Modification of Central Catecholaminergic Systems by Stress and Injury

Elizabeth D. Abercrombie and Michael J. Zigmond

Central neuronal systems utilizing the catecholamines norepinephrine (NE) and dopamine (DA) play an important role in hypotheses concerning the biological bases of many neurological and psychiatric disorders, as well as the mechanisms of action of drugs used in their treatment. This has helped to stimulate much of the basic research focused on better understanding the roles of these transmitters and to promote the development of specific pharmacological tools for the manipulation of catecholaminergic function. Because the details of recent advances in our understanding of the physiological, pharmacological, and behavioral characteristics of the major central catecholaminergic systems are presented in other chapters of this volume, only brief comments on these topics are included here. The primary goal of this chapter is to review the nature of cellular responses brought about in central catecholaminergic neurons in response to conditions of chronically increased demand and to consider the functional significance of such changes.

It is not surprising that catecholamines have been implicated in many cellular and behavioral phenomena (see Mesocorticolimbic Dopaminergic Neurons: Functional and Regulatory Roles, Pharmacology and Physiology of Central Noradrenergic Systems, Modification of Central Catecholaminergic Systems by Stress and Injury: Functional Significance and Clinical Implications, and Physiological and Anatomical Determinants of Locus Coeruleus Discharge: Behavior and Clinical Implications). In the mammal, the neuronal systems that utilize NE or DA as their transmitter comprise widespread networks, usually innervating multiple target structures. Consistent with this diffuse anatomy are the relations between the activity of several catecholamine systems and arousal. Most prominent is the relation between the firing rate of NE neurons of the locus coeruleus (LC) and the sleep–wake cycle, with LC active during the waking state and relatively inactive or even silent during stages of sleep. Furthermore, LC neurons emit a burst of activity when the organism is presented with a novel environmental stimulus and show further tonic elevations in firing rate during exposure of the organism to an environmental stressor (1, 2, 3, 4). The tonic level of activity of the DA cells in the substantia nigra (SN) also varies between wakefulness and sleep, although far less than does activity in LC (5). Evoked responses are obtained from DA cells in response to phasic sensory stimuli, and, similar to the case for LC neurons, these responses lessen in magnitude with decreasing arousal level and increase in magnitude if the stimulus has acquired behavioral significance (6, 7, 8). Postmortem biochemical determinations as well as measurements of extracellular levels of NE and DA obtained using in vivo microdialysis are consistent with the electrophysiological data in demonstrating arousal-related alterations in catecholamine efflux at a number of forebrain sites (9, 10, 11, 12, 13).

Postsynaptically, NE and DA appear to play key roles in modulating cellular excitability at their diverse targets, usually via relatively slow second-messenger-mediated mechanisms. In the case of NE, this modulation typically is manifest as an increase in evoked responding of the target neuron relative to the spontaneous level of activity, whereas postsynaptic modulation by DA apparently can be either facilitatory or inhibitory depending on the momentary state of the target cell membrane (14, 15, 16, 17). Specificity of signaling in such systems can be achieved by the simultaneous activation of the diffuse, relatively slow catecholaminergic inputs along with the pathways that rapidly transmit primary information (18).

Based on the general attributes described above, it would seem that the central catecholaminergic systems are best viewed as important in determining the overall state of arousal of the organism, the concomitant efficiency of ongoing sensorimotor processing, and, ultimately, the ability of the organism to react flexibly and appropriately to continuously changing environmental demands. In this regard they play a role in the broad task of maintaining homeostasis, as do the catecholaminergic cells in the periphery. This view should be contrasted with the types of explanatory models that emerged during the 1960s and the 1970s, positing a primary role for one or another catecholamine system in subserving specific behavioral processes.

The concept of homeostasis is relevant not only to what catecholamines do but also to how they do it. Catecholamine neurons themselves exhibit a wide range of homeostatic properties. For example, when these neurons are exposed to conditions of increased demand, cellular changes occur that enhance their capacity to synthesize and release transmitter (19). Two examples are given in this chapter to illustrate this phenomenon: (i) the alterations in the NE-containing neurons of the LC that take place in response to conditions of chronic stress and (ii) the alterations in the DA-containing neurons of the SN that occur following partial injury to that system.

A wide variety of acutely presented stressful stimuli, including footshock, cold environment, and immobilization, have been shown to affect indices of activity in NE-containing neurons originating in the LC (20, 21). Among the changes that occur as a result of acute stress are increases in the tonic firing rate of NE neurons in LC (1), decreases in brain NE content (22, 23), increases in NE turnover (24, 25), and increases in the extracellular level of NE (9, 26). Following chronic exposure to stress, the reductions in brain NE levels that occur after acute stress no longer are observed (27, 28). Indeed, brain NE levels may actually be increased in response to chronic stress exposure (23, 25, 29, 30, 31). Based on such results, it has been hypothesized that chronic stress leads to a compensatory increase in the biosynthesis of NE which then permits a sustained increase in NE release such that NE content does not decline and may even increase.

