|Additional related information may be found at:|
|Neuropsychopharmacology: The Fifth Generation of Progress|
Intracellular Messenger Pathways as Mediators of Neural Plasticity
Intracellular Messenger Pathways as Mediators of Neural Plasticity
Eric J. Nestler and Ronald S. Duman
Previous Chapters have described the post-receptor, intracellular messenger pathways that mediate signal transduction in the brain (see Chapters by Aghajanian; Duman and Nestler; Piomelli; and Sanders-Bush and Canton, this volume). These pathways subserve three major aspects of signal transduction. First, they mediate certain short-term aspects of synaptic transmission: those rapid actions of neurotransmitters on ion channels that do not involve ligand-gated channels are achieved through intracellular messengers. Second, they play the central role in mediating other actions of synaptic transmission: virtually all other effects of neurotransmitters on target neuron functioning, both short-term and long-term, are achieved through intracellular messengers (see Duman and Nestler, this volume). This includes those long-term actions of neurotransmitters that are mediated through alterations in neuronal gene expression. This role for intracellular messengers is not limited to actions of neurotransmitters mediated via G-protein-linked receptors. Although activation of ligand-gated ion channels leads to initial changes in membrane potential independent of intracellular messengers, activation of ligand-gated ion channels also leads to numerous additional (albeit slower) effects that are mediated via intracellular messengers. Third, by virtue of numerous interactions among various intracellular messenger pathways, these pathways play the central role in coordinating a myriad of neuronal processes and adjusting neuronal function to environmental cues (24).
The short- and long-term modulatory effects that neurotransmitters exert on their target neurons via regulation of intracellular messenger pathways can be viewed as the basis of neural plasticity. Environmental factors of virtually every type produce long-term changes in brain function by influencing these processes. In this Chapter, we demonstrate how psychotropic drugs can serve as unique tools to study the mechanisms by which the brain adapts to chronic perturbations in its signaling pathways by environmental factors. This is because drugs provide bridges between preclinical and clinical situations. Indeed, because most types of psychotropic drugs must be given for several weeks before their therapeutic effects are apparent, and because drugs of abuse induce addiction gradually and progressively with continued drug exposure, the clinically relevant actions of most psychotropic drugs can be considered drug-induced neural plasticity. We present evidence that the brain's intracellular messenger pathways are themselves targets of long-term regulation and contribute prominently to drug-induced neural plasticity. In addition, the role of intracellular messengers as general mediators of neural plasticity is discussed.
GENERAL MODEL OF DRUG-INDUCED NEURAL PLASTICITY
A model of drug-induced neural plasticity is shown in Figure 1. According to this scheme, drugs produce acute and short-term changes in brain function by influencing the brain's signal transduction pathways. This includes drug regulation of the amount of neurotransmitter available at the synapse and of neurotransmitter activation of plasma membrane receptors and post-receptor signaling pathways (e.g., second messengers and protein phosphorylation). (These short-term processes are discussed in greater detail in Chapters by Duman and Nestler; and Sanders-Bush and Canton, this volume.) This is true regardless of whether the drug interacts initially with proteins located extracellularly or intracellularly, presynaptically or postsynaptically. While these transient and readily reversible effects are occurring in response to drug treatment, drug perturbation of signal transduction pathways are also initiating longer-term effects, which involve altered levels and altered types of proteins expressed in target neurons. These changes gradually build up over time in response to continued exposure to the drug and eventually become quantitatively significant so as to lead to long-term changes in brain function.
There are three general types of mechanisms by which a drug could alter levels of a protein: regulation of gene transcription, regulation of RNA translation and turnover, or regulation of protein turnover (Figure 1). Most attention to date has focused on drug regulation of gene expression. This attention is based on the many demonstrated cases of equivalent drug-induced changes in various proteins and their mRNAs, and of drug regulation of transcription factors (see 24). Transcription factors are proteins that bind to specific sequences of DNA contained within the regulatory regions of certain genes and thereby increase or decrease the rate at which those genes are transcribed (see Morgan and Curran, this volume). Most genes contain binding sites for multiple types of transcription factors, such that their transcriptional rates are probably determined by unique combinations of transcription factors that interact cooperatively.
