White KJ, Walline CC, Barker EL. Serotonin Transporters: Implications for Antidepressant Drug Development. AAPS Journal. 2005; 07(02): E421-E433. DOI:  10.1208/aapsj070242

Correspondence to:
Eric L. Barker
Tel: (765) 494-9940
Fax: (765) 494-1414
Email: ericb@pharmacy.purdue.edu

Received: March 9, 2005;  Accepted: May 30, 2005;  Published: October 5, 2005

Abstract

Due to the complexity of the disease, several hypotheses exist to explain the etiology of depression. The monoamine theory of depression suggests that disruptions in the serotonergic and noradrenergic systems result in depressive symptoms. Therefore, the serotonin transporter (SERT) has become a pharmacological target for treating these symptoms. This review will discuss what is known about the molecular interactions of antidepressants with SERT. The effects of antidepressants on SERT regulation and expression in addition to the receptors that may be involved in mediating these effects will be addressed. Specifically, how changes to SERT expression following chronic antidepressant treatment may contribute to the therapeutic benefits of antidepressants will be discussed. Furthermore, the effects of SERT gene polymorphisms on antidepressant efficacy will be examined. Finally, a brief overview of other hypotheses of depression will be addressed as well as factors that must be considered for future antidepressant development.

Keywords: SSRIs, antidepressant, serotonin transporter, depression, reuptake

Monoamine Theory of Depression

Functional deficiencies in serotonin (5-hydroytryptamine, 5-HT) and norepinephrine (NE) have been implicated in the pathophysiology of depressive syndromes, and restoring the normal function of 5-HT- and NE-associated signaling pathway has been the target of antidepressants. Restoration of monoamine deficiencies to normal levels as a therapeutic strategy is based on the monoamine hypothesis of depression.¹ The oldest antidepressants, the monoamine oxidase inhibitors (MAOIs), increase synaptic levels of 5-HT and NE by inhibiting the enzymatic degradation of these neurotransmitters. The tricyclic antidepressants (TCAs), as well as newer selective 5-HT reuptake inhibitors (SSRIs) and 5-HT/NE reuptake inhibitors (SNRIs), all increase synaptic levels of 5-HT or NE by inhibiting reuptake via the 5-HT transporter (SERT) or NE transporter (NET), respectively. Emerging evidence indicates that the monoamine hypothesis of 5-HT and NE modulation fails to explain the whole mechanism of antidepressants. Other hypotheses, including the cytokine hypothesis of depression, the hypothalamic-pituitary-thyroid hypothesis of depression, as well as the role of brain-derived neurotrophic factor and cyclic AMP response element binding protein will be considered later in this review.

Antidepressant Development

The development of antidepressants began in the early 1950s with the fortuitous discovery that the antitubercular drug iproniazid, which functioned as an inhibitor of monoamine oxidase the enzyme responsible for breaking down 5-HT, had mood-elevating effects.² Subsequently, compounds with antihistamine activity such as imipramine and chlorpromazine were discovered to confer mood-elevating effects.² These discoveries led to the development of the first 2 classes of drugs used to treat depressed patients: the MAOIs and the TCAs. The first generation MAOIs (iproniazid, tranylcypromine, and phenelzine) were nonselective, irreversible inhibitors of MAO leading to increased levels of dopamine, 5-HT, and NE in patients. However, these drugs had dangerous and life-threatening side effects including hepatotoxicity and hypertensive crisis due to interactions with food high in tyramine. The risks of side effects and necessity for strict patient compliance led to the eventual disuse of these drugs. Nonetheless, MAOIs have seen a resurgence in recent years, particularly for the treatment of atypical and drug-resistant forms of depression.³

