I tinglg ll”!lllllllllllllmlllllllllllllllllfll"Hill"!!! ‘ 3 This is to certify that the thesis entitled RELATIONSHIP OF AMYGDALOID 5-HYDROXYTRYPTAMINE- CONTAINING NEURONS TO ANTICONFLICT EFFECTS OF BENZODIAZEPINES IN THE RAT presented by Clinton Donald Kilts has been accepted towards fulfillment of the requirements for Ph .D . Pharmacology & Toxicology flake; M/fl/ Major professor degree in Date a” /j /7]7 0-7639 —'—'_._.-..__.. __’~__.." ' - M- “4 ‘4.—'——.~———v ——%‘fl~w‘—'—f——‘ RELATIONSHIP OF AMYGDALOID 5-HYDROXYTRYPTAMINE-CONTAINING NEURONS TO ANTICONFLICT EFFECTS OF BENZODIAZEPINES IN THE RAT By Clinton Donald Kilts A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology & Toxicology 1979 ABSTRACT Relationship of Amygdaloid 5-Hydroxytryptamine-Containing Neurons to Anticonflict Effects of Benzodiazepines in the Rat by Clinton Donald Kilts Despite widespread clinical use of the benzodiazepines, the mecha— nisms by which these drugs affect central neurons remain obscure. Determining how benzodiazepines produce their anxiolytic actions may afford insights into the neuronal substrates of clinically—defined anxiety. However, the concept that the anxiolytic effect is a specific action on a given neuronal system appears to be an oversimplification, as studies of these drug effects have implicated virtually every known putative neurotransmitter as well as the cyclic nucleotides. A conflict procedure was deve10ped based on intermittent electri— fication of a drinking tube set in a box housing a thirsted rat. This method allowed for rapid acquisition of stable baseline performance, a high output and identification of side effects; it also yielded anti- conflict profiles for the benzodiazepines correlated with their rela— tive clinical antianxiety potency. Past studies have associated anti- conflict effects with disruption of brain 5-hydroxytryptamine functions. If reduced central 5—hydroxytryptamine activity is causally related to anticonflict efficacy, other pharmacological manipulations that decrease 5—hydroxytryptamine activity should influence conflict behavior. Clinton Donald Kilts However, in the shocked drinking tube procedure 5—hydroxytryptamine antagonists showed poor anticonflict activity and failed to modify the anticonflict effect of diazepam. Elevation of brain 5—hydroxytryptamine following 5—hydroxytryptophan (with a peripheral decarboxylase inhibitor), potentiated the anticonflict effect of a submaximal dose of diazepam in the shocked drinking tube paradigm. Amitriptyline, which alone had no effect on suppression of drinking, potentiated the anticonflict effect of a submaximal dose of diazepam. prhlorophenylalanine yielded an equivocal anticonflict effect on the drinking behavior not well related to 5-hydroxytryptamine depletion. Depletion of brain 5-hydroxytryptamine by intracerebroventri— cular or intra-amygdaloid 5,7-dihydroxytryptamine produced a significant increase in punished responding, but only in the modified Geller— Seifter procedure using food reinforcement. Thus, while changes in brain S—hydroxytryptamine mechanisms may influence some types of condi— tioned suppression, reduction of brain 5-hydroxytryptamine activity is probably not the major mechanism of the anticonflict effect. Previous anticonflict studies using food reinforcement may have been misleading in this regard. The effects of acute and repeated doses of diazepam on brain 5- hydroxytryptamine activity were estimated from the rate of decline of S-hydroxyindoleacetic acid following monoamine oxidase inhibition by pargyline. Five consecutive daily doses of 1.8 mg/kg diazepam, but not an acute dose, reduced 5-hydroxytryptamine activity in the amygdaloid complex and anterior hypothalamus. A larger acute dose of diazepam (5.6 mg/kg) similarly reduced 5-hydroxytryptamine turnover, suggesting that chronic diazepam may cause an accumulation of the drug and/or Clinton Donald Kilts active metabolites. Anticonflict activity of diazepam may be asso— ciated with reduced 5-hydroxytryptaminergic input to the amygdala and hypothalamus, but perhaps not as a direct influence of the drug. It may be indirectly mediated, compensatory to a drug-induced enhancement of activity of postsynaptic tryptaminergic neurons. Both single and repeated doses of diazepam reduced dopamine turn- over, as measured by decline of the amine levels following d-methyl— tyrosine in some amygdaloid nuclei. This effect was less pronounced than that observed for S—hydroxytryptamine and insufficiently charac— terized to attempt a rigid correlation with behavioral effects. Nevertheless, the effect on dopamine turnover was less after chronic, as compared to acute, drug treatment, suggesting the development of tolerance. Since the anticonflict effect became more prominent during the chronic drug regimen, the effects of diazepam on brain dopamine systems and on behavior appear to be poorly associated. It can be concluded that effects of benzodiazepines on brain 5— hydroxytryptamine neurons are not related causally to anticonflict efficacy. Effects in the shocked drinking tube paradigm indicated that increased activity in brain S-hydroxytryptamine neurons may actually release punished responding. Since benzodiazepines release and d7 amphetamine enhances conditioned suppression similarly in the Geller— Seifter, Estes-Skinner (conditioned emotional response) and shocked drinking tube procedures, the latter method appears to be a valid measure of drug effects on conflict behavior. ACKNOWLEDGEMENTS I would like to acknowledge the contribution of my wife, Susan, to this effort. Her encouragement and compassion made this dissertation possible. It is with considerable respect and pride that I would like to thank my parents who gave me the ideals and encouragement to under— take this work. I would also like to express my gratitude to the members of my committee: Drs. Brody, Moore and Bennett with special thanks to Dr. Rech for giving me the opportunity to pursue and develop my research interests and to Dr. Rickert for the unselfish gift of his time and experience. ii ACKNOWLEDGEMENTS LIST OF TABLES LIST OF FIGURES INTRODUCTI N A. B. C. STATEMENT MATERIALS A. B. TABLE OF CONTENTS Page ii vi viii Objectives 1 Anxiety: Problems in Quantitation 4 Effects of Benzodiazepines on Avoidance and Suppressed Behavior 7 l. Avoidance behavior 7 2. Suppressed responding: Response-contingent - punishment (conflict) and conditioned suppression- 8 3. Chronic gs, acute benzodiazepine administration: The "initial treatment phenomenon" 13 Effects of Benzodiazepines on the Dynamic Equilibrium of Putative Neurotransmitters in the Central Nervous System 14 l. Catecholamines 15 2. S-Hydroxytryptamine (serotonin) 18 3. y-Aminobutyric acid and glycine 22 4. Acetylcholine and cyclic nucleotides 29 "Site of Action" of Antianxiety Drugs: The Limbic System 32 1. Effects of limbic structure lesions on avoidance behavior ~ ' 33 2. Effects of limbic structure lesions on suppressed behavior 38 The Amygdaloid Complex: Anatomical, Neurochemical and Electrophysiological Aspects 41 1. Anatomical aSpects of the amygdala 41 2. Neurochemical aspects of the amygdala 43 OF PURPOSE 48 AND METHODS 50 Animals 50 Body Temperature Recordings 50 iii TABLE OF CONTENTS (continued) C. RESULTS A. B. Page Behavioral Procedures 51 l. Suppressed responding: Response-contingent shock (conflict) 51 2. Suppressed responding: Noncontingent shock (con- ditioned emotional response) 53 Chemical Lesioning Techniques 55 Biochemical Procedures 57 1. Dissections 57 2. Methods of quantitation: Dopamine 6O 3. Methods of quantitation: 5-Hydroxytryptamine and 5—hydroxyindoleacetic acid 64 Estimation of D0pamine and 5-Hydroxytryptamine Turnover 71 Drugs 1 74 Statistics 75 78 Effects of Benzodiazepines on Suppressed Responding---- 78 Effects of Drug—Induced Alterations in Brain S—Hydroxy— tryptamine Activity on Conflict Behavior 97 Effects of S-Hydroxytryptamine Depletion on Conflict Behavior 113 Effects of Single and Repeated Administration of Diaze- pam on Body Temperature 123 Effects of Diazepam on DOpaminergic Neuronal Pathways—- 124 1. Effects of acute diazepam administration on .dopamine turnover in the amygdala, olfactory tubercle, nucleus accumbens and caudate nucleus——— 127 2. Effects of repeated diazepam administration on dopamine turnover in the amygdala, olfactory tubercle, nucleus accumbens and caudate nucleus--- 130 3. Effects of picrotoxin on diazepamrinduced changes in dOpamine turnover in the amygdala and olfactory tubercle 132 Effects of Diazepam on 5—Hydroxytryptamine Neuronal Pathways 137 1. Effects of acute and repeated diazepam administra— tion on steady state 5-hydroxytryptamine and 5- hydroxyindoleacetic acid concentrations in the amygdala, hypothalamus and hippocampus 137 2. Effects of acute diazepam.administration on 5- hydroxytryptamine turnover in the amygdala, hippo- campus and hypothalamus 142 3. Effects of repeated diazepam administration on 5— hydroxytryptamine turnover in the amygdala, hippo- campus and hypothalamus 145 iv ——_._. hue-.- w‘ . TABLE OF CONTENTS (continued) Page DISCUSSION 149 A. Utility of Conflict (Punishment) Procedures in Esta- blishing Relevant Correlates of the Therapeutic Action of Antianxiety Drugs 149 B. Effects of Benzodiazepines on Experimentally-induced Conflict Behavior 153 C. Feasibility of Pharmacological Manipulation of Brain 5-Hydroxytryptamine Activity: Effects on Conflict Behavior 155 D. Methodological Considerations Relevant to the Estima- tion of Dopamine and 5—hydroxytryptamine Turnover---——- 165 E. Effects of Acute and Repeated Diazepam Administration on Dopamine Turnover in the Amygdala, Olfactory Tu- bercle, Nucleus Accumbens and Caudate Nucleus ---------- 169 F. Effects of Acute and Repeated Diazepam Administration on 5-hydroxytryptamine Turnover in the Amygdala, Hypo- thalamus and Hippocampus 171 SUMMARY AND CONCLUSIONS 177 BIBLIOGRAPHY 179 Table 10 11 LIST OF TABLES Distribution of neurotransmitters among the amygdaloid nuclei Effects of various drugs on an experimentally-induced conflict situation involving the punished suppression of water consumption Effects of various drugs on an experimentally-induced conflict situation involving the punished suppression of a food-reinforced conditioned response Effects of various agents on response suppression re— sulting from noncontingent shock Brain 5-hydroxytryptamine concentrations following 5- hydroxytryptoPhan, 5,7—dihydroxytryptamine or pfchloro— phenylalanine Effect of diazepam (1.8 mg/kg) on rectal temperature (°C) in rats DOpamine concentrations in various nuclei in the rat brain Effects of acute diazepam administration on the d- methyltyrosine—induced depletion of dopamine in fore— brain areas Effects of repeated diazepam administration on the a- methyltyrosine-induced depletion of dopamine in fore— brain areas Effects of picrotoxin on the diazepam—induced attenua— tion of the u-methyltyrosine-induced depletion of Page 44 84 94 96 107 125 126 131 133 dopamine in the amygdala and olfactory tubercle--—--—~— 136 Effects of acute and chronic diazepam administration on the steady state concentration of 5—hydroxytrypta— mine in the amygdala, hippocampus and hypothalamus ————— vi 138 LIST OF TABLES (continued) Table Page 12 The effects of an acute dose of diazepam (1.8 mg/kg) on the rate of decline of the 5-hydroxyindoleacetic acid concentration in the amygdala, hippocampus and hypo- thalamus after i.p. injection of pargyline (75 mg/kg)-— 140 13 The effects of repeated diazepam administration (1.8 mg/kg) on the rate of decline of the 5-hydroxyindole- acetic acid concentration in the amygdala, hippocampus and hypothalamus after i.p. injection of pargyline (75 Ins/kg) 141 14 Effects of diazepam (5.6 mg/kg) on the pargyline-induced depletion of 5—hydroxyindoleacetic acid in the rostral amygdala and anterior hypothalamus 147 vii Figure 10 LIST OF FIGURES Page Structural formula of the 1,4—benzodiazepine nucleus and a derivative, diazepam 3 Schematic drawing of frontal sections of the rat brain 4620 and 3750 uM anterior to the frontal zero plane according to the stereotaxic atlas of Konig and Klippel (1963) 58 Partial mass spectra of dopamine (upper) and dOpamine— 2,5,6—d3 (lower) as pentafluoroproPionyl (PFP) deri- vatives with the proposed fragmentation patterns ------- 62 Partial mass spectra of 5-hydroxytryptamine (5-HT) (upper) and d—d2,B-d2-5—HT (lower) as pentafluoropro- pionyl (PFP) derivatives with the proposed fragmenta- tion patterns 66 Time course of the formation of the products of 5— hydroxyindoleacetic acid (5-HIAA) reacted with penta— fluoropropionic anhydride (PFPA) and pentafluoropro- panol (PFPOH) 69 Partial mass Spectra of S—hydroxyindoleacetic acid (5— HIAA) (upper) and d—dz—S—HIAA (lower) as pentafluoro— propyl ester pentafluoropropionyl derivatives with the prOposed fragmentation patterns 72 Dose—effect curve for diazepam on punished and un- punished water consumption 79 Time course of the effects of a submaximal dose (1.8 mg/kg) of diazepam on punished and unpunished water consumption 82 Evaluation of chlorpromazine for anticonflict activity- 86 Initial treatment effects of diazepam (5.6 mg/kg) on punished and unpunished responding in conflict— trained, drug—naive rats 89 viii LIST OF FIGURES (continued) Figure Page 11 Initial treatment effects of diazepam (1.8 mg/kg) on punished and unpunished responding in conflict- trained, drug—naive rats 91 12 Evaluation of methysergide for anticonflict activity-—— 98 13 Effects of peripheral 5-hydroxytryptamine antagonists on the anticonflict activity of a submaximal dose of diazepam.(1.8 mg/kg) 100 14 Evaluation of cyproheptadine for anticonflict activity— 102 15 Evaluation of cinanserin for anticonflict activity ————— 105 16 Effects of 5-hydroxytryptophan (S-HTP), alone or in combination with a peripheral decarboxylase inhibitor (MKr486), on the anticonflict activity of a submaximal dose of diazepam (1.8 mg/kg) 109 17 Effects of amitriptyline on the anticonflict activity of a submaximal dose of diazepam (1.8 mg/kg) 111 18 Effects of dflysergic acid diethylamide (d-LSD) on punished and unpunished responding 114 19 Effects of single and repeated administration of pr chlorOphenylalanine (p—CPA) on experimentally—induced conflict behavior 117 20 Effects of intraventricular and intraamygdaloid admini— stration of 5,7—dihydroxytryptamine (5,7—DHT) on experi— mentally-induced conflict behavior 120 21 The logarithmic decline of dopamine concentration in various brain nuclei following the administration of d— methyltyrosine (300 mg/kg, i.p.) — 128 22 The logarithmic decline of 5-hydroxyindoleacetic acid (5-HIAA) concentrations in various brain nuclei follow— ing the administration of pargyline (75 mg/kg, i.p.)——— 143 INTRODUCTION A. Objectives Derivatives of the 1,4—benzodiazepine nucleus (Figure 1) are among the most widely used drugs in clinical medicine in the United States. More than 57 million prescriptions and refills for diazepam, the most frequently prescribed member of this class of compounds, were filled in the twelve month period extending from March, 1976, to April, 1977. Until 1960 the pharmacotherapy of distressing anxiety symptoms con— sisted of little more than nonspecific sedation. It was with the introduction of the benzodiazepines (Randall §£_§1., 1960; Randall, 1960) that an effective means of attenuating anxiety which was reason— ably devoid of sedative liabilities became available. The mechanisms by which the benzodiazepines exert their antianxiety activity have been sought, with dubious success, by a number of research scientists. This is understandable in light of our lack of knowledge of the cellular basis of anxiety and the basis for its therapeutic management. Eluci— dation of the processes involved in mediating the anxiolytic actions of the benzodiazepines would greatly improve our concept of emotional states in general and afford additional insight in the areas of diag— nosis and pharmacological manipulation. The objectives of this research were two-fold, both of which related to the actions of diazepam (7—chloro—l,3-dihydro—1dmethy1—S- . R . phenyl—ZH—l,4—benzodiazepine—2-one; Valium ) in the rat. The 1 2 structural formula of this 1,4—benzodiazepine derivative is illustrated in Figure l. The first objective was to examine the effects of benzo— diazepines on a behavioral model of anxiety and to assess the value of such a model in establishing relevant pharmacological correlates of clinical antianxiety activity. As a second objective the potential of diazepam to alter the activity of neuronal systems utilizing several putative neurotransmitters was studied in an attempt to define the role of these transmitters in the mechanism of action of the benzodiaze- pines. To accomplish this, biochemical estimates of the effects of diazepam on the functional activity of selected neurotransmitters were utilized. Salient variables, i,§,, doses and pretreatment intervals, were drawn from the time course and dose—effect relationship of the previously determined diazepam-induced behavioral alterations. An important corollary in attempts to identify the neurochemical substrate(s) of a drug's action on the central nervous system is the identification of its anatomical substrate as well. The effects of diazepam on the concentrations and dynamics of the neurotransmitters 5—hydroxytryptamine (serotonin) and dopamine (3,4-dihydroxyphenyl- ethylamine) in the amygdala of the rat was examined. This collection of telencephalic nuclear groups was selected in light of the reported electrophysiological effects of the benzodiazepines on the amygdala (Schallek g£_a1,, 1964; Chou and Wang, 1977; Haefely, 1978) as well as data regarding the interaction between selective lesions of this structure and conditioned behavior (Kellicutt and Schwartzbaum, 1963; Spevack_etlal., 1975). The experimental data presented in this dissertation support a diazepam—induced depression in the functional activity of dOpamine- and .ammmmmao .m>flum>flumo m was muoHUSG odflmmnmwooucoplq.a ofiu mo wasauom HmHSHUSHum .H osswflm ziflso 330:2 uzamu<80~zuués 4 5—hydroxytryptamine—containing neurons with terminals in the amygdala as well as in other areas receiving a moderate to rich input. Whether the observed depression is the result of a direct action on presynaptic receptors or represents a compensatory or reflex response to an action on postsynaptic sites cannot be discerned from the present evidence alone. The relationship between the benzodiazepines and S—hydroxy- tryptamine has been interpreted heretofore as being an interference with 5—hydroxytryptamine-mediated synaptic mechanisms by the drug. However, over the course of these behavioral investigations into the effects of relatively selective 5—hydroxytryptaminergic agents and their interactions with diazepam, it became clear that a re-evaluation of this relationship may be in order. B. Anxiety: Problems in_guantitation Available methods for quantitation of clinical anxiety have serious limitations that make the accurate clinical assessment of drug effects on anxiety difficult. It is ironic that anxiety neurosis, by far the most common indication for benzodiazepine pharmacotherapy, is the disorder for which their clinical efficacy is most difficult to demonstrate. An understanding of this problem is apparent when one considers the nature of this emotional state. Anxiety is an elusive syndrome that defies definition and quantitation. It describes a phasic or episodic psychophysiological re3ponse resembling fear, though inappropriate to the reality of any perceived threat. Direct measurement of the effects of psychotropic agents on anxiety is not possible due to the fact that no single satisfactory method for quantifying anxiety has been devised. Results from numerous controlled 5 and uncontrolled clinical trials using physician—rated and/or patient self—rated scales testify to the difficulties involved in unequivocally delineating the relative effectiveness of an agent in ameliorating anxiety. However, the fact remains that no other antianxiety agents, regardless of how new or exotic, have proven superior to the benzo— diazepines (Greenblatt and Shader, 1974,1978). The use of animal behavioral models of clinical anxiety offers several indispensable advantages in evaluating theories regarding the mechanism of anxiolytic action of the benzodiazepines as well as of anxiety itself. First, ethical reasons preclude the invasive study of the physical and psychological make-up of the human subject. The selection and regulation of experimental parameters, so important to the formation of sound empirical conclusions, is difficult or often not permissible in the clinical setting. Second, animal models may prove useful because the phenomenon under study (anxiety) is simply too complex in the clinical setting for the experimental and conceptual tools at hand. Clinical diagnosis relevant to anxiety suffers from a reliance on verbal reports of feelings. While such feelings do indeed exist, the problem relates, as Skinner (1975) has suggested, to the lack of a suitable language for reliably conveying such an emotional state. It can be argued with some vigor that anxiety represents a unique- ly human condition and therefore no direct animal model is possible. However, animal behavioral models are not necessarily true models of human psychological or emotional disorders (Dews, 1976). Such proce— dures are tools that provide data correlating with the therapeutic as well as the undesirable effects of drugs in man (Cook and Davidson, 1973). These models are best viewed as experimental compromises in which a possibly simpler and available system.is used to gain insight into the more complex and less readily accessible system. Attempts to readily extrapolate from the animal pharmacology of the benzodiazepines to their effects on human anxiety neurosis result in behavioral inter- pretations that may be more in the eye of the beholder than in the brain of the animal. Heroic acts of faith should not be necessary in accepting underlying assumptions or interpretations of data if one adheres to the true value of the animal model as a predictor of clini— cal antianxiety activity. Tedeschi (1969) has preposed that such predictive value depends on the fulfillment of necessary but not sufficient criteria: I l. Tests should be selective enough to differentiate false positives and to distinguish therapeutic activity from side effects. 