Modulation of the Induction or Expression of Psychostimulant Sensitization by the Circumstances Surrounding Drug Administration

Abstract

ROBINSON, T.E., K.E. BROWMAN, H.S. CROMBAG, A. BADIANI. Modulation of the induction or expression of pyschostimulant sensitization by the circumstances surrounding drug administration. NEUROSCI BIOBEHAV REV 22(2), 347–354, 1998. The conditions necessary to induce psychomotor sensitization and to promote its expression are not well understood. Two examples are reviewed here of how the circumstances surrounding drug administration (“set and setting”) can powerfully modulate the sensitization produced by psychostimulant drugs, such as amphetamine or cocaine. In the first example it is suggested that repeated exposure to psychostimulant drugs may induce “neural sensitization” (i.e., produce relevant adaptations in the nervous system). The circumstances surrounding drug administration may determine, however, whether neural sensitization is expressed in behavior. In the second example it is suggested that the circumstances surrounding drug administration may determine whether sensitization is induced at all, or at least the rate and extent of sensitization produced by a given dose of a drug. It is concluded that psychomotor sensitization is not an inevitable consequence of exposure to psychostimulant drugs, but is the result of interactions amongst the pharmacological actions of drugs and the circumstances surrounding drug administration.

Introduction

When psychostimulant drugs, such as amphetamine or cocaine, are administered repeatedly their effects often change. Some effects may decrease (show tolerance) while at the same time other effects may increase (show sensitization) 34, 35, 39. The effects of psychostimulants that increase, which are the topic of this paper, include their psychomotor activating effects and their rewarding effects. For example, the repeated intermittent administration of amphetamine or cocaine results in a progressive increase in their ability to produce locomotor hyperactivity, stereotyped behaviors or rotational behavior 34, 39, and can facilitate the later acquisition of a conditioned place preference or a drug self-administration habit 21, 22, 26, 31. It is especially interesting that once sensitization has been induced animals may remain hypersensitive to the psychomotor activating or rewarding effects of subsequent drug exposure for very long periods of time: in some instances for months to years 29, 35, 46.

The acute neuropharmacological actions of amphetamine and cocaine have been relatively well characterized. These drugs are thought to produce most of their actions by increasing the synaptic concentrations of monoamines via an interaction with monoamine transporters 9, 16. In particular, their ability to produce psychomotor activation and reward are thought to be primarily due to their ability to increase dopamine (DA) neurotransmission. This has led to the idea that their ability to induce adaptations in these same systems may be responsible for the development of behavioral sensitization. Consistent with this idea, sensitization-related adaptations have been reported in a number of neural regions thought to be involved in psychomotor activation and reward, including the mesolimbic DA system and related circuitry in the ventral striatum/nucleus accumbens 23, 34, 49. A review of this literature is beyond the scope of the present paper, but suffice it to say that both presynaptic and postsynaptic adaptations have been described in catecholamine systems in animals that have been sensitized to amphetamine or cocaine (see 23, 34, 49, for reviews). For example, presynaptic adaptations include a persistent increase in the ability of amphetamine or cocaine to enhance the overflow of DA in the caudate and nucleus accumbens, and postsynaptic adaptations include a sensitization-related increase in the responsiveness of DA D1 receptors in the nucleus accumbens. Nevertheless, the neural basis of sensitization is still not well understood and is the topic of considerable research and debate.

Sensitization of DA/accumbal systems has attracted much of the attention in this field, in part because of the well-known role of these neurons in mediating the rewarding effects of drugs of abuse and of natural rewards 25, 51. The fact that drugs of abuse can produce long-lasting alterations in a neural system involved in reward suggests that “neural sensitization” may play a role in the process of addiction. Indeed, given that repeated exposure to drugs renders DA/accumbal systems persistently hypersensitive to activating stimuli, including drugs themselves, it is difficult to think that such neuroadaptations would not alter the process of reward, and thus in some way contribute to addiction. It remains to be seen, however, how sensitization of brain reward systems might contribute to addiction, although this has been the subject of some speculation and discussion 7, 35.

Despite considerable research on both the behavioral pharmacology and neurobiology of sensitization, the exact conditions that lead to sensitization and promote its expression have not been well characterized, and there is considerable debate in the field regarding the fundamental nature of the phenomenon [1]. For example, it is often tempting to think of sensitization as an inevitable consequence of the pharmacological actions of a drug on a sensitive neural substrate; that is, when a neural substrate is repeatedly exposed to a drug (or other ligand) the substrate adapts. It is well established, however, that the behavioral effects of drugs can be powerfully modulated by the circumstances surrounding drug administration. Indeed, there are large individual differences in the acute subjective effects of addictive drugs, and much of this variability is due to the context in which the drug is administered. The importance of environmental factors in modulating drug effects has been emphasized by many investigators 5, 15, 24.