It is well established that the activity of tyrosine hydroxylase (TH) is regulated in response to the demand for catecholamines. Electrical stimulation of both central and peripheral catecholamine neurons results in an apparent activation of TH which persists following the termination of the stimulation period (32, 33). Furthermore, an activation of TH occurs in response to other stimuli known to increase the activity of central NE neurons, including administration of catecholamine-depleting drugs, partial injury to the LC system, and various forms of stress (33, 34, 35, 36). In addition to these increases in TH activity, which represent immediate adaptation to increased transmitter utilization, a second process may come into play after prolonged increases in the activity of NE neurons. This latter mechanism is reflected by an apparent increase in the maximal velocity of TH, which appears to be due to an actual increase in the number of active enzyme molecules (34, 37, 38, 39). Presumably, the increase in TH protein permits a further increase in the capacity for catecholamine biosynthesis and release, although direct support for this hypothesis currently is lacking.

Recent experiments have examined the impact of stress-induced alterations in LC neurons on their response to a subsequent stressor (40, 41)). Extracellular NE in the dorsal hippocampus was measured under resting conditions and in response to 30 min of intermittent tail-shock using in vivo microdialysis methods. Two groups of rats were studied, a naive control group and a group that previously had been housed for 3–4 weeks at 4°C, a condition known to produce an increase in maximal TH activity within the cell bodies of the LC (39, 42, 43). The basal extracellular concentration of NE in hippocampus was found to be the same in the naive and chronically cold-stressed rats. However, 30 min of exposure to an intermittent tail-shock paradigm produced a 52% greater elevation of extracellular NE in the cold-stressed animals compared to the naive control animals ().

To examine changes in NE biosynthesis after chronic stress, subsequent experiments employed a method in which an inhibitor of aromatic amino acid decarboxylase (AADC), NSD-1015, is administered via the dialysis probe and the resulting accumulation of DOPA in the extracellular fluid is measured (44). In rats previously exposed to chronic cold, NE synthesis in hippocampus did not differ from that observed in naive controls. However, acute tail shock produced a significantly greater and more prolonged increase in hippocampal NE synthesis in chronically stressed rats than in control animals (). This issue also was examined by utilizing the more traditional method of measuring the postmortem accumulation of DOPA in tissue after inhibition of AADC (45). In agreement with the in vivo microdialysis data, these results revealed that NE synthesis was elevated to a greater extent in hippocampus of chronically stressed rats exposed to acute tail shock than in naive controls ().

Taken together, these data suggest that under basal conditions NE release and ongoing tyrosine hydroxylation are not altered by prior exposure to chronic stress. However, the increases in release and synthesis of NE evoked by a novel stressor are larger in magnitude in hippocampus of chronically cold-stressed rats than in that of controls. These changes, then, represent alterations in LC–NE neurons in response to conditions of increased demand.

There is reason to believe that the phenomenon described above is not limited to LC neurons or to the particular combination of stimuli used in the experiments outlined. For example, similar effects of long-term stress on subsequent responsiveness are also observed in the mesolimbic DA system (46) and with chronic injection of saline instead of exposure to cold (Abercrombie and Zigmond, unpublished data). Moreover, these observations may be related to a larger group of findings in which exposure to a stimulus (often a drug) leads to an enhanced responsiveness to a subsequent exposure to the same stimulus ("sensitization") or to another stimulus ("cross-sensitization") (reviewed in refs. 47 and 48).

Injury to the nigrostriatal DA system does not lead to gross neurological deficits unless the extent of the injury is very great. In the adult rat, for example, few permanent behavioral deficits are observed in the resting state unless the loss of tissue DA content in striatum exceeds 95%. There is a considerable body of evidence to suggest that this phenomenon reflects compensatory changes occurring in the remaining DA neurons—including elevated rates of transmitter synthesis and release (49, 50, 51, 52), as well as decreased reuptake of released transmitter (52, 53, 54). These events seem adequate to maintain a normal level of dopaminergic function in the resting state despite extensive neuronal degeneration (55). Up to a certain point, compensatory modifications in DA neurons may even permit continued responding under conditions where stimulus-evoked increases in DA release are brought about, such as stress (13).