2. The relative potency of reference agents in the animal test should compare favorably with their relative potency in man. 3. Tolerance should not develop to the measure presumed to reflect therapeutic activity. 4. Tests should be sensitive enough to detect the activity of reference agents within a reasonable dose range. A point worth reiterating is that animal behavioral models are useful tools when used to establish relevant pharmacological correlates of the clinical properties of the benzodiazepines. Their utility does not imply the ability to precipitate and/or identify anxiety in ani- mals. With this orientation in mind the following represents a brief —7—_mmlw_m_m-. I" 7 review of the literature concerning the most consistent alterations in central nervous system (CNS) physiology and pharmacology elicited by the benzodiazepines. The behavioral effects of selective lesions of various limbic structures will be reviewed to illustrate potential functional similarities to the effects of systemic administration of the benzodiazepines. C. Effects of Benzodiazepines on Avoidance and Suppressed Behavior 1. Avoidance Behavior Experiments in which subjects have to refrain from reaponding to avoid punishment (passive avoidance) may be influenced by benzo— diazepines. Response latency or passive avoidance percentage is generally taken as an indirect measure of emotionality in passive avoidance situations, an increased latency and percentage of witheld responses being indicative of an ongoing "fear" reaction. Pretreatment with benzodiazepines produces a significant decrease in both parameters (Oishi §t_§1,, 1972; Fuller, 1970). Rats can also be taught to asso— ciate the termination of a previously neutral conditioned stimulus with footshock so that they actively avoid the shock by climbing a pole, jumping a hurdle or entering an alternate chamber on presentation of the conditioned stimulus (Greenblatt and Shader, 1974). Treatment with benzodiazepines prior to teSting reduces the active avoidance responses evoked by presentation of the conditioned stimulus in trained rats, although the deficit is less consistent and requires a higher dose than that necessary for disinhibition of passive avoidance (Greenblatt and Shader, 1974). Some authors (Cannizzaro g5 a1., 1972; Delini—Stula, 1971; Molinengo and Gamalero, 1969) claim that active avoidance ———‘ ”mt—“WW * *e 8 inhibition is a resultant of nonspecific sedation and neuromuscular impairment rather than fear attenuation. However, the benzodiazepines have been shown to produce active avoidance deficits at doses well below those which produce significant central nervous system depression (Cole and Wolf, 1962; Chittal and Sheth, 1963; Heise and Boff, 1962). The degree of the response deficit for any given benzodiazepine is, in part at least, a function of the manual difficulty involved in the avoidance response. Results from active avoidance studies are compli— cated by a number of factors which modulate the effects of benzodiaze— pines, g,g,, strain (Kuribara gt a1., 1976) and species differences and rates of baseline shock acceptance (Dantzer, 1977). Additional varia— bility arises from the uncertain role conditioned fear plays from a motivational point of view and its postulated attenuation by benzodia- zepines (Haefely, 1978). 2. ' Suppressed Responding: Regponse-contingent Punishment (Conflict) and Conditioned Suppression The restoration of behavior suppressed by punishment (anti— conflict or disinhibition effect) is the most consistently observed sequel of benzodiazepine treatment. The virtually unanimous finding that antianxiety agents attenuate the behaviorally suppressant effects of punishment in an immensely diverse array of experimental designs is testimony to the remarkable predictability of this effect. Typically, behavior reinforced to a high frequency of occurrence is subsequently suppressed by experimenter—introduced manipulation. The great majority of experimental approaches utilize a multiple schedule format. Geller and Seifter (1960) introduced the first "multiple schedule" procedure for which the most salient feature was a two component, repetitive cycle: a punishment component signalled by the onset of a previously neutral conditioned stimulus used to assay selective drug effects, .iflg., potential anxiolytic activity, and an unpUnished segment used to evaluate nonspecific drug effects, such as general depressant activity. In experimental designs in which signalled periods of re- sponding are simultaneously rewarded (food or water) and punished (electrical shock), interposed on a background of appetitive behavior, the benzodiazepines consistently increase the rate of punished re- sponding. Such procedures utilizing response-contingent punishment, referred to as conflict situations, have their basis in the prototype described by Geller and Seifter (1960). The alternation of periods of positively rewarded behavior with signalled periods of punishment— induced suppression of that behavior is considered to represent a conflict situation since the animal's response tendency is controlled by the consequences of both positive and negative reinforcement. The behavioral baseline often takes the form of conditioned operant re— Sponses that are positively reinforced on a variety of schedules although the use of unconditioned, innate responses such as food or water intake yields similarly stable baselines (Bertsch, 1976). Benzodiazepines appear to effectively attenuate the suppressive effects of punishment on consummatory as well as on operant behavior (Miczek, 1973a; Vogel §£U§1., 1971). The specific benzodiazepine—induced diminution of the sup— pressant effects of response—contingent punishment (anticonflict effect) is characteristically accompanied by an unchanged rate of unpunished responding (Goldberg and Ciofalo, 1969; Miczek, 1973a; 10 Geller gt_§13, 1962; Geller, 1964; Blum, 1970; Ts'o and Chenoweth, 1976; Robichaud §£_§1,, 1973), for chlordiazepoxide. Equivalent anti— conflict effects were obtained for chlordiazepoxide when the intensity of the conflict situation was progressively increased over successive trials by increasing the intensity of shock resulting from incorrect lever presses and by making the discrimination (light intensity) between punished and nonpunished sessions more difficult (Bremner g£_§1,, 1970). A significant restoration of punishment-suppressed reSponding has been reported for diazepam (Geller, 1964) and oxazepam (Babbini, 1975; Geller, 1964) at doses producing no appreciable nonspecific debilitation. In a modification of the conflict paradigms utilizing re— sponse-contingent punishment cited above, chlordiazepOXide has been shown to increase responding on an intermittent punishment schedule (fixed ratio) over a wide range of doses (2.2 to 35.6 mg/kg i.p., Cook and Davidson, 1973; 1.25 to 40 mg/kg p.o., Cook and Sepinwall, 1975). Decreases in the rate of unpunished behavior in the nonshocked segment (variable interval of reinforcement) of the two—schedule design was seen only at the higher doses tested. Such a disinhibition of punish— ment-suppressed responding by chlordiazepoxide was corroborated by Miczek (1973a) in a similar concurrent variable interval food rein- forcement/fixed ratio punishment schedule. Diazepam exhibits a greater and oxazepam a lesser potency relative to chlordiazepoxide in releasing behavior suppressed by intermittent punishment (Cook and Davidson, 1973). While the aforementioned anticonflict effects of the benzo- diazepines have dealt exclusively with rats, these effects appear to 11 generalize across a number of species including squirrel monkeys (Miczek, 1973a; Sepinwall gt a1., 1978), pigeons (McMillan, 1973), and pigs (Dantzer and Roca, 1974). Operant responses for a food reward may be suppressed during the presentation of a conditioned stimulus associated with and pre— ceding an aversive unavoidable stimulus (shock), even if the shock is not contingent upon response (Estes and Skinner, 1941). The suppres— sion of responding during the conditionedstimulus presentations is interpreted by Millenson and Leslie (1974) to represent an indirect measure of a conditioned emotional response (GER) which has seen considerable, though highly controversial, use as a behavioral baseline for the identification and characterization of antianxiety drugs. Pretreatment with chlordiazepoxide (Lauener, 1963; Miczek, 1973b; Bainbridge, 1968) attenuates this response suppression, again at doses producing insignificant effects on the rate of unpunished behavior. The findings with antianxiety drugs in conditioned emotional response paradigms appear less consistent than with response-contingent shock. In an attempt to define the role played by the relationship between the animal's behavior and the shock delivery, Huppert and Iversen (1975) found that response-contingent shock produced greater suppression of the behavior than did noncontingent shock. More signi— ficantly, they showed that chlordiazepoxide was more effective in releasing behavior from suppression when shock was response-contingent (controllable) than when shock was uncontrollable. The reported variability in the effects of benzodiazepines on the conditioned emotional response may be attributed to methodological inconsistencies 12 as well (Millenson and Leslie, 1974). For whatever the reasons, it would appear that conflict procedures (response-contingent punishment) exhibit certain advantages in terms of flexibility of design and unanimity of findings over methods using noncontingent shock (gag., conditioned emotional response) as reliable methods capable of yielding relevant pharmacological correlates of clinical antianxiety activity. In terms of qualitative discriminations, conflict procedures seem to be quite accurate in distinguishing minor tranquilizers from other classes of compounds. For instance, punished and unpunished behavior are not appreciablyaltered by neuroleptics until high dose levels are reached, at which point both are depressed simultaneously (Cook and Sepinwall, 1975; Schallek gt a1., 1972; Geller 3; a1., 1962). While an alleviation of suppression is seen for various barbiturates at low doses, the therapeutic index (median behavioral debilitating dose to median conflict—attenuating dose ratio) is characteristically much larger for the benzodiazepines (Blum, 1970). Ethanol significantly increased the rate of punished responding at a dose (1000 mg/kg, p.o.) which minimally affected the rate of unpunished responding (Cook and Davidson, 1973). Amphetamine (Geller, 1960; Wilson, 1977) and morphine (Geller gt a1., 1963; Cook and Davidson, 1973) appeared to be ineffec- tive in restoring punishment—suppressed responding over a wide range of doses. In fact, dramphetamine is well known to enhance this suppression. Conflict methods have the added advantage of being able to detect qualitatively different profiles within the class of benzo— diazepines. Different benzodiazepines can be seen to differ in the magnitude and dose range of the anticonflict effect as well as the dose 13 at which significant depression of unpunished responding first occurs (Cook and Sepinwall, 1975). 3. Chronic vs. Acute Benzodiazepine Administration: The "ini— tial treatment phenomenon" All of the behavioral results described in the previous discussion were obtained after the animals had been exposed to the benzodiazepines on several prior occasions, i,§,, after the animals were already drug-experienced. Margules and Stein (1968) observed that the degree and even the direction of the modification induced in the punishment-suppressed behavior by antianxiety drugs could vary depen— ding on whether the animals had a prior history of drug exposure or were being given the agent for the first time ("drug-naive"). Upon initial administration to conflict-trained, drug-naive animals, the benzodiazepines often produce a marked nonspecific depression of behavior, visible in a reduction of unpunished behavior with either decreased or relatively unchanged rates of punished responding. With additional eXposure to the drugs at daily or longer intervals charac— teristic increases in punished behavior occur while unpunished behavior returns to predrug control rates (Margules and Stein, 1968). A quali- tatively similar development of tolerance to the depressant effects and simultaneous unmasking or development of their antipunishment effects has been reported for chlordiazepoxide (Cook and Sepinwall, 1975; Sepinwall §t_a1,, 1978; Goldberg 3; £13, 1967), diazepam (Sepinwall E; El}, 1978), flurazepam (Cannizzaro g£_§1,, 1972), and oxazepam (Mar- gules and Stein, 1968). 14 Sepinwall_g£_a1. (1978) also found the effects of benzodiaze- pines on unpunished and punished behavior to follow different courses with their repeated administration in squirrel monkeys. These effects were found to be dose—related and to reflect individual differences in drug sensitivity. A possible clinical correlate to the "initial treatment phenomenon" defined in animals may exist in that after benzodiazepines have been administered to humans it has been reported that sedative effects predominate at first but then diminish in inten- sity and disappear after a few days of treatment in anxious patients (warmer, 1965) as well as normal volunteers (Hillestad gt g1,, 1974). The initial sedation was replaced by feelings of relaxation and well— being after several daily treatments. An additional similarity between the animal and clinical findings is that the effect of diazepam in drug-naive humans also appears to be dose-related, i,§,, marked seda— tive effects occur initially at a daily dose level of 30 but not 15 mg (Hillestad gg.a1., 1974). D. Effects of Benzodiazepines on the Dynamic Equilibrium of Putative Neurotransmitters in the Central Nervous System Various drugs acting on the central nervous system appear to do so either by regulating the synthesis, storage, release or metabolism of a neurotransmitter or by facilitating or inhibiting the action of a transmitter at postsynaptic receptors. Such a simplified rationale has not been shown to exist for the benzodiazepines or central nervous system depressants in general and there is no unified theory capable of correlating their pharmacological effects with a specific action on a given neuronal system. To the contrary, the collective findings of a 15 recent symposium (Costa and Greengard, 1975) devoted to the study of the benzodiazepines appear to implicate virtually every known chemical family of neurotransmitters as well as the cyclic nucleotides. The following section is a brief review of attempts that have been made to relate some of the pharmacological actions of the benzodiazepines to their effects on various possible neurochemical substrates. It is in keeping with the overall theme of this dissertation that certain neuro— chemical influences of the benzodiazepines, perhaps localized to a Specific area of the brain, underlie their efficacy in the treatment of anxiety. Therefore, this review will be primarily limited to their anxiolytic activity as predicted by effects of these drugs on suppressed responding in animal models. The scope of this review will be further restricted to those neurotransmitters that appear to be the most promising candidates in light of the current state of the art, i,§,, catecholamines, 5-hydroxytryptamine and y—aminobutyric acid. 1. Catecholamines Neither chlordiazepoxide (Corrodi §E_§1,, 1967) nor diazepam (Consolo gt a1., 1975) have been shown to consistently alter the whole brain content of norepinephrine or dopamine. However, static tissue concentrations, especially whole brain values, do not necessarily reflect the effects of drugs on neurotransmitter function. Estimates of transmitter turnover rate (i,§,, the rate at which the renovation of the pool of a transmitter in a given area proceeds) have supplanted static measurements as viable approximations of neuronal activity. Using both biochemical and histochemical techniques Lidbrink §E_§1, (1973) reported a decrease in whole brain depamine turnover by 16 benzodiazepines and barbiturates in animals pretreated with d-methyl— tyrosine. The doses used were associated with a clear-cut depression and, in the case of barbiturates, a hypnotic action. More signifi— cantly, Fuxe 32 a1. (1975) have shown a similar retardation in the rate of disappearance of d0pamine fluorescence following a-methyltyrosine. This was found in the nucleus accumbens, olfactory tubercle and head of the caudate nucleus by lower, behaviorally effective anticonflict doses of diazepam and chlordiazepoxide (1 and 10 mg/kg, respectively). The actions of diazepam and chlordiazepoxide on dopamine turnover appear to be more definitive in the limbic forebrain as compared to the caudate due to the high variability in the response in the latter. In an attempt to characterize the interaction between benzo— diazepines and neuroleptics at central dopamine neurons, Fuxe gt_§1, (1975) demonstrated the ability of diazepam (1 and 5 mg/kg) to selec— tively counteract the pimozide—induced increase in dopamine turnover in the "limbic forebrain" (nucleus accumbens and olfactory tubercle) but not in the neostriatum. Keller g£.§13 (1976) have reported the reduc— tion by diazepam (10 mg/kg) of the haloperidol or chlorpromazine- induced increase in 3-methoxy-4-hydroxyphenylacetic acid (homovanillic acid, HVA) to be similar in striatum and limbic forebrain. Diazepam (10 mg/kg) also reduces DA turnover in the entorhinal cortex (Fuxe gt_ _§1., 1975), a terminal of the mesocortical dopaminergic projections (Hokfelt at 31., 1974). Furthermore, chlordiazepoxide and diazepam decrease striatal dopamine turnover as estimated from the rate of decline of the specific activity of 3H—dopamine following its intra- ventricular administration, i,§,, labelling of the endogenous stores of catecholamines (Taylor and Laverty, 1973). 17 High doses of diazepam and chlordiazepoxide (10 and 25 mg/kg, respectively) reduce cortical norepinephrine turnover, an observation attributed to a reduction of neuronal activity in the ascending ceru- locortical noradrenergic pathway (Corrodi §t_al,, 1971). Norepine- phrine turnover in subcortical structures appears to be relatively unaffected. However, chlordiazepoxide (Corrodi g£_§1,, 1967) and diazepam can counteract the stress-induced increase in norepinephrine turnover in all parts of the rat brain (Lidbrink g; g1., 1973). Thus, when activity in noradrenergic pathways is generally increased, selec- tivity is lost. The significance of such an effect of the benzodiaze— pines in regard to their antianxiety action is not supported by the inability of lower, nonsedative doses (at least in the cortex) to attenuate the stress-induced increase in norepinephrine turnover (Fuxe 35 331;, 1975). Behaviorally, intracerebroventricular (Stein §t_a1,, 1975) and systemic (Robichaud gt_§1,, 1973; Sepinwall §t_a1,, 1973) admini— stration of propranolol, a B-adrenergic receptor antagonist, failed to release punishment—suppressed behavior in the rat conflict test. Phentolamine, an u-adrenergic receptor antagonist, was similarly inactive. Intracerebroventricular injections of lfnorepinephrine increased rather than decreased the punishment-lessening activity of systemically administered benzodiazepines (Stein §t_a1,, 1973). These findings contradict the hypotheses that the anxiety—reducing activity of the benzodiazepines depends on a reduction of transmitter function at noradrenergic synapses. The inactivity of the dopamine receptor blocking agents in conflict procedures similarly weakens any role that the benzodiazepine—induced decrease in dopaminergic activity (if the 18 decreased turnover is truly reflecting this) may play in their anti— conflict effects. Furthermore, the central catecholamine destruction produced by the intraventricular administration of 6—hydroxyd0pamine (6—OHDA) neither produces an anticonflict effect nor interferes with the anticonflict activity of diazepam (Lippa 23 31,, 1977a). Intraventricular lfnorepinephrine antagonized the depressant effect of oxazepam (10 mg/kg) on nonpunished responding in the rat conflict test (Stein gt al., 1973), suggesting that the generalized depressant activity of benzodiazepines may be mediated by a reduction of noradrenergic activity. These same authors preposed that the initial decrease in unpunished behavior on the first eXposure of conflict—trained, drug-naive animals to benzodiazepines ("initial treatment phenomenon") is related to a transient benzodiazepine-induced decrease in norepinephrine turnover. They have subsequently reported that a decrease in norepinephrine turnover in the midbrain-hindbrain region, as estimated by the reduced rate of decline of 3H-norepine— phrine following its intraventricular administration, was no longer detectable after six daily doses of oxazepam (20 mg/kg, i.p.). While this correlates with the disappearance of the generalized depressant action of this dose of oxazepam on unpunished responding in the con— flict test following six daily injections, Cook and Sepinwall (1975) have found the correlation to be much less clearcut. 2. S-Hydroxytryptamine Studies with various benzodiazepines have yielded ambiguous results concerning their effects on brain 5-hydroxytryptamine con— centration. Lidbrink 2E.