Current perspectives of the roles of the central norepinephrine system in anxiety and depression


  • Conflict of interest: Professor Andrew Goddard has a conflict of interest plan with Indian University School of Medicine (IUSOM) in reference to consulting and clinical trial work (the clinical trial was closed) for Orexigen Therapeutics, Inc. He is not currently consulting, nor conducting a trial, with Lilly or its affiliates. Professors Andrew Goddard and Anantha Shekhar are from IUSOM, and the other authors are Lilly employees.


Norepinephrine (NE) is a major monoamine neurotransmitter that has widespread effects across multiple brain areas to regulate arousal and stress responses. The underlying function of the NE cortical system is to balance vigilance/scanning behavior with focused attention on novel environmental stimuli and the state of arousal. The central NE system is involved intrinsically with the stress response system, and dysregulation within the NE system has been implicated in the pathogenesis of anxiety and depressive disorders. Central NE activity paradoxically has either anxiogenic or anxiolytic effects, depending on whether the time course of the stress is acute or chronic, whether the stress is predictable or unpredictable, and which underlying brain regions are affected. Under conditions of chronic stress, NE system activity dysregulation of the hypothalamic-pituitary-adrenal system may turn a homeostatic stress response into a pathological stress response. Data suggest that the NE interplay with the serotonin system may exert neurobiological normalization of the pathophysiological state of anxious depression. Accordingly, pharmacological interventions targeting the NE system can result in anxiolytic, rather than anxiogenic, outcomes when used to treat patients with anxiety and depression. Depression and Anxiety, 2010. © 2009 Wiley-Liss, Inc.


As an integral part of the stress response system in an organism, norepinephrine (NE) serves as an essential neurotransmitter to regulate arousal and adapt to environmental and internal stressors. Although a facilitated functioning of the NE system has been associated with anxiogenic effects, the central NE system can more accurately be described as a modulator with anxiogenic or anxiolytic effects that vary according to acute or chronic conditions of stress. This review will provide an overview of preclinical and clinical findings associated with NE transmission during adaptive and maladaptive responses to stress. Specifically, chronic stress may contribute to enduring dysregulation of the NE system, possibly associated with key symptoms related to anxiety and depressive disorders. Accordingly, facilitated NE transmission can potentially be anxiolytic via normalization of underlying NE system dysregulation in anxiety and depression. Besides the NE system, there are a number of other targets being affected at different stages of anxiety and depression. Interested readers can refer to the review of this topic published elsewhere.1 Moreover, we acknowledge that central NE also has a role in panic disorders, and this specific topic has already been reviewed.2

This review focuses on neurobiological mechanisms of NE actions in anxiety and depression, as well as the clinical significance of an NE adaptation. This adaptation includes pharmacological interventions that modify NE system functioning to maximize treatment outcomes for patients with anxiety and depression. The objective of this review is to provide an overview of preclinical and clinical findings associated with NE transmission during adaptive and maladaptive responses to stress. We propose here that facilitating NE transmission in chronic stress-induced anxious depression can be anxiolytic via normalization of the underlying NE system dysregulation.


The location and distribution of NE neurons illustrate their intrinsic involvement in global cortical activity. Most noradrenergic neurons are clustered in separate small groups in the brainstem and designated as A1 to A7 cell groups, which include the locus coeruleus (LC) (A4, A6 cell groups). Subgroups of noradrenergic nuclei project their axons diffusely throughout the brain, including the prefrontal cortex (from A1/A2 cell groups), hypothalamus (from A1/A2 cell groups), thalamus, hippocampus, and amygdala (from LC: A4, A6 cell groups), and spinal cord (from LC: A4, A6 cell groups) (Fig. 1).3, 4 The diversity of targets that receive NE input suggests that changes in NE activity can globally influence a wide range of psychobiologic functions. These include executive processes involved in decision-making and attention mediated by the prefrontal cortex, stress response mediated by the hypothalamus, arousal/sleep mediated by the thalamus, memory encoding in the hippocampus, and fear-learning processed by the amygdala.

