Satiety threshold regulates maintained self-administration: comment on Lynch and Carroll (2001).

In this commentary the authors argue that the satiety threshold is the only mechanism that is sufficient and necessary to explain the regulation of maintained self-administration. The other mechanisms have been proposed mainly because of 2 sources of confusion surrounding the self-administration paradigm: the failure to distinguish between separate phases of a self-administration session and the assumption that positive reinforcement underlies drug self-administration. The authors of this commentary emphasize that the direct effects and aversion mechanisms have been proven to be untenable for the reasons reviewed by W. J. Lynch and M. E. Carroll (2001) and that the "ascending limb" of the dose-response curve is an experimental artifact. These ideas have persisted only to salvage a role for positive reinforcement in drug self-administration. The authors conclude that reinforcement is not relevant to the regulation of maintained self-administration.

Satiety threshold during maintained cocaine self-administration in outbred mice.

Male Swiss Webster mice maintained cocaine self-administration in a regular and dose-dependent manner. These characteristics made it possible to apply the satiety threshold model of drug self-administration developed recently for cocaine self-administration in rats. Non-linear regression analysis revealed that cocaine satiety threshold was 1.3 +/- 0.6 mg/kg and the functional half-life of the cocaine was 8.1 +/- 2.2 min. Whether the self-administration of cocaine was maintained by lever presses or nose pokes did not influence the inter-injection intervals. The results are consistent with the pharmacological model of maintained cocaine self-administration. The ability to determine addiction-relevant phenotypes (the satiety threshold and functional half-life of cocaine) in inbred strains of mice may help to identify the genetic determinants of cocaine self-administration behavior.

Satiety threshold: a quantitative model of maintained cocaine self-administration.

The intervals between self-injections of cocaine by rats are defined by an equation that contains only three parameters: the dose of cocaine administered, the elimination half-life of cocaine, and an amount of cocaine in the body, which we have termed the cocaine satiety threshold. This latter parameter is defined as the maximal level of cocaine at which the probability of self-administration approximates one and above which the probability of self-administration is low. The mathematical model generated mean values for the satiety threshold and the functional elimination half-life of cocaine of approximately 1.7 mg/kg (i.v.) and 8.2 min, respectively. Therefore, the simple equations presented here permit the measurement of the pharmacokinetics and pharmacodynamics of cocaine using self-administration behavior as a bioassay. Our satiety model predicts that when cocaine levels are maintained above the satiety threshold, rats would not self-administer cocaine. The elimination rate of cocaine at the satiety threshold was calculated to be approximately 2 microg kg(-1) s(-1). Therefore, an infusion of cocaine at this rate should maintain cocaine levels fractionally above the satiety threshold. A continuous infusion of cocaine at this rate prevented cocaine self-administration for the duration of the infusion, thereby confirming the validity of the satiety model. These equations provide a quantitative description of cocaine self-administration and contain no subjective terms, implying that concepts such as "craving", drug "wanting" and "liking" and "reinforcement", used in psychologically oriented models, are not necessary for descriptions of this behavior in rats.

Priming threshold: a novel quantitative measure of the reinstatement of cocaine self-administration.

The intravenous injection of cocaine has been reported to reliably reinstate (prime) the self-administration of cocaine in animals. We report herein that there is a cocaine priming threshold in rats trained to self-administer cocaine. The cocaine priming threshold is defined as the minimum level of cocaine in the body that will reinstate maintained cocaine self-administration. The mean cocaine priming threshold in rats was calculated to be approximately 186 to 212 microg kg(-1). Therefore, any injection, series of injections or continuous infusion that result in a level of cocaine equivalent to that produced by a single intravenous injection of this range of doses, will reinstate cocaine self-administration. The priming threshold was significantly increased by the D(1) dopamine receptor antagonist SCH23390 (10 microg kg(-1), i.v.), indicating a role for dopaminergic neurotransmission. The priming threshold, but not the inter-injection interval of maintained self-administration, was increased following withdrawal from a 7-day infusion of D-amphetamine. In addition, there was no correlation between the cocaine priming threshold and the inter-injection intervals of maintained cocaine self-administration. Therefore, the mechanisms underlying the reinstatement of cocaine self-administration are distinct from the mechanisms underlying the maintenance of cocaine self-administration and they are differentially regulated. It is possible that the priming threshold may represent a distinct target for medications development.

Transient amelioration of the sensitization of cocaine-induced behaviors in rats by the induction of tolerance.

Intermittent administration of cocaine produced a progressive increase in the stereotypy response of rats to a challenge dose of cocaine (7.5 mg/kg, i.p.). Continuous infusion of cocaine (80 mg/kg per day) via osmotic pumps for 7 days into the sensitized rats produced tolerance to the behavioral responses to the challenge dose of cocaine 1 day after the removal of the pumps. Therefore, tolerance can mask the expression of behavioral sensitization in rats. However, by 10 days after the removal of the pumps, the behavioral tolerance was reversed and the rats again displayed a sensitized response to cocaine. Therefore, the tolerance to cocaine was temporary while the underlying sensitization persisted. The development of tolerance did not alter the underlying sensitization demonstrating that these represent independent phenomena. The relationship between sensitization and tolerance observed in these studies may provide a model relevant to the progress in humans of addiction to psychomotor stimulants.

Mixed D2/5-HT2A antagonism of cocaine-induced facilitation of brain stimulation reward.