§l- (1974) and Consolo_g5.al. (1975) have 19 found chlordiazepoxide and diazepam to produce no reproducible changes in cortical or whole brain steady state 5-hydroxytryptamine values. In contrast, a significant increase in whole brain 5—hydroxyindoles and tryptophan has been reported for diazepam (Jenner §£_al,, 1975). However, the observed increase could be due, at least in part, to the influence of benzodiazepines to competitively inhibit the binding of tryptophan onto serum albumin, thus increasing the concentration of free serum tryptophan and thereby increasing the amount available to the brain (Bourgoin g; a1., 1975). Whereas a reduction of brain 5-hydroxytryptamine turnover associated with the acute administration of benzodiazepines is ge— nerally agreed upon, considerable variability exists in terms of the benzodiazepine studied, dose, site of maximum effect and technique used to estimate turnover. Little or no work has been done in investigating the time course of this effect. Using radioisotopic tracer techniques, a decreased rate of decline of injected radiolabelled 5—hydroxytrypt— amine has been reported for chlordiazepoxide (20 mg/kg; Lippmann and Pugsley, 1974), oxazepam (Wise §t_§1,, 1972) and diazepam (Chase 35 _al., 1970). Furthermore, diazepam (5 and 10 mg/kg) and chlordiaze— poxide (20 mg/kg) slow the rate of formation of 3H-S-hydroxytryptamine from its radiolabelled precursor (3H-tryptophan) following the intra- venous injection of the latter (Dominic 2; a1., 1975). Sedative doses of chlordiazepoxide and diazepam decrease the rate of decline of cortical 5—hydroxytryptamine following the inhibition of the rate— limiting enzyme involved in its synthesis (Lidbrink gt El-: 1974). A single injection of diazepam (10 mg/kg) decreased in_yi§£9_midbrain tryptOphan hydroxylase activity (Rastogi §t_a1,, 1977). Although the 20 benzodiazepines appear to decrease 5—hydroxytryptamine activity, little is known concerning the level of regulation at which this effect is "exerted or even whether it is perhaps secondary or compensatory to a drug—induced increase in serotonergic activity. If the anxiety—reducing activity of benzodiazepines is related to their ability to reduce brain serotonin turnover then agents that reduce serotonergic activity should also counteract the suppressive effects of punishment. This contention (Stein gt_a13, 1975) has been the impetus for a large number of studies dealing with the interaction between putative antagonists of 5—hydroxytryptamine or inhibitors of its synthesis and conflict behavior or other forms of conditioned suppression. Increases in the rate of punished responding have been reported following the administration of the peripheral 5—hydroxytrypt— amine antagonists methysergide (Stein §t_a1,, 1973; Graeff, 1974; Cook and Sepinwall, 1975), cinanserin (Geller gt_a1., 1974; Cook and Sepin— wall, 1975) and cyproheptadine (Graeff, 1974; Schoenfeld, 1976). Miczek and Luttinger (1978) have reported methysergide to be ineffec— tive in attenuating the behaviorally suppressive effects of noncontin— gent punishment (conditioned suppression or conditioned emotional response). The utility of these peripheral antagonists as blockers of central 5—hydroxytryptamine receptors is compromised by what appears to be an uncertain relationship with 5—hydroxytryptamine receptors in the brain (Haigler and Aghajanian, 1974; Jacoby g; g1., 1978; Bfirkipg£_§1., 1978). prhlorophenylalanine (p—CPA) selectively depletes 5—hydroxy— tryptamine consequent to the inhibition of tryptophan hydroxylase, the 21 rate—limiting enzyme involved in S-hydroxytryptamine synthesis (Jequier g£_§1., 1967). The administration of pfchlorophenylalanine has been associated with a significant anticonflict effect (Robichaud and Sledge, 1969; Geller and Blum, 1970; Stein gt_§1,, 1973) as well as an ”attenuation of a conditioned suppression of the conditioned emotional response type" (Hartmann and Geller, 1971). However, Blakely and Parker (1973) and Cook and Sepinwall (1975) have been unable to obtain consistent anticonflict effects with pfchlorophenylalanine. The intra— cerebroventricular or intracerebral injection of the neurotoxic deriva- tives of 5—hydroxytryptamine, 1,2,, 5,6- and 5,7—dihydroxytryptamine (5,6- and 5,7—DHT), have been observed to produce a marked reduction in brain S—hydroxytryptamine. When 5,6-dihydroxytryptamine was admini— stered by an intracerebroventricular route to c0nf1ict—trained rats a profound increase in punished responding was observed (Stein §£_§1,, 1975; Lippa §E_§l., 1977b). This anticonflict effect reportedly paralleled the pharmacological time course of 5—hydroxytryptamine depletion, while the rate of unpunished responding was little affected. Bilateral injection of 5,7—dihydroxytryptamine into the ventral medial tegmentum, the main pathway taken by ascending serotonergic fibers, prevented the acquisition of behavioral suppression in a conflict paradigm and released punished behavior in previously trained animals (Tye §£_§1,, 1977). In addition, the behaviorally suppressive effects of intracerebroventricular 5—hydroxytryptamine or chemical stimulation of the dorsal raphe nucleus (concentration of 5-hydroxytryptamine- containing cell bodies in the midbrain that supply the serotonergic innervation to areas of the forebrain and diencephalon) by carbachol can be reversed by oxazepam (Stein §£_§1,, 1973). 22 These findings, together with those obtained from neuro— chemical studies, have led to the suggestion that a benzodiazepine- induced decrease in functional 5—hydroxytryptamine activity may be causally related to their anxiety—reducing actions. However, the relationship between the benzodiazepines and central 5-hydroxytrypt— amine activity may be less simplistic than this. The activation of brain 5—hydroxytryptamine receptors by 5-hydroxytryptamine precursors or by 5-hydroxytryptamine agonists has been found to induce head twitches (Jacobs, 1976; Corne gg‘al., 1963). Nakamura and Fukushima (1977,1978) have feund the head twitches induced by intracerebroventri— cular 5—hydroxytryptamine or the systemic administration of 5—hydroxy— tryptamine precursors and agonists to be potentiated by benzodiaze— pines. Thus, gross neurological imbalances in volitional motor control may be induced by the increase in central serotonergic activity, which may then indirectly influence the pattern of punished responding. It should also be emphasized that the preposed role of 5—hydroxytryptamine as the neurotransmitter implicated in conditioned suppression was developed from results involving conflict tests utilizing food rein— forcement. Since brain 5—hydroxytryptamine pathways appear to be implicated in the motivation of feeding behavior (Samanin gt al., 1977), an indirect effect on the pattern of conflict behavior may be exerted via this mechanism. 3. _y:Aminobutyric Acid and Glycine Various data suggest that some of the biochemical (Fuxe g5_ 31,, 1975; Biggio §£_§1,, 1977; Mao gt a1., 1975; Costa §E_§l,, 1975), pharmacological (Haefely §t_§1,, 1975; Costa et_§1,, 1975) and IIIIIIIIIIIIIIIIIIIIIIIIIIIIIll"':::_______________fi ‘ "'_"'”"V '9: "“'*”I" ' "MIN" 23 behavioral (Soubrie §£_§1,, 1976; Soubrie and Simon, 1977) effects of benzodiazepines result from an as yet undefined interaction with y— aminobutyric acid mechanisms. Electrophysiological studies of the nature of this interaction have yielded conflicting results since a facilitation of y—aminobutyric acid—mediated synaptic transmission (Polc _e_t;_a1., 1974; Haefely at. 3;” 1975; Polc and Haefely, 1976; Costa §£_§1,, 1975) as well as an antagonism of y-aminobutyric acid-mediated inhibition (Steiner and Felix, 1976; Gahwiler, 1976; Curtis 3.; 531;, 1976; MacDonald and Barker, 1978) have been reported for the benzo— diazepines. The relationship of these findings to the pharmacological profile of the benzodiazepines has led to the suggestion that several actions of the benzodiazepines (muscle relaxation, ataxia, anticon- vulsant effects) may be mediated by an enhanced function of y-amino— butyric acid—containing neurons. Costa gghal. (1975) suggest a similar 5 mechanism to be involved in their antianxiety effects based on a con- cept of functional economy or efficiency. However, when one asso- ciates the main therapeutic actions and side effects of benzodiazepines with accepted or presumed functions of y-aminobutyric acid, an involve— ment of y-aminobutyric acid in their anxiolytic action is not presently apparent (Haefely g5flal., 1975). Along the same line of investigation, quite obvious connections for y-aminobutyric acid-mediated pathways { with the muscle relaxant, ataxic and anticonvulsant effects of the j benzodiazepines appear to be substantiated. A Studies of the central neurotransmitter role of yéaminobu- tyric acid and effects of benzodiazepines have led to 2 major proposals: 1) The depolarization of afferent terminals by the putative transmitter —7—_—' ~— ---.—- ‘-—n'- - : --«~—»—~ M——~. ; 4..--..—-. . . , ,, ._._._...... .. -... . , 24 and the enhancement of this process of presynaptic inhibition by the benzodiazepines may be of significance for the muscle relaxant pro— perties of these drugs (Polc_ggflal., 1974). 2) The postsynaptic hyper— polarization of y-aminobutyric acid in cerebellar inhibitory processes may be facilitated by benzodiazepines and may account in some part for the ataxia and motor incoordination elicited by this drug class (Costa §E_§13, 1975; Haefely 35 al., 1975). A similar enhancement of post— synaptic inhibition at higher levels of the neuraxis has been reported for benzodiazepines, i.g,, the medulla (Dray and Straughan, 1976), substantia nigra (Haefely 35 a1,, 1975) and cerebral cortex (Zakusov_g£ a1,, 1975). It has been suggested that information concerning the parti— cipation of y—aminobutyric acid in mediating some of the central actions of the benzodiazepines may be obtained through a study of the selectivity and potency of antagonism against convulsions elicited by perturbations of y-aminobutyric acid function (Costa EE.§l-: 1975). Benzodiazepines preferentially and Specifically antagonize the seizures associated with a reduction in the biosynthesis of y-aminobutyric acid elicited by isoniazid (Costa ggflal., 1975), thiosemicarbazide and 3— mercaptopropionic acid (Haefely 35 31,, 1975). Their ED50 anticon— vulsant dose was approximately 7— to 10-fold higher against seizures elicited by y—aminobutyric acid receptor blocking agents, i,§,, picro— toxin and bicuculline (Mao gt El}: 1975). Barbiturates and diphenyl— hydantoin weakly and equipotently antagonized the convulsions induced by y-aminobutyric acid synthesis inhibition or receptor blockade. The action of aminooxyacetic acid, a compound that increases the concen— tration of y—aminobutyric acid in the central nervous system by 25 inhibiting y-aminobutyric acid—transaminase, parallels that of the benzodiazepines on convulsions induced by inhibition of y—aminobutyric acid synthesis (Haefely §§_§1,, 1975). Various y-aminobutyric acid— mediated systems appear to be affected in a similar fashion by the benzodiazepines, aminooxyacetic acid and exogenous y—aminobutyric acid: enhancement of presynaptic inhibition in the spinal cord (Polc g; g1., 1974) and cuneate nucleus (Polo and Haefely, 1976), increase in firing rate of ponto-geniculo-occipital waves induced by depletion of brain 5— hydroxytryptamine (Ruch-Monachon_g£fla1., 1976), and enhancement of the cataleptic effect of neuroleptics (Keller 35 a1,, 1976). It should be mentioned that the majority of studies con— cerning the interaction between y-aminobutyric aciddmediated systems and benzodiazepines have so designated such functions based upon the ability of picrotoxin and/or bicuculline to antagonize them. There— fore, conclusions drawn from such studies are only as valid as the selectivity and specificity of these agents for y—aminobutyric acid receptors. Similar caution should be exercised when using inhibitors of the synthesis or catabolism of y—aminobutyric acid. y—Aminobutyric acid content in the lumbosacral cord of spinal cats was significantly elevated by diazepam (3 mg/kg, i.v.). Saad (1972) reported a significant increase in mouse cerebral hemisphere y-aminobutyric acid content following doses of 5 and 10 mg/kg diazepam. Higher doses (30 mg/kg, i.p.) were required to increase mouse and rat whole brain concentrations (Haefely §t_§1,, 1975). Studies concerning the effect of diazepam on the estimated y—aminobutyric acid turnover rate in various brain regions suggest that the increased concentrations l:- j”. 26 of y—aminobutyric acid following diazepam are not due to an accelera— tion of its synthesis. Diazepam decreased the turnover rate of y— aminobutyric acid in rat whole brain (Fuxe_§§”§l., 1975), various cortical regions (Pericic g£“§1., 1977) and the cerebellum (Guidotti, 1978), as well as the nucleus accumbens and caudate nucleus (Mao_§£ 31,, 1977). The y-aminobutyric acid receptor agonist, muscimol (Krogs— gaard—Larsen g£_§1,, 1975), similarly decreases y—aminobutyric acid turnover (Guidotti, 1978; Mac 95 gl., 1977). This fact further supports a positive influence of the benzodiazepines on y-aminobutyric acid activity and suggests that the observed decrease in y—aminobutyric acid turnover elicited by diazepam may reflect a reflex or compensatory decrease in y—aminobutyric acid synthesis in response to an enhanced y-aminobutyric acid receptor activity. The molecular mechanisms by which benzodiazepines enhance y—aminobutyric aciddmediated inhibition is the subject of much debate (Haefely, 1977), the scope of which exceeds the intent of this review. Behavioral evaluation of the effects of manipulations in central y—aminobutyric acid neuronal activity on punishment—suppressed responding (conflict) generally yield negative results. An evaluation of aminooxyacetic acid for anticonflict activity revealed no increase in the rate of punished responding in doses ranging from 2.5 to 25 mg/kg, i.p., whether given 40 or 240 minutes prior to a session (Cook and Sepinwall, 1975). Furthermore, the response to a subthreshold anticonflict dose of diazepam was unaffected by aminooxyacetic acid pretreatment. The elevation of brain y—aminobutyric acid produced by another inhibitor of its catabolism, ethanolamine—O-sulphate, was 27 similarly ineffective (File and Hyde, 1977). The y-aminobutyric acid receptor agonist, muscimol, exhibited no anticonflict activity and did not alter the anticonflict activity of diazepam (Sullivan 35 31., 1978). Unpunished responding was dose-relatedly depressed. Considerable controversy exists concerning the effects of y— aminobutyric acid receptor blockade on the anticonflict profile of the benzodiazepines. The anticonflict effects of oxazepam (Stein ggnal., 1978) and chlordiazepoxide (Billingsley and Kubena, 1978) were antago- nized by picrotoxin in a dose-related manner without appreciably affecting the rate of unpunished responding. However, Cook and David- son (1978) have found picrotoxin to be ineffective in this regard. Moreover, Lippa ggual. (1977b) have shown the effects of picrotoxin on the ataxic and anticonflict pr0perties of chlordiazepoxide to be dissociable. Subconvulsant doses of picrotoxin produced a dose—related reversal of the ataxic effects of chlordiazepoxide while failing to appreciably alter its anticonflict effects in untrained rats. Thus, the y-aminobutyric acid-mediated properties of the benzodiazepines may play a role in the sedative and motor-incoordinating effects of the benzodiazepines whereas the evidence for a similar role in their anti— conflict (antianxiety) activity is much less convincing. Further support for this contention is suggested by the antagonism exhibited by y-aminobutyric acid receptor blockers of the behaviorally depressant effects of diazepam (Soubrie and Simon, 1978) and chlordiazepoxide (Billingsley and Kubena, 1978). Furthermore, chronic diazepam admini— stration resulted in the development of tolerance to the y—aminobutyric aciddmediated properties of this benzodiazepine, as measured by the 28 ability of diazepam to protect against the convulsions produced by the y—aminobutyric acid receptor blocker, bicuculline (Lippa and Regan, 1977). A similar tolerance deveIOpment is not characteristic of the anxiolytic actions of benzodiazepines and thus minimizes the importance of y—aminobutyric acid mechanisms in mediating this effect. In addition to y-aminobutyric acid, glycine is a prominent inhibitory neurotransmitter in the central nervous system. The distri— bution of glycine is predominantly in the brainstem and spinal cord and, like y—aminobutyric acid,this transmitter hyperpolarizes the neuronal membrane of postsynaptic cells. However, unlike y-aminobu— tyric acid, it does not depolarize presynaptic terminals, i,§,, does not evoke presynaptic inhibition (Costa §t_§1., 1975). Young 25 a1. (1974) have proposed that benzodiazepines may exert their antianxiety and muscle—relaxing effects by mimicking glycine at its postsynaptic receptor sites. 3H—Strychnine, a potent glycine antagonist, binds to synaptic membrane preparations of spinal cord and brain stem in a selective fashion, indicating an interaction with postsynaptic glycine receptors (Young and Snyder, 1973). The regional localization of strychnine binding in the central nervous system correlates with endo— genous glycine concentrations, i,§,, spinal cord > medulla > midbrain. Additionally, the displacement of strychnine by glycine and other amino acids parallels their glycine—like neurophysiologic activity. Benzodiazepines have been shown to be very potent inhibitors of the in_vitro specific binding of labelled strychnine to glycine receptors. Moreover, the potency of a series of 21 benzodiazepines as inhibitors of 3H-strychnine binding correlates closely with their 29 potency in pharmacological and behavioral tests considered relevant to clinical efficacy (Young §t_§1,, 1974). While the above findings would suggest an interaction of benzodiazepines with the neurotransmitter glycine at its central nervous system receptor sites, Hunt and Raynaud (1977) have not found the in_yi£rg_receptor binding activity of the benzodiazepines to be reflective of differences in ip_yiyg_pharmacolo— gical activity and suggest that differences in their liposolubility may well be an important factor in the displacement of 3H-strychnine binding. Evidence from electrophysiological investigations further weakens the significance of an interaction between benzodiazepines and glycine receptors, as diazepam neither inducedinhibitory phenomena characteristic of glycine nor blocked inhibitory processes in which glycine is claimed to be involved (Curtis E; él-’ 1976). Furthermore, the rank-order potency of 10 benzodiazepines in displacing 3H-strych- nine and in producing anticonflict effects did not correlate signifi— cantly (Cook and Sepinwall, 1975). The disinhibitory effect of oxaze— pam on punished behavior (anticonflict effect) was not selectively antagonized by strychnine (Stein gg_al., 1975). Finally, doubt con- cerning the possible role of glycine involvement derives from the observation that the protection afforded by benzodiazepines against convulsions induced by strychnine is relatively weak (Costa 93.31., 1975) and tolerates out with repeated diazepam administration (Lippa and Regan, 1977). 4. Acetylcholine and Cyclic Nucleotides Acetylcholine (ACh) is unique among neurotransmitters in that the steady state concentration of brain acetylcholine is consistently 30 increased by diazepam (Ladinsky g; 31., 1973; Cheney ggngl., 1973; Consolo §£_§1,, 1975). The concentrations of choline as well as choline acetyltransferase and acetylcholinesterase activities ig_yitgg were unaffected (Consolo §£_§1,, 1974). Based on these findings it has been suggested that diazepam may be acting centrally by inhibiting the presynaptic release of acetylcholine. Although this hypothesis would also explain the decreased turnover of brain acetylcholine by diazepam (Cheney gt g1,, 1973), a similar decrease can be obtained upon admini— stration of dOpamine or y-aminobutyric acid agonists,_egg., apomorphine and muscimol, respectively (Zsilla_§E.§1., 1976). Therefore, these findings do not exclude the possibility that theobserved effects on cholinergic activity may be indirectly mediated. In the rat the increase in acetylcholine produced by diazepam (5 mg/kg, i.v.) was limited to hemispheric structures, 123., the striatum, hippocampus, and hemispheric residuum after removal of the hippocampus and striatum (Cansolo EE.§l:’ 1975). The increase attained a maximum within 15 minutes and waned to control values by 60 minutes. Among diazepam's effects, the anticonflict, anticonvulsant or hypo— thermic actions cannot be temporally correlated with the increase in hemispheric acetylcholine. However, study of more specific areas of the brain or the use of more specific measures of cholinergic activity may indicate a stronger correlation between the biochemical effect on acetylcholine mechanisms and the pharmacological actions of benzo— diazepines. A variety of evidence has been amassed which implicates the Cyclic nucleotides in translating the release of neurotransmitters into 31 a postsynaptic response (Daly, 1976). If the central actions of the benzodiazepines can be attributed to an alteration of neurotransmitter function, then their activity may be reflected in changes in cyclic nucleotide concentrations or the activity of the enzyme systems con— trolling the steady state concentrations. Decreases in rat cerebellar 3',5'—cyclic guanosine monophosphate (cGMP) (Costa §t_al,, 1975), increases in 3',5'—cyclic adenosine monophosphate (CAMP) in brain slices (Hess §t_a1,, 1975) and inhibition of in_yit£9_phosphodiesterase activity (Beer §£_§1,, 1972) have been reported for the benzodiaze— pines. The possibility of a dopaminergic or cholinergic link is sug— gested by the ability of dopamine and acetylcholine receptor agonists or blockers to alter rat cerebellar 3',5'-cyc1ic guanosine monophos— phate content (Burkard gt_al,, 1976). Alternatively, diazepam can prevent the increase of cerebellar 3‘,5'—cyclic guanosine monophosphate content elicited by a decreased y—aminobutyric acid synthesis or a blockade of y—aminobutyric acid receptors (Costa §E_§1,, 1975; Biggio gg.al., 1977). The consistency of the relationship between cerebellar 3',5'—cyc1ic guanosine monophosphate concentrations and y—aminobutyric acid inhibitory transmission has prompted the measurement of concen— trations of cerebellar 3',5'—cyclic guanosine monophosphate to monitor ig_yiyg_drug interactions with y—aminobutyric acid inhibitory mecha— nisms (Mao §£_a1., 1975). The concentrations of y—aminobutyric acid and 3',5'—cyclic adenosine monophosphate in the rat cerebellum were unaltered by diazepam pretreatment. However, while the cerebellum may play a significant role in some actions of the benzodiazepines, i,g,, ataxia and motor incoordination, it is doubtful that it functions in 32 their antianxiety effects. The effects of benzodiazepines on non— cerebellar cyclic nucleotides are poorly characterized. In contrast to the findings of Beer §£_al. (1972), an evaluation of potent phospho— diesterase inhibitors, i.e., caffeine and theophylline, for anticon— flict activity showed them to be inactive; nor did theophylline alter the effect of a subthreshold dose of chlordiazepoxide (Cook and Sepin— wall, 1975). Moreover, anticonflict potency and in vitro phosphodies— terase inhibitory activity for a series of benzodiazepines appeared to be unrelated, further weakening the possible participation of the phosphodiesterase inhibitory activity of many benzodiazepines in their antianxiety effects. E. "Site of Action" of Antianxiety Drugs: The Limbic System An important corollary in attempting to associate benzodiazepine— induced behavioral alterations with their biochemical effects is the elucidation of the anatomical substrate in the central nervous system which, upon interaction with benzodiazepines, produces the response characteristically seen in the whole organism. Our primitive under— standing of the cellular basis of anxiety greatly complicates attempts to relate structure to function. However, the limbic system has for years been implicated by neurophysiologists as the seat of emotion and its behavioral, autonomic and endocrinological sequelae. For the sake of brevity, the "limbic" system will be c0nsidered to represent a func— tionally interconnected composite of subcortical nuclear groups,_i.g., the amygdala, septum, hippocampus and hypothalamus. However, in its broadest definition, based on the numerous projections and interconnec- tions of this core, the limbic system contains cortical and olfactory 33 components in addition to other subcortical structures (Isaacson, 1974). A concept involving anxiety as an emotional manifestation of limbic system function and its subsequent attenuation by a benzodiaze- pine-induced alteration in the activity of limbic neurons is plausible. Such an association is no less warranted than attempts to correlate ataxic or sedative effects of these drugs with alterations in motor function of the cerebellum or sensory processing in the spinal cord, respectively. Moreover, the term "chemical amygdalectomy" has been used to describe electrophysiological evidence which focuses the action of various benzodiazepines on the limbic system, particularly the amygdala (Schallek gg_§1,, 1964; Morillo, 1962; Chou and Wang, 1977; Haefely, 1978). The placement of relatively Specific, circumscribed lesions in a particular structure has seen widespread usage in an attempt to gain information concerning the function of the area in the expression of a behavioral pattern. When considering the effects of lesions of the amygdala, septum or hippocampus on behavioral paradigms previously shown to be affected in a consistent fashion by benzodiazepines, the behavioral effects of amygdaloid lesions more closely parallel those seen following benzodiazepine administration. The following is a brief review of the effects of bilateral lesions of the amygdala, septum and hippocampus on avoidance behavior and suppressed responding. 