Although agents that modify neurotransmitter action have become the primary therapies for depression and although numerous abnormalities in neurotransmitters have been uncovered in depression, the attempt to establish primacy of any single neurotransmitter or of neurotransmitters over hormones has been unsuccessful. As emerging technologies permit further examination of new systems, additional perturbations have been noted, but findings and formulations of hypotheses have necessarily reflected methods available at the time.
For over four decades, tricyclic antidepressants and monoamine oxidase inhibitors have been known to be effective treatments and show serotonergic, norepinephrinergic, and/or dopaminergic activities. These observations provide the so-called pharmacological bridge to the monoamine hypothesis of depression (108), which has guided much research to elucidate the role of the monoamine neurotransmitters, serotonin (5-HT), norepinephrine (NE), and dopamine (DA), in the pathophysiology of depression. Further development of more specific agents including selective serotonin reuptake inhibitors (SSRIs), norepinephrine reuptake inhibitors (NERI), and dopaminergic reuptake inhibitors, has reinforced the importance of monoamine systems for the treatment of depression. Thus, the monoamine hypothesis continues to encourage investigation of the biological basis of depression. Such, investigations are now focusing on additional components of monoamine action such as postsynaptic receptors, presynaptic autoreceptors and heteroreceptors, second messengers, and gene transcription factors. For example, several antidepres-sants have been noted to downregulate 5-HT1A receptor activity reducing negative feedback of 5-HT1A in the raphe nuclei resulting in greater 5-HT release (109). Such findings support the possibility of adding a 5-HT1A antagonist to an SSRI to potentiate antagonist effects (110). Alternatively, it has been argued that postsynaptic 5-HT1A receptors may be a target for antidepressant therapy although existing evidence suggests that full agonists may have too narrow a therapeutic index in humans to test the hypothesis (111).
In addition to their independent effects, the monoamines interact with all of the systems described here and elsewhere in this paper. For example, the glia have postsynaptic 5-HT and NE receptors on their cell bodies and processes (112, 113) that could be expected to affect concentrations of glutamate and neurotrophic factors. In addition, substance P is co-expressed with 5-HT in ascending dorsal raphe neurons (114) and substance P modulates mesolimbic DA activity and is involved in stress-induced activation of the ascending norepinephrine projection from the locus coeruleus (115). For the sake of clarity, we will briefly consider each monoamine by itself, recognizing that in vivo there are, among them, complex regional and structure-specific interactions.
A role for 5-HT in depression was established with the use of SSRIs in its treatment. That SSRIs really do depend on 5-HT has been elegantly tested by showing that depletion of tryptophan, a precursor of 5-HT synthesis, leads to return of depressive symptoms in patients with recent response to SSRIs (116). Women have lower rates of 5-HT synthesis and thus may show even greater relapse rates in response to depletion than men (117).
Evidence of reduced serotonergic function has been found in untreated depressed patients. Investigators have assessed [3H] imipramine and [3H] parox-etine binding in platelets from depressed and healthy subjects as a possible peripheral marker of the brain serotonin transporter (SERT) with some, but not all, studies showing reductions in depressed patients (118, 119). Postmortem studies show similarly decreased SERT binding in hypothalamus (120) as have imaging studies (121). One might expect more consistent findings in postmortem samples from suicides, since low serotonin metabolite concentrations in cerebrospinal fluid may be associated with reduced impulse control that might predispose depressed subjects to commit suicide (122). On the other hand, there is no obvious causal relationship between a measure of 5-HT turnover in cerebrospinal fluid and density of the transporter.
Mechanistically, 5-HT and the HPA axis are linked. Figure 1 shows the normal interaction of 5-HT, NE, and DA with the endocrine system. 5-HT can stimulate CRF release mediated by 5-HT2, 5-HT1A, and 5-HT1C receptors. Glucocorticoids tend to enhance 5-HT function, possibly as a compensatory effect in chronic stress. The extent to which this input exerts major influences in humans remains to be established. Nonetheless, preclinical studies point to several potentially important relationships. For instance, acute stress increases 5-HT release transiently, but continued stress leads to 5-HT depletion. Chronic stress may also increase production of 5-HT1A autoreceptors that further reduce 5-HT transmission. 5-HT1A knockout mice demonstrate increased stress-like behaviors indicative of "increased anxiety" (123, 124). These mice show increased mobility in response to stressors which is used as a model for antidepressant drugs (125). These data provide a basis for the association of anxiety symptoms with depressed mood.
Imaging studies have investigated the relationship of 5-HT and hippocampal atrophy. Positron emission tomography studies of 5-HT2A binding of [(18)F] altanserin has, however, led to disparate findings (121). It is to be anticipated that development of additional and more selective imaging ligands will clarify many of the suggestive, but variable, findings in depression that tend to overlap with both normal populations and those with other conditions (e.g., Alzheimer's disease).
The role of NE has also been demonstrated in clinical trials as well as in depletion studies. Alpha-methyl paratyrosine (AMPT), which blocks NE synthesis, does not alter the rating on the Hamilton Depression Rating Scale (HAMD) in normal subjects, but produces a depressive relapse (increases the HAMD rating) in patients who remitted to an NE antidepressant such as desipramine or mazindol (126). In contrast, there was no return of depression after AMPT in patients who had remitted on a serotonergic antidepressant (126).