The thesis explored in this paper is that the circumstances surrounding drug administration not only modulate acute drug effects, but also modulate the ability of drugs to change the nervous system and, in particular, to induce or express sensitization. Thus, in the remainder of this paper we will consider two examples of how the circumstances surrounding drug administration can alter the induction or expression of sensitization. In the first example it is suggested that when drug administration induces neural sensitization the circumstances surrounding drug administration and readministration can determine whether neural sensitization is expressed in behavior. In the second example it is suggested that the circumstances surrounding drug administration can modulate the induction of sensitization, that is, whether neural sensitization is induced at all. We believe these examples demonstrate that sensitization is not an inevitable consequence of the pharmacological actions of psychostimulant drugs, but is the result of interactions amongst the pharmacological effects of drugs and the circumstances surrounding drug administration [15].

There are many reports that the expression of sensitization can be context-specific 30, 41. In such studies typically the animals in one group are transported from their home cage to a unique test environment, where they receive drug treatments. This group is usually called the Paired group because drug treatments are paired with exposure to the test environment (these animals usually also receive an injection of saline in their home cage). Animals in a second group (the so-called Unpaired group) receive saline in the test environment and the drug later, in their home cage. Control animals receive saline in both the test environment and at home. The influence of drug treatment (and environment) is assessed on a test day, when all animals receive a challenge injection of the drug in the test environment. In many studies of this kind, sensitization is expressed in the Paired group, but not the Unpaired group (i.e., the Paired group shows a greater response to a drug challenge in the test environment than the Saline-pretreated control group, but the Unpaired group does not). Thus, sensitization is said to be context-specific because it is expressed only in the group that previously experienced the drug in the test environment. It is not necessary, however, to explicitly “unpair” drug treatments with exposure to the test environment by giving saline injections in that environment to observe context-specific sensitization. For example, Fig. 1 shows that if a group of animals is transported each day from their home cage to a “third world” (an environment that is distinct from both the home cage and the test environment), where they receive daily injections of amphetamine, they also later fail to express sensitization when given a challenge injection in the test environment [1]. This suggests that, at least under some conditions, distinct pairing between a particular environment and drug administration may be important to the behavioral expression of sensitization.

The data shown in Fig. 1 illustrate the powerful control that contextual stimuli can exert over the expression of sensitization (see 30, 41for reviews). There are two reasons to believe, however, that even though animals may not express behavioral sensitization they may undergo neural sensitization (i.e., exposure to the drug may induce relevant neuroadaptations). First, animals that receive drug administration in an environment other than the test environment may develop behavioral sensitization in their drug treatment environment [1]. Fig. 2 shows, for example, that the “third world” group discussed in Fig. 1 developed robust behavioral sensitization when observed in their drug treatment environment (i.e., in the “third world”). They just did not express sensitization when tested in a new environment, the usual test environment, as illustrated in Fig. 1. This suggests that the neuroadaptations responsible for behavioral sensitization were induced in these animals, although “neural sensitization” was not expressed behaviorally when later the animals were given a drug challenge in the test environment. Second, examples of neural sensitization have been described under conditions that preclude the influence of contextual stimuli on the neurobiological expression of the drug response, for example, in striatal tissue slices in vitro and in anaesthetized animals 17, 33.

In summary, studies on context-specific sensitization indicate that the expression of sensitization is not simply the result of unconditional neurobiological adaptations produced by the actions of drugs on a neural substrate. This is because the major determinant of whether sensitization is expressed at any particular place or time is the context in which the drug is readministered. We know, however, very little about how contextual stimuli and other circumstances surrounding drug administration and readministration modulate the expression of sensitization. There are a number of possibilities, and the reader is referred to Anagnostaras and Robinson [1]for a discussion of some of these.

In the previous section we presented an example of how the circumstances surrounding drug administration can determine whether neural sensitization is expressed in behavior. In this section we will present examples of how the circumstances surrounding drug administration may determine whether neural sensitization is induced in the first place, or at least the rate and extent of sensitization induced by a particular dose of a drug.