Figure 1.

The distribution and functions associated with the brainstem NE neuronal network that includes the LC. The LC projections from the brainstem ascend in a widespread manner to most of the cortical and subcortical areas of the brain to modulate diverse functions including mood, cognition, and sleep. The A1/A2 NE neuronal cell groups project to the paraventricular nucleus (PVN) of the hypothalamus that regulates the HPA axis in stress responses. Dotted lines linking LC with the hypothalamus indicate an indirect nature of this connection. The LC-NE neurons receive reciprocal serotoninergic input from the raphe nuclei and glutamatergic and corticotrophin-releasing factor (CRF) and glutamatergic (GLUT) input from the nucleus paragigantocellularis (PGi). In turn, the orbitofrontal cortex (OFC) and the anterior cingulated cortex (ACC), which have interconnections within the prefrontal cortex (PFC), exert “top–down” feedback control of the LC activity.

Cortical influences on the ascending brainstem NE system can provide important feedback regulation and play a crucial role in emotional stress response and stress-related psychiatric disorders. Task-related decision processes occurring in the prefrontal cortex can drive accurate task-relevant NE-dependent behavioral responses. Within the prefrontal cortical regions, the orbitofrontal cortex is known to process the evaluation of rewards, whereas the anterior cingulate cortex monitors performance errors and general negatively valenced stimuli, such as pain, monetary loss, and social exclusion. Thus, direct input from both the orbitofrontal and anterior cingulate cortical areas can exert strong “top–down” cognitive and emotional feedback influence on brainstem NE neuronal functioning (Fig. 1).5


The brainstem NE neurons from the A1 (ventrolateral medulla) and A2 (nucleus of the solitary tract) cell groups that directly innervate the paraventricular nucleus (PVN) of the hypothalamus have a prominent role in mediating stress responses.6, 7 Only a small subgroup of NE neurons from the LC innervates the PVN. Instead, the LC-NE neurons project and conduct stress signals primarily to the forebrain areas involved in the organization of stress responses.4

The PVN and its output are key components of the classic hypothalamic–pituitary–adrenal (HPA) axis that regulates neuroendocrine responses to stress.4 Direct LC-NE input (from a small subgroup) regulates corticotropin-releasing hormone (CRH)-containing neurons in the parvocellular portion of the PVN,8–10 which then project to the brainstem nuclei that mediate autonomic responses.11, 12 These CRH-containing neurons are the principal sources of stressor-induced CRH delivery to the hypophyseal portal system and stimulate production of pituitary adrenal corticotrophic hormone (ACTH). In turn, circulating ACTH stimulates corticosterone release in the adrenal cortex as part of the metabolic and adaptive response to acute stress.4, 13


The LC-NE neurons that project primarily to the forebrain regions are involved in the organization of stress responses. The LC-NE innervation of cortical higher cognitive processing regions plays a greater role in affective and cognitive components of stress and anxiety responses. Within the cortical brain regions, LC-NE input has a primarily modulatory role to “boost” or “dampen” the signals mediated by other neurotransmitters, such as serotonin, dopamine, glutamate, and γ-amino butyric acid (GABA).14

Neuronal firing activity of NE neurons from the LC correlates positively with the organism's arousal and attention states.15–17 The LC-NE neurons fire in transitional modes that are behavior-dependent. First, steady-state or tonic firing of LC neurons occurs as arousal level increases, with the lowest frequency occurring during quiescence, moderate firing rates during automatic behaviors, and highest rates upon the initiation of alertness (e.g., wakening). Second, phasic burst firing in LC neurons correlates with focused attention to optimize correct task-related performance.18 Presumably, these two distinct modes of LC-NE neuronal firing also lead to differential levels of NE-release at the terminal regions.