1. Effects of Limbic Structure Lesions on Avoidance Behavior Bilateral limbic lesions, whether septal, hippocampal or amygdaloid, are generally associated with poor passive avoidance performance. The deficit obtained with amygdaloid lesions is charac- teristic for a number of qualitatively different experimental designs 34 (see below), all relying on the animal's suppression of a prepotent motor response as an indication of passive avoidance. The character— istic pattern seen with bilateral ablation of the amygdala is a de— creased latency, when compared to shamsoperated controls, in performing a response previously associated with an aversive stimulus. Amygdaloid lesions have been reported to disrupt passive avoidance, i,§,, de— creased latency to respond and increased rate of shock-contingent responding, in a shocked drinking tube procedure (Pellegrino, 1968; Kemble and Tapp, 1968) or shock prod (Blanchard and Blanchard, 1972). Similar results were obtained when measuring latency in entering a shocked second compartment of a two—compartment design (Suboski gt_§1,, 1970; Gaston and Freed, 1969) or in stepping down onto a shocked grid floor (Russo gt §l~i 1976). Although lesions of the corticomedial division of the amygdala impair passive avoidance performance to some extent, the basolateral lesions produce a more severe deficit (Kemble and Tapp, 1968; Pellegrino, 1968). While passive avoidance deficits produced by amygdaloid lesions are not inconsistent with a lesioneinduced inability to asso- ciate fear or anxiety reactions with neutral stimuli paired with pain, Blanchard and Blanchard (1972) have reported a similar deficit for rats in the passive avoidance of unconditioned threatening stimuli (an immobile cat). The decreased freezing and crouching and increased approach behavior parallel that seen in response to a conditioned stimulus, i,§,, shock prod. Moreover, the decreased avoidance and enhanced approach tendencies to both unconditioned and conditioned threat stimuli do not support a lesion-induced increase in general motor activity or an alteration in response perseveration as 35 explaining the behavioral effects. No appreciable difference between amygdaloid—lesioned and control animals was seen for post—operative, home cage, food and water consumption (Dacey and Grossman, 1977) or in bar-pressing rate on either a continuous reinforcement schedule or gradually increasing fixed ratio schedule for a food reward (Pelle— grino, 1968). As flinch-jump and galvanic skin responses shock thresholds are not increased by amygdaloid lesions, impairments in shock motivated behavior are probably not due to elevated pain thresholds (Blanchard and Blanchard, 1972). The effects of bilateral amygdaloid lesions in passive avoidance behavior must be interpreted in terms of the size and loca— tion of the lesion in the amygdala and the nature of the inhibited response. The inclusion of extra—amygdaloid structures, i,§,, internal capsule, caudate—putamen, pyriform cortex, does not appear to be a significant factor in the lesion—induced impairment of passive avoi— dance (Kemble and Tapp, 1968; Blanchard and Blanchard, 1972). The passive avoidance deficits appear to generalize to a variety of inhi— bited responses and aversive stimuli (Nagel and Kemble, 1976). The most consistent behavioral manifestation of amygdaloid lesions is a taming effect; placidity is increased and responsiveness to normally noxious stimuli is markedly reduced. Amygdaloid lesions can even produce a calm demeanor in animals made rageful and hyper— reactive by lesions in the septal area. Slotnick (1973) found that animals with amygdaloid lesions, in contrast to c0ntrols, continue to Show shock-contingent exploratory behavior and little or no "fear” behavior (freezing). Such a deficit in fear behavior is reflected by a ... 36 reduction in species-specific defensive reactions and an altered reac— tivity to fearful stimuli (Blanchard and Blanchard, 1972) and has been attributed to a lesion—produced decrement in "fear or anxiety arousal". Bilateral lesions of selected limbic structures have been repeatedly shown to alter performance in a variety of active avoidance tasks. Rats with bilateral septal (Schwartzbaum EEH§1., 1967; Krieck- haus gt g1., 1964) or hippocampal (Isaacson gt gl., 1961; Copobianco 33 _al., 1977) lesions are facilitated in the acquisition of a conditioned avoidance response. These findings are in good agreement with a postulated role of septal and hippocampal areas in reSponse inhibition (Pellegrino, 1968) and are further supported by the poor passive avoidance performance seen in animals similarly lesioned. In contrast, rats with bilateral amygdaloid lesions are generally deficient in acquiring conditioned active avoidance reSponses. Bilateral amygda— lectomy produces deficits in the acquisition of a shuttle-box condi— tioned avoidance response (Bush §E_gl,, 1973; Suboski 25 gl., 1970). In addition to a retarded acquisition of an active avoidance response, Robinson (1963) reports a lesion-induced impairment in the acquisition of a new response (wheel turning) motivated by the acquired fear rather than shock itself. Similar deficits appear to exist for active avoi— dance responses other than avoidance performance in a shuttle-box: avoiding an approaching shock prod (Blanchard and Blanchard, 1972) and bar—pressing (Campenot, 1969). While the unconditioned stimulus is usually in the form of electric shock, the conditioned stimulus has taken many forms, e.g., tones of varying intensity, light, different colored compartments, and a combination of a tone and a change in 37 illumination. A similar lesion—produced deficit in responding sig— nalled by a diverse array of conditioned stimuli argues against a lesion-induced alteration in any specific sensory modality involved in the recognition of a fearful stimulus. Lesion—induced deficits in the performance of active avoi— dance tasks are consistent with amygdalectomy curtailing the registra- tion of habituation to the conditioned stimulus. If such were the case, the unconditioned suppressive quality of the conditioned stimulus would not habituate out with repeated presentation and the response rates would remain low in subsequent conditioning, i,§,, conditioned stimulus—unconditioned stimulus pairings. However, Spevack e£_al. (1975) have failed to demonstrate any reliable effects of amygdalectomy on habituation. The dual findings of poor passive avoidance but enhanced or unaltered active avoidance learning with septal and hippocampal lesions is consistent with a lesion-induced impairment in response inhibition. The poor performance of amygdalectomized animals in both types of avoidance tasks weakens the suggestion that destruction of this latter structure interferes with the ability to withhold responding. While the amygdala in the rat may play a significant role in the animals' response to aversive situations, extrapolation of an analogous function to the amygdaloid complex of humans is tenuous. However, the decreased emotionality resulting from temporal lobe lesions (Kluver— Bucy syndrome) and the feelings of fear or anger reported by conscious patients undergoing stimulation of the amygdaloid complex, which often occurs at the start of an epileptic seizure emanating from the temporal 38 lobe, implicate limbic structures as the neurological substrate of emotions in humans (DeGroot, 1975). 2. Effects of Limbic Structure Lesions on Suppressed Behavior Evaluation of lesion—induced alterations in the postoperative conditioning of a conditioned emotional response (CER) affords further differentiation between the effects of specific limbic lesions. The conditioned emotional response was previously mentioned in terms of its disinhibition by various benzodiazepines and can be defined as a be- havioral pattern, elicited by the presentation of a conditioned stimu— lus predicting onset of an unavoidable aversive stimulus, which is incompatible with previously learned operant responding for food reward. Responding during the conditioned stimulus is not shock— contingent: termination of the conditioned stimulus is associated with the unconditioned stimulus, usually footshock, regardless of the subject's response. In studying the effects of lesions located in circumscribed areas of the hippocampus, Nadel (1968) found the acqui— sition of a conditioned emotional response to be unaffected or slightly enhanced by ventral or dorsal hippocampal lesions, respectively. Similarly, rats with bilateral septal lesions show levels of response suppression (conditioned emotional response) during the conditioned stimulus not unlike that seen in control animals (Hobbs, 1976). Lesions of the amygdala characteristically disrupt the forma— tion of conditioned emotional responses, i,g,, a lack of response suppression with successive conditioned—stimulus-unCOnditioned stimulus pairings. The consistent and reproducible nature of the lesion—induced impairment has led Kellicutt and Schwartzbaum (1963) to postulate the 39 utilization of the conditioned emotional response technique as a sensitive index of amygdaloid damage. While sham—operated controls require only 2—3 daily sessions to attain a reasonable level of re— sponse suppression during the conditioned stimulus presentation, amygdalectomized animals failed to meet an identical criterion of suppression within the preset limit of 15 sessions (Kellicutt and Schwartzbaum, 1963). Moreover, the lesioned group as a whole exhibited no pattern of suppression exceeding that attributable to chance. DOubling the intensity or duration of the shock proved ineffective in promoting the suppression of bar-pressing during the conditioned stimulus. Similar deficits in a qualitatively similar conditioned emotional response paradigm have been reported for bilateral amygdaloid lesions by Spevack §E_§1. (1975). The licking response of water— deprived amygdalectomized rats was relatively unaffected by successive presentations of a compound conditioned stimulus, the termination of which was associated with footshock. In sham—operated and unoperated controls the conditioned stimulus presentations initiated a significant conditioned suppression of licking behavior. The slight adipsia reported for amygdalectomized animals makes unlikely the possibility that the lesion—produced conditioned emotional response deficits can be attributed to their inability to inhibit drinking. Spevack g£_§1, (1975) conclude that the observed deficits in response suppression are reflective of an interference with the arousal of fear rather than lesion-produced deficits in habituation or response inhibition. 4O Similarities in the behavioral effects of bilateral amygda- loid lesions and the systemic administration of benzodiazepines suggest that a benzodiazepine—induced decrease in amygdaloid function may be somehow related to their actions observed in the whole animal. Evi— dence in support of such a "pharmacological amygdalectomy" comes from electrOphysiological studies which report a benzodiazepine—induced reduction in the electrical activity recorded from the amygdala and an alteration by low doses in the influence of this structure on various other neuronally connected systems. Diazepam and chlordiazepoxide suppress the spontaneous single unit activity of the amygdala and hippocampus in a dose-related manner (Chou and Wang, 1977). Umemoto and Olds (1975) found that these same two benzodiazepines reduce both the background rate of discharge and the neuronal responses correlated with the presentation of a condi— tioned stimulus signalling aversive stimulation (conditioned emotional response paradigm), but only in the amygdala. Benzodiazepines exert a profound inhibitory action upon the hippocampal response to low fre— quency stimulation of the ipsilateral amygdala (Morillo §E_§1., 1962). Moreover, the hippocampal potential evoked by amygdaloid stimulation in the curarized cat is reduced by antianxiety drugs with an order of potency reflecting therapeutically effective doses, whereas other psychoactive agents, such as tricyclic antidepressants and neuroleptics, are either inactive or modify the induced activity in the opposite direction (Haefely, 1978). Tsuchiya and Kitagawa (1976) have extended the influence of the benzodiazepines (using evoked potentials) to include the various neuronal connections of the intra-limbic as well as 41 midbrain-limbic systems. It should be emphasized that an alteration of spontaneous or evoked electrical events in specific regions of neuronal organization by a psychoactive drug does not necessarily prove that the therapeutic action of the drug in humans is mediated through this particular brain area. A more direct indication for a limited site of action may be obtained by intracerebral injections of drugS. Thus, diazepam injected in minute amounts into the amygdala, but not the hippocampus, was found to produce changes in electroencephalographic activity (EEG) and to inhibit the carbachol—induced hypothalamic rage reaction in a manner similar to systemic administration (Nagy and Decsi, 1973). F. The Amygdaloid Complex: Anatomical, Neurochemical and Electro— physiological Aspects ' 1. AnatOmical Aspegts of the Amygdala The amygdala is a collection of nuclei found in the anterior portions of the temporal lobes in the brains of primates. Subdivision of the amygdaloid nuclei into groups has produced counts of from 5 to 22 in various species and by various authors. The amygdaloid complex of the rat is generally accepted to be composed of eight major dissec— table nuclei: the central, lateral, medial, cortical, basal, medial posterior, basal posterior and posterior (Figure 2). The basal nucleus is further subdivided into medial and lateral parts by some authors (Isaacson, 1974). A consensus shows that the amygdala is phylogene— tically, physiologically and anatomically divided into a corticomedial and basolateral division. The corticomedial complex is generally thought to be composed of the cortical, medial and central nuclei and 42 the nucleus of the lateral olfactory tract and, at least in the rat, is associated with olfactory structures. The basolateral division is considered to be made up of the basal, lateral and accessory basal nuclei. This group is associated with neocortical systems and is the most prominent nuclear group of the amygdala in the human. The projections from the amygdala have not been completely mapped but two main efferent systems are recognized: the stria termi— nalis and ventral amygdalofugal pathways. The stria terminalis pre— sumably receives the majority of its fibers from the corticomedial division and a lesser contribution from the basolateral nuclei. DeOlmos (1972) further subdivides the stria terminalis into dorsal, ventral and commissural components and includes among their termina— tions the lateral septal nucleus, nucleus accumbens, olfactory tubercle, anterior olfactory nucleus, preoptic area, ventromedial nucleus of the hypothalamus, premammillary area, habenula and the bed nucleus of the stria terminalis. The ventral amygdalofugal pathways are thought to receive input from both nuclear divisions and a considerable contri— bution from periamygdaloid cortex. In addition to innervating many of the structures receiving input from the stria terminalis, the ventral pathway fibers project to the prepyriform, pyriform and entorhinal cortex, the caudate—putamen, and the thalamus. In light of the inter— connections between cortex and amygdala, Kemble and Tapp (1968) have argued that the behavioral manifestation of amygdaloid lesions may reflect the destruction of fibers that originate in the periamygdaloid cortex and course through or end in the amygdaloid cortex. They have subsequently shown a similar impairment of passive avoidance responding '2‘ 43 to result from bilateral lesions of the pyriform cortex or basolateral amygdaloid nuclei. Afferents to the amygdala arise from many sources. The mesencephalon, diencephalon and telencephalon are all known to directly or indirectly supply input to the amygdala (Veening, l978a,b). 2. fieurochemical Aspects of the Amygdala The monoamines, acetylcholine and yaminobutyric acid are well represented in the amygdala. Drug-induced changes in the concen— tration of these putative transmitters in the amygdala in Egtg offers little insight into the potential functional importance of these changes. This is due to their uneven distribution among the nuclei which make up the amygdala. Table 1 summarizes the available informa— tion concerning the distribution of depamine, norepinephrine, 5— hydroxytryptamine and acetylcholine in the amygdaloid nuclei of male rats. The in yiggg_activity of L—glutamate decarboxylase, the rate— limiting enzyme in the synthesis of y—aminobutyric acid from glutamate, is included as a marker of y—aminobutyric.acid-containing neurons. The olfactory tubercle and nucleus accumbens are included as representa— tives of the rostral limbic nuclei. The caudate nucleus is included as a reference structure. Dopamine is well represented and unevenly distributed among the amygdaloid nuclei of the rat (Table l), in agreement with the density and distribution of catecholaminergic fluorescent nerve termi— nals in the rat amygdala (Jacobowitz and Palkovits, 1974). The central and lateral nuclei contain the most dopamine; the medial, medial posterior and posterior nuclei contain the least. The high dopamine 44 TABLE 1 Distribution of Neurotransmitters Among the Amygdaloid Nuclei Brain Nuclei DA NE S-HT ACh GAD Amygdaloid Nuclei Anterior amygdaloid 45.9i 5.1 9.5i0.9 19.0il.9 54.0: 7.6 -—— area Central 16.9i 2.8 13.4i2.5 16.8t2.2 ——— 422140 Lateral 22.0i 3.3 12.0il.0 20.7i2.7 23.4i 1.6 333i38 Basal 8.6i 0.5 9.1i0.9 17.5il.6 45.3i 6.4 34li34 Basal posterior 5.4: 1.1 6.3i0.9 23.0i2.1 ——- --— Medialposterior 1.6: 0.4 4.3:1.0 17.6i2.l ——— ——— Posterior 1.9i 0.3 2.5:0.5 15.9i1.6 —-— ——— Cortical 5.1: 1.0 7.5il.7 17.3il.3 7l.5il4.2 349:33 Medial 3.6i 0.6 ll.4il.2 19.9i4.0 48.2: 6.3 310i15 Rostral limbic Nuclei Nucleus accumbens 67.5i 4.0 14.1il.5 14.7i1.2 87.6i 8.9 574i70 Olfactory Tubercle 114.4i15.8 7.2i0.9 26.112.6 29.2t 2.5 526i28 Caudate nucleus 96.6: 6.5 l.li0.2 5.8i0.5 87.6i 4.7 270110 Dopamine (DA), norepinephrine (NE) and 5—hydroxytryptamine (5—HT) con— centrations are expressed as ng/mg protein i S.E.M. (Brownstein et_al., 1974; Saavedra §t_al., 1974; Demarest 23 al., 1979; W. Lyness, personal communication). Acetylcholine (ACh) concentrations are expressed as ng/mg protein :tS.E.M. (Cheney et_§l,, 1975). L—Glutamate decarboxy— lase (GAD) activity is expressed in pmole 002 produced/ g Protein/hr i S.E.M. (Tappaz gt_al., 1976). 45 content of the central and lateral amygdaloid nuclei is consistent with a reported pathway from midbrain dopaminergic cell bodies terminating in the central (Ungerstedt, 1971) and lateral (Kizer, 1976) amygdaloid nuclei. Moreover, Hokfelt ggngl. (1974) described the presence of dopaminergic terminals in the amygdala, with the lateral and basal areas exhibiting a particularly dense localization. Tyrosine hydroxy- lase, the enzyme that catalyzes the rate—limiting step in the synthesis of dopamine, also shows a high content and uneven distribution among the amygdaloid nuclei (Saavedra and Zivin, 1976). A similar pattern is seen for both isoenzymes of monoamine oxidase (type A and B), a major enzyme in the catabolism of depamine, norepinephrine and 5—hydroxy— tryptamine (Hirano §£_§l,, 1978). MicroiontOphoretically applied dopamine has been reported to depress the firing rate of spontaneously active and glutamate-excited amygdaloid cells of the cat (Ben—Ari and Kelly, 1976; Straughan and Legge, 1967) and rat (McCrea §§_§l,, 1973). Moreover, the inhibitory action of dopamine was antagonized by iontophoretic applications of the neuroleptic a—flupenthixol (Ben-Ari and Kelly, 1976). A dopamine- sensitive adenylate cyclase in the amygdala (Clement-Cormier and Robison, 1977), the stimulation of which is inhibited by neuroleptics, may mediate the inhibitory effects of depamine receptor stimulation. Thus, various lines of investigation suggest that dopamine may function as a neurotransmitter or modulator of the total synaptic response in the amygdala as has been demonstrated for the better characterized dOpaminergic systems. 