The first study to report that an environmental manipulation alters the induction of sensitization that we are aware of was by Badiani et al. [2]. In these experiments rats received repeated i.p. injections of amphetamine or cocaine [3]under one of two conditions. One group was transported each day from distinct cages in which they lived to test cages where they then received drug treatments. Another group lived in the test cages, where they received the drug. Thus, both groups received drug treatments in physically identical environments, but for one group this was “home” and for the other group it was a relatively “novel” environment (i.e., it was explicitly paired with drug administration). In these experiments the rate of psychomotor sensitization, as indicated by changes in either locomotor activity [2]or rotational behavior 2, 3(Fig. 3A), was greater when amphetamine was given in the “novel” situation, relative to the “home” situation. Also, when given a drug challenge after a period of withdrawal the magnitude of sensitization was greater in animals pretreated in the “novel” condition (Fig. 3B). Finally, the influence of environment on the magnitude of sensitization was also evident by a greater shift to the left in the dose-effect curve for psychomotor activation in animals pretreated in the “novel” condition (Fig. 4).

As mentioned above, these were the first studies we are aware of to demonstrate an influence of environment on the induction of sensitization, when the physical characteristics of the environments were held constant. It is surprising, therefore, that it has been suggested previously that psychostimulant sensitization is enhanced if animals receive drug injections in a novel environment, relative to a home environment. For example, Willner et al. [50](p. 43) state: “sensitization is greater in animals that experience the drug in the testing environment than in animals that experience the drug in their home cage 13, 14, 18, 27, 28, 32, 38, 40, 42, 45, 47”. Einat and Szechman [12](p. 599) state: “psychomotor stimulants produce greater enhancement of locomotion when administered repeatedly in the test environment (activity monitors) than when administered in the home cage 10, 20, 32, 41, 48”. These statements are surprising because, although Willner et al. [50]cite 11 studies and Einat and Szechman [12]cite five studies, not a single one of these studies actually compared the influence of home vs. novel environments on the development of sensitization. The effect of environment in all of these studies is actually an example of contextual control over the expression of sensitization, as discussed above.

It needs to be emphasized that in all the previous studies demonstrating context-specific sensitization the rate and extent of sensitization in the home environment was not assessed (the animals were observed in a test environment after drug pretreatment, not at home during drug pretreatment). This design does not allow one to make any inferences about the relative rate or extent of sensitization in “home” vs. “novel” environments. It only allows one to make inferences about the influence of contextual cues previously associated (or not associated) with drug administration on the subsequent expression of sensitization. That is, the studies cited by Willner et al. [50]and Einat and Szechman [12]represent examples of how the circumstances surrounding drug administration and readministration modulate the expression of sensitization. They do not provide any information about whether the circumstances surrounding drug administration also modulate the induction of sensitization. We think the distinction between modulation of induction versus modulation of expression of sensitization is an important one because there is considerable evidence to suggest that the neurobiological mechanisms involved in the induction of sensitization are different from those involved in its expression [23]. Thus, the mechanisms by which the circumstances surrounding drug administration modulate these two processes could be quite different as well.

It is worth noting that there have been a few studies in which a drug response was assessed in both a home and a novel environment, but in these studies the two environments were physically different (e.g., 12, 36). This precludes any direct comparison of either the acute response to the drug, or the development of sensitization, because the physical characteristics of an environment have a large effect on the quantity and quality of behaviors expressed in response to a drug 6, 12, 13, 19, 36, 37, 43, 44, 50. Different physical features of an environment can elicit different behaviors. Thus, our recent studies 2, 3, 4are the first we are aware of to directly assess the influence of environment (“home” vs. “novel”) on the induction of sensitization, when the physical characteristics of the two environments were held constant.

Although the two environments used in the experiments by Badiani et al. 2, 3were physically identical, they did differ in a number of ways that could contribute to their effects on susceptibility to sensitization. One major difference was that in the “novel” condition there were many salient cues that reliably predicted drug administration (i.e., the entire context). When animals received drug injections at “home”, however, context did not reliably predict drug administration, because the drug was absent for most of the time spent in that context. Nevertheless, in the “home” condition there were a number of other cues that reliably predicted drug administration, including the appearance of an experimenter, handling and a needle prick, as well as the interoceptive cues produced by the drug itself. We hypothesized, therefore, that to the extent environmental cues predictive of drug administration are important for the induction of sensitization, eliminating them should further attenuate sensitization.