In general, the LC-NE system provides an optimal and flexible mechanism for balancing vigilance/scanning with focused attention based on the novelty of environmental stimuli and the underlying state of arousal. During an acute stress response, heightened level of readiness for action engages the LC-NE system. Additionally, activating the A1/A2 NE neurons connected to the HPA axis leads to a concerted effort to mediate cognitive, affective, neuroendocrine, metabolic, global, emotional, and behavioral response to manage stress allostatically.19


Many NE receptor subtypes, such as α1, α2, and β subtypes, are particularly relevant for the modulatory activities of the NE system. Postsynaptic α1- and β-adrenergic receptor subtypes mediate postsynaptic actions of NE. When present on non-NE axons, the α1 and β receptors can influence presynaptic release of other neurotransmitters.20 The α2-adrenergic receptor functions as a presynaptic autoreceptor, as well as a postsynaptic receptor. The α2 autoreceptors are distributed at somatodendritic sites and axonal terminals of NE neurons. Binding of NE at the α2 receptors on NE axonal terminals inhibits NE-release, but blockade of α2 somatodendritic or axonal terminal sites facilitates NE-release. Non-NE nerve terminals also contain α2 heteroreceptors on which NE can modulate the synaptic release of these non-NE mediators and neurotransmitters.21 The α2 receptors, located postsynaptic to the NE axonal terminals, have been shown to have a significant role in mediating the antidepressant effects of tricyclic antidepressants, which can elevate brain NE to activate these postsynaptic α2 receptors.22 Thus, the interactions among these different adrenoreceptor subtypes enable a highly plastic and dynamic system that can respond to daily adaptive demands.



Acute stressors associated with a high level of threat demand prompt attention. Functioning as an attention gatekeeper, the LC-NE neuronal pathway is poised to respond to acute stressors. Although definitions of stressors (i.e., causative factors) and stress (i.e., outcome) abound, most conceptualizations of stress involve internal or external changes to which the organism must adapt to maintain homeostasis.23 For humans, stressors include both physical and psychological threats, and the perception of threat is as capable of triggering a biological response as an actual threat alone.4 Stress responses involve physiological adaptations, notably neuroendocrine and autonomic, designed to support behavioral defense behaviors—such as escape (i.e., flight), aggressive defense (i.e., fight), or avoidance (i.e., freezing behavior)—that can neutralize threats.

In response to acute stress, LC-NE neuronal firing activity occurs in phasic bursts when attending to the threat cue while tonic activity increases arousal.24, 25 In preclinical studies, exposure to different types of stressors (e.g., immobilization, foot-shock, tail-pinch, and conditioned fear paradigms) results in increased NE turnover in the LC, hypothalamus, hippocampus, and amygdala. Depending on the type of acute stress, increased NE activity in the PVN (e.g., more in acute pain and immobility, but less so for hypoglycemia and hemorrhage) also stimulates CRH-release from the PVN of the hypothalamus, which in turn activates the HPA axis to mediate a variety of peripheral endocrine and metabolic effects (Fig. 2).4 Moreover, the CRH-containing neurons projecting from the PVN to the brainstem NE neurons have functional and reciprocal interactions.26–29 Thus, activation of the NE system during acute stress also increases sympathetic outflow to multiple peripheral organ systems (e.g., cardiovascular and respiratory), which leads to glucocorticoid- and adrenaline-release as endocrine responses. A sudden surge of these autonomic and neuroendocrine responses can cause acute panic attacks in humans (Fig. 2).30 Furthermore, acute stress and anxiety can also be elicited when existing environmental cues are associated with a previous stress-inducing adverse experience; this learned association leads to avoidance behavior.31 Under homeostatic adaptive conditions, the system responds acutely through negative feedback loops mediated by NE, autonomic pathways, and glucocorticoids to reestablish the basal activities of the NE circuit and the PVN.

Figure 2.