46 S—Hydroxytryptamine exists in fairly high concentrations, and, unlike d0pamine, is rather uniformly distributed among the amygdaloid nuclei (Table l). The relatively moderate concentrations of tryptophan hydroxylase in the amygdaloid nuclei (Saavedra, 1977) suggests that this structure has some capacity to synthesize 5-hydroxytryptamine from its precursors. The amygdala receives a dense, uniform 5—hydroxytrypt— amine input from the midbrain raphe (Aghajanian §E_al,, 1973). Both microiontoPhoretically applied 5-bydroxytryptamine and electrical stimulation of the dorsal raphé nucleus markedly inhibit the sponta— neous firing rate of amygdaloid cells (wang and Aghajanian, 1977). The inhibitory influence of 5-hydroxytryptamine on the amygdaloid cells appears to be of a tonic nature as the depletion of brain 5—hydroxy— tryptamine by 5,7—dihydroxytryptamine or pfchlorophenylalanine results in a significantly increased discharge rate in the amygdala in addition to preventing the inhibitory effects of dorsal raphé nucleus stimula— tion. Norepinephrine exhibits less regional differences among the amygdaloid nuclei than dopamine. In general, the more basal and caudal nuclei are poorer in norepinephrine than the more dorsal and rostral ones. The concentrations of acetylcholine exhibit considerable variability among the amygdaloid nuclei (Table 1). The iontophoretic application of acetylcholine produces an excitatory effect on amygda— loid firing rate (Wang and Aghajanian, 1977). L-glutamate decarboxy— lase activity is rather evenly distributed throughout the amygdaloid nuclei, with the greatest activity in the central nucleus (Table 1). Ben—Ari §t_al, (1976) report that L—glutamate decarboxylase activity is 47 higher in the rostral than in the caudal part of the central nucleus. Spontaneously active and glutamate—excited amygdaloid cells are inhi- bited by the iontOphoretic application of y—aminobutyric acid (Ben-Ari and Kelly, 1976). In summary, the amygdala is biochemically as well as anatomi- cally heterogeneous. Any one nucleus receives a multiplicity of inputs. STATEMENT OF PURPOSE The validity of studies attempting to correlate drug-induced behavioral alterations with their biochemical effects requires that these effects share similar dose—effect and temporal relationships. Although the concept of the benzodiazepines as potent anxiety—allevia— ting agents has received extensive behavioral consideration, the great majority of the biochemical studies, whether by design or interpreta— tion, has revolved around properties other than that of an anxiolytic, 1,3,, as anticonvulsants, muscle relaxants and sedatives. The doses of the benzodiazepines characteristically used in studying their effects 0n the dynamics of putative neurotransmitters in the central nervous system are usually of sufficient magnitude to elicit pronounced seda— tion and muscle relaxation and are well in excess of doses that are effective as anticonflict (antipunishment) agents in animal models. While it has been argued that biochemical determinations are generally less sensitive than functional tests to reveal changes in neuronal activity, it is in most cases improper to attempt to correlate low—dose behavioral effects with biochemical effects that are obtained with relatively much higher doses. This investigation characterized the effects of diazepam on a quantitative behavioral index of drug action (experimentally induced conflict), the relevance of which to clinical anxiety has been gene— rally accepted by the foremost investigators in this area of 48 IIIIIIIIIIIII'llll'...-----l___ _ "Pammflwflwir 49 experimental pharmacology. The salient characteristics of diazepam administration so derived (e,g,, dose- and time—effect relationships and differences resulting from acute Kg, chronic dosing regimens) were then applied to a study of the effects of diazepam on biochemical measures seeking putative neurotransmitter correlates in limbic and extralimbic structures of the anticonflict activity. MATERIALS AND METHODS A. Animals Male Sprague-Dawley rats (Spartan Research Animals, Inc., Haslett, Mich.) weighing 200—320 g were maintained in air conditioned rooms with room—lights alternated on a 12—hour dark—light cycle. Subjects used in all aspects of this research were group-housed three or four to a cage, except during periods of training, testing or controlled feeding or drinking. Animals classified as "food—deprived" were gradually reduced to 75-80 percent of their original body weight and maintained at these weights by limited feeding after each testing session. "Water—de— prived" animals were denied access to water for a 48-hour period prior to conditioning and were subsequently limited to water obtained during daily experimental sessions. B. Body_Temperature Recordings The rectal temperature of male rats was measured by inserting a thermistor probe 6 cm into the rectum and displayed on a telethermo— meter (Model 4lTA, Yellow Springs Instrument Co.). Temperature readings for diazepam and vehicle—injected control subjects were made at 15—minute intervals for 60 minutes post—injection. 50 51 C. Behavioral Procedures l. Suppressed Responding: Response—contingent Shock (Conflict) Two forms of conflict behavior were used which differed in the nature of the response used to generate the behavioral baseline. In the first case, the behavior took the form of an unconditioned, consummatory response (tube licking) and in the second a conditioned, instrumental response (bar pressing) was employed. The response suppression consequent to the punishment of consummatory behavior, here termed the conditioned suppression of drinking, will be described first. The experimental chamber used in the conditioned suppression of drinking conflict procedure consisted of a rectangular box (30x56x28 cm) with plexiglass sides and a stainless steel floor and ceiling. A 6.5 cm drinking tube protruded through the center of one wall at an angle of approximately 75 degrees to the wall plane and at a height of 15 cm from the floor. A speaker attached to a tone generator was affixed to the ceiling. Contact between the floor and drinking tube by the rat completed a circuit during the latter period of tone presen- tation resulting in an electric shock being delivered through the tube, with each shock automatically recorded. The drinking tube was coated with Insl—X insulating paint up to the tip so as to limit the comple— tion of the circuit to contact with the floor and the water in the tube. A calibrated polyethylene tube was attached to the drinking tube to allow for monitoring the volume of water consumed. Water—deprived rats were allowed two consecutive daily 10- minute periods of unpunished access to the drinking tube. Subsequent placement in the chamber initiated a lO—minute daily session with 52 alternating periods of unpunished and punished consummatory responding distinguished by the presence of a tone. A 7—second tone was intro- duced at varying intervals of time (K'= 21 sec) which signalled the delivery of a shock on tube contact during the last 5 seconds of the tone. Tube contact was not punished during the initial 2 seconds of the tone and during the silent intervals. A shock intensity of 30 uamps was found to produce a low, stable baseline of shocks received following approximately two weeks of daily sessions. More signifi— cantly, an intensity of 30 uamps produced a level of suppression amenable to rate—decreasing as well as rate-enhancing effects of drugs. The measurable parameters in the punished and unpunished components, 1,3,, number of shocks received and water consumption, respectively, were analyzed separately. The method has been used in another study, the results of which have been published (Ford_g£flal., 1979). The second form of experimentally-induced conflict behavior represents a modification of the conflict procedure of Geller and Seifter (1960). The apparatus consisted of standard sound insulated environmental chambers, each containing a lever, an automatic feeder for the delivery of food reinforcement (45 mg Noyes food pellets), a loudspeaker to provide auditory stimuli, and a grid floor which could be electrified. Electro-mechanical programming equipment was used to control experimental contingencies and record data. Food—deprived rats were trained to respond on a fixed ratio schedule of 40 lever presses for each food pellet reinforcement (FR— 40). Following stabilization of this operant response a non-aversive tone of 17 seconds duration was randomly presented on an average of once every three minutes (habituation trials). When performance had 53 again stabilized a punishment contingency was added in which every response during the last 15 seconds of the tone was simultaneously rewarded (continuous reinforcement) and punished (scrambled footshock). An important corollary concerning the effect of benzodiazepine admini— stration on suppressed responding is that the magnitude of the dis- inhibition may be dependent on the level of baseline or control sup— pression, which in turn is a function of the intensity of the punishing stimulus (McMillan, 1973), magnitude of deprivation and inter-animal differences. To normalize the inherent differences between animals and to maximize the drug related differences, a similar level of reSponse suppression in each animal was maintained by titrating with varying shock intensities (0.4—1.1 mA, 25 msec duration). After training (40— 60 days), subjects made very few punished responses while the tone was on. Testing sessions were of one hour duration with total responses during the alternating silent, unpunished period (fixed ratio—40) and tone—signalled, punished (continuous reinforcement) components collected and analyzed separately. 2. Suppressed Responding: Noncontingent Shock (Conditioned Emotional ReSppnse) In contrast to the response—contingent (controllable) nature of shock presentation characteristic of conflict procedures, the response suppression engendered by noncontingent shock is characterized by the uncontrollable quality of the aversive stimulus, i,§,, the animal is shocked irrespective of its own actions. The "conditioned suppression" or "conditioned emotional response" paradigm employed was a modification of that reported by Estes and Skinner (1941). *t—uhhflh M" “7"" ' " 4 54 The experimental apparatus was the same as that used in the above-mentioned bar-press conflict procedure. Additional commonalities shared by both methods include the shaping of the operant response (bar pressing) to a stable performance on a fixed ratio—40 schedule of reinforcement and the subsequent habituation trials to remove the unconditioned effects of the tone. However, the conditioned emotional response paradigm differed from the bar—press conflict method in two important ways: 1) the fixed ratio—40 reinforcement schedule was maintained throughout the alternating signalled and unsignalled periods; and 2) the termination of the tone (15 sec in duration with a variable intertrial interval averaging 3 min) was associated with an unavoidable shock delivered to the grid floor through a shock scrambler. Testing sessions were of one hour duration with total responses during the silent intervals and the tone—signalled periods being collected and analyzed separately. These three behavioral models share several features essen— tial to their utility. First, all three procedures utilize conditioned stimuli which forewarn of shocks, as well as silent intervals or "safe periods" in which shocks never occur. Drug-induced increases in the rate of suppressed responding are interpreted as an index of anti— anxiety activity. Comparison of behavior in the safe periods before and after the introduction of the drug provides a within-sessions control for nonspecific side effects on locomotor activity, sensory function and overall motivation. Secondly, a selective manipulation of parameters is used to produce a partial, but not total, suppression of the conditioned or unconditioned behavior. Thirdly, experimental sessions in all three procedures were conducted 7 days a week with drug W, 55 testing sessions conducted at 5—day intervals. In evaluating drug effects each subject was used as its own control. The average number of responses in each component on the 3 days immediately preceding drug treatment were compared to the number on the treatment day. D. Chemical Lesioning Techniques Past efforts to develop agents which exhibited neurotoxic specifi— city towards indoleamine—containing neurons in the central nervous system culminated in the synthesis of the 5—hydroxytryptamine deriva— tives 5,6-dihydroxytryptamine and 5,7—dihydroxytryptamine. The spec— trum of biochemical effects attributed to these dihydroxylated trypta— mines (Baumgarten ggflgl., l973a,b) is considered to result from the degeneration of serotonergic preterminal axons and axon terminals. The cytotoxic effect of 5,7—dihydroxytryptamine on 5-hydroxytryptamine— containing neurons was used to investigate the role of 5-hydroxytrypt— amine in the expression of conflict behavior in rats. 5,7—Dihydroxy— tryptamine was selected over its 5,6-dihydroxylated congener in light of the reported advantages in terms of stability, potency and toxicity (Sanders—Bush and Massari, 1977). In the present study intracerebral injections of 5,7—dihydroxy— tryptamine were used to produce selective lesions of the amygdala, an area which receives a dense and uniform serotonergic input from the midbrain raphé (Aghajanian_gg.§l., 1973). Rats weighing 270-320 g were anesthetized with equithesin (an anesthetic mixture containing chloral hydrate, pentobarbital, magnesium sulfate, propylene glycol, and ethyl alcohol) and mounted in a KOpf stereotaxic instrument with the 56 incisor bar set 5.0 mm above the interaural plane. The creatinine sulfate salt of 5,7—dihydroxytryptamine (Regis Chemical Co.) was dissolved in 0.9% NaCl with ascorbic acid added (0.2 mg/ml) to protect against autooxidation. Stereotaxic injections were made through a 31 gauge stainless steel cannula connected to a 10 ul Hamilton syringe. The coordinates (Pellegrino and Cushman, 1967) were for the amygdala: 1.0 mm posterior to bregma, i4.5 mm lateral to midline, 8.0 mm ventral to the surface of the cortex. The 5,7—dihydroxytryptamine (calculated as the free base) was injected bilaterally in the amygdala (7 ug in 3.5 ul on each side of the brain). The injection speed was 0.75 ul/min and the cannula remained in the brain for an additional 4 minutes to permit the drug to diffuse away from the injection site. Subjects were pretreated with desmethylimipramine (25 mg/kg, i.p.) 45 minutes before surgery to antagonize the cytotoxic effect of 5,7-dihydroxytryptamine on noradrenergic neurons (Breese and Cooper, 1975). In addition to these intracerebral injections 5,7—dihydroxytrypt- amine (100 ug in 20 pl) was injected into the lateral ventricle via a stereotaxically guided 31—gauge cannula. The injection site was 1.2 mm posterior to bregma, 3.0 mm lateral to midline and 3.0 mm below the dorsal surface of the brain. Control rats were treated in an identical manner as the 5,7- dihydroxytryptamine rats, except that only the vehicle was injected into either amygdala (3.5 ul, bilaterally) or lateral ventricle (20 ul). H 57 E. Biochemical Procedures 1. Dissections: After decapitation rat brains were rapidly collected and the hindbrain removed by a midcollicular cut. The remainder of the brain was placed base—down on the stage of an A0 microtome (model 880, A0 Instrument Co.) and frozen using a Histo—freeze (Scientific Products). Consecutive 500 pm thick frontal sections were sliced and transferred to glass slides on dry ice. With the aid of a stereomicroscope, using the decussation of the anterior commissure as a point of reference, the desired brain nuclei were bilaterally punched out from the unfixed brain slices (Palkovits, 1973) using stainless steel tubing (0.8 or 1.0 mm i.d.). The atlas of Konig and Klippel (1963) served as a general guide, although a different cutting angle from that of the atlas was used. Figure 2 represents a schematic drawing (modified from Konig and Klippel, 1963) of the amygdaloid nuclei as seen in frontal sections of the rat brain at different distances (in pm) from the anterior commissure. The circles represent the size and location of tissue pellets removed by punch. Palkovits §£_al. (1974) afforded additional aid in the dissection of extra—amygdaloid areas as well the amygdaloid nuclei. Tissue samples were homogenized in 0.5 ml glass homogenizers (Kontes Glass Co.) in 150 pl of 0.1 M HCl containing 50 uM ascorbate. The resulting homogenates were transferred to microcentrifuge tubes and centrifuged for 3 minutes in a Beckman Microfuge at 4°C. The clear supernatants were transferred to silanized glass vials while the homo— genate pellet was assayed for protein content according to the method of Lowry §£_§l. (1951) using bovine serum albumin as a standard. [ll 58 Figure 2. Schematic drawing of frontal sections of the rat brain 4620 and 3750 uM anterior to the frontal zero plane according to the stereotaxic atlas of Konig and Klippel (1963). The amygdaloid nuclei and hippocampus are designated by numbers as follows: 1 = central amygdaloid nucleus; 2 = lateral amygdaloid nucleus; 3 = medial amyg- daloid nucleus; 4 = basal amygdaloid nucleus; 5 = cortical amygdaloid nucleus; 6 = medial posterior amygdaloid nucleus; 7 = basal posterior amygdaloid nucleus; 8 = posterior amygdaloid nucleus; 9 = hippocampus Abbreviations: cp = caudate putamen, cl = claustrum, cai = capsula interna, mfb = medial forebrain bundle; vt — ventral thalamus; cae = capsula externa. 60 2. Methods of Quantitation: Dopamine Supernatants from tissue homogenates were analyzed for endo— genous dopamine content by Selected ion monitoring (SIM) (Koslow_§£ £13, 1972; K0 EE.§lra 1974; Kilts §£‘§1,, 1977) with computer control. Selected ion monitoring, or mass fragmentography, c0nsists of the alternate focusing of the mass spectrometer on identified fragments of the compound of interest and on the corresponding fragments of a stable isotopically labelled or structural analog as they are eluted from a gas chromatographic column. An internal reference standard of deuterated dopamine (75 pmole) was added to all samples to be assayed prior to homogenization. Trideutero—B,4—dihydroxyphenylethylamine (d3-dopamine) was prepared from dopamine by an acid catalyzed exchange reaction (Lindstrom §£_al., 1974), the aromatic hydrogen atoms at positions 2, 5 and 6 having been replaced by deuterium atoms. Mass spectral analysis of the deuterated dopamine demonstrated that contamination by the protium form was less than 1% with a typical deuterium isotope enrichment, ca1Cu1ated from the pentafluoropropionyl derivative of d —dopamine to be 75%) of the perfluoroacylated 5-hydroxyindole— acetic acid in the form of the partially reacted monOpentafluorOpro— Pionyl derivative (PFP -5—HIAA) which, upon mass spectral analysis, 2 revealed a parent ion at m/e 469. The fully derivatized product (PFPB— 5—HIAA) showed a parent ion at m/e 615. Study of the time course of the reaction between 5-hydroxyindoleacetic acid and pentafluoropro— Pionic anhydride—pentafluoro-N-propanol revealed that the formation of l the fully derivatized product had reached a steady state after approxi— mately 4 hours at 75°C (Figure 5). Thereafter, vials were sealed and heated overnight at 75°C. The samples were then cooled, dried and redissolved in 1% pentafluoroprOpionic anhydride in ethyl acetate (20 Ml) prior to injection into the gas chromatograph—mass spectrometer. 69 Figure 5. Time course of the formation of the products of 5-hydroxy— indoleacetic acid (5—HIAA) reacted with pentafluoropropionic anhydride (PFPA) and pentafluoropropanol (PFPOH). Forty ng of 5—HIAA was reacted with 40 p1 of PFPA and 10 pl of PFPOH at 75°C and the reaction was terminated at varying times by evaporation of the reagents under nitrogen. The parent ion of the pentafluoropropyl ester of 1,5- dipentafluoropropionyl-S-HIAA (PFP3—5-HIAA) at m/e 615 and that of the pentafluoropropyl ester of 5-monopentafluoropropionyl-S-HIAA (PFP2‘5’ HIAA) at m/e 469 were identified by selected ion monitoring. m/e 615: . 3 m/e 469, 0 Values represent means and vertical lines i l S.E.M. of 5 determinations. 7O .1. in. IO /1 \ \ o.\.. \ \/ 10\ lo I... ./. s x . / p1! F F P m B m u ..meczu 20. auhugwm uO 5.508?— A. Ilmo(hours) Figure 5 71 Quantitation of 5—hydroxyindoleacetic acid was as described for dopamine using solutions containing a fixed amount of d2—5—hydroxy— indoleacetic acid (52 pmole) and varying amounts of 5—hydroxyindole— acetic acid (0—104 pmole). For routine analysis the fragments m/e 438 and 615 and m/e 440 and 617 were monitored (Figure 6). The ion at m/e 438 (and 440) results from cleavage between the d—carbon and the esterified carboxyl group. The ion at m/e 615 (and 617) represents the parent or molecular ion (loss of an electron) of the pentafluoropro— pionyl derivative of S—hydroxyindoleacetic acid (and dZ-S—hydroxy— indoleacetic acid). Instrument conditions were as described for 5—hydroxytrypt— amine except that gas chromatrographic separation was achieved at an isothermal column temperature of 165°C with a carrier gas (helium) flow rate of 15 ml/min. F. Estimation of Dopamine and 5—Hydroxytryptamine Turnover Nonsteady state methods were used to estimate the relative rates of turnover of both dopamine and 5—hydroxytryptamine. While these techniques have proven to be of considerable empirical value, the number of substantiated as well as unsubstantiated assumptions in— trinsic to this methodology limits their utility to a purely compara— tive application (Weiner, 1974). The turnover rate of dopamine in the present study was estimated by following the rate of decline of the endogenous dopamine concentra— tion subsequent to synthesis inhibition with a—methyltyrosine (Brodie et al., 1966). Dopamine concentrations were determined immediately 72 Figure 6. Partial mass spectra of 5—hydroxyindoleacetic acid (S—HIAA) (upper) and d—dz—S—HIAA (lower) as pentafluoropropyl ester penta- fluoropropionyl derivatives with the proposed fragmentation patterns. Minor fragments (<5% relative abundance) have been omitted. Fragmen- tation of 500 ng of each derivative was accomplished by electron impact and spectra obtained at a scanning rate of 20 a.m.u.s.'l. The retention time for elution of PFP-S—HIAA and PFP—a—dz-S—HIAA was 1.9 min. See text for details of derivatization and instrument conditions. 