise, R.A & Colle, L.A. (1984). Pimozide attenuates free feeding: Best scores analysis reveals a motivational deficit. Psychopharmacology, 84. 446-451. Wise, R.A. & Schwartz, H.V. (1981). Pimozide attenuates acquisition of lever-pressing for food in rats. Pharmacology. Biochemistry & Behavior. 15. 655-656. Wise, R.A., Spindler, J., deWit, H. & Gerber, G.J. (1978). Neuroleptic-induced "anhedonia" in rats: Pimozide blocks reward quality of food. Science. 201, 262-264. Woods, J.W. (1964). The behavior of chronic decerebrate rats. Journal of Neurophysiology. 27. 635-644. Woodworth, R.S. (1918). Dynamic Psychology. New York: Columbia University Press. - 118 -Woolf, P.D., Akowuah, E.S., Lee, L., Kelly, M. & Feibel, J. (1983). Evaluation of the dopamine response to stress in man. Journal of Clinical Endocrinolgy and  Metabolism. 56, 246-250. Wyrwicka, W. & Doty, R.W. (1966). Feeding induced in cats by electrical stimulation of the brain stem. Experimental Brain Research. 1, 152-160. Yamato, B.K. & Freed, CR. (1982). The trained circling rat: A model for inducing unilateral caudate dopamine metabolism. Nature. 298. 467-468. Yamato, B.K. & Freed, CR. (1984). Asymmetric dopamine and serotonin metabolism in nigrostriatal and limbic structures of the trained circling rat. Brain Research. 297. 115-119. Yokel, R.A. & Wise, R.A. (1975). Increased lever pressing for amphetamine after pimozide: Implications for a dopamine theory of reward. Science. 187. 547-549. Zigmond, M.S. & Strieker, E.M. (1972). Deficits in feeding behavior after intraventricular injection of 6-hydroxydopamine in rats. Science. 177, 1211-1214. - 119 -


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items