The acute stress response is illustrated by stressors activating the HPA axis and the LC-NE pathways (A4, A6 divisions), which results in the release of stress hormones from the paraventricular nucleus (PVN) and the dorso-medial hypothalamus (DMH). Corticotrophin-releasing factor (CRF) neurons from the PVN also interact reciprocally with LC-NE neurons during stress, resulting in an elevation of stress hormones and sympathetic outflow. Stress hormones, such as adrenocorticotrophic hormone (ACTH) and cortisol, facilitate autonomic reactivity in multiple systems, including increased respiration, cardiovascular output, and perspiration. The glucocorticoids and autonomic compensatory reflexes also regulate the overall response via negative feedback to the PVN and LC neurons. CRH, corticotropin-releasing hormone.

At the level of adrenergic receptors, preclinical studies have shown that acute stress (e.g., immobilization stress and acute cold-restraint stress in rodents) reduces the number of α2-adrenergic binding sites in the brainstem LC-NE area and other brain areas.3 Over the short-term, NE levels in the brain elevate without this α2 receptor feedback regulation,32 but subsequently return to normal through negative feedback loops by restoration of autoreceptor functions. However, repeated activation of this LC-NE system with exposure to chronic stress may lead to different outcomes.


Although activation of the NE system in response to acute stress has been fairly consistent across different preclinical studies, findings from chronic stress paradigms in animal studies showed varied outcomes, depending upon the characteristics of the stressor and the individual (e.g., gender and genetic vulnerability).33, 34 Prolonged, repeated, and uncontrollable stress may lead to the development of anxiety. Specifically, repeated expectation of uncertain threats may cause hyperarousal, hypervigilance, and distressed states. Humans with high anxiety strongly bias their attention toward threat-related stimuli. The amygdala and prefrontal connections, which are both innervated by LC-NE, have crucial roles as the neural substrates of these fear and anxiety attention states in animals and in humans.35, 36

Stressors of a more “psychological” nature, such as chronic psychosocial stress, have greater NE reactivity and HPA axis activation than stressors of a predominantly physical nature, such as immobilization and cold-restraint stressors.3, 4 Rodents subjected to the psychosocial stressor of repeatedly living in a habitat with a dominant male for a prolonged period (i.e., weeks) demonstrated adaptation in NE reactivity that resulted in a decreased release of brain NE and increased α2–binding sites, despite having higher plasma NE concentrations.32, 37, 38 Chronic stress may also cause postsynaptic β-adrenergic receptor upregulation (i.e., an increase in the number of β receptors), possibly due to reduced NE activity after prolonged chronic stress. Additionally, at the neuronal level, chronic stress has been associated with a retraction or atrophy of NE axonal projections, consistent with a reduced NE transmission.39, 40

Genetic vulnerabilities also play a role in response to chronic stress.41 When subjected to chronic stress caused by intermittent cold exposure (7 days, 4 hr/day, 4°C), inbred Wistar–Kyoto rats showed greater sensitization of both the brain NE system and the HPA axis, when compared with the outbred Sprague–Dawley rats.32, 42 Microdialysis studies of animals from various stress paradigms suggest that NE changes associated with chronic stress can be mediated by an upregulation of α2 autoreceptors. This enhanced α2 autoreceptor inhibition of NE-release results in a potential deficit in NE activity.43



Chronic stress alone can have a cumulative effect on the noradrenergic and neuroendocrine responses. A comparative study evaluated mothers of children with cancer and mothers of healthy control subjects for “allostatic load,” an index of the degree to which the body must adapt to stress to maintain homeostasis. A high allostatic load is associated with chronic stress and may be reflected by changes in physiological variables (e.g., blood pressure, cardiovascular measures, and hormonal indices), which combined make up a single index score.19 Among mothers of children with cancer, the rank order of high allostatic index is mothers who had post-traumatic stress disorder (PTSD)>mothers without PTSD>mothers of healthy children.44 Similarly, in a study observing elderly caregivers of patients with Alzheimer's disease, caregivers reporting higher levels of stress, as compared with those not reporting high stress levels, had greater increases in lymphocytic β-adrenergic sensitivity. However, the predictive power of stress depended upon interactions with other variables, such as age and gender.45