73 ‘00 ’ 119 9 o at?" s F: 1' a o c 3 I : MW=615 25- 6 5 ll , ’160 260 ' 360 T «50 650 "V0 1001' 119 ‘5’. 9 15~ c' ‘l—jah “ ”2°25” 2 : f1:; '1‘ 2 WE 0 3 so ,, c215 0 MW'M? > 32' 0 25 _ 617 V 100 266 r 360 ' 460 ’ "V. Figure 6 74 before (zero time) or 45 and 90 minutes after the administration of d—methyltyrosine methyl ester HCl (300 mg/kg of free amino acid, i.p.). The calculated rate constants for the decline of dopamine concentra- tions were compared among groups to determine treatment differences. As an alternative index of dopamine turnover, the extent of depamine depletion was determined at one time point, 90 minutes, following a— methyltyrosine. Differences in turnover were determined by comparing the subsequently depleted dopamine concentrations in various treatment groups. S-Hydroxytryptamine turnover was estimated by observing the rate of decline of endogenous 5—hydroxyindoleacetic acid concentration following the inhibition of its synthesis with the monoamine oxidase inhibitor pargyline (Tozer e£_§l,, 1966; Morot—Gaudry 23.3139 1974). 5-Hydroxyindoleacetic acid concentrations were determined immediately before (zero time) or 20 and 40 minutes after the administration of pargyline HCl (75 mg/kg of the free base, i.p.). Treatment differences were determined by comparing the calculated rate constants for the decline of 5-hydroxyindoleacetic acid concentrations. Alternatively, the extent of 5—hydroxyindoleacetic acid depletion at one time point, 40 minutes, after pargyline was used as an index of 5-hydroxytryptamine turnover. Treatment differences were determined from a comparison of the percent depletion at this time. G. Drugs The following drugs were administered as a suspension in 0.5% methylcellulose: diazepam (Ro 5—2807), flunitrazepam (Ro 5-4200) and 75 desmethyldiazepam (R0 5-2180) were obtained through the courtesy of Dr. W.E. Scott, Hoffmann—LaRoche Inc., Nutley, N.J.; oxazepam, obtained from Wyeth Laboratories, Philadelphia, Pa.; D,L-5—hydroxytryptophan, purchased from Regis Chemical Co., Morton Grove, 111.; B-(3,4-dihy— droxyphenyl)-d-hydrazine—d-methyl—DL-propionic acid (MK—486), obtained from.Merck, Sharp and Dohme Research Laboratories, West Point, Pa.; methaqualone, obtained from W.H. Rorer, Inc., Fort Washington, Pa. The following drugs were dissolved in saline: methysergide maleate, obtained through the courtesy of Sandoz Pharmaceuticals, E. Hanover, N.J.; cinanserin HCl, obtained through the courtesy of E.R. Squibb and Sons, Inc., Princeton, N.J.; pfchlorophenylalanine methyl- ester HCl and pargyline HCl purchased from Regis Chemical Co., Morton Grove, 111.; D,L—a-methyltyrosine methylester HCl, purchased from Aldrich Chemical Co., Milwaukee, Wis.; chlorpromazine HCl, obtained from Smith Kline and French Laboratories, Philadelphia, Pa.; caffeine citrate, obtained from K and K Laboratories, Plainview, N.Y.; desi- pramine HCl, obtained from Merrell National Laboratories, Cincinnati, Ohio; amitriptyline HCl, obtained from Merck, Sharp and Dohme; sodium barbital, purchased from Sigma Chemical Co., St. Louis, Mo.; sodium amobarbital and sodium pentobarbital purchased from Ganes Chemical Co. Picrotoxin (purchased from Sigma Chemical Co., St. Louis, Mo.) was dissolved in a hot saline solution. Cyproheptadine HCl (obtained from Merck, Sharp and Dohme Research, West Point, Pa.) and dflysergic acid diethylamide bitartrate (d—LSD; obtained through the courtesy of Sandoz Pharmaceuticals, E. Hanover, N.J.) were dissolved in distilled water. 76 Drug solutions of varying concentrations were prepared to allow the administration of a constant volume (1 mllkg). All drugs were administered by the intraperitoneal route. 5,7-Dihydroxytryptamine creatinine sulfate (purchased from Regis Chemical Co., Morton Grove, 111.) was dissolved in 0.9% saline with 0.2 mg/ml ascorbic acid immediately prior to intracerebral or intraventri— cular injection. H. Statistics Statistical analysis of the behavioral effects of each given treatment was performed using a two-way analysis of variance (ANOVA; Sokal and Rohlf, 1969) with subjects as rows and days as columns. Data from 4 days were used for each ANOVA: the 3 control days before a treatment plus the treatment day. The lowest tested dose of a given compound that produced a significant treatment F ratio was designated the minimum effective dose (MED). The highest tested dose which produced a significant antipunishment effect without significantly altering unpunished responding was termed the highest effective dose (HED). The significance of the differences in the behavioral effects of diazepam alone and in combination with various agents was tested by a Mann—Whitney U—test. Student's t-test was used to examine the be— havioral effects of chronic pfchlorophenylalanine administration and of 5,7—dihydroxytryptamine injections. The effects of diazepam on body temperature were also tested by Student's t—test. 77 The significance of the differences in the fractional rate con— stants, obtained by a least—squares regression analysis (Sokal and Rohlf, 1969), was examined by Student's t—test. The effects of diaze— pam on the pargyline—induced depletion of 5—hydroxyindoleacetic acid and the d—methyltyrosine—induced reduction of dopamine in selected brain regions were analyzed using Student's t—test (dopamine) or a Mann—Whitney U—test (5—hydroxyindoleacetic acid). RESULTS A. Effects of Benzodiazepines on Suppressed Responding The effects of benzodiazepines on behavior are dominated by their striking ability to release conditioned and unconditioned behaviors previously suppressed by punishment. of the three behavioral proce— dures employed, the use of an uncenditioned, consummatory response (tube—licking) to establish the behavioral baseline offered several noticeable advantages (see Discussion). Obvious advantages such as the relatively short time necessary to attain stable baseline performance (14 days) and the substantial number of animals that can be tested on a daily basis led to the almost exclusive use of this procedure for the evaluation of anticonflict (antipunishment) activity. Figure 7 illustrates the dose—effect profile of diazepam on punished responding and background water consumption. Treatment effects are expressed as a percentage of the mean control baseline. The actual levels of response in control trained subjects ranged from 15—25 shocks taken per session and 13—17 m1 of water consumed per session. Diazepam produced graded, dose—related increases in punished responding with the effect becoming mixed and diminished with higher doseS. This "inverted U—shaped" dose—effect curve is characteristic of behaviorally active benzodiazepines and closely resembles that obtained with methods using more complex, operantly conditioned responses (Cook and Sepinwall, 1975). The minimum effective dose and the highest 78 79 .Houuooo Eowm Amo.ovmv haunmoflwfiowflm HoMMfiU SUHLB mooam> omonu moumoflvofl « .Honekm ago mo moflcmo osu ammo mmoa mH moam> msu woumoflwofl mH .E.m.m on open: .mdoaumofla luoumc mumummom HH on m mom .Z.m.m H H wooHH HMUHuHo> pom momma onu unommpmmp maonfihm .ANQOH mm woumowflmowv unoaumouu Sumo wdflcoomum mkmc Houunoo m ms“ mom mDHm> some emu mo mwmunooumm .mofiumou ou posse mononflfi ma omOHoHHoo m an wommopmxo mH mzmp unoaummuu do poowwm noumoflwuo pommumfloflavm mums Emmmsmflw mo mmmom .ham>flu IHkSuma Nm.o GH deflmnmmmom m mm kHHmmoouHHommHuoH Iommmou .posowooo “mums mo oESHo> osu pom soxMu mxoonm mo Henson one mo manou ow deflunaomooo no ammonmflp mom o>noo noowmolmwom .m ouomflm HmumB A v wmnmwnomoo pom A V woemficoa 80 OpfiofioaOfiomdmd pd m ouswflm A: niac— 4 d1 J. H J _ 1 o1\a11o1laltoVo 42. .OOm 10l1NOD :IO 1N3333d IIIIIIIIIIIIIIIIIIIIIIII-lllll---———______l 81 effective dose of diazepam that increased punished responding were 0.3 and 18.0 mg/kg, respectively. The highest effective dose refers to the largest tested dose which significantly increased the number of shocks accepted without altering the volume of water consumed (unpunished responding) relative to control values. Unpunished responding (water consumption) was significantly decreased (23% of control) at a dose of 30 mg/kg. Such a decrease in unpunished responding is generally interpreted as an index of the dose at which nonspecific sedative activity occurs (Geller and Seifter, 1960). The relatively short duration (10 min) of each experimental session of this conflict procedure permits an assessment of the time course of drug effects and, more importantly, the time of peak anti— punishment effect. Evaluation of the time course of the antipunishment effect of a submaximal dose (1.8 mg/kg) of diazepam devoid of appre— ciable effects on unpunished responding as indicated in Figure 7 shows a peak antipunishment effect (275% of control) 15 minutes following its administration (Figure 8). The rate of punished responding had waned to control values 60 minutes following diazepam injection. The conditioned suppression of drinking paradigm, as with more elaborate conflict techniques, appears able to detect qualitatively different profiles within the class of benzodiazepines. Flunitrazepam, the 7—nitro, 5-phenyl ortho fluorine congener of diazepam, exhibits a greater anticonflict potency than diazepam over a narrower range of doses (Table 2). The minimum and maximum doses of flunitrazepam that increased punished responding were 0.1 mg/kg and 1.0 mg/kg, respec— tively. An obvious criterion to be fulfilled by animal tests to have predictive value as models for a particular human condition is that 82 Figure 8. Time course of the effect of diazepam on punished ( . ) and sumption. Diazepam was administered suspension at varying times prior to s of a submaximal dose (1.8 mg/kg) unpunished ( C) ) water con- intraperitoneally as a methylcellrllose the initiation of behavioral testlngo of 3 to 11 separate Same as in Figure 7. Abscissa: Interval rug administration and time of testing. * indicates h differ significantly (p<0.05) from control. Each time point represents the mean i l S.E.M. determinations. Ordinate: between time of d those values whic 83 300» l g) .1/ \, 200- \‘l , \1 45......¢—-~_5___a\l 1¢)(: £3 ‘__. “" "" :=:::4i1.... llulose sting. PERCENI OF CONTROL L [L 1 ll '71 :- § 15 30 45 so 120 minutes cflor dloxopom Figure 8 84 .O.H .wx\wfi mm commouaxo mmmoox .omop Hod Ooumou mHmaHdm mo gonads 1 z .cdeEmMm momop mo omens ofiu Ho>o wsHOoommoH HouHm hHuomoHMHome ou OoHHmm Omumou UnsomSoo onu AIIIIV OmumoHOaH mH oon> on onmfiz .onOooamoH OoSmHosmo: deHouHm kHamHomHOem usoguHB wcHanmmou OosmHosa powonUGH hHuomoHMHame umau Ossomaoo omwa m mo omop Owumou umoman ozu mmumoapaa omop o>auoommo umoans Ho 9mm .oUGMHHm> mo mHmkaom ha Oooaauouov mm Awsapaommou Oonwaoomoov Ooaomsoo woume mo 0.000H 1111 1111 OH Ale OOONIO.Om Hoomsuo O.wH 1111 1111 OH q O.Om1O.m oaHHmomoe 0.0m 1111 1111 om q1m o.om1o.m mnammmmo O.m 1111 1111 Om 01m O.m1H.O ooHNMSOHOMOHso e.m 1111 1111 om s1m o.oauo.a mafiaauaopuaaw 0.0H 1111 1111 om m1m o.oH1o.H meaawpaaaa mpoaoaaoo o>Huode m.qH O.OH O.O OH wlm m.wH1q.q oGOHmovauoE OOH m.mO 0.0¢ Om m OOHImN HmuHonmn om o.mm m.NH OH R omnmm.e Hmuaepmeoamea OH 0.0H O.m OH R O.Om1O.H HmanumnoEo OH 0.0H O.H OH Ale O.wH1m.O Hmannmnouood em 0.0m e.m ma a1m o.emno.m amamnmxo 1111 o.wa O.H ma s1m o.om1m.o amaowmaeasepmamme1z O.H O.H H.O mH mlm O.m1mO.O smamumuuHoon o.om o.wa m.o ma calm o.om1a.o amamnaae mwosomaoo o>Huo¢ am: am: OMS AGHEV mEHu smmmow mo name mos «wdHOooammm OoSdeomaD «onwoommom OmsmHnsm unmaummpuoum z ownmm u H :H mmmouoom oH ommouunH GOHumasmnoo nouns mo GOHmmoHOmsm Omanoom ozu wdH>Ho>nH GOHumouHm uoHHmooo OooovnwthHmuooaHummxm am no mwdup mDOHMm> mo muoowmm N mafia. 85 they be able to distinguish therapeutically effective agents from other classes of compounds. In agreement with the findings of Schallek e£_§l, (1972), the phenothiazine-type neuroleptic, chlorpromazine, was found to be devoid of anticonflict activity (Figure 9). In fact, chlorpro- mazine significantly enhanced the behavioral decrement produced by response—contingent shock. Table 2 summarizes the effects of various agents on a conflict situation involving the punishment of a consummatory response (tube— licking). This procedure, conditioned suppression of drinking, appears sensitive to differential drug actions and selective for clinically- defined antianxiety agents. All of the benzodiazepine derivatives examined increased the rate of punished responding at doses lacking any measurable nonspecific sedative actions. Barbiturates, long the drug of choice for the treatment of distressing anxiety symptoms (Berger, 1963) also exhibit significant antipunishment activity (Table 2). However, the difference between the minimum effective dose which in— creases punished responding and that which decreases unpunished re— Sponding is considerably less than that seen with the benzodiazepines, an observation that agrees quite well with the findings of Blum (1972). The anticonflict activity of the oxoquinazoline derivative, methaqua— lone, in this conflict procedure may be indicative of an anxiolytic component of activity which has been reported in clinical studies (Cromwell, 1968; Duchastel, 1962). The dose-response data summarized in Table 2 were obtained after the rats had been exposed to various benzodiazepines on several occa— sions i e , after they were "drug-experienced". An additional , _.—. 86 Figure 9. Evaluation of chlorpromazine for anticonflict activity in conditioned suppression of drinking. Punished responding ( . ); unpunished responding ( O ). Doses of chlorpromazine HCl (expressed as the salt) were administered intraperitoneally in saline 30 min prior to testing. Symbols represent the mean i l S.E.M. of 3—6 separate determinations. Ordinate: Same as in Figure 7. 7" indicates a significant (p<0.05) decrease in responding compared to control values- 87 200 —l 2 . p. z 0 U 8 100 __ 1‘3 T \ ” \ E \\\\\\\\\\\ l ‘.\l_§‘l > _ I \6- 0.1 Q3 Eli/kg i.p. Figure 9 88 characteristic of the interaction between benzodiazepines and conflict behavior, termed the "initial treatment phenomenon", is seen when these drugs are administered for the first time to conflict-trained, drug- naive animals. Initial treatment with benzodiazepines produces a qualitatively different effect, compared to that seen in drug—experi— enced subjects, which changes over the course of the first several drug administrations (Margules and Stein, 1968; Cook and Sepinwall, 1975). Characteristically, upon first exposure to 5.6 mg/kg of diazepam, rats trained to a stable baseline of response suppression in the condi— tioned suppression of drinking conflict paradigm exhibited a signifi— cant decrease in unpunished water consumption while punished responding was submaximally increased (Figure 10). During two additional treat— ments the anticonflict effect further increased and reached asymptote while background water consumption returned to predrug values. Appa— rently the general depressant or nonspecific sedative effects, repre— sented by the initial decrease in unpunished re5ponding, rapidly undergoes tolerance while the anticonflict activity fails to show tolerance. The fact that a similar pattern was seen when the same dose of diazepam was administered at 5-day intervals (Figure 10, right-hand panel) rather than daily (Figure 10, left—hand panel) argues against an explanation of this phenomenon involving an accumulation of drug and/or active metabolites. Figure 11 depicts the same patterns when testing a lower dose (1.8 mg/kg) of diazepam. However, unpunished water consumption was unaltered from pre—drug control values at any time after treatment with the 1.8 mg/kg dose. Interanimal differences in drug sensitivity are reflected in the variability of the responses to '11 89 . WHOHUGOU wouoomnH mHoH£m> Eocfl Am0.0vmv kHuomUHmHome nmmmHO SUHOB moonS mmocu mmumoHHEH a .Hm>uounH monoHc 30mm on woumou who: mums q mo mmoouo .AmM\HE Hv muooeumouu Houuoou oHoHcmS muoomonaos O "ommHong .m osomHm CH mm wHHmumm noumsflwuo .AummHuv mHmSHouoH amp m on no Auonv .AHHmfi HmfiuHm dounmommmom omOHoHHmoHrEuoE m an kHHmooounonmnuoH woumumHoHEwm mos EmaonHQ .AononMp mo GOHmmosmmom OmoOHunooov mums mammalwshfi .OwchHuluoHHmnou dH mcHOoommoH A O V OmemHosa L5 pom A . v Omcdeon no va mucmmwumon O "mmmHomn< .m ouome EH mm mHHmqu "oumoHOHo .mGOHumoHapouoO mumhmmom AmHm>HouoH how m on usoEumopoV q Ho Awudofiumonu AHHmOV .E.m.m H H once one munomoumow oon> nomm .AummHuv me>HouoH mow m up so AumoHv kHHmO HuoH OonumHnHEOm mp3 EmaonHm .Aononnp HwauHo :0Hmnmmmom mmOHoHHmonsuoE m mm zHHmmoouHHomm Ho GOHmmmuamom OoGOHuHOnoov mums o>HmnlmoHO .vmonHuluoHHmnoo CH moHpoommmH A nu v Omcdeoa 1a: mom A . v Omanoom no va ou pmwwano compo mxootm HH ouanm 305.80% abutmfobo i r U .1111. N. H p .1 H 1H 1 11 )\\w// pfi 1)) 1111.11 41114114 1 1* H 11 as 92 -—o l I 1 . "—0 \ 101111403 10 1113311311 l‘ —" 93 both single and repeated diazepam administration (Figures 10 and 11). Thus, in agreement with the findings of Sepinwall 35 pl. (1978) in squirrel monkeys, it is possible in drug—naive rats, at certain doses, to obtain a pattern of anticonflict activity similar to that seen in rats with a prior history of benzodiazepine eXposure. The fact that the antipunishment (anticonflict) and general de— pressant actions of the benzodiazepines follow different courses during their repeated administration, with the anticonflict effects failing to show tolerance and even increasing with repeated doses, may be of value in evaluating possible synaptic mechanisms involved in the anxiety- reducing action of benzodiazepines. Thus, one may assume that a change in neurotransmitter dynamics that persists following the prolonged administration of diazepam may reflect synaptic mechanisms related to its antianxiety action. Similarly, those changes elicited by diazepam which show tolerance may be unrelated to its antianxiety actions or perhaps related to the general depressant effects of the drug. The validity of this assumption obviously rests on the power of this conflict procedure for predicting the therapeutically desirable pro— perties of benzodiazepines in the treatment of human anxiety. While response—contingent shock has been extensively used to generate low rates of suppressed responding, the relative intensity of the suppression and, by inference, its subsequent attenuation by benzodiazepines, may be dependent on the nature of the response, 133:, unconditioned consummatory responding pp, Operantly conditioned, instrumental responding (Bertsch, 1976). Table 3 summarizes the effects of various agents on a conflict procedure utilizing an Operant l1 1111‘ 1 . we .Q.H .wx\wa mm Ommmonmxo momomx .omop pom woumou mHmaHam mo Hommon mumou WMHMMMOU omoc eon um momdoomou OmzmHssa mo amass: Maw omnowomwamqumewwme”smwumMHwoWOmoemou woJmHoom o 111 oumoH s mH on m> on whom: . onno was won H . . . wwmmwuoWHnaHuGNWHWHome menu ensconco oo>Hw a mo omop Omummu umowan can mounMMMMwHomMMmeMMMMHWM ummann so Qmm .Aoalmmv unmamouomnHoH Mo mHDOofium ouMHOoEumudH mfiu do mom mmmhuqm unmoamadwam unease o:u oH «monsoon o no Amommoum “any noncommou posdeom Ho nomads can oH mm o emaassa.wm an: m woodwoua unto canoaaoo do>Hm m mo omop woumou umoBOH ofiu moumoHvaH owow 0>Hu mm . . . 1 . wuomznume . 1111 111 on mlm O OH O H opH W.MM 1111 111 OO m1q O.Om1O.OH sHuomoooHo e . m 1111 111 8 sum 0 .310 A mfiafififi e . m 1111 111 8. 01m 0 . Sue . H 1.513235 mundomfioo o>HuomoH OO.wH 0.0H O.m OH wlm O.OH1O.H Hmanumnouomm 1.... 00.0m 0.0m ©.m OH wlm 0.0mlo.m EQONNNO 9 o . 