Studies of drug regulation of gene expression to date have focused almost exclusively on two families of transcription factors: CREB (cAMP response element binding protein) and related proteins mediate many of the effects of cAMP and probably Ca2+ on gene expression (24,33). CREB's transcriptional activity is regulated primarily via its phosphorylation by cAMP-dependent and Ca2+-dependent protein kinases. Increasing evidence demonstrates that psychotropic drug treatments can regulate CREB function in the brain, presumably by influencing these intracellular pathways. Although CREB expression is believed to be constitutive and not subject to physiological regulation, recent evidence challenges this view (40).
c-Fos, c-Jun, and products of related immediate early genes (IEGs) are regulated in the brain by diverse types of stimuli, including numerous drug and other treatments (35; see also Morgan and Curran, this volume). Extracellular stimuli are thought to regulate these transcription factors primarily by regulating their expression, possibly mediated via the cAMP- or Ca2+-dependent phosphorylation of CREB or CREB-like proteins. However, Fos- and Jun-like proteins are also known to be phosphorylated by many protein kinases, and this serves to further regulate their transcriptional activity.
While there is little doubt that these transcription factors play an important role in mediating the effects of certain drugs and other extracellular signals on gene expression, there is also little doubt that the effects of a drug on gene expression are probably mediated through regulation of enumerable transcription factors, with different mechanisms operating in different neuronal cell types. Moreover, proteins that control the packaging of chromatin and the accessibility of certain genes to transcription factors are also highly regulated and possibly subject to drug effects (see 16).
Current emphasis on gene expression as a long-term target of drug action should not detract from the likely importance of post-transcriptional mechanisms of drug-induced neural plasticity. Indeed, increasing evidence indicates that RNA processing, transport to the cytoplasm, assembly into polysomes, stability, and rate of translation are also highly regulated in neurons (see 20,49). More attention should be given in future studies to drug regulation of these parameters. Similarly, more attention should be given to regulation of protein turnover, including rates (and possibly sites) of proteolysis, subcellular trafficking, and association with specific cellular organelles and macromolecular complexes. Changes in the processing and degradation of the b-adrenergic receptor induced by long-term exposure to agonist (see Duman and Nestler, this volume) emphasize the probable importance of post-translational regulatory mechanisms in drug effects on the nervous system. Finally, it should be emphasized that regulation of levels of a particular protein often involves all three of the processes shown in Figure 1, as the cell gradually adapts to drug exposure or other environmental inputs.
EXAMPLES OF DRUG-INDUCED NEURAL PLASTICITY
The complexity of these regulatory systems will make it exceedingly difficult to delineate the precise mechanisms underlying neural plasticity. However, studies of drugs offer the key advantage of investigating these regulatory mechanisms within a functional context by studying drug action in anatomically well-defined brain regions known from behavioral pharmacological studies to mediate important effects of the drugs. The goal of such studies is threefold. First, to identify molecular and cellular adaptations that drugs induce in those regions. Second, to relate the altered biochemical phenotype of cells in those regions to the altered electrophysiological phenotype of the cells and to the altered behavioral phenotype of the organism. Third, to elaborate the detailed molecular mechanisms by which the drugs produce the altered biochemical and electrophysiological phenotypes. One experimental system in which this approach has been used with considerable success is opiate action in the locus coeruleus (LC).
Studies of Opiate Addiction in the LC
The LC is the major noradrenergic nucleus in brain, located on the floor of the fourth ventricle in the rostral pons. A variety of pharmacological and behavioral studies have demonstrated that regulation of LC neuronal excitability plays an important role in physical aspects of opiate addiction, namely, physical dependence and withdrawal (see 27,37).
Acute opiate action. The electrophysiological effects of opiates in the LC are well-established (Figure 2). Acutely, opiates inhibit LC neurons via activation of an inward-rectifying K+ channel and inhibition of a Na+-dependent inward current (3,5,41). Both actions are mediated via pertussus toxin-sensitive G-proteins (i.e., subtypes of Gia and/or Goa). It is believed that activation of the K+ channel occurs through direct coupling of the opioid receptor to the channel via a G-protein. In contrast, inhibition of the Na+-dependent current appears to be indirect. The current is normally activated by cAMP-dependent protein kinase, through either the phosphorylation of the responsible channel or pump itself or some associated protein (4). Opiate inhibition of the current appears to be mediated via reduced levels of cAMP and of activated cAMP-dependent protein kinase. Biochemical studies have confirmed that opiates acutely inhibit adenylyl cyclase activity in the LC, as seen in many other brain regions, as well as inhibiting cAMP-dependent protein phosphorylation (37,40).