Repeated acute traumatic stress-induced elevation of central NE and cortisol can interact in the amygdala in the consolidation of lasting forms of negative emotional memory.46, 47 The negative emotional memory can be reactivated by external cues. This leads to re-experiencing memories of the original trauma and causes early hyperaroused psychophysiological responses and, with time, turns into a late “numbing” experience (when circulating NE and cortisol are actually low perhaps due to a “burn-out” state).48 The efficacy of the α1 antagonist prazosin, the α2 agonist clonidine, and the β antagonist propanolol in the clinical treatment of early PTSD provide the evidence of an involvement of a dysfunctional central NE system in early trauma-related syndromes.48–51 There is substantial evidence to show that the sleep problems and nightmares that are seen in PTSD may be secondary to hyperactivity of the noradrenergic systems during the earlier acute stages of the trauma and thus prazosin, a selective α1 noradrenergic antagonist, was shown to provide clinical benefit.52–55 Taken together, these findings suggest that traumatic stress results in a hyperactive central NE system and exaggerated central nervous system responses (i.e., sensitization) to noradrenergic activation.48, 49, 56


Major life events, which can be acute or chronic stressors, are consistent precipitating factors associated with the onset of symptoms of anxiety and mood disorders.57 There is a considerable overlap between anxiety and depression in humans. Sixty-two percent of patients with major depressive disorder (MDD) have high levels of psychic anxiety,58 and these patients typically show a poorer response to antidepressant treatments than patients having only MDD.59 The co-morbidity of anxiety and depression has led to the concept of an anxious depression,60 which is clinically defined as depressed patients with baseline 17-item Hamilton Rating Scale for Depression anxiety/somatization factor scores ≥7. Anxious depression is recognized as a recalcitrant subtype of depression, which is associated with poorer remission rate even with antidepressant treatment.61, 62 In response to antidepressant treatment, patients with anxious depression either take longer to achieve a response or lack a response to the treatment. The patients with MDD having lifetime general anxiety disorder, compared with patients having uncomplicated MDD, usually have worse symptom severity.63, 64 Our current views on the importance of the central NE system in anxious depression come from clinical studies showing efficacy of selective norepinephrine reuptake inhibitors (NRI) and the dual serotonin/NE reuptake inhibitors,61, 65, 66 although there is a lack of differential efficacy in these classes of antidepressants.62 Although it is known that the central NE system interacts with the HPA system to mediate stress and anxiety,67, 68 the underlying neurobiological mechanisms causing chronic stress-induced anxious depression are not well understood.

The early monoamine hypothesis suggested that depression was due to insufficient central monoamines. However, the results from clinical studies, which typically measured the primary metabolite of NE, 3-methoxy-4-hydroxyphenylglycol (MHPG) in urine and cerebrospinal fluid, failed to conclude that the source was central,69 and these results could not conclusively support the monoamine hypothesis of depression.70 In a small study, venoarterial NE and MHPG levels measured from internal jugular veins of nine depressed patients were significantly lower than levels from 19 healthy control subjects.71 The internal jugular vein receives the superior sagittal sinus blood directly from the brain. The low brain NE levels normalized after acute intravenous treatment with desipramine, a tricyclic antidepressant and an NE transporter blocker, thus supporting a lower central NE hypothesis in depression.