The first tricyclic antidepressant, imipramine, elicited antidepressant effects by inhibiting NE and 5-HT transport (Table 1). These compounds act as antagonists for SERT or NET and block neurotransmitter uptake, thus increasing synaptic concentrations of 5-HT or NE, respectively. However, imipramine was not without unfavorable side effects due to antihistaminic, antiadrenergic, and anticholinergic effects.^5,6 The development of second generation TCAs led to more selective inhibitors of NE uptake (desipramine, nortriptyline, maprotiline) and 5-HT uptake (clomipramine), but demonstrated no significant improvements in side effect profiles. More recently, the development of the SSRIs has consumed the mental health market (Table 1). The introduction of fluoxetine in 1988 provided clinicians with a safer treatment alternative to the MAOIs and the TCAs since the side effect profile was considerably improved. A meta-analysis of 36 double-blind clinical trials of TCAs and SSRIs found that patients taking a TCA were more likely to drop out of a study due to an adverse side effect (22.4%) than those prescribed an SSRI (15.9%).⁷ However, despite the improvement in side effects and drug interactions, the SSRIs are generally no more clinically efficacious at treating depression than any previous therapy.⁸ Increasing response rates to the SSRIs are paralleled with increases in the placebo effect.⁵ For example, response rates as measured by a 50% or greater decrease in the total depression score on the Hamilton Rating Scale for Depression were 79% and 60% for sertraline and fluoxetine, respectively. However, the corresponding placebo effect was 48% for sertraline, compared with only 33% for fluoxetine indicating a similar (~30%) response rate to these SSRIs.⁵ Additionally, approximately 30% to 45% of patients fail to attain an adequate response to their initial pharmacotherapy.⁹

Table 1. Structures of Commonly Prescribed Antidepressants and Ki Values for Inhibition at hSERT and hNET*

Structure Antidepressants hSERT Ki (nM), [3H]5-HT Uptake Inhibition hNET Ki (nM), [3H]NE Uptake Inhibition Amitriptyline 36 ± 1 102 ± 9 Nortriptyline 279 ± 20 21 ± 0.77 Desipramine 163 ± 5 3.5 ± 0.6 Imipramine 20 ± 2 142 ± 8 Venlafaxine 102 ± 9 1644 ± 84 Sertraline 3.3 ± 0.4 1716 ± 151 Fluoxetine 20 ± 2 2186 ± 142 Citalopram 8.9 ± 0.7 30,285 ± 1600 Paroxetine 0.83 ± 0.06 328 ± 25 * Data were Obtained in HEK cells Heterologously Expressing hSERT or hNET.4

The failure of the SSRIs to treat some depressed patients led to the development of less selective agents. These SNRIs are inhibitors of both 5-HT and NE uptake. The SNRIs may have increased clinical utility over the SSRIs at treating refractory depression and anxiety indicating the need to target both the serotonergic and noradrenergic uptake systems in some patients.¹⁰ Additionally, the SNRIs may have efficacy in reducing the physical symptoms associated with depression, including muscle tension, fatigue, appetite changes, and body aches, in addition to an alternative use as a treatment for fibromyalgia.¹¹ Drugs in this class include venlafaxine and duloxetine and although they are less selective than the SSRIs with regard to targeting specific transporters, they have limited affinity for the muscarinic, adrenergic, and histaminergic receptors, resulting in a favorable side-effect profile.¹² Finally, the shift from focusing completely on the 5-HT system to the NE system was complete with the development of 2 selective NE reuptake inhibitors, reboxetine and atomoxetine. Reboxetine is indicated for major depressive disorder, whereas atomoxetine is prescribed for attention-deficit/hyperactivity disorder.

The discussion of drugs used to treat affective disorders would not be complete without mentioning the atypical antidepressants. These drugs do not fall into a specific class and affect a wide range of physiological targets. Clinically used agents include the 5HT_2A receptor antagonists such as trazadone and its second generation compound nefazodone, the NE/dopamine reuptake inhibitor bupropion, and the 5-HT_1A receptor agonist and D2 dopamine receptor antagonist buspirone.

Molecular Interactions of Antidepressants with SERT

Antidepressants are represented by many structurally distinct classes of compounds each of which may interact with SERT in a unique manner (Table 1). Although the 3-dimensional structure of SERT and other monoamine transporters such as the dopamine transporter and NET has not been solved, experimental assays correlating functional potency data to structural changes provide some clues into the binding sites of these different drug classes as well as possible distinct binding. Species-scanning mutagenesis has also been employed to identify residues that contribute to inhibitor binding. This method exploits the differences in inhibitor potencies between variants of SERT (ie, human versus rat). Similarly, cross-species chimeras can be generated to determine the influences of specific transmembrane domain (TMD) regions on drug potencies. For example, Barker and colleagues¹³ generated cross-species chimeras between the human and Drosophila SERT (hSERT and dSERT, respectively) homologs to determine the regions of the SERT responsible for the divergent pharmacology of SERT inhibitors at the Drosophila species as compared with hSERT. The inhibitor potency data from human/Drosophila SERT chimeras led to the observation that residues distal to the second TMD conferred potency specificities for most inhibitors including paroxetine and fluoxetine. Conversely, the region including TMDs I and II contained key sites for species discrimination of citalopram and mazindol. Further investigation revealed that mutation of tyrosine 95 in the hSERT to the corresponding dSERT phenylalanine residue altered mazindol and citalopram potency values nearly equivocal to those for dSERT.¹² Furthermore, computational modeling of the SERT-citalopram interaction suggests a dipole-dipole interaction of the nitril group of S-citalopram with Y95 providing further support for a specific molecular interaction with this amino acid.¹⁴