Chronic opiate action. Chronically, LC neurons develop tolerance to these acute inhibitory actions as neuronal firing rates recover toward control levels. The neurons also become dependent on opiates after chronic exposure in that abrupt cessation of opiate treatment, such as through administration of an opiate receptor antagonist, leads to a marked elevation in LC firing rates above control levels in vivo and in vitro (1,26,45).
Given the role of G-proteins and the cAMP pathway in the acute actions of opiates on the LC, it was investigated whether long-term adaptations in these intracellular messengers could be involved in the tolerance, dependence, and withdrawal that occurs with chronic opiate exposure. Indeed, over the past several years, chronic administration of opiates has been shown to increase, in the LC, levels of Gia and Goa, adenylyl cyclase, cAMP-dependent protein kinase, and several phosphoprotein substrates for the protein kinase, including tyrosine hydroxylase (TH), the rate-limiting enzyme in the biosynthesis of catecholamine neurotransmitters (Figure 2) (37,40). Up-regulation of the cAMP pathway occurs in the absence of alterations in several other major protein kinases (e.g., Ca2+/calmodulin-dependent protein kinases, protein kinase C, protein tyrosine kinases) in this brain region (53).
Direct evidence for a functional role of an up-regulated cAMP pathway in opiate addiction in the LC. Up-regulation of adenylyl cyclase and cAMP-dependent protein kinase can be viewed as a compensatory, homeostatic response of LC neurons to persistent opiate inhibition of the cells (Figure 2) (37). According to this view, up-regulation of the cAMP pathway increases the intrinsic excitability of LC neurons and thereby accounts, at least in part, for opiate tolerance, dependence, and withdrawal. In the opiate-dependent state, the combined presence of the opiate and the up-regulated cAMP pathway would return LC firing rates to control levels. Removal of the opiates would leave the up-regulated cAMP pathway unopposed, which would lead to withdrawal activation of the neurons. This scheme is similar to one proposed previously based on studies of cAMP levels in cultured neuroblastoma x glioma cells (48).
There is now direct evidence for a role of the up-regulated cAMP pathway in opiate dependence and withdrawal: 1) The spontaneous firing rate of LC neurons is dependent on the activity of the cAMP protein phosphorylation pathway and of the Na+-dependent inward current activated by this pathway (4); 2) The spontaneous firing rate of LC neurons from morphine-dependent rats, in brain slices where most synaptic connections of the neurons have been severed, is more than two fold greater compared to neurons from control rats, and shows a greater maximal excitatory response to cAMP analogues (26); 3) Excitation of LC neurons during withdrawal is both necessary and sufficient for producing many of the behavioral signs of physical opiate withdrawal (see 28,31,37,45); and 4) The time course by which the up-regulated cAMP pathway reverts to normal during opiate withdrawal parallels the time course by which withdrawal activation of LC neurons, and various behavioral signs of withdrawal, recover (45). Taken together, these findings indicate that up-regulation of the cAMP pathway is a likely mechanism of opiate dependence in the LC. While there may be many other mechanisms of opiate dependence in the LC and elsewhere, up-regulation of the cAMP pathway represents one of the few examples where a behavioral manifestation of neural plasticity (in this case physical opiate dependence) can be linked to electrophysiological and biochemical adaptations that occur in specific neurons.
The mechanisms underlying tolerance remain less certain, but probably overlap with those underlying dependence (Figure 3). The up-regulated cAMP system likely contributes to tolerance by making it more difficult for opiates to acutely inhibit the Na+-dependent inward current via inhibition of the cAMP system. It is also possible that the up-regulated cAMP system, through phosphorylation of the opioid receptor, could result in greater levels of receptor desensitization. This possibility is based on recent observations that brief exposure to met-enkephalin desensitizes the m-opioid receptor in the LC (2,19), and the evidence that agents that activate the cAMP pathway promote this desensitization (2). The up-regulated cAMP system in the tolerant state, by promoting desensitization, could lead to a reduced ability of opiates to activate G-proteins and subsequently regulate the K+ channel and Na+-dependent inward current regulated by opiates acutely. Chronic opiate-induced alterations in bARK's (b-adrenergic receptor kinases), or other protein components of this system (e.g., the b-arrestins), could also conceivably be involved in tolerance. This possibility is based on the role of bARKs and b-arrestins in mediating ligand-induced desensitization of other G-protein coupled receptors (see Duman and Nestler, this volume). This is believed to occur via the following scheme: ligand binding to the receptor renders it a good substrate for bARK; phosphorylation of the receptor then triggers its binding to b-arrestin, which inhibits coupling of the receptor to its G-protein. Consistent with a role for bARK in opioid tolerance is the recent finding that chronic morphine increases levels of one form of bARK in the LC (53), which might be expected to increase opioid receptor desensitization. Cloning and biochemical characerization of the m-opioid receptor expressed in LC neurons is needed to study these various possibilities directly.