In postmortem studies, patients who had depression or committed suicide showed reduced NE activity as demonstrated by a decrease in NE transporter binding in the LC (which suggests a loss of NE-containing neurons) and in other limbic areas (which suggests a loss of NE axonal innervations). The results of these studies, however, have also been mixed.72, 73 In unmedicated depressed suicide victims, α2-adrenergic receptor density was increased (by 31–40%) in NE terminal sites, such as the prefrontal cortex. These postmortem data, thus, support the hypothesis suggesting an upregulation of supersensitive α2-adrenergic receptors in major depression.73 Findings from these clinical results show that depression is not simply caused by an increase or decrease in NE activity per se in the brain, because central NE functionally primarily acts as a modulator that dynamically enhances the signal-to-noise ratio of major excitatory and inhibitor transmitter input (e.g., glutamate and GABA).74

Instead of focusing on absolute differences of NE levels, other studies have examined whether there is a disturbance in the functional sensitivity or responsiveness of adrenergic receptor subtypes modulated by NE activity in patients with anxiety and depression. In humans, pharmacologic challenge agents can be used to examine neuroreceptor sensitivity. Clonidine, an α2-adrenergic agonist, increases the release of growth hormone (GH) in healthy control subjects.75 A blunted GH response in anxiety and depression, therefore, supports the concept of a defective α2 heteroreceptor-mediated NE transmission.76 In one challenging study, patients with both anxiety and co-morbid depression had a blunted clonidine-induced GH response. In contrast, patients with MDD alone and healthy control subjects exhibited healthy GH response to clonidine challenge, thus suggesting that hyporeactivity of the adrenergic receptors is associated with anxious features in the patient population with mood/anxiety disorders.67, 77


The discovery of neurogenesis in the adult human brain and its regulation by brain-derived neurotrophic factor (BDNF) and other growth factors responsible for neuronal survival, proliferation, and differentiation may have significant implications for the pathophysiologic theories of mood and anxiety disorders.78–80 For example, corticosteroid elevation due to stress-activation of the HPA axis may suppress adult neurogenesis, especially in the hippocampal dentate gyrus, to disrupt plasticity-dependent learning and memory in the hippocampus.81–83 Additionally, persistent elevation of circulating corticosteroids can also cause atrophy of other important brain regions for cognitive and emotional processes, such as the prefrontal cortex and amygdala.84, 85 As depletion of central NE leads to a decline in adult hippocampal neurogenesis,86, 87 it is likely that an elevation of NE (e.g., by antidepressants) has an indirect role in regulating BDNF-dependent adult hippocampal neurogenesis.88 This elevation of BDNF may be an important part of the cellular processes that underlie antidepressant efficacy.89 Indeed, antidepressants that enhance NE and/or 5-HT neurotransmission reverse the stress-induced suppression of BDNF-dependent neurogenesis.34, 90–93 Given that a failure in the central NE regulation of HPA activity and the subsequent inability to regulate the stress response can contribute to the pathogenesis of mood and anxiety symptoms, the neuroplastic changes after antidepressant treatments suggest that NE and/or 5-HT can have a direct role in regulating BDNF expression, and a disruption of NE and/or 5-HT functions in the brain can contribute to the cellular mechanisms underlying anxiety and depression.34, 91–97 Furthermore, nonpharmacological approaches (e.g., exercise, diet control, and life-style changes) also increase BDNF and neurogenesis.78, 98–100 These approaches may constitute an indirect way to modulate NE functions, which, in turn, regulate BDNF and neurogenesis–dependent mechanisms to normalize the state of anxious depression.101, 102


The effectiveness of selective serotonin reuptake inhibitors (SSRIs) in depression clearly implicates the central 5-HT system in antidepressant therapeutic mechanisms.103 However, emerging clinical experimental studies using a 5-HT or catecholamine depletion approach to probe for neurotransmitter specificity in depression concluded that depression is not simply caused by a depletion of 5-HT or catecholamine alone.104 Depletion of tryptophan, a key amino acid precursor for the synthesis of brain 5-HT, was used to reduce brain 5-HT, whereas α-methyl-para-tyrosine was used to inhibit catecholamine synthesis. Human subjects were given compounds that temporarily deplete the synthesis of 5-HT, or catecholamines. In healthy subjects with no prior personal or family history of depression and in unmedicated patients with MDD who previously responded to antidepressants, 5-HT or catecholamine depletion treatment did not result in depressive symptoms.105 Conversely, healthy control subjects with a family history of mood disorders, as compared with healthy control subjects without family history of mood disorders, showed a mild depressive response to tryptophan depletion.106