Previous behavioral, neurochemical and neurophysiological experiments have shown that selective 5-HT2A and mixed D2/5-HT2A antagonists can attenuate some, but not all, responses to amphetamine. The generality of these findings were determined in the present experiment by assessing the effect of mixed D2/5-HT2A antagonists on cocaine-induced facilitation of ventral tegmental area self-stimulation in rats. Although amphetamine and cocaine influence activity in monoaminergic neurons through different mechanisms, our previous research has shown that selective D2 and 5-HT2A antagonists have similar effects on behavioral responses to these psychostimulants. Therefore, we expected a similar pattern of results using mixed D2/5-HT2A antagonists. As shown previously, cocaine decreased self-stimulation threshold in a dose-dependent manner. Haloperidol and the mixed D2/5-HT2A antagonists risperidone and MDL 28, 133A antagonized cocaine-induced facilitation of self-stimulation, but only at doses that increased baseline self-stimulation threshold. There was a significant correlation (r = 0.87, p < 0.001) between antagonist-induced change in baseline threshold and attenuation of cocaine's effect on threshold. Taken together, the results of this and previous experiments support the importance of D2 receptors in the mechanisms of brain stimulation reward. 5-HT2A receptors appear not to be involved in mediation of both brain stimulation reward and amphetamine- and cocaine-induced facilitation of brain stimulation reward.

Tolerance to effects of high doses of ethanol: 1. Lethal effects in mice.

Male Swiss Webster mice were injected with ethanol doses ranging from 6.5-10.5 g/kg (20% w/v, IP). Survival time distribution revealed three waves of deaths with peaks around 5 min, 300 min, and 33 h. There were two windows with very low density of probability of death between 30-130 min and between 22-25 h following lethal injections. This time structure of the probability density function did not significantly depend upon ethanol overdose, novelty of the experimental environment, or prior injections of saline and/or 3.5 g/kg ethanol. Injections of high doses of ethanol in BALB/c mice showed that this strain of mice was more sensitive to ethanol-induced lethality (LD50 = 6.6 g/kg) and over 99% of deaths occurred between 5-200 min following injections of the doses from 5.5-7.5 g/kg. Preexposure to ethanol increased tolerance to ethanol-induced lethality. LD50 increased from 8.1 g/kg (at 24 h following lethal injections in ethanol-naive Swiss Webster mice) to 8.5 and 9.0 g/kg in mice following four and eight injections of 3.5 g/kg ethanol, respectively. In BALB/c mice, eight prior injections of 3.5 g/kg ethanol increased LD50 also slightly but significantly to 7.15 g/kg. The results suggest that: a) Ethanol-induced lethality is not a unitary phenomenon and that deaths that occurred within distinct waves probably have different causes; b) mice strains have different susceptibility to different causes of ethanol-induced deaths; c) preexposure to 3.5 g/kg ethanol results in significant but small increase in tolerance to ethanol-induced lethality.

Role of environmental cues as Pavlovian-conditioned stimuli in enhancement of tolerance to ethanol effects: 1. Lethal effects in mice and rats.

Twice daily for 4 days, Swiss Webster or BALB/c mice were injected with 3.5 g/kg ethanol (20% w/v, IP) immediately after moving their home cages from the colony room to the experimental room. On day 5, half the mice were moved to the same room and the other half to a novel room with different lighting, acoustic, and olfactory stimuli. All mice were injected with ethanol overdoses ranging from 4.5-10.0 g/kg. LD50 for ethanol increased following ethanol preexposure as compared to control ethanol-naive mice tested in the same experimental room. However, LD50 was lower in both Swiss Webster and BALB/c mice tested in a novel environment than in the familiar environment. Novelty increased sensitivity to the effect of low and moderate but not the highest lethal doses of ethanol. This effect of novelty occurred only in ethanol-experienced, but not ethanol-naive mice. In the following experiments, using a balanced design, Swiss Webster mice and Wistar rats were exposed to ethanol and saline alternatively in two distinct experimental rooms. On the final day, we found that there was no difference between animals tested in the room previously associated with administration of ethanol and animals tested in a saline-associated room in terms of LD50 for ethanol. These results suggest that: a) Environmental stimuli do not play a role as Pavlovian conditioning stimuli in the development of tolerance to ethanol-induced lethality; and b) novelty acts as an unconditioned stimulus that increases ethanol's lethal effects by unspecific disruption of conditioned compensatory responses to internal conditioned stimuli, such as irritation of peritoneal cavity, smell, and taste of ethanol.

REM-sleep deprivation, stress and emotional behavior in rats.

Eighty-eight adult white rats were divided into 9 groups. Groups 1 and 2 served as controls. The rats of Group 3 were repeatedly aroused during 4 days at the very onset of each REM-sleep period by direct midbrain reticular formation stimulation. This deprivation decreased the daily amount of REM-sleep by 70%, while slow-wave sleep was reduced by 10% only. In Group 4, the animals were given food and water for 1 h a day only. Groups 5 and 6 were subjected to immobilization and cold stress, respectively. Groups 7, 8 and 9 were deprived of REM-sleep on platforms of 15, 11 and 6.5 cm in diameter, respectively. Stress was estimated by the classical Selye's triad: weight of adrenals and thymus and gastric ulceration. Emotionality was measured in the open-field and also by self-stimulation of the lateral hypothalamus. Neither emotional behavior disturbances nor Selye's stress features were found after REM-deprivation in Group 3. Moreover, arousal deprivation induced a slight, though significant, reduction in adrenal weight. Also, no changes in emotional behavior were noted in stress-exposed groups (5 and 6). Only the interplay between REM-sleep deprivation and stress on the platforms (Groups 7, 8 and especially 9) led to a considerable shift in emotionality.