Additionally, an aspartate in the first TMD of the dopamine transporter has been implicated as a key residue for dopamine transport and cocaine analog recognition.¹⁵ To determine the role of this residue in SERT, several mutations were generated in rat SERT (rSERT). The carboxylic acid of the D98E mutant was identified as an essential point of interaction for the amine in 5-HT, indicating this residue was important for substrate recognition. Furthermore, this mutation caused a significant decrease in citalopram potency for inhibiting 5-HT transport.¹⁶ Therefore, this aspartate may be important for antidepressant recognition as well. In addition to the D98 residue in TMD I, key residues in the third TMD of hSERT appear critical for the recognition of cocaine, amphetamine, and antidepressants (C. Walline, unpublished data).

Other key residues have been implicated in antidepressant recognition in NET within the TMD V-IX region.^17-19 Although discussion of other transporters within the monoamine family is outside the scope of this review, 2 of these amino acids (P418 and S375) are sites of sequence divergence between hSERT and dSERT. Therefore, these residues were ideal candidates for species-scanning mutagenesis in order to confirm their importance in antagonist recognition. Both P418 and S375 were independently mutated to the equivalent dSERT amino acid (serine and alanine, correspondingly).²⁰ Unfortunately, both mutants failed to show any statistically significant potency differences as compared with hSERT for a panel of TCA and SSRI antidepressants suggesting that these amino acids may not be critical sites for antidepressant recognition by SERT.²⁰

Residues in SERT have also been discovered that confer high potency interaction with tricyclic antidepressants. Using species-scanning mutagenesis, F586 in hSERT was mutated to the corresponding valine in rSERT. The F586V hSERT mutant demonstrated a phenotype similar to the rSERT with regard to TCA potency.²¹ The complementary rSERT mutation V586F showed hSERT-like potencies for the tricyclics imipramine, desipramine, and nortriptyline.²¹ Taken together, these studies suggest that SERT TMD XII may contain distinct molecular determinants for binding of the TCAs. A more complete understanding of antidepressant recognition by SERT at the molecular level will be revealed by future efforts focused on elucidation of SERT structure.

Acute Regulation of SERT Expression by Antidepressants

Alterations to the serotonergic system are thought to partially explain the therapeutic benefits of antidepressants. Since the discovery that antidepressants work in part by potentiating the actions of 5-HT within the serotonergic system (for review see Schloss and Williams²² and Vetulain and Nalepa²³), the effects these drugs elicit on the SERT protein have been an area of active research. Specifically, antidepressants are thought to initially function by acting as SERT antagonists resulting in an immediate increase in the synaptic 5-HT concentration.²² Changes to the expression level of SERT in response to drugs could contribute to neuroadaptive changes associated with chronic antidepressant administration (see below). Surprisingly, the effects of antidepressants on regulation of SERT expression remain largely unevaluated.

Regulation of SERT activity by kinases seems plausible as SERT contains putative phosphorylation sites for protein kinase C (PKC), cAMP-dependent protein kinase (PKA), and cGMP-dependent protein kinase (PKG). Early studies in human embryonic kidney (HEK) cells heterologously expressing hSERT (HEK-hSERT) identified SERT as a phosphoprotein sensitive to rapid internalization following activation of PKC concurrent with a decrease in SERT activity.^24,25 Further studies in the HEK-hSERT cells showed that antidepressants alone could not affect the ability of PKC to phosphorylate SERT. The ligand 5-HT, however, caused a reduction in PKC-induced phosphorylation.²⁶ Therefore, in the presence of substrate, SERT would not become phosphorylated, the signal necessary for internalization. Interestingly, acute treatment of HEK-hSERT cells with 5-HT in the presence of the SERT-selective antidepressants, imipramine, paroxetine, and citalopram, but not the NET-selective nisoxetine, resulted in SERT phosphorylation by PKC.²⁶ The ability of antidepressants to counteract the action of 5-HT suggests a mechanism that would further increase the synaptic concentration of 5-HT by reducing the number of transporters at the cell surface available to eliminate 5-HT from the synapse.²⁶

