M.E. Olds

Reinforcing effects of morphine in the nucleus accumbens

The possibility that morphine exerts in reinforcing effects via an action in the nucleus accumbens was investigated in the rat using the self-administration model and the method of intracerebral injection. Male, adult rats implanted with a chronic cannula aimed at this nucleus were tested for self-administration in a rectangular chamber equipped with two levers, one at each end. Each subject was given several sessions in blocks of two sessions for a given substance, the first session with the active lever at one end of the chamber, the next with the active lever at the other end. During a session, responses on the inactive lever were without consequences. Depression of the active lever was coupled with the presentation of a tone to facilitate discrimination between the two levers. All subjects received a sequence of 5 sessions, the first for habituation, the next two to test for effects of artificial cerebrospinal fluid, and the next two to test for effects of morphine. A few subjects in which the cannula had not been dislodged or plugged were further tested for effects of morphine and then for effects of morphine mixed with naloxone. Each response on the active lever resulted in an injection of 0.02 microliter. During the morphine sessions, this volume contained 0.2 microgram of the opiate; during the sessions of morphine and naloxone, this volume contained 0.2 microgram morphine mixed with 0.2 microgram naloxone. The results show that self-administration was induced and maintained when morphine was injected inside the nucleus accumbens, not in an area outside this region but at the same brain level. The rate or responding on the active lever for morphine was higher than for artificial cerebrospinal fluid and it showed a pattern of gradual increase over time. In contrast, the rate of responding on the inactive lever during the morphine sessions showed a decline. This pattern was not evident in these animals during control sessions and in the animals in which the injections were outside the n. accumbens. These results are interpreted to indicate that the n. accumbens may be part of a system of structures mediating the reinforcing effects of opiates.

Hypothalamic substrate for the positive reinforcing properties of morphine in the rat.

Abstract The question of whether morphine has reinforcing effects when applied at brain-stimulation reinforcing sites of the hypothalamus in the rat was studied with the method of self-administration. Adult, male rats were implanted with a combination cannula/probe for tests of selfstimulation and selfadministration. The test chamber had a lever at each end; one delivered 0.02 μl of a solution when it was depressed, the other nothing. In the first experiment, the active lever remained on the same side for all tests. The subjects were low to moderate selfstimulators. Morphine was tested at two concentrations, 5 μg/μl and 10 μg/μl. The vehicle was artificial cerebrospinal fluid with a pH of 5.7. Results showed significantly higher rates of responding on the active lever for morphine at the 10 μg dose than for artificial cerebrospinal fluid, and higher rates on the active lever than on the inactive one for morphine at the 10 μg dose, as compared with rates for the 5 μg dose. In the second experiment, high selfstimulators learned successfully to reverse repeatedly when the active lever delivering morphine (10 μg/μl) was changed from one side of the test chamber to the other in successive sessions. Results with a mixture of naloxone and morphine or with naloxone given intraperitoneally and morphine intrahypothalamically produced partial or total blockade of lever-pressing. These results are indicative of a specific action of morphine at sites that support selfstimulation behavior, and of these hypothalamic sites as neural substrate for the reinforcing effect of morphine.

Separation of associative from non-associative short latency changes in medial geniculate and inferior colliculus during differential conditioning and reversal in rats

Abstract Results obtained in a recent experiment in this laboratory drew attention to the possible contribution of subtle non-associative processes to changes in the early latency unit responses obtained in MGB and IC of the rat during the learning of a differential appetitive conditioning task. An important possible source of such non-associative effects was shown to arise from differences in the interval between a previous food pellet presentation and the following CS+ or CS−. To determine whether previously reported changes in MGB and IC were due to associative processes or to non-associative ones, the present experiments were carried out using a counterbalanced stimulus presentation schedule which placed the CS+ and CS− at the same average time following a previous food pellet. Criteria adopted for characterizing a change as associative required that the response to the CS+ be enhanced relative to that to the CS− throughout repeated reversal sessions. The results obtained support previous findings of early latency associative changes in MGB, and are consistent with the idea that such changes are most likely to occur in the medial division of this nucleus. The findings in the IC thus far do not support the notion that this nucleus participates in this process. However, in view of the complex structure of the IC and the fact that not all of its subdivisions were sampled, these negative findings must be viewed with caution. The results also support the view that the anatomical distribution of neurons participating in the elaboration of associative changes may be more restricted than that of neurons participating in non-associative ones.