In a follow-up study to this series of experiments, drug-naïve patients with depression were first treated with and responded to either an NRI desipramine or the SSRI fluoxetine. They then underwent a tryptophan depletion challenge that reduced plasma tryptophan concentrations by up to 80% from baseline. Patients treated with the SSRI experienced a transient return of depressive symptoms, whereas patients treated with the NRI did not show a depressive relapse.107 These results not only suggest that 5-HT- and NE-specific antidepressants depend on the availability of endogenous 5-HT, but they also suggest that it is effective in treating a 5-HT depletion–induced depression by elevating central NE.104, 105, 108

Based on the considerable interconnections between LC-NE neurons and raphe-5-HT neurons in the brainstem,109 a functional “cross-talk” between NE and 5-HT systems would affect cellular plasticity between these two systems and their outputs. As an example of this type of interaction, activity of the NE axons exert facilitatory effects on 5-HT axonal regeneration, whereas activity of 5-HT axons exert inhibitory influence on NE axon regeneration.110, 111 Significant interactions between these two systems may underlie the clinical efficacies of antidepressants when one or both monoamines are elevated by the SSRIs or the selective NRI antidepressants.70 Local elevation of NE in the raphe serotoninergic neurons is likely to activate the α2 autoreceptor on NE terminals locally and the α2 heteroreceptor on the raphe serotoninergic neurons.112 This results in an overall local reduction of NE inhibitory influence of serotonin neurons, as shown by the ability of reboxetine to enhance raphe neuron firing and serotonin release at the serotoninergic terminals in the prefrontal cortex.113 Chronic treatment with dual-acting reuptake inhibitors (SNRIs) or SSRIs in anxious depression may restore central NE transmission through normalizing the upregulated α2 adrenoreceptors and increasing raphe-5-HT transmission, probably via 5-HT1A receptor desensitization.70 Given this important relationship between NE and 5-HT, the involvement of treatment strategies for both systems could help to normalize the deficits resulting from chronic stress.


The evidence reviewed above suggests that the central NE system undergoes homeostatic adaptive modifications to acute stress, and further plasticity changes occur under conditions of chronic stress. The findings within chronic stress conditions vary depending upon factors associated with the stressor (e.g., type, duration, and intensity) or with the individual (e.g., gender, age, and genetic predisposition). The overall literature, however, suggests that pathological anxiety and depression are associated with time-dependent plastic changes within the NE system. Hence, pharmacological alteration of NE transmission may be critical to treatment response.70


In the month or so after a major life event, HPA axis and NE systems in at-risk patients would be hyperactive. Acute treatments that target CRF overproduction or NE hyperactivity may have a special role at this early stage of treatment/decompensation. Based on the proposed network model of LC-NE changes in the pathophysiology of anxiety and depression, findings from the preclinical studies and treatment efficacy studies suggest that blocking or attenuating the hyperactive NE system resulting from acute and repeated traumatic stress would be beneficial (e.g., the use of α1 antagonist, β blocker, with benzodiazepine as an adjunct medication to turn down hyperactivity of the NE system).114 In contrast, in treating chronic stress-induced anxious depression, a facilitation of NE and/or 5-HT functioning (e.g., the use of NRI and SNRI) would be beneficial to normalize NE hypoactivity and this is not associated with exacerbation of anxiety symptoms (Fig. 3).115 Clearly, further research into augmentation strategies with NE medications to determine the anxiolytic benefits appears warranted.

Figure 3.