We tested this hypothesis recently by preparing rats with a chronically-implanted intravenous (i.v.) catheter [11]. The animals in one group lived in the test chambers, and their catheter was connected via a liquid swivel to a syringe mounted on a syringe pump. The syringe pump could be activated remotely to give an i.v. infusion of amphetamine (or saline) in the absence of any environmental cues predictive of drug administration (no handling, no needle prick, no experimenter). Of course, this procedure does not eliminate the interoceptive cues produced by the drug itself. In contrast, the animals in a second group were transported from the colony room to the test cages, where they received an i.v. infusion of amphetamine or saline immediately after placement in the test chamber. Again, for the two groups the drug treatment environments were physically identical, but for one group drug administration was unsignalled (because the animals lived in the test chamber) and for the other group drug administration was signalled by placement into a novel context [11].

The results of this experiment are shown in Fig. 5. The major findings were, (a) there were large group differences in the acute response to amphetamine. The group that received a signalled infusion of amphetamine showed a much greater psychomotor response to the first infusion than the group that received an unsignalled infusion. This difference was not due to the behavioral activating effects of simple exposure to a “novel” environment, because an injection of saline in the “novel” environment produced negligible psychomotor activation using this measure, as we have reported with i.p. injections 2, 3. (b) There were marked group differences in the development of sensitization. In animals that received signalled drug administration there was a significant increase in drug response (sensitization) between the first and last test session, but there was no change in drug response between the first and last test session in animals that received unsignalled drug administration (Fig. 5). (c) When all animals (both saline- and amphetamine-pretreated) were given a challenge injection of amphetamine after 5 days of withdrawal there were also marked group differences in the expression of sensitization (Fig. 5). Animals that received signalled pretreatment with amphetamine were hypersensitive to an amphetamine challenge relative to saline-pretreated control animals. Animals that received unsignalled amphetamine pretreatment, however, did not differ significantly from saline-pretreated controls. Indeed, on the challenge test day the behavioral response to an amphetamine challenge varied up to 23-fold, even though all four groups received the same dose of amphetamine (0.5 or 1 mg/kg) in the same physical environment. These groups differed only in whether they had any previous experience with amphetamine, and where they had that experience. In summary, this experiment demonstrates that the circumstances surrounding drug administration, in this case whether it is signalled or unsignalled, determines whether a given dose of amphetamine is capable of inducing sensitization.

Indeed, the results illustrated in Fig. 5 [11]lead one to question whether it is possible to induce sensitization with unsignalled infusions of psychostimulants given in the environment where animals live. This would be quite remarkable, and would require rethinking many assumptions regarding the nature of the sensitization phenomenon. It seems more likely that the effect of environment is to shift the dose-effect curve for inducing sensitization, and that unsignalled drug administration is just less effective. We recently tested this hypothesis using cocaine [8]. The initial results from this experiment suggest that the latter is true. That is, the effect of environment (signalled versus unsignalled cocaine administration) was to alter the dose necessary to induce sensitization (Fig. 6). Sensitization was induced at lower doses when drug administration was signalled by placement into the test chamber than if drug administration was unsignalled. If a high enough dose was used, however, it induced sensitization even when drug administration was unsignalled (Fig. 6).

As in the case of contextual control over the expression of sensitization, it is not known how environmental and psychological factors gain access to the neural substrate that is sensitized by psychostimulant drugs to then modulate the induction of sensitization. Indeed, it is not even known what the critical variables are. We have speculated that exposure to a relatively “novel” environment may promote sensitization either because of some action as a stressor, or because it facilitates associative learning, but these hypotheses remain to be tested 2, 3. Whatever the mechanism, to fully understand the phenomenon of sensitization (and other drug-induced neuroadaptations) it will be important to determine how seemingly simple manipulations of the circumstances surrounding drug administration can have such profound effects on the ability of drugs to change the nervous system.

In humans the experience associated with the administration of drugs of abuse varies enormously. This variation exists because a given drug effect is not a simple consequence of the pharmacological actions of a drug, but is due to complex interactions amongst pharmacological, environmental and psychological factors; for example, whether drugs are taken in a drug-associated environment, what expectations the person brings to the situation, etc. [15]. It is suggested here that similar complex interactions determine the probability that psychostimulant drugs will produce adaptations in neural systems involved in psychomotor activation and reward, and in this way may contribute to susceptibility to sensitization and addiction. The two examples discussed above of how the circumstances surrounding drug administration can modulate the expression or induction of psychostimulant sensitization may provide, therefore, a powerful animal model to study the nature of the environmental and psychological factors that influence the ability of drugs to change the nervous system, as well as a way to begin to explore the neurobiological mechanisms by which they act.