Short-term changes in the firing pattern of hypothalamic neurons during pavlovian conditioning

Summary The neuronal activity in the region of the hypothalamic medial forebrain bundle was studied in the chronic preparation, using a classical conditioning paradigm with food as the reward. Changes in the rate of firing after the auditory stimuli (conditioned stimuli) were assessed in terms of the time of their occurence during the CS-US interval and in terms of the number of trials before their appearance. A similar analysis was made of the movements of the animals which were recorded electronically andfunctioned as behavioral indices of learning giving a basis for relating neuronal changes to behavior changes. The statistically significant modifications of the spontaneous firing rate after the CS during the conditioning were called learned responses and characterized according to latencies. The modifications during the sensitiaation tests to the same stimuli were called innate responses. The findings for the hypothalamic region were mainly 3: (1) there were many innate neuronal responses with latencies of 20 msec or less measurable with the methods used in these experiments. All of the innate responses occured in the time interval before 80 msec after the stimuli presentation; (2) there were no learned neuronal responses with latencies of 20 msec or less. The majority of the learned responses occurred during the 40–80 msec intervals after the CS, that is before the occurence of the behavioral learned responses which had latencies of 80–120 msec, but later than many of the innate responses; (3) neuronal learning occured some time during the 10–30 trial sequence whereas learned behavior occured later in the sequence. These changes correlated with the acquisition of a skill rewarded by food, using a Pavlovian paradigm, are viewed as signifiing a participation of the hypothalamic region in the cognitive process, not as the primary ‘informational processor’ but as a region where input-output sequences are related to positive reinforcement.

Effects of intraventricular 6-hydroxydopamine and replacement therapy with norepinephrine, dopamine, and serotonin on self-stimulation in diencephalic and mesencephalic regions in the rat

Although 6-hydroxydopamine (6-OHDA) reduced the rate of hypothalamic self-stimulation, it is not known whether the same effect is produced at other reinforcing sites remote from the hypothalamus. This was investigated in rats implanted with rewarding probes in the hypothalamus, in the substantia nigra, in the midbrain, and in the pontine region. Two patterns of self-stimulation emerged in each subject. One, characterized by short, medium and long-stimulus-train durations, was seen in the hypothalamus, in the substantia nigra, and in selected sites in the pons. The other, which comprised only short stimulus-trains, was found in the medial midbrain and in the pontine region. 6-OHDA (250 mug intraventricular route) reduced the rate of responding in the regions where responding was for short, medium, and long stimulus trains. It has a minor effect on responding in regions where the brain rewards were exclusively of short duration. Thus, self-stimulation in the hypothalamus, in the substantia nigra and in the pons was suppressed after 6-OHDA, while self-stimulation in the medial midbrain and at sites in the pons, where selection was for short trains, was only slightly below control levels after 6-OHDA. L-Norepinephrine (L-NE) (10, 20 and 30 mug) injected into the lateral ventricle of the 6-OHDA-treated rats temporarily reinstated self-stimulation in the lateral hypothalamus but not in the substantia nigra and not in the pontine region where the pattern of selection had been for long brain rewards. Dopamine (DA) was not effective as an antagonist of the suppressant action of 6-OHDA. Serotonin (5-HT) reinstated self-stimulation behavior in the lateral hypothalamus but not in the other positive regions. Its action was less than that of NE and did not take place in all animals tested.