3 o . S o . a S are $10.0 58.2.3 mpooo Eoo o>Huo< am: can OMS ASHEV mEHu «momop wo sonvdoammm vosdedmsD «wcdeommmm OmflmHodm unmeummuuoum z wwnmm unmaumeH oH ommwuoon aH owmouucH AHouwHomluoHHou OonHUoEv amoommou OoGOHuHOooo hHuanmao no mo GOmeonaom pothdsm opp on>Ho>aH EOHumsuHm uoHHmooo woonOoHlkHHmunoEHuoaxo do no mmsuw mSOHum> mo wuoommm m mHawH 95 response (bar pressing) for food maintained on either a fixed ratio 40 (unpunished component) or continuous reinforcement (punished component) schedule. This is the modified Geller—Seifter procedure (1960). While subtle differences exist in terms of the magnitude and breadth of the anticonflict effect, the dose—effect profiles are qualitatively similar to those obtained with the conditioned suppression of drinking method despite obvious differences in the nature of the response and type of reinforcer. Although benzodiazepines and barbiturates appear to effec— tively attenuate the suppressive effects of punishment on consummatory responses as well as complex operant responding, the formidable metho— dological problems encountered in comparing consummatory and instru— mental behavior (Bertsch, 1976) limits comparisons of relative drug effects on these behaviors. Another factor influencing the intensity of the response suppression engendered by electric shock is the relationship between the shock and the animal's behavior, 132:: response—contingent pg. noncontingent shock (Huppert and Iversen, 1975). The decrease in operant response rate, referred to as the conditioned emotional response, obtained by the repeated pairing of a signal with unavoidable (noncontingent) shock is considered to represent a behavioral analogue of anxiety (Millenson and Leslie, 1974). The effects of various agents on responding in this behavioral paradigm are summarized in Table 4. The disinhibitory effect of benzodiazepines on suppressed responding appears to general— ize to experimental designs in which the subject is shocked irrespec- tive of response it makes. The use of the conditioned emotional response as a behavioral baseline susceptible to alteration by l1 96 .Am..m nmx\wa mw mmmmua . p ..w.wxm mmmooe once one pounce meaHem Ho “when: u z .emummu mmmoe mo omens one pm>o AoeHmpommnume V wUOHumo vwxoocmno: one oH noncommon mo nomads can announce on no onpooamon pommouaaom mo comm .mHonuco o 0 own a oncoH aHucmonHome on poHHmm woumou pa:omfioo one A1111v woumoHpoH mH mon> on manna nawmwwwmwomm muoomnom onmo woomHum> mo mHmmHmom do %n posHEMQuoO mums Qmm pom ems can mom momwww Oommmumoa huowHHm or“ CH wdHOnoomou wSHuouHm ozonuHB xoosm usomcHuoooson mp Oommoummam wnHOnmo o panama; no. Huoonmmome umnu wouoafioo om>Hm m Ho once Omommu umowumH can moumoHOoH moon MbHMmnon can mo oOHuuo wowuonw unwoamaowam m m Hones: wAu dH ommouoop n so acorn unomnHunoodoo an Oommoumaom onOncmmms GH 0 . . . . mongoose noon poooaaoo oo>Hm a mo omen ummBOH wen mnemoHan omow o>HuommHo esaHnHa no om: O.Om 11111 1111 mH mlm O.Om1O.m oHoucmwhaHhaoamHO llll lllll Illl CM 0.1..» O .OHIO .H mflflwhmwhfluma w .H 11111 1111 3 m4 o.m1m.o «damages O.ON 11111 1111 mH mlm O.O¢1m.u HmuHonmoooona mwnoo Boo m>HuoocH wuww O.OH O.m mH 01m O.om1O.H Hmanumooocom o .H 0.2 o . ... fl Tm o .310 . N 58388 9m 35 3.0 2 arm ono .o databases O.m om.o mH wlo 0.0H1wH.o amnHmNmHHc 11111111111111111 mpdnomfioo o>Huo< 11111111111111mmwM1111111111. omm om: «waspsoawwm Aoaav oEau «momow mo . oqlmm moHHommm we a . . unoaummHH oH mmmoHqu « Hwomamwmmwwwmwummom unmaumowuoum z omomm Ammsommmu HmGOHuoEm pooOHunooov Xuo m one a wnHuoooooo Eouw moHuHowmu GOmemHmmom uncommou so muoowm mSOHnm> mo muommmm q MHMEH IIIIIIIIIIIIIIIIIIIIII-llll---——_______l 97 benzodiazepines proved to be a fragile design dependent upon rigorous control of the temporal and incremental aspects of the variables (i.e,, conditioned stimulus duration, shock intensity, degree of food deprivation, schedule of reinforcement) which interact to produce the conditioned suppression. The effects of benzodiazepines on responding in the conditioned emotional response procedure were considerably more variable than that in the conflict tests involving response—contingent punishment. B. Effects of Drpgfinduced Alterations in Brain S—Hydroxytryptamine Activity on Conflict Behavior A seemingly convincing body of evidence obtained from animal studies implicates 5—hydroxytryptamine systems in the anxiety—reducing actions of benzodiazepines (see Introduction and Stein pp p1,, 1977). The peripheral S—hydroxytryptamine antagonists methysergide (Gyermek, 1961), cinanserin (Rubin ep_pl,, 1964) and cyproheptadine (Stone pp 31:, 1961; van Riezen, 1972) were evaluated for anticonflict activity using the conditioned suppression of drinking conflict procedure. Methysergide had no significant effect on punished responding whether administered one or 30 minutes prior to behavioral testing (Figure 12). The combination of a submaximal anticonflict dose of diazepam (1.8 mg/kg) with a dose of methysergide (3.0 mg/kg, one minute pretreatment, which alone had no significant anticonflict effect) produced a greater anticonflict effect than that of diazepam alone (Figure 13). The injection of various doses of cyproheptadine, 30 minutes prior to testing, also failed to significantly alter the rate of punished responding (Figure 14). Pretreatment with Cyproheptadine (l and 3 mg/kg) did not significantly alter the anticonflict effect of diazepam 98 .opHmuomhsumE mo .wx\wa wH .mmov ummwumH one Houmm oHS H on hHso wommmhomp hHunmoHMHdem me onpoommoH OmstdomoD .HOHudoo Eoum AmO.OvmV kHuomonHome hmHMHw noHss moSHm> omonu moumoHpoH a .A onome nH mm mHHmUoQ "mumoHOnO .mEOHumoHEHouov ouwummom w1m Ho .E.m.m H H some map unomohmou mHonakm .onummu ou “OHHO AustuV cHE om Ho AummHV ADHmm msu mm commouame mumonE owHw :HE H uofluHo ooHHMm nH mHHmooouHHoamuucH wououMHoHEOm mums Ipomzzumfi mo momom .A mu V wnHOnommOH OmamHnomoo mA Av V wandommmn OmsmHoom .AononHO mo EOHmmmHmmom ONGOHuHOnooV huH>Huom uUHHmoooHuom How oOHmemkaumE mo oOHumon>m .NH whome 99 3.08.0320... 0...an on NH ouome 9:9: owl 9 o...” o... a // L a 17 71....\\\ /fi11111q\ 2.05.0030...“ 0.355 w ._oon 108 INOD $0 1N33‘3d Illl 100 .mCOHm Emmmmme Cu pmummfioo COHuQECmCoo Houme memHCoaCo CH mommmuoop Am0.0vaV quonHCme wouMUHUCH « .m oHCme CH mm wHHmuoO "oumCHOHO .wCOHuMCHEHoumO oumnmaom Ole Ho .E.m.m H H moCHH Ho0HuHm> pCm Cmoa onu muComemoH oCHm> Comm .wCHumou on HOHHC CHE mH common IHCHBOm mp3 EmmonHm .ouCOH HmoCouHHmmmHuCH man he OonoumHCHawm mums OCm uHmm osu mm pommoumxo mum momoo .wCHummu ou HOHHQ CHE OO Hum CHHomCmCHu OCm CHE om Hum mCHOmuamJOHm%0 .CHE H Omnwum IHCHEvm mp3 mummHmE oOHwHomA£umE .mumn OomHHum .wCHOCommmH wormHCCmCC ”mama Como .wCHOCommoH OmamHCCm .wa\wE w.HV ammoanO mo omOO HmEmeanCm m Ho AwCHxCHHO Ho COHmmonmom meOHuHOCooV kuH>Huom uoHHwCooHqu mam Co mumHCowmqu oCHEmua%Hu%NOHO%:1m HmHoanHHom Ho muoowwm .MH mHCmHm .2 ma mpsmea 0m OM mw 1 a 1 1 stamens—U .. 1 1 O.H O.—. .. 1 oc_$o.no.._osa>u 1 1 1 1 .. 1 o .90». Eon—onua 101 1081NOD :IO "133834 100 .oCOHm SmamNmHO ou posmmfioo COHuOECmCoo HmumB OmCmHCCaCC CH mommwuoop AmO.OvmV quonHCme moumoHOCH « .m onstm CH mm mHHmqu ”oumCHOHO .mCOHumCHEHouoO oumnmmom Ole Ho .E.m.m H H moCHH HmoHuHo> pom Coma can muComonoH oCHm> zoom .wCHumou on HOHHa CHE mH ponoum IHCHEOm one emmwNmHm .ouoou HooCouHHommHuCH osu kn pououmHCHEOm mums OCm uHmm onu mm Oommoumxo mum momom .wCHummu ou HOHHO CHE OO Hum CHHmmCMCHo OCm CHE om Hum oCHOmumononmmo .CHE H ponmum IHCHSOm mm? oumonE oOHwHomksumz .men ponHHum .MCHOCoamoH OoSmHCCCCC momma Como .wCHwCoammH OofimHCom .Amx\ma O.HV EmmonHO mo mmow HmemeEnCm m Ho AwCHxCHHO Ho COHmmmHmmCm OmCOHuHOCooV mquHuom uoHHmCooHqu on“ Co mumHCowmqu oCHEmuthuhxouwzfilm HmsoamHHmn Ho muoommm .MH oHCme 101 ma opsmHm 2.. on 2 - - - - 53.2.50 1 1 1 QM Oi 1 1 o:.vu.no;9&>u 01.9.0150: Eamonua 'IOULNOD :IO 1N33834 102 Figure 14. Evaluation of cyproheptadine for anticonflict activity. Punished responding( . ); unpunished responding ( O ). Doses of cyproheptadine HCl (expressed as the salt) were administered intraperi- toneally in distilled water 30 min prior to testing. Each value repre— sents the mean i l S.E.M. of 3—7 separate determinations. Ordinate: Details as in Figure 7. * indicates a significant (p<0.05) decrease in unpunished water consumption. 103 \ ‘\ U\\?l k. 10 1.0 10 mg/kg 1% 1e" HOF—ZOU no .—ZuU¢ mm Figure 14 1111.1 :46 rt. 11 104 (1.8 mg/kg; Figure 13). In agreement with Geller §£.§l: (1974), cinanserin increased punished responding, but only at a single dose (56 mg/kg) which was associated with a significantly decreased water con— sumption (Figure 15). However, in contrast to the findings of Cook and Sepinwall (1975), no anticonflict activity was observed with lower doses of cinanserin. The effect of various doses of cinanserin on the anticonflict activity of diazepam was examined (Figure 13). Pretreat- ment with either 18 or 30 mg/kg cinanserin did not appear to change the response to diazepam (1.8 mg/kg). Combination of the active dose of cinanserin (56 mg/kg) with diazepam (1.8 mg/kg) yielded an anticonflict effect, the magnitude of which suggested the interaction to be of an additive nature. However, the response to this combination was mixed and was accompanied by a large decrease in unpunished water consump— tion. The following study was undertaken to investigate the effects of an enhanced 5-hydroxytryptamine—mediated activity on the anticonflict effects of diazepam in the conditioned suppreSSion of drinking. The brain concentration of 5—hydroxytryptamine was elevated by the peri— pheral administration of the immediate precursor of 5—hydroxytryptamine, 5—hydroxytryptophan (Moir and Eccleston, 1968). In order to minimize any peripheral side effects of the S—hydroxytryptamine formed from the exogenously administered 5-hydroxytryptophan, a peripherally acting L— aromatic amino acid decarboxylase inhibitor (MKr486, 60 mg/kg, i.p.) was administered 30 minutes prior to 5-hydroxytrypt0phan. The 5— hydroxytryptophan, given in conjunction with the decarboxylase inhibi— tor, produced a dose-related increase in whole brain 5-hydroxytrypta— mine (Table 5). When administered without prior MK-486 treatment, 105 Figure 15. Evaluation of cinanserin for anticonflict activity (conditioned suppression of drinking). Punished responding ( . ); unpunished responding ( C) ). Doses of cinanserin HCl (expressed as the salt) were administered intraperitoneally in saline 60 min prior to testing. Symbols represent the means and vertical lines indicate i l S.E.M. as determined from 3-6 separate determinations. Ordinate: Details as in Figure 7. *Significantly different (p<0.05) from control. PERCENT OF CONIROL 106 \\ 6\ o \ 1e 30 so 160 Inc/ks Figure 15 107 TABLE 5 Brain 5-Hydroxytryptamine Concentrations Following 5-Hydroxytryptophan, 5,7-Dihydroxytryptamine or p7Chlorophenylalanine Administration Treatment pg/g5;H§.E.M. % of Control Control 3 0.53i0.02 100 5—HTP (18) + MKr486 (60) 3 2.13i0.03 402.7 5-HTP (30) + MK-486 (60) 3 2.77i0.10 524.0 intravent. 5,7-DHT 3 0.39i0.02 73.8 acute p—CPA 3 0.30i0.03 57.5 chronic p—CPA 3 O.21i0.01 40.2 Telencephalic 5-hydroxytryptamine (5-HT) concentrations were determined by fluorometric detection after pfphthalaldehyde reaction (Curzon and Green, 1970). Excitation and emission wave-lengths were 360 and 470 nm respectively. Standard solutions containing 25—1250 ng 5—HT and tissue samples were assayed in duplicate. Blank solu- tions were run in triplicates. Animals receiving 5-hydroxytryptophan (5-HTP) were pretreated with the peripheral aromatic amino acid decarboxylase inhibitor MK—486 (60 mg/kg, i.p.) 30 minutes prior to 5-HTP treatment and were sacrificed 60 minutes later. prhloro- phenylalanine (p-CPA) treated animals were sacrificed either 5 days following a single injection of 400 mg/kg of p-CPA ("acute p—CPA") or 14 days from the initiation of a regimen of 200 or 300 mg/kg p-CPA every 2nd or 3rd day respectively (”chronic p-CPA"). The assay of brain 5—HT content of rats receiving 5,7-dihydroxytryptamine (5,7—DHT) (100 pg) intracerebroventricularly was performed 10 days after the injections. Numbers in parentheses designate doses eXpressed as mg/kg, i.p. 108 5—hydroxytryptophan (18 mg/kg) produced a significant decrease in punished responding and a trend to decrease, although not signifi— cant, the anticonflict effect of a single dose of diazepam (1.8 mg/kg, Figure 16). This enhancement of the behaviorally suppressant effects of response—contingent shock by 5-hydroxytryptophan administration was no longer observed when the animals were pretreated with the peripheral decarboxylase inhibitor. Surprisingly, the anticonflict effect of diazepam (1.8 mg/kg) was significantly increased in animals pretreated with 5—hydroxytryptophan and MK—486 (Figure 16). Inhibition of the major mechanism involved in the termination of the synaptic actions of 5-hydroxytryptamine, neuronal reuptake from the synaptic cleft, enhances 5—hydroxytryptamine-mediated synaptic trans— mission by prolonging the effect of 5-hydroxytryptamine at the synapse. The clinically useful tertiary amine tricyclic antidepressants have been reported to inhibit the reuptake of 5-hydroxytryptamine at nerve endings ip_yiyp_(Carlsson ep_§l,, 1969; Von Voigtlander and Losey, 1976). As already mentioned, the tertiary amines imipramine and ami— triptyline exhibited no measureable anticonflict activity (Table 2). Additionally, a reportedly specific blocker of neuronal 5-hydroxytrypt— amine uptake mechanisms, fluoxetine (3 mg/kg, 120 minute pretreatment; Fuller e£_§l,, 1975), was found to be similarly ineffective in in- creasing the rate of punished responding. However, amitriptyline (5.6 mg/kg) enhanced the anticonflict effect of diazepam (1.8 mg/kg) in the drinking procedure; doses of 3 and 10 mg/kg amitriptyline Were ineffective in this regard (Figure 17) . While this observation is in general agreement with the findings of Babbini ep el. (1976), the 109 Figure 16. Effects of 5—hydroxytryptophan, alone or in combination with a peripheral decarboxylase inhibitor (MK—486), on the anticon- flict activity (conditioned suppression of drinking) of a submaximal dose of diazepam (Diaz.; 1.8 mg/kg). Punished responding, open bars; unpunished responding, striped bars. 5—Hydroxytryptophan was admini— stered 60 min, MK—486 90 min and diazepam 15 min prior to testing. Where no treatment dose is indicated the subjects received injections (1 ml/kg) of the vehicle, 0.5% methylcellulose. Each value represents the mean i S.E.M. of 3—10 determinations. Ordinate: Details as in Figure 7. *Significant (P<0.05) decrease in punished responding compared to methylcellulose injected controls. °Significant (P<0.05) increase in punished responding compared to subjects receiving diazepam alone. PERCENT OF CONTROL 500 400 300» 200- 100~ 110 \\\\J~ chx. 5 "HTP MK" 486 18 Figure 16 1.8 18 3" \\\\1— 1.8 18 60 111 Figure 17. Effects of amitriptyline on the anticonflict activity (conditioned suppression of drinking) of a submaximal dose of diazepam (1.8 mg/kg). Punished responding, open bars; unpunished responding, striped bars. Doses of amitriptyline HCl (expressed as the salt) were administered intraperitoneally 30 min prior to testing. Diazepam was administered 15 min prior to testing. Where no treatment dose is indicated the subjects received injections of the appropriate vehicle: Saline or 0.5% methylcellulose in lieu of amitriptyline or diazepam, respectively. Each value represents the mean i S.E.M. of 3—6 deter- minations. Ordinate: Details as in Figure 7. *Significant (P<0.05) decrease in punished or unpunished responding compared to saline injected controls. °Significant (P<0.05) increase in punished responding compared to subjects receiving diazepam alone. 112 soul .001 FL and 1O G .— 3 H U 300} l IL 0 ["1 .— z Ill 9.: 200*- I.“ A. 1001' ‘ . dluz. 1.3 ' 1.8 " 1.8 " 1.3 an’mrip. - 3.0 3.0 5.6 5.6 10 10 Figure 17 113 significance of this interaction is obscured by the uncertain relation- ship between the tricyclic antidepressants and the synaptic activity of 5—hydroxytryptamine (Fuxe_epfl§l., 1977). The systemic administration of very small doses (10-20 pg/kg) of dflysergic acid diethylamide (LSD) produces a marked, reversible inhi- bition of activity of 5—hydroxytryptamine neurons (Aghajanian epflel., 1968) presumably as a consequence of a direct inhibitory effect on cell bodies comprising the raphé nuclei. An evaluation of the effects of lysergic acid diethylamide on conflict behavior revealed no signifi— cant anticonflict activity whether the hallucinogen was administered one or 30 minutes prior to behavioral testing (Figure 18). This finding is in seeming contradiction with that of Schoenfeld (1976), who reported that lysergic acid diethylamide significantly attenuated the suppressive effect of punishment on licking behavior in untrained rats. It must be emphasized, however, that important methodological differences exist between the two paradigms. In fact, lysergic acid diethylamide (10, 30 and 100 pg/kg) significantly decreased the rate of punished responding when given one minute prior to testing (Figure 18). A one-minute pretreatment time was employed in light of the very rapid short-lived depressant effect of lysergic acid diethylamide on 5- hydroxytryptamine-containing midbrain raphé neurons (Aghajanian_ep.§l., 1968). C. Effects of 5—Hydroxytryppamine Depletion on Conflict Behavior A decrease in brain 5—hydroxytryptamine content was effected by an inhibition of 5—hydroxytryptamine synthesis by pfchlorophenylalanine or by the administration of 5,7—dihydroxytryptamine, which has a neurotoxic 114 .moCHm> HouuCoo on OmHmmaoo OoEDmCoo House mo oBCHo> Ho Coxmu oxoonm mo HoOECC one CH wmmohoop quonHCme m moumoHOCH H .n oHCme CH mm mHHMqu "oumCHOHO .mCOHumCHEHouoO oumumaom NHle EOHH OouMHsono mm .E.m.m H H moCHH HmoHuHo> OCm mCmoE uComoHaoH wHOLEAm .wCHumou HMH0H>mCo£ mo COHumHuHCH osu ou HOHHQ AuonV CHE Om Ho AuanuV CHE H HoCuHo OoHoumHCHEOm mums Aommn ooum on“ on Oommoum Ime ouMHuHMuHL QmHIO mo mowom .mCHxCHHO mo COHmmoHaaCm OoCOHquCoo CH wCHUCommoH A AV V OoCmHCCmCC OCm A . V OmamHCCa Co AmmHIUV oHcHEMHtEumHO OHom UHwHomkHlfl mo 39de .OH 0.3me 115 Ens—«09.7.... 0.35:; ma magmas a; \9« 8. on 9 on 3 3 q a . fi 1 s O / / / 1111.0 I I! 1 \§ 306.3335 035:. CM CON “IOTILNOD £0 “43383:! 116 effect on 5-hydroxytryptamine—containing neurons (Table 5). In agree— ment with the findings of Koe and Weissman (1966), pretreatment with 400 mg/kg pfchlorophenylalanine, i.p. (120 hours) caused a considerable (57.5 percent of control) depletion of whole brain 5—hydroxytryptamine (Table 5). The effects of this single dose of pfchlorophenylalanine on experimentally—induced conflict behavior are shown in Figure 19. Punished responding was significantly increased 48 hours (day 2) following pfchlorophenylalanine administration, actually decreased significantly below control levels on the next day, and attained control values by day 5. Unpunished water consumption was signifi— cantly decreased on all testing days subsequent to pfchlorphenylalanine administration. As the effects of pfchlorophenylalanine on brain 5— hydroxytryptamine and tryptophan hydroxylase activity are reversible (Koe and Weissman, 1966), pfchlorophenylalanine was chronically admi— nistered (200 or 300 mg/kg i.p. every 2 or 3 days, respectively, for a total 0f 14 days) to another group of conflict—trained rats to effect a more uniform decrease in S—hydroxytryptamine concentrations and enzyme activity. Although the chronic administration of pfchlorophenylalanine appeared to increase the rate of punished responding, only the increase on days 5 and 7 attained statistical significance (p<.05; Figure 19). Unpunished water consumption tended to be decreased, reaching signi- ficance on several days, throughout the course of drug administration. Subjects sacrificed upon termination of the study exhibited a marked decrease (to 40 percent of control) in whole brain 5—hydr0xytryptamine (Table 5). In agreement with Blakely and Parker (1973), the findings of this experiment fail to support the contention of Geller and Blum 117 .maosucoo aoum Amo.ovmv udmumeHw mauamoflwflawflme .wououdeHEwm was manna udmmmumou mHonahm .HH can w .0 .q .N mkmw do wawumwu wdHBOHHom commands new GOHmmmm Houudoo Hodwm osu wdflsoaaom commauwdfl hasmaflsflm mmB Auzwfls mmx\wa com no oomv coaumuumflsfiawm o osms wdflvcommon wmsmfldsmas was wosmfiasm do muomwmo mSu paw scammmm Houudoo Hmcfim msu wdeoaHom kaoumwwmaafl wououmflcflavm mm: Aumma mmx\wa oo¢v smolm mo coauomhcfl wudom a4 .Awdflxdflnw mo dosmmmHQQSm woGOHuflwdoov H0H>mnmn uoflamnoo nooswdfilhaamucmafluodxm no A¢mUImv wdfidmama%dm£mouoaaotm mo mGOHumuuchfiawm wmuwommh can onnfim mo muommmm .mH muawflm 118 ma muamam uhfl‘ can 8a 8a 8a SN OOH 8v .meeflihfallem m .r M mun wilfilfl‘Olm/0\fl\4,%‘mwlfi/O\ 4 fl Wlfl WI. _l g 6 10mm: so mama «/ / ... / Walk KW» e _\ \ a a 717K l. 02.0.2.0 ‘1 4 / Bugpuodsaa" poqswnd 119 (1970) and Robichaud and Sledge (1969) that depletion of brain 5— hydroxytryptamine by pfchlorophenylalanine produces an attenuation of the behaviorally suppressive effects of punishment qualitatively similar to that reported for various antianxiety agents. However, numerous methodological differences in the paradigms limit the number of meaningful comparisons. The intracerebroventricular injection of 5,7-dihydroxytryptamine (100 ug) produced a transitory release of punishment—suppressed beha— vior in the modified Geller—Seifter procedure (Figure 20). Qualita— tively similar, though less long—lasting, anticonflict effects have been reported by Stein EE.§£- (1975) following the injection of 5,6- dihydroxytryptamine (100 pg) by the same route. Unpunished responding regained control rates within 5 days of 5,7—dihydroxytryptamine injec— tion but was significantly depressed prior to this time (Figure 20). Vehicle injected controls exhibited a similar initial depression, suggesting that this effect may be the resultant of the after-effects of the anesthesia and/or the trauma of the surgical procedure. Intra— cerebroventricularly injected 5,7—dihydroxytryptamine produced a 26% decrease in whole brain 5—hydroxytryptamine compared to controls in animals sacrificed 10 days following the injection of the neurotoxin or vehicle (Table 5). Bilateral injections of 5,7—dihydroxytryptamine (7 pg) into the amygdala produced a profound, transitory anticonflict effect in the Geller-Seifter paradigm of more rapid onset and shorter duration than that observed following intracerebroventricular injection (Figure 20). Rates of unpunished responding had reached control values by day 3. The 5—hydroxytryptamine content of the amygdala (dissected free in_toto 120 .mmsam> Houuaoo GOHuoomafimum Scum Amo.ovmv .n ouswflm dfi mm mHkumm "mumcflwso .mGOflumsHapmump mumnmmom m unmmmudmu maonahm .doauoondfl wasp pam hummusm mo man wouMflquH wdfluwmu HQHOH>m£on was mums pmnflmsu mamwwhaw mnu was Aumma mm: OOHV mHoHsudo> annouma v mdflpdommms votmflasm pamumnmas sanamoamaawama mom .2.m.m H H mmsHH Hwowuuo> can manna msu maaMuMp How uxou mom .AH %mwv hmw waHBOHHow IuoHHmGoo mo Auswfls whaamsmumafln .w: NV 93 once wmuoomaw mma Hmmlmfi .A O v wdfiwsommos wosmfldsmns “A . .Amuswoooum HmHMmeIHmHHmU meMHwoav H0H>m£ma uoflamdoo wooswufilhaamuaoaflummxm do Aammlm.mv odflam lum%uu%xouwksfiwlm.m mo doaumuumwcwawm pacamcwmammsucfi was “wasowsuco>muucfl mo muoommm .om muswflm 121 an. .14 1.0—ovunEuatfi ow magmas shout.» to.“ 101 m|IWIIOI IOIIOII@\ \W‘ \ \ x4 _ » _\7 a. _\ .\ ..a_:u_...:o>o.__:_ 8 N TOULNOD iO 1N 3383:! 