A model that summarizes NE activity in different stress levels, as well as the time-dependent adaptation of the central NE system under acute and chronic stress. Via α1-, α2-, and β-adrenergic receptor subtypes, LC-NE neuronal activity can result in changes in NE turnover, as well as in receptor sensitivities. During acute stress, effective treatment includes a blockade of α1- and β-adrenergic receptor subtypes, with adjunct benzodiazepine treatment to dampen LC-NE neuron activity. Chronic stress and genetic vulnerabilities contribute to further dysregulation in NE activity associated with pathological anxiety and depression. Chronic stress-induced anxious depression can be normalized after chronic enhancement of serotonin (5-HT) and NE neurotransmission by antidepressant treatment in the chronic stress stage (but not in the acute stress stage) to restore adrenoceptor functions (including α2 autoreceptor and α2 heteroreceptors). α1R, α1 adrenoceptor; α2R, α2 adenoreceptor; Auto, α2-autoreceptor; Post, postsynaptic; βR, β adenoreceptor; AD, antidepressant; ANX, anxiety; DEP, depression; LC, locus coeruleus; NE, norepinephrine.


In preclinical studies, healthy animals that are pretreated acutely with an NRI reboxetine show an ineffective block of the typical stress response to an acute stressor. However, animals that are chronically pretreated with reboxetine show an attenuated stress response, suggesting that chronic facilitation of NE neurotransmission is necessary for a therapeutic response.116 Noradrenergic agents, including reboxetine, decrease LC neuron-firing rates. Chronic treatment with antidepressants that elevate brain NE levels results in decreased β-adrenergic receptors.117, 118 This adaptive β-adrenergic receptor down-regulation has been used as a marker for antidepressant activity,74 as the timing of this biological change correlates with the onset of therapeutic effects.

Clinically, the use of selective NRIs to elevate brain NE levels to achieve efficacy in depression and anxiety supports the important role of NE in anxiety and depression.119–121 Normalization of NE dysregulation is associated with a therapeutic response, as demonstrated by pre- and post-treatment clonidine/GH challenges. In a study comparing efficacy response to the SSRI fluoxetine or the tricyclic NRI amitryptiline, there was no overall difference in the outcome between the treatment groups. However, patients who demonstrated a blunted GH response before treatment demonstrated significantly preferential response to amitryptiline, as compared with fluoxetine.122 It is likely that the SNRI properties of amitryptiline123, 124 account for the preferential GH response in the amitryptiline-pretreated subjects. Because anxiety and depression disrupt normal cognition,125–127 an elevation of brain NE after antidepressant treatment could stimulate β1- and α2-adrenergic receptors in the brain to therapeutically enhance cognition.128

Given the permissive role of NE in antidepressant effectiveness, there is growing interest in incorporating NE interventions into the treatment plan. The results from a recent meta-analysis in patients with MDD showed that SNRIs have a small but significant advantage in effectiveness compared with SSRI treatments.115 Along these lines, augmentation treatment with an NRI for patients with partial response to an SSRI is commonly used when treating MDD.129, 130 Several studies have shown that SNRIs that elevate both 5-HT and NE (e.g., venlafaxine and duloxetine) have been shown to be efficacious in treating anxious depression.63, 131–133 The duration required for these treatments to achieve remission depends highly on successfully minimizing the residual symptoms.134 Antidepressant treatments of anxious depression that extend to months or years may be the best approach to achieve a higher remission rate.135, 136


The role of NE is increasingly recognized not only for its facilitatory effects in acute stress, but also for helping changes of receptor sensitivity and plastic adaptations that occur under conditions of chronic stress. Interplaying with genetic vulnerability, dysregulation in the NE system appears to be a mechanism involved in the occurrence of pathological anxiety and depression. The use of pharmacological interventions that facilitate NE-release may promote adaptations that restore the regulatory control of NE and, thus, can enhance anxiolytic relief for patients.


Authors Contributions: Each author predicated in the conceptualization of the review article and search strategies, interpretation, drafting, and critical revision of the article. The final version of the article has been approved by each author.