Unit responses in the medial forebrain bundle to rewarding stimulation in the hypothalamus

Abstract The effects of different intensities of stimulation were studied on self-stimulation behavior and on the neural responses in the medial forebrain bundle correlated with it. Of the two neural responses observed, an excitatory type was more diffusely distributed throughout this pathway than an inhibitory type, and its properties varied more as a function of the rate of self-stimulation at the various intensities tested than the properties of the inhibitory responses did. The latter were more localized to the caudal portion of the medial forebrain bundle studied, and their threshold was generally lower than the threshold of self-stimulation behavior. Also, the inhibitory responses appeared to be more sensitive indicators of stimulation applied to the medial forebrain bundle, whereas the excitatory responses were more sensitive indicators of variations in the rate of responding for brain reward. The neural responses in the MFB to rewarding stimulation are evaluated in terms of the possibility that they are specific to the positive reinforcement mechanism in the diencephalon.

Subthalamic responses to amphetamine and apomorphine in the behaving rat with a unilateral 6-OHDA lesion in the substantia nigra.

The activity of neurons in the subthalamic nucleus (STN) of the behaving rat, before and after a unilateral 6‐OHDA lesion of the substantia nigra, was recorded with the extracellular technique to determine whether it was altered following systemic amphetamine, 5 mg/kg, apomorphine, 3 mg/kg, and apomorphine, 0.3 mg/kg, and whether in cases of altered activity, it was related to the drug‐induced motor response expressed concurrently. Activity in the STN of intact rats increased dramatically after amphetamine, 5 mg/kg. This excitatory response had the same latency, similar magnitude, and the same duration as the motor response expressed in terms of locomotion and oral stereotypy. Motor and unit responses were also induced by amphetamine after the lesion with 6‐hydroxydopamine (6‐OHDA), but now the excitatory response was attenuated while the motor response was not. The effects of the 6‐OHDA lesion were the same in all animals with loss of the nigra dopamine neurons, regardless of whether they were rotators or non‐rotators. Activity in the STN of intact rats also increased after apomorphine, 3 mg/kg, and again, this increase was correlated with the increase in motor behavior, but both responses were of shorter duration than the responses to amphetamine. The increases in unit activity and motor behavior induced by apomorphine in the 6‐OHDA‐lesioned rats had the same magnitude but lasted longer than in the intact rats. Treatment with apomorphine, 0.3 mg/kg, of the intact rats produced small and very brief increases in the activity of the STN and in motor behavior. The same treatment given the 6‐OHDA‐lesioned rats produced responses of larger magnitude but no change in duration. These findings demonstrate a role for STN neurons in the mediation of the motor behaviors induced by stimulation of the dopamine receptor. The results also show that a unilateral lesion of the substantia nigra with 6‐OHDA did not block these responses but altered them in a manner consistent with a dopaminergic deafferentation of the basal ganglia. The increased activity in the STN during the expression of dopamine‐dependent motor behavior conflicts with the current model of basal ganglia function that assumes prejudicial effects of excessive STN activity on the expression of motor behavior. An explanation for this conflict suggests that it is more apparent than real. Synapse 34:228–240, 1999.

Patterns of conditioned unit responses in the auditory system of the rat

Abstract Short-latency unit changes in the auditory pathway of the rat during learning of a differential classical conditioning task with tones as the stimuli and food as the reward have been reported. It is possible that the changes in individual animals at the beginning of the conditioning trials could have been produced by abrupt shifts in the state of the animal. The present experiment was therefore carried out with the aim of finding out (i) whether acquisition-related unit changes take place in the medial geniculate, inferior colliculus, and posterior nucleus, and if so, what kind of changes; (ii) which types of changes are likely to be associative, using criteria more restrictive than those used previously; (iii) the latency of the changes; and (iv) the pattern of change in the different brain regions. Using the present method of analysis, the findings indicate the presence of acquisition-related increments of an associative nature with latencies between 20 and 50 ms. Changes with shorter latencies that met all the criteria for associativeness were not found, but could have been masked. Different types of increments were found in the different regions sampled and in some subdivisions within the medial geniculate, perhaps indicating activity of different cell groups.