122 on ice) was decreased by 27% (1.20i0.l7 ug/g) relative to controls (1.66:0.20 ug/g) 10 days following treatment with the neurotoxin. Neither intra-amygdaloid nor intracerebroventricular injections of the vehicle yielded consistent effects on conflict activity (data not shown). Thus, interference with normal S—hydroxytryptamine functioning can produce an effect on conflict behavior quantitatively similar to that observed with benzodiazepines. However, the manifestation of this effect does not appear to be directly related to the degree to which whole brain 5—hydroxytryptamine content is depleted, as the depletion produced by pfchlorophenylalanine was substantially greater than that produced by intracerebroventricular 5,7-dihydroxytryptamine. Rather, the results suggest that the discrepancy between the behavioral effects of pfchlorOphenylalanine and 5,7-dihydroxytryptamine-induced 5-hydroxy— tryptamine depletion may be related to a differential effect on the integrity of 5—hydroxytryptamine—containing nerve terminals. These are destroyed by 5,7-dihydroxytryptamine but remain structurally intact following pfchlorophenylalanine treatment. Moreover, Sloviter §E_al, (1978) have reported that pfchlorophenylethylamine (prPEA), a decar- boxylation product of pfchlorophenylalanine, displaces 5-hydroxytrypt— amine from intact terminals. Alternatively, the measurement of 5- hydroxytryptamine concentrations in more circumscribed areas of the brain may reveal a closer correlation between the degree of depletion and effects on punishment suppressed responding. These experiments involving the administration of 5,7-dihydroxy- tryptamine yielded results in apparent contradiction to the findings depicted in Figures 16, 17 and 18. That is, reducing activity of brain 123 5—hydroxytryptamine pathways by treating with the neurotoxin showed a clear anticonflict effect in the Geller-Seifter test. However, a presumed reduction in forebrain serotonergic tone by LSD caused an increased suppression in the drinking procedure, while enhanced 5- hydroxytryptamine activity by combining S—hydroxytrypt0phan or ami— triptyline with diazepam actually increased the anticonflict effect. It must be emphasized that the Geller-Seifter procedure utilized food reinforcement, whereas the conditioned suppression of drinking utilized the motivation of thirst. These different reinforcements did not seem to affect the anticonflict spectrum of the benzodiazepines alone. On the other hand, the alteration in brain serotonergic function may be expected to influence food-motivated behaviors on the basis of the reinforcement since increased 5—hydroxytryptamine activity is anorectic and decreased activity tends to increase food consumption (Samanin gt .§l°: 1977). This topic will be expanded in the Discussion. In addition, the administration of S—hydroxytryptophan is not necessarily selective in producing changes in brain 5—hydroxytryptamine. The amino acid may also enter brain catecholamine neurons and influence levels of these neurotransmitters. D. Effects of Single and Repeated Administrations of Diazepam on Body Temperature Bartholini ggflal. (1973) have suggested that the effects of high doses of diazepam (10 mg/kg) on the dynamics of various neurotrans— mitters (gag,, dopamine) in the central nervous system may be secondary to a hypothermic effect of this benzodiazepine. It was thus of in— terest to determine whether diazepam, at an effective anticonflict dose (1.8 mg/kg, i.p.), devoid of measurable nonspecific sedative effects 124 (Figure 7), similarly decreased the rectal temperature of rats when given acutely or following repeated administrations. In agreement with the findings of Carpenter 35 a1. (1977) a single injection of diazepam (1.8 mg/kg) failed to significantly alter the rectal temperature of rats (Table 6). Furthermore, when chronically administered (1.8 mg/kg/day for 5 consecutive days), diazepam produced no significant effects on rectal temperature compared to vehicle injected controls (Table 6). It would thus appear unlikely that the effects of diazepam on neuronal activity as related to the anticonflict effect, at least at this dosing regiment, are secondary to alterations in body temperature. In support of this contention Fuxe §t_§l, (1975) have found the benzo— diazepine—induced alterations in rat brain dopamine turnover to be unchanged when the animals' body temperature is maintained by an elevated environmental temperature. E. Effects of Diazepam on DOpaminergic Neuronal Pathways The steady state dopamine concentrations and average protein content of the amygdaloid nuclei, nucleus accumbens, olfactory tubercle and caudate nucleus are listed in Table 7. The anterior amygdaloid area, while representing the most rostral aspect of the amygdala, contains a number of fibers and cell bodies which subserve neuronal systems other than those involving the amygdala (Palkovits §t_§l,, 1974). The relatively uniform protein content values attest to the reproducibility of the microdissection techniques. The depamine con- centration of the amygdaloid nuclei are essentially the same as, though somewhat higher than, those reported by Brownstein §t_§l, (1974). Corresponding values for the caudate nucleus, nucleus accumbens and 125 TABLE 6 Effect of Diazepam (1.8 mg/kg) on Rectal Temperature (°C) in Rats Time After Injection (min) Day Treatment 15 30 45 60 1—5 vehicle 38.6i0.2 38.3i0.3 38.3i0.2 38.2i0.2 l diazepam 38.7i0.3 38.5i0.4 38.5i0.2 38.5i0.2 2 diazepam 38.5i0.1 38.4i0.2 38.5i0.1 38.4i0.l 3 diazepam 38.7i0.1 38.6i0.1 38.5i0.l 38.5i0.1 4 diazepam 38.5i0.2 38.3i0.2 38.3i0.2 38.3i0.l 5 diazepam 38.4iO.2 38.0i0.1 37.9i0.2 37.9iO.2 Diazepam or vehicle (0.5% methylcellulose) was administered i.p. Temperature recordings for any given time point in vehicle-injected controls were not significantly altered by repeated vehicle admini— strations and were subsequently combined. Each value represents the mean i l S.E.M. of 15 (vehicle-injected controls) or 5 (diazepam) separate determinations. 126 TABLE 7 Dopamine Concentrations in Various Nuclei in the Rat Brain Brain Nuclei N pg of Protein Dopamine . per sample ng/mg protein amygdaloid nuclei anterior amygdaloid area 13 62f 3 39.7i4.1 Central 17 99ill 26.3i5.2 Lateral 18 113i 7 17.9i2.7 Basal 15 96i10 10.1i1.9 Cortical l6 77f 9 7.2i0.8 Medial l6 89$ 6 4.4i0.7 Posterior l9 40i 6 2.2i0.6 Basal posterior l9 61f 3 6.6i0.9 Medial posterior 18 42: 5 1.8i0.5 rostral limbic nuclei nucleus accumbens 11 108i 8 79.7i7.3 olfactory tubercle 12 79: 9 53.9i4.0 Caudate nucleus 11 87: 2 83.6i4.3 Values represent means i 1 S.E.M. N = number of samples assayed. 127 olfactory tubercle are similar to those reported by other investigators (Koslow §£_al., 1974; Cheney g£_al., 1975). The endogenous dopamine concentration of the brain is a reflection of the finely regulated processes of synthesis and release at a steady state. The characteristically consistent maintenance of this steady state ensures a relatively unchanging endogenous dopamine concentration in spite of physiologically— and pharmacologically-induced alterations in dopaminergic neuronal activities. Hence, the study of the steady state dopamine content of various brain regions is of little value by itself when estimating the activity of neuronal pathways involving these areas. The function of the central nervous system involves a dynamic interaction within and between neurons. Therefore, function, both basal and drug—induced alterations thereof, may be best described in terms of biochemical correlates by studying the dynamics of the neurotransmitters in various brain nuclei in terms of their estimated rate of turnover. Similar arguments pertain to S-hydroxytryptamine and other neurotransmitters as well. 1. Effects of Acute Diazepam Administration on Dgpamine Turnover in the Amygdala, Olfactory Tubercle, Nucleus Accumbens and Caudate Nucleus With the exception of the medial posterior amygdaloid nucleus and the posterior amygdaloid nucleus, the c0ncentrations of dopamine declined exponentially after administration of a—methyltyrosine (300 mg/kg, i.p.) in all brain areas examined. Representative examples of the logarithmic nature of the a—methyltyrosine—induced depletion of dopamine are illustrated in Figure 21. The failure of the two afore— mentioned amygdaloid nuclei to exhibit a similar pattern most probably 128 Figure 21. The logarithmic decline of dopamine concentrations in various brain nuclei following the administration of a-methyltyro— sine (300 mg/kg, i.p.). Animals were sacrificed immediately before (0 time) or 45 and 90 minutes after treatment with demethyltyrosine. Symbols represent means and vertical lines 1 1 S.E.M. as determined from 5-8 animals. Dopamine (ng/mg prolein) 45 mlnulu alhr d MT Figure 21 n. condoms ollcdory lubcrclo n. cmyg. coma-all: n. amyg. corllcclls n. amyg. modiclls T 8.. 130 reflects methodological difficulties involved in reproducibly measuring depleted dopamine concentrations in tissue samples with a calculated approximate wet weight of 400 pg (assuming protein content represents 10-12% of the wet weight) and possessing a relatively low intrinsic dopamine content. Diazepam (1.8 mg/kg, 15 minutes prior to sacrifice) failed to alter the steady state dopamine concentration in any of the brain areas examined before the administration of d—methyltyrosine (1,2,, zero time concentration). However, when given in conjunction with d— methyltyrosine, a single dose of diazepam (1.8 mg/kg, i.p.) reduced the rate constants for the d-methyltyrosine-induced decline of dopamine in amygdaloid nuclei: the lateral, central, cortical and basal posterior, as well as in the olfactory tubercle (Table 8). Acute diazepam admini- stration failed to significantly alter the rate constant for the @- methyltyrosine—induced decline of d0pamine in the basal and medial amygdaloid nuclei or the nucleus accumbens and caudate nucleus. As the clinical (Warner, 1965) and animal behavioral (Margules and Stein, 1968; Sepinwall_9£fl§l., 1978) effects of the benzodiazepines appear to undergo significant qualitative and quantitative changes with their repeated administration, it was of interest to determine whether the diazepam-induced change in dopamine turnover exhibits a similar alteration over the course of repeated administrations. 2. Effects of Repeated Diazepam Administrations on Dopamine Turnover in the Amygdala, Olfactory Tubercle, Nucleus Accumbens and Caudate Nucleus Rats were injected with diazepam (1.8 mg/kg, i.p.) daily for 5 days. The last dose was given 15 minutes before the administration . ..‘x. 131 TABLE 8 Effects of Acute Diazepam Administration on the a—Methyltyrosine- Induced Depletion of DOpamine in Forebrain Areas Steady State Fractional Calculated Rate , Concentration Rate Constant of Formation Nuclei (N) ng/mg protein h”1 i S.E.M. ng/mg protein/h i S.E.M. amygdaloid nuclei Lateral vehicle (15) 16.6il.1 O.28i0.05 4 6 diazepam (23) 17.7i2 O O.15i0.03* 2 6 Central vehicle (15) 28.0i3.7 O.39i0.05 10.9 diazepam (26) 25.5i2.9 0.27i0.06* 6.8 Cortical vehicle (14) 7.0:1.0 0.26:0.02 1.8 diazepam (22) 7.7:0.6 O.20:0.03* 1 5 Basal vehicle (15) 8.8il.3 0.26i0.04 2.2 diazepam (22) 9 2:0.9 0.22:0.03 2.0 Medial vehicle (15) 3.9i0.7 0 2810.02 1.1 diazepam (21) 4.6:0.8 0.27:0.06 1.2 Basal Posterior vehicle (16) 6.9iO.8 0.30i0.03 2.1 diazepam (21) 2.7il.2 O.21i0.03* 1.2 rostral limbic nuclei Olfactory tubercle vehicle (13) 59.2:3.8 0.32:0.03 19.1 diazepam (17) 61.1i4.4 0.21i0.04* 12.8 Nucleus accumbens vehicle (14) 84.4i2.9 0.29:0.01 24.3 diazepam (l6) 82.3i3.7 0.26i0.03 21.4 Caudate nucleus vehicle (14) 80.6i4.2 0.25i0.03 20.1 diazepam (l6) 88.8:6.3 0.22i0.03 19.3 Diazepam (1.8 mg/kg) or vehicle (1 m1/kg) was injected 15 minutes before a-methyltyrosine (300 mg/kg, i.p.). Groups of 4-11 rats were sacrificed at O, 45 and 90 minutes after damethyltyrosine. Rate of formation of dopamine was estimated by multiplying the fractional rate constant of the decline of dopamine concentrations times the steady state dopamine content in each brain region. Steady state values for subjects receiving vehicle or diazepam were determined 90 minutes after injection of saline (N = 6—8). *Indicates those values of the diazepam-pretreated animals which differ significantly (p<.05) from the vehicle-pretreated animals. 132 of d—methyltyrosine. The steady state dopamine concentrations of the brain nuclei examined were similar in animals chronically pretreated with diazepam or vehicle (Table 9). The rate constant of d0pamine loss following synthesis inhibition was decreased following prolonged diazepam treatment in only two of the amygdaloid nuclei: the lateral and basal amygdaloid nuclei (Table 9). The rate of depamine formation was similarly decreased in the olfactory tubercle and nucleus accum- bens. These results suggest that, at least in the amygdala, some degree of tolerance may develop to the effects of diazepam on dOpamine turnover over the course of its repeated administration. The rate- decreasing effect of chronic diazepam treatment on d0pamine turnover in the olfactory tubercle and nucleus accumbens, but not in the caudate nucleus, closely resembles the results obtained by Fuxe g; 51. (1975) in these brain regions with higher doses of diazepam (5 and 10 mg/kg). Thus, the differential effects of acute vg, chronic diazepam admini- stration on depamine turnover in these latter areas may be partly the result of an accumulation of diazepam or its active metabolites with repeated injections. Whether the observed effects of acute diazepam administration on dopamine turnover rates represent a direct action of the drug on dopaminergic neurons or are secondary to actions on other neurotrans— mitter systems cannot be discerned from the present evidence. 3. Effects of Picrotoxin on Diazepam-induced Changes in DOpamine Turnover in the Amygdala and Olfactory Tubercle A number of investigators have postulated that benzodiaze— pines may exert a primary action on y-aminobutyric acid—containing 133 TABLE 9 Effects of Repeated Diazepam Administration on the d—Methyltyrosine- Induced Depletion of Dopamine in Forebrain Areas Steady State Fractional Calculated Rate Nuclei (N) Concentration Rate Constant of Formation ng/mg protein h_1 i S.E.M. ng/mg protein/h i S.E.M. amygdaloid nuclei Lateral vehicle (12) 15.3i1.6 0.31:0.05 4.8 diazepam (17) 15.9i1.1 0.19i0.03* 3.1 Central vehicle (14) 23.7i2 9 O.38i0.04 8.9 diazepam (22) 25.0i3.3 0.32i0.06 7.9 Cortical vehicle (13) 6.7:0.9 0.28i0.04 1.9 diazepam (19) 6.7i0.5 O.26i0.03 1.7 Basal vehicle (14) 8.1:0.9 0.25:0.02 2.0 diazepam (19) 8.3i1.2 0.17i0.04* 1.4 Medial vehicle (13) 3.7i0.3 O.26i0.05 1.0 diazepam (20) 4.1i0.7 O.3li0.05 1.3 Basal Posterior vehicle (13) 5.8i0.8 0.29i0.04 1.7 diazepam (18) 6.3il.7 0.27i0.05 1.7 rostral limbic nuclei Olfactory tubercle vehicle (12) 52.6i4.1 0.29i0.02 15.1 diazepam (l6) 59.1i5.4 0.18i0.03* 10.8 Nucleus accumbens vehicle (13) 90.2i7.0 0.32i0.02 29.0 diazepam (l6) 88.8i3.8 O.26i0.01* 23.1 Caudate nucleus 1 vehicle (13) 86.6i6.2 0.27i0.04 22.9 ‘ diazepam (15) 77.9:8.3 0.23io.o4 18.1 ‘ Diazepam (1.8 mg/kg) or vehicle was injected i.p. daily for 5 ‘ consecutive days. d—Methyltyrosine(300 mg/kg, i.p.) was administered 15 minutes following the last injection. Groups of 3—9 rats were sacri— ficed at O, 45 and 90 min later. Other details of the calculations ‘ are the same as for Table 8 (N=7—9). *Indicates those values from the chronic diazepam—pretreated animals which differ significantly (p<0.05) from their vehicle—pre- treated controls. 134 neurons which in turn modulate the activity of other neuronal pathways utilizing different neurotransmitters (Costa 9E“§1°’ 1975; Haefely, 1977). If the diazepam-induced reduction in dopamine turnover is secondary to an enhanced y—aminobutyric acid—mediated transmission, then the administration of a y—aminobutyric acid receptor blocking agent, such as picrotoxin (Curtis and Johnston, 1974), should antago— nize the effects of diazepam on d0pamine turnover. Diazepam (1.8 mg/kg, i.p.) was administered 15 minutes, and picrotoxin (l or 2 mg/kg, i.p.) 5 minutes, prior to d—methyltyrosine or saline. Animals were sacrificed 90 minutes after the administration of d-methyltyrosine and dopamine content was measured in those brain nuclei which exhibited a significant reduction in the rate constant for the decline of d0pamine and, consequently, the calculated rate of formation of dopamine follow— ing acute diazepam treatment (Table 8). Ninety minutes after a- methyltyrosine treatment the dopamine concentration in the amygdaloid nuclei and olfactory tubercle of vehicle—pretreated rats was reduced to approximately 63% of the steady state value. A single injection of diazepam (1.8 mg/kg, i.p.) attenuated this u-methyltyrosine-induced depletion of dopamine concentrations to approximately 78% of the steady state value. Picrotoxin, at a dose of 1 mg/kg, did not significantly alter the effect of diazepam on the d-methyltyrosine—induced depletion of dopamine concentrations in the central, lateral, cortical and basal posterior amygdaloid nuclei or the olfactory tubercle. However, a doubling of the dose of picrotoxin was found to counteract the de- pressant effect of diazepam on the d-methyltyrosine—induced reduction 135 of dopamine concentrations in most of these brain regions (Table 10). That is, the dopamine concentrations 90 minutes after a—methyltyrosine treatment were the same in animals receiving either vehicle or the combination of diazepam and picrotoxin. Neither dose of picrotoxin appreciably altered the steady state dopamine concentration in any region examined. The concentration of dopamine in the cortical amygda— loid nucleus after d—methyltyrosine administration was not signifi— cantly increased in diazepam—pretreated animals compared to those receiving vehicle. The results suggest that a picrotoxin—sensitive mechanism may mediate the effects of acute diazepam administration on dopamine turnover in the amygdala and olfactory tubercle. However, although the steady state dopamine concentrations in these brain regions were un~ altered by this large dose of picrotoxin, the direct effect of picro- toxin on the dopamine depletion evoked by d—methyltyrosine was not examined. Nevertheless, both the behavioral (Waddington and Longden, 1977) and biochemical (Keller §£_§l., 1976) expression of the inter— action between benzodiazepines and dopaminergic systems have been shown to be negatively influenced by the y—aminobutyric acid receptor blocking agents. Moreover, a low dose of diazepam (1 mg/kg) has been reported to decrease Y—aminobutyric acid turnover in subcortical nuclei (Mao EL 31,, 1977) and the limbic cortex (Pericic §£_§l,, 1977). The amygda— loid nuclei are rich in y-aminobutyric acid and its synthesizing enzyme glutamate decarboxylase (Ben—Ari §t_gl., 1976; Tappaz §t_§l,, 1976). Thus, a diazepam—induced facilitation of y—aminobutyric acid—mediated transmissiOn could potentially influence the activity of dopamine— containing neurons terminating in the amygdala by a presynaptic action 136 TABLE 10 Effects of Picrotoxin on the Diazepam-induced Attenuation of the a—Methyltyrosine—induced Depletion of Dopamine in the Amygdala and Olfactory Tubercle Treatment Nuclei . a . Pretreatment N Saline a Methyltyr031ne ng/mg protein i S.E.M. Lateral amygdala vehicle 5 11.4il.2 (66) diazepam 7 17.2il.3 14.2i0.9 (83)* diazepam + picrotoxin 6 12.7il.6 (74) Central amygdala vehicle 5 15.1il.l (56) diazepam 8 27.2i3.6 l9.5il.8 (72)* diazepam + picrotoxin 6 16.0i1.7 (59) Cortical amygdala Vehicle 5 4.7i0.6 (65) diazepam 8 7.2i0.9 5.7i0.8 (79) diazepam + picrotoxin 6 4.5:0.8 (63) Basal posterior amygdala vehicle 4.0:0.5 (64) diazepam 8 6.3i0.6 5.4i0.6 (85)* diazepam + picrotoxin 6 3.7i0.3 (59) ’ Olfactory tubercle vehicle 5 34.6i3.9 (61) diazepam 5 56.6:4.6 46.3i5.5 (82)* diazepam + picrotoxin 6 40.2i6.1 (71) Animals received vehicle (0.5% methylcellulose) or diazepam (1.8 mg/kg) intraperitoneally 105 minutes prior to sacrifice. An additional group of rats received picrotoxin (2 mg/kg, i.p.) 10 minutes following diazepam treatment (95 minutes before sacrifice). Ninety minutes before sacrifice, half of the animals in each of the three groups received a—methyltyrosine (300 mg/kg, i.p.) and the other half received saline (1 ml/kg). Numbers in parentheses represent the d-methyltyrosine values ex— pressed as a percentage of the combined pretreatment values obtained in saline injected animals for each brain region. aAs the dopamine concentrations in the various brain nuclei of rats administered saline were similar regardless of the pretreatment these values were combined. *Designates those values which differ significantly (p<.05) from vehicle—injected controls. N = number of samples assayed. 137 on midbrain cell bodies or via pre— and postsynaptic inhibitory mecha— nisms at the terminals. F. Effects of Diazepam 0n 5—Hydroxytryptamine Neuronal Pathways The effects of diazepam on the steady state concentration of 5- hydroxytryptamine and 5—hydroxyindoleacetic acid were determined in the amygdaloid nuclei as well as other nuclei which have been shown by histochemical fluorescent mapping techniques to contain a major sero— tonergic input (Aghajanian §£_al., 1973). The microdissected hippo— campal tissue samples contained all the layers of the hippocampus. Hypothalamic samples represent tissue collected from the region of the anterior hypothalamic nucleus. 5—Hydroxytryptamine and 5-hydroxy- indoleacetic acid were simultaneously assayed in each brain region examined. The effects of single (1.8 mg/kg) and repeated (1.8 mg/kg/ day for 5 days) intraperitoneal diazepam injections were evaluated since several actions of the benzodiazepines deemed relevant to their properties as anxiolytics undergo appreciable change over the course of repeated administration (see Introduction). 1. Effects of Acute and Repeated Diazepam Administration on Steady State 5—Hydroxytryptamine and 5—Hydroxyindoleacetic Acid Concentrations in the Amygdala, Hypothalamus and Hippocampus The endogenous 5-hydroxytryptamine concentrations of the amygdaloid nuclei, hippocampus and hypothalamus from animals receiving vehicle or either acute (1.8 mg/kg) or chronic (1.8 mg/kg/day for 5 days) diazepam administration are listed in Table 11. Animals (n=8) in each of the three groups were sacrificed 30 minutes following the last pretreatment and tissue samples of pairs of rats in each grOup were 138 .mHosuaou tmuomde mHoHnm> ou pmummfioo mGOHumuuamodou Balm uH Amo.ovav mummsoaH unnoHMHdem m mounoHtcHa .mGOHumaHahmumw mumummmm a now saw H H manna mtu unmmmummu mmsHm> .meoom mums mHmaHdm mo mHHmm aonm pmuomHHoo mMHmamm mummHH .moHMHHUMm ou HOHMQ moundHa om Amhmw m>Hudommnoo m so one now wx\ma w.Hv ammmnan so AomOHdHHmlosumE Nm.ov mHoH£m> wm>HmooH mHmqum m.mHo.mN H.NHm.wH n.HHN.mH masmHmsuom%: s.one.s s.one.m o.oa~.q msaamooaaae m.NHH.oH m.HHw.NH N.NHm.mH MOHnmumom m.NHm.mH o.NHm.oH H.mHH.mH MOHHmumom Hmema w.ma~.o~ H.mso.om o.NHN.wH MOHHmumom Hmmmn o.mao.H~ o.NHm.sH m.NHo.wH Hmema m.HHN.sH ¢.Ham.oH m.HHm.qH Human «m.NHH.qN m.osH.oH H.HHo.mH HmoHuuoo w.mnm.HN w.HHw.oH w.HHm.mH Hmuuumo o.mam.Hm H.NHm.wH o.NHw.wH HMHMumH HmHUSd wHOHmvwma< aHmuoum wE\Hmlm ma ammmNmHm ammmumHm oHcosno muso< mHUH£m> HoHonz udmfiummHH wsamHm3uomhm can mnmfimooamHm .meww%a< mnu GH AHmImv maHamumksuhxouwhmlm mo aOHumsuumocoo mumum hummum map so GOHuwHumHnHaw< EmmmumHm oHaouzo can musu< mo wuommmm HH HHm