Although phosphorylation of SERT by acute PKC activation results in rapid changes in SERT activity and cellular distribution, regulation of SERT expression at the plasma membrane by PKA or PKG does not occur after acute kinase activation despite an increase in SERT phosphorylation.²⁵ One explanation for these results would be that although phosphorylation of SERT occurs through numerous kinase pathways, changes to SERT activity and expression due to PKA or PKG phosphorylation occur over a longer period of time than PKC regulation. Another explanation is that HEK cells do not contain the proteins necessary to affect SERT expression following phosphorylation by PKA or PKG. Interestingly, chronic treatment with antidepressants increases the production of intracellular cAMP due to an increased coupling between the G protein Gs_α and adenylyl cyclase (for review see Donati and Rasenick²⁷). Elevation of intracellular cAMP would result in increased translocation and activity of PKA suggesting a potential mechanism for PKA regulation of SERT after extended antidepressant administration. This effect could occur through changes to unidentified protein interactions facilitated by alterations in the phosphorylation state of SERT. Studies in human placental choriocarcinoma (JAR) cells, a cell line that endogenously expresses SERT, suggest that increases in cAMP should lead to an increase in SERT activity due to an increase in both SERT mRNA and protein levels.²⁸ This study is in conflict, however, with numerous studies that have reported a decrease in SERT levels following chronic antidepressant administration (see below) despite the increase in cAMP production explained above. Further investigation will be required to determine if this conflict is due to cell type-dependent variability in the role of PKA phosphorylation in SERT regulation under normal and antidepressant-treated conditions.

Not surprisingly, kinase pathways that increase the activity of SERT have also been reported. Studies have identified a mechanism for up-regulation of SERT expression through PKG-dependent pathways.²⁹ Previously, Miller and Hoffman³⁰ reported that the A₃ adenosine receptor regulated SERT in rat basophilic leukemia (RBL-2H3) cells through cGMP and nitric oxide, subsequently increasing the V_max value for 5-HT transport. This regulation of SERT could occur under physiological conditions as RBL-2H3 cells endogenously express both the A₃ adenosine receptor and SERT. Interestingly, the increase in 5-HT transport was not associated with an increased expression of SERT at the cell surface.³⁰ One caveat of these experiments is that the cell-surface binding was performed using a membrane permeable ligand, thus, limiting the conclusions that can be drawn. Zhu et al,²⁹ however, reported that an up-regulation of SERT surface expression occurred when activating A₃ adenosine receptors in RBL-2H3 cells as measured by specific binding of the cocaine analog [¹²⁵I]RTI-55. Because SERT protein levels are low in RBL-2H3 cells, Chinese hamster ovary (CHO) cells were transiently transfected with A₃ adenosine receptor and SERT cDNAs to confirm these results. Similar increases in SERT protein surface expression levels following A₃ adenosine receptor activation occurred in the transfected CHO cells suggesting a common signaling pathway for SERT up-regulation within the 2 cell types.²⁹ Moreover, these studies extended those of Miller and Hoffman³⁰ by showing that activation of the PKG pathway required the upstream activity of phospholipase C, calcium, and guanylyl cyclase.²⁹ Additionally, investigation into other proteins involved in the PKG-signaling cascade led to the identification of p38 mitogen-activated protein kinase (MAPK) regulation of intrinsic SERT activity. Inhibition of p38 MAPK did not affect A₃ adenosine receptor-activated changes to SERT plasma membrane expression levels in RBL-2H3, thus suggesting that changes to SERT by the PKG activation of p38 MAPK occur independently of changes caused directly by PKG.²⁹ Interestingly, the studies of Samuvel et al³¹ showed that application of the p38 MAPK-specific inhibitor PD169316 resulted in decreases in the activity, phosphorylation, and plasma membrane expression of SERT isolated in rat synaptosomes. Reduction of p38 MAPK by siRNA in HEK cells transiently expressing hSERT produced a reduction in SERT at the cell surface, and inhibition of p38 MAPK by PD169316 decreased delivery of SERT to the plasma membrane. Moreover, activation of either p38 MAPK or PKC also inhibited interaction of SERT with syntaxin 1.³¹ Syntaxin 1 is a SNARE protein that has previously been found to interact with and regulate SERT’s expression.^32-34 Therefore, p38 MAPK appears to influence SERT expression via regulation of plasma membrane