Dopaminergic basis for the facilitation of brain stimulation reward by the NMDA receptor antagonist, MK-801

MK-801 (dizocilpine maleate), an antagonist of the NMDA receptor, was given alone or in combination with dopamine D1 and D2 receptor antagonists to rats self-stimulating in lateral hypothalamus to determine whether the dopamine neurons play a role in mediating the effects of MK-801. MK-801 given at a dose of 0.1 mg/kg i.p. to self-stimulators induced a prolonged facilitation of lever-pressing, but given to non-self-stimulators, the drug had no effects. Pretreatment of self-stimulators with the dopamine D1 receptor antagonist Schering 23390 (SCH 23390), 0.2 mg/kg given i.p. 15 min before MK-801, prevented the facilitation seen with MK-801, but did not suppress self-stimulation. SCH 23390 given alone suppressed self-stimulation. Pretreatment of self-stimulators with the dopamine D1/D2 receptor antagonist, haloperidol, 0.2 mg/kg given i.p. 15 min before MK-801, also prevented the facilitation of self-stimulation induced by MK-801 yet did not suppress self-stimulation. Haloperidol given alone suppressed self-stimulation. Pretreatment of self-stimulators with both SCH 23390 and haloperidol 15 min before MK-801 prevented the facilitation seen with MK-801 and suppressed self-stimulation. The combined treatment with SCH 23390 and haloperidol (without MK-801) suppressed self-stimulation, and the suppression lasted longer than the suppression seen when the two dopamine receptor antagonists were given as pretreatment, before MK-801. Pretreating self-stimulators with the combination of SCH 23390 and haloperidol 15 min before amphetamine (2 mg/kg) prevented the facilitatory response and suppressed responding for the brain reward. The suppression was of shorter duration than the suppression seen after the injection of SCH 23390 plus haloperidol. The treatment of self-stimulators with both MK-801 and amphetamine resulted in a greater and longer-lasting facilitation than the increase in responding produced by either drug alone. The similarity between the effects of MK-801 and those of amphetamine and between the effects of pretreatment with the dopamine receptor antagonists before MK-801 and before amphetamine suggests that dopaminergic activity played a significant role in the action underlying the effects of MK-801 on brain stimulation reward.

Effects of acute and chronic fenfluramine on self-stimulation and its facilitation by amphetamine

DL-Fenfluramine (20 mg/kg) releasing serotonin and amphetamine (2 mg/kg) releasing dopamine were given to adult rats trained to bar press for electrical stimulation to the medial forebrain bundle. Amphetamine treatment enhanced lever-pressing for 1-2 h. A single fenfluramine treatment rapidly suppressed self-stimulation with slow recovery in 5-7 days to a rate below the initial basal rate. A second treatment a week later again suppressed response rate and rates returned to a still lower baseline. Combined fenfluramine-amphetamine treatment at this time transiently abolished lever pressing for 1-3 h followed by 9-11 h of enhanced responding. The serotonin antagonist, ketanserine (0.1 mg/kg), but not cyproheptadine (0.1 mg/kg i.p.), partially protected against the effects of fenfluramine. The serotonin agonist, quipazine (0.5 mg/kg), but not dimethoxyiodophenylisopropylamine (DOI) (2.2 mg/kg), partially substituted for fenfluramine in the combined treatment. Fenfluramine markedly depleted serotonin and 5-hydroxyindoleacetic acid in frontal cortex, hippocampus, and caudate putamen. Ventral and midline midbrain regions were less affected. Combined fenfluramine and amphetamine treatment elevated dopamine levels in frontal cortex, hippocampus and caudate-putamen, but not in midbrain. These findings support a serotonin-dopamine interaction in self-stimulation behavior and suggest that repeated fenfluramine treatment results in chronic low level serotonergic stimulation and diminished serotonin storage capacity.