Recently, the relation between the loss of striatal DA terminals and extracellular DA level in striatum has been directly determined by using in vivo microdialysis to monitor extracellular DA in intact rats and in rats sustaining depletions of DA in striatal tissue (56; see also refs. 57 and 58). The depletions were produced by central administration of the catecholaminergic neurotoxin 6-hydroxydopamine (6-OHDA). The concentration of DA measured in the extracellular fluid of striatum was not significantly reduced by prior 6-OHDA treatment unless the depletion of DA in striatal tissue exceeded 80% (). Importantly, this also is the approximate level of DA loss required for behavioral deficits to be observed (59, 60).

Because the concentration of DA in extracellular fluid was decreased by a lesser amount than was the concentration of DA in tissue, the ratio of these two measures increased with increasing lesion size (). These and other results (see above) suggest that after the partial degeneration of central dopaminergic terminals, those terminals that remain compensate to maintain the extracellular concentration of DA at a normal level. When functional impairments eventually do occur, it is assumed that it is because these compensatory processes are limited. Data compatible with this conceptual framework also have been obtained in studies of central serotonergic and noradrenergic systems and in the peripheral sympathoadrenal system (34, 61, 62, 63, 64, 65).

It appears that alterations in the demands placed on a catecholaminergic neuron can significantly alter the subsequent characteristics of that neuron. We have illustrated this with two different examples. In the instance of NE and chronic stress, the number of neuronal elements was held constant and the environment was altered by exposure to the chronic stressor. The result was enhanced responsivity of the neurons to a subsequent novel stressor. In the example of DA and partial injury, we illustrated how biochemical adaptations in catecholamine neurons can occur when the environment is constant and the number of neuronal elements is reduced. These two cases involve apparent increases in function, but downward regulation also is possible. For example, compensatory down-regulation of a number of cellular parameters has been noted to occur in NE neurons of the LC after chronic treatment with antidepressant drugs (66, 67).

It is proposed that a fundamental property of catecholaminergic neurons is the capacity to adjust levels of transmitter synthesis and release as a function of past history of activity or of ongoing changes in activity requirements. This capacity for cellular homeostasis in catecholaminergic neurons can account for the phenomena described in this chapter as well as similar phenomena described for responses to other stimuli or for other catecholaminergic systems (see above).

The functional significance of stimulus-induced modifications in the responsiveness of catecholaminergic neurons is not yet clear. It seems likely that the underlying mechanisms evolved to serve some adaptive function. For example, enhanced catecholamine synthesis during chronic stress may protect against the depletion of stores, permitting an organism to react more effectively to a novel challenge. However, it also may be that under certain conditions overresponsiveness may result in maladaptive behavior, as occurs during clinical anxiety or in post-traumatic stress disorder. Might genetic or acquired differences in the regulation of transmitter dynamics during conditions of altered use underlie the normal range of individual differences in the ability to successfully respond to environmental challenge? Furthermore, might certain disease states represent "neuroregulatory disorders" of compensatory responding in these neurons, leading to behavioral disorders of arousal and mood such as schizophrenia and affective illness?

A consideration of the properties of biochemical adaptation inherent to catecholaminergic neurons may also be relevant to an understanding of the dissociation between neuropathology and symptomatology that appears to occur in many disorders, including Parkinson's disease. Parkinson's disease is a progressive neurological disorder caused by the degeneration of DA neurons in the nigrostriatal system. As in the case of 6-OHDA-lesioned animals, patients with Parkinson's disease show marked clinical deficits only after the near-total loss of the dopaminergic innervation of the striatum (50). Presumably, the increase in DA synthesis and release that has been documented in the animal model (see above) also develops in parkinsonism and underlies the extensive preclinical phase of the disease. In support of this hypothesis is the observation that the concentrations of DA metabolites are less dramatically affected by the disease than is DA itself, leading to a ratio of DA metabolites to DA whose value sometimes reached more than 10 times that observed in control tissue (50, 68). A related hypothesis deriving from such data is that perhaps age-related deficits in the homeostatic capacities of catecholaminergic neurons results in the accelerated rates of behavioral decline seen in senescence (69, 70, 71, 72) (see Biological Markers in Alzheimer’s Disease, Cognitive Impairment in Geriatric Schizophrenic Patients: Clinical and Postmortem Characterization, and Parkinson’s Disease).

In summary, exposure to chronic environmental stress or partial injury to the neurons in a particular cell group can alter the functional characteristics of catecholaminergic neurons, leading to changes in transmitter synthesis and release. These changes may be maladaptive, as may occur under certain conditions of chronic stress, or they may be adaptive, as appears to be the case in Parkinson's disease. Further examination of these phenomena and their functional implications may provide some insights into a variety of neurological and psychiatric conditions as well as their treatment.

This work was supported by United States Public Health Service grants MH-43947, MH-00058, and NS-19608, the American Parkinson's Disease Association, and the National Alliance for Research on Schizophrenia and Depression. We are indebted to the many colleagues and students who importantly contributed to these efforts and whose work is cited in this review.

published 2000