NE neurons in the locus coeruleus (LC) project to almost all major brain regions and serve an important role in regulating and focusing additional and other responses to external stimuli (127, 128). Not surprisingly, NE systems are involved in responses to stress since there are multiple interactions between the HPA axis and NE. For instance, under experimental conditions, CRF secretion increases LC neuronal firing, resulting in enhanced NE release. NE release stimulates CRF secretion in the paraventricular nucleus (PVN) which leads to ACTH secretion. Increased ACTH leads to increased cortisol which provides a negative feedback to decrease CRF and NE in the PVN.
During behavioral stress, LC neuronal firing is increased (129) in association with increased release of NE. This LC responsiveness is enhanced with a novel stress after chronic or prior stress. When stress exposures are repeated in situations that prevent the animal from escaping, the animal exhibits learned helplessness which is associated with the depletion of NE (130). It is thought that this depletion is due to inability of the animal to synthesize sufficient NE to replace that which is released (131). Whether such depletion occurs in human brain following the chronic stress of depression is not known.
Various approaches to evaluating NE function in depression have been pursued over the last three decades from quantifying its metabolites in urine to radiolabel isotope dilution techniques to track its "spillover" in plasma (132). Taken together, there is evidence of a shift toward elevated turnover and release in unipolar depression, although values overlap with those observed in age and gender-matched healthy volunteers. One hypothesis to explain these increases emerges from consideration of reports on altered sensitivity of the platelet a2 receptor in depression. Subsensitivity of a2 receptors and/or their coupling mechanism to downstream intracellular events could be responsible for the exaggerated release of NE observed in depressed patients subjected to acute physiological or psychological manipulation (reviewed in Manji and Potter (133)).
Despite some evidence for an association of elevated HPA and NE function, relevant and consistent relationships have not yet been established among the available peripheral (blood and urine) measures. It is conceivable that availability of appropriate methods to simultaneously quantify CRF and NE in all brain regions would lead to demonstration of tight relationships. It is equally conceivable that elevations of CRF and NE can occur relatively independently as a function of different subtypes of depression. The result that emerges from such studies will have implications for the treatment and understanding of depression pathophysiology.
As noted above, antidepressant treatment, including that with NE and 5-HT uptake inhibitors, has effects on the NE signal transduction pathway and increases BDNF by blocking stress-induced decreases of BDNF in the hippocampus (134). Furthermore, stimulation of 5-HT2A receptors increases BDNF mRNA (135, 136). Consequently, various elements along the NE and 5-HT pathways, including a receptors, G proteins, cyclic adenosine monophosphate (c-AMP), c-AMP regulatory element binding protein (CREB), and brain-derived neurotropic factor (BDNF), are being evaluated as targets for antidepressant medications (67). As an example, focused on the NE cascade, chronic administration of norepinephrine antidepressants, such as desipramine and reboxetine, causes desensitization of the b-adrenoceptor-coupled adenylate cyclase system. Nuclear phosphorylated CREB (CREB-P) decreases in rat frontal cortex after chronic administration and in fibro-blasts after incubation, suggesting that norepinephrine antidepressants exert direct effects beyond b-adrenoceptors. This would be consistent with deamplification of the NE-mediated signal transduction cascade, resulting in "normalization" of increased norepinephrine activity, which is an evolving hypothesis (137).
Deficiencies in DA have been tied to depression and DA is tied to the regulation of the endocrine system. Moreover, there is a long-standing case for a role of enhancing DA function in the treatment of depression, particularly that not relating to monotherapy, as reflected in the special role of MAOIs or bupropion as an adjunctive therapy (138-140). CSF levels of homovanillic acid (HVA), a major DA metabolite (141-143), and urinary dihydroxyphenylacetic acid (DOPAC), another major DA metabolite (142), are reduced in a proportion of depressed patients. Consistent with the tendency for depressed patients to have decreased DA metabolites, imaging (single photon emission computed tomography, SPECT, with the high affinity D2 ligand 123I-iodobenzamide) studies of D2 receptor binding have demonstrated 10% more basal ganglion activity in depressed patients than in controls. This may be due to decreased dopaminergic transmission since decreased intrinsic D2 occupancy would tend to lead to upregulation of D2 receptors (144). When antidepressant treatment is instituted, the D2 activity decreases in the striatum
(145, 146), consistent with this hypothesis. Most recently, decreased density of the DA transporter and increased density of D2/3 receptors were found in the amygdala in a study of postmortem brain samples from subjects with depression (147). Finally, in some brain regions, DA is transported by the NE transporter into the presynaptic neuron (148). Therefore, in brain areas with NE nerve terminals, a norepinephrine reuptake inhibitor may act like a DA reuptake inhibitor and avoid the peripheral dopaminergic effects, as well as cocaine-like effects, that might be seen if the DA transporter itself were universally inhibited.
Increased glucocorticoid activity leads to altered or decreased prefrontal cortical DA metabolism (149, 150) and increased mesolimbic DA activity (149). DA is also known to be a prolactin (PRL) release-inhibiting factor since it is released by the arcuate nucleus of the hypothalamus where it binds to the D2 receptor inhibiting the activity of the acidophilic cells of the anterior pituitary, thereby blocking PRL and also growth hormone release. Thus, the blunted PRL response to a serotonergic agent, seen in depressed patients, could involve a dopaminergic component, especially in light of well-known 5-HT-DA interactions (96).
Was this article helpful?