Molecular mechanisms of opiate action in the LC. Regulation of gene expression may be involved in opiate up-regulation of the cAMP pathway in the LC. Up-regulation of the individual protein components of this pathway is associated with equivalent changes in their mRNAs. In addition, recent studies in transgenic mice have shown that chronic morphine increases the expression of the chloramphenicol acetyltransferase gene fused to 4.8 kb of 5' regulatory sequence of the TH gene (see 40).
As a first step in understanding how opiates might alter the genetic expression of these proteins, a systematic investigation of opiate regulation of transcription factors in the LC has been undertaken in recent years. To date, short- and long-term opiate exposure, and precipitation of opiate withdrawal, have been shown to influence: 1) CREB phosphorylation and the DNA-binding activity of this and other cAMP-regulated transcription factors (e.g., CCAATT-enhancer binding proteins); and 2) expression and DNA-binding activity of Fos and Jun-like proteins (see 40).
While still relatively preliminary and phenomenological, these studies highlight the utility of the LC as a model system in which to study the mechanisms underlying molecular plasticity to psychotropic drugs. This is a system in which altered expression of specific target genes has been shown to have physiologically important consequences. The next step in these studies is to relate regulation of a specific transcription factor to altered expression of a specific target gene and to a functional effect in LC neurons. Such studies, while very difficult methodologically, will gradually delineate the precise steps by which long-term exposure to opiates induces addiction in these neurons.
Studies of Other Psychotropic Drugs
Studies of chronic opiate action in the LC have the unique advantage that the behavioral consequences of such action, e.g., physical opiate dependence and withdrawal, can be readily and accurately quantified in laboratory animals. This is much more complicated for most other psychotropic drugs, where the target behavior (e.g., anxiolysis, mood elevation, reduction in psychosis) cannot be studied in animals in as straightforward a manner. Nevertheless, recent studies of neural plasticity induced by some other psychotropic drugs have been promising.
Drug reinforcement. The rewarding properties of drugs of abuse, thought to be a core feature of their addictiveness, are also amenable to detailed molecular investigations. This is because it is possible to quantify aspects of drug reward in laboratory animals by use of various experimental procedures, such as drug self-administration and conditioned place preference (see 27). This makes it possible to identify brain areas that mediate drug reward and to study the functional relevance of molecular events that occur in those areas in response to chronic exposure to drugs of abuse. Most attention to date has been given to dopaminergic neurons in the ventral tegmental area and their various projection regions, such as the nucleus accumbens and medial prefrontal cortex. Recent studies have provided increasing evidence that adaptations in G-proteins and other intracellular messenger proteins contribute to the long-term effects of drugs of abuse in these brain reward pathways. This work has been reviewed recently and is not presented in detail here (see 27,37,40,47,52).
Antidepressant drugs. Until relatively recently, studies of antidepressant drugs have focused almost exclusively on drug regulation of the metabolism of monoamine neurotransmitters and of binding sites of monoamine receptors in various brain regions. More recent studies have extended this work by attempting to identify the molecular basis of drug-induced changes in monoamine metabolism and receptor binding, as well as characterization of other molecular actions of antidepressant treatments.
Antidepressant regulation of catecholamine levels has been related to long-term changes in the phosphorylation and expression of TH in the LC and other brain regions (see 29,39). Decreased expression of TH in the LC by chronic antidepressant treatment is accompanied by down-regulation of cAMP-dependent protein kinase and decreased firing rates of LC neurons (see 32). Antidepressant regulation of TH and cAMP-dependent protein kinase could represent a compensatory response to antidepressant-induced augmentation of noradrenergic function in projection areas such as cerebral cortex.
Antidepressant treatments are also known to regulate b-adrenergic and 5HT2-serotonin receptor binding sites in cerebral cortex and other brain regions, but the mechanisms underlying regulation of these receptors have remained elusive. Recent studies demonstrate that regulation of receptor binding sites is associated with regulation of receptor mRNA levels (8,23). While the regulation of receptor mRNA is sometimes associated with an equivalent change in levels of binding sites, in other cases regulation of receptor mRNA is opposite to that observed for receptor binding. This suggests that either the turnover of receptor mRNA and/or protein is also influenced by these treatments (see 23). The complex mechanisms by which antidepressants likely regulate the b-adrenergic and other receptors is highlighted by the many known ways in which the receptor is regulated by long-term exposure to agonist in cultured cells, with alterations seen in receptor affinity for ligand, receptor coupling to G-proteins, receptor sequestration from the plasma membrane, receptor degradation, and receptor synthesis (see Duman and Nestler, this volume). Antidepressant regulation of these various processes must now be investigated.
The mechanisms by which antidepressant treatments regulate mRNA expression of the b-adrenergic and 5-HT2 receptors in brain have not been identified. However, studies in cultured cells indicate that agonist regulation of b-adrenergic receptor mRNA is mediated by the cAMP system (12,18,22). In vivo studies have shown that chronic antidepressant treatments induce an apparent nuclear translocation of cAMP-dependent protein kinase (38), as well as a change in its enzymatic activity (44), in cerebral cortex. Acute and long-term antidepressant treatment has also been reported to influence Fos and related IEG transcription factors in this and other brain regions (15,21,51,54). Antidepressant regulation of the protein kinase and of these and other transcription factors could conceivably be related to altered expression of the b-adrenergic receptor.
In addition to changes in the expression of the bAR, antidepressant-induced nuclear translocation of the protein kinase and regulation of transcription factors would be expected to result in altered expression of many additional target genes in the brain, which would underlie many of the long-term consequences of antidepressant exposure. Such target genes are only now beginning to be identified. Chronic antidepressant administration has been shown to alter the expression and functional activity of specific G-protein subunits and adenylyl cyclase in brain (11,30,42), changes that would further influence b-adrenergic and other receptor-mediated signal transduction. Antidepressants have been reported to alter the expression of glucocorticoid receptor in cultured cells in vitro (43). If replicated in vivo, this action could contribute to the ability of antidepressants to influence the brain's responses to stress. Preliminary studies demonstrate that antidepressants increase the expression of neurotrophin mRNA in limbic brain regions (36). Increased expression of neurotrophins could contribute to antidepressant-induced plasticity in the brain and to some of the drugs' long-term actions. Future studies will undoubtedly identify many additional target genes of antidepressant treatments.
One critical obstacle in studies of antidepressant drugs is that no specific target brain area has been identified as the major substrate of antidepressant action. This results, in part, from the fact that it is exceedingly difficult to assay clinically relevant effects of the drugs in laboratory animals. This limitation also makes it difficult to study the functional importance of drug-induced adaptations in the brain's signal transduction pathways at the behavioral level. However, antidepressant regulation of the expression of specific genes provides functional endpoints by which drug-induced effects in the brain can be assessed for the first time at the molecular and cellular levels.
GENERAL MECHANISMS OF NEURAL PLASTICITY
Mechanisms underlying drug-induced neural plasticity may be representative of general ways in which neural systems adapt to physiological and behavioral stimuli. Adaptations that drugs and other stimuli induce in the brain can often be classified into two broad categories: negative or positive feedback mechanisms.
Negative feedback can be defined as up- or down-regulation of a system, which compensates, respectively, for decreased or increased stimulation of the system. One example is the up-regulation of the cAMP pathway in the LC in response to chronic opiate exposure, as outlined above. Another example is agonist-induced down-regulation of the b-adrenergic receptor, which involves agonist regulation of the receptor at numerous levels, as mentioned above and covered in detail in Duman and Nestler, this volume. It is likely that similar mechanisms operate for other neurotransmitter receptor systems.
Positive feedback can be defined as enhanced responsiveness of a system to the same stimulus. One example is long-term potentiation, characterized by enhanced synaptic efficacy (postsynaptic potentials) elicited by high frequency stimulation of a presynaptic pathway. Another example is locomotor sensitization (increased locomotor activation) observed in response to repeated administration of cocaine or other stimulants. The mechanisms underlying these sensitization phenomena have not yet been established with certainty, but appear to involve presynaptic and postsynaptic adaptations: increased presynaptic release of the stimulating neurotransmitter as well as increased responsiveness of the receptor-intracellular signal transduction pathways for the stimulating neurotransmitter (7,25).
In addition to negative and positive feedback processes, certain chronic perturbations can result in different types of responses to the original stimulus, not just up- or down-regulation. In other words, chronic peturbation could produce qualitative, as well as quantitative, changes in the brain's signaling pathways. One example is the ability of chronic electroconvulsive seizures to alter the types (not just the amounts) of Fos-like proteins expressed in the brain (21). Altered Fos-like proteins in the acute versus chronic state would be expected to mediate different types of effects of a seizure on neural gene expression.
A critical need is to identify the factors that determine whether a negative or positive feedback response, or different response, will occur following a particular perturbation. It is likely that this depends, not only on the stimulus, but also on the specific receptor systems, intracellular messenger pathways, and nuclear regulatory mechanisms active in a cell at a given point in time. Such adaptations are the subject of tremendous interest since these types of mechanisms at the molecular and cellular level presumably accumulate and interact to form the basis of complex forms of learning and memory at the behavioral level.
Such processes of neural plasiticity are in constant operation as the brain receives external stimuli via neuronal pathways as well as endocrine and cytokine substances in the circulation. These inputs influence multiple signal transduction pathways, which ultimately control the function and activity of the neural pathways which make up the brain. Other than drug-induced neural plasticity, the ways in which these processes could operate in response to environmental changes have been best studied for stress.
Stress-Induced Neural plasticity
Numerous studies have documented the effects of short- and long-term exposure to stress on the brain. Particular attention has been given to the hypothalamic-pituitary-adrenal axis which controls glucocorticoid secretion and to each of the major monoamine neurotransmitter systems in brain. Many neurotransmitter and neuropeptide systems in the brain are known to be influenced by acute and chronic stress, and more recent studies have begun to extend this work by examining stress-induced adaptations in intracellular messenger pathways.
The influence of stress on the brain's catecholamine systems is one of the best-studied and is discussed here for illustrative purposes. Similar types of adaptations to stress probably occur in many other brain neurotransmitter systems. Acute stress leads to activation of LC neurons, in part via activation of corticotrophin releasing hormone (CRH) pathways that innervate the LC (possibly from the paraventricular nucleus of the hypothalamus). Chronic stress leads to a sustained activation of LC neurons and to increases in the expression of TH, adenylyl cyclase, and cAMP-dependent protein kinase (33,46). Up-regulation of the cAMP pathway may contribute to the increased firing of these neurons in response to chronic stress via similar mechanisms as discussed above for opiate addiction. These biochemical adaptations to stress, which can be viewed as a positive feedback mechanism, could result from stress-induced activation of LC neuronal firing and the increased demand for norepinephrine. Increased firing could also be driven by a combination of sustained positive inputs, such as CRH activation, and loss of negative inputs, such as a2-adrenergic autoreceptor feedback inhibition, both of which would contribute to increased activation in the cAMP system.
Chronic stress also leads to adaptations of norepinephrine-stimulated cAMP formation and related signal transduction pathways in cerebral cortex. This effect is mediated in part by sustained elevation of adrenal glucocorticoids in response to stress (14,34,50). Norepinephrine-stimulated cAMP formation in cerebral cortical brain slices is mediated by b-adrenergic receptors, which directly couple to adenylyl cyclase, and a1-adrenergic receptors, which alone have little effect but enhance direct-acting agonists. Chronic stress decreases norepinephrine activation of cAMP formation by decreasing the a1-adrenergic enhancement of b-adrenergic receptor action. The exact mechanism for a1-adrenergic receptor enhancement is not known, but is dependent on extracellular Ca2+ and may involve activation of protein kinase C or release of free G protein bg-subunits which can activate certain forms of adenylate cyclase (see Duman and Nestler, this volume).
It is likely that chronic stress alters the expression of specific genes in the brain. Altered levels of TH, AMP-dependent protein kinase, and receptor function could be mediated, at least in part, at the transcriptional level. Indeed, acute stress has been shown by several groups to induce the expression of c-fos and related IEG transcription factors in specific brain regions, including the LC, the mesolimbic dopamine system, amygdala, hippocampus, and certain layers of cerebral cortex (6,9,10,13,15,17). Interestingly, repeated application of the stressful stimulus leads to down-regulation of the c-fos response, whereas exposing an animal to a neutral stimulus previously associated with stress leads to IEG induction (9). Stress regulation of these and other transcription factors, which could mediate some of the long-term effects of stress on brain function, provides novel tools with which the detailed effects of stress on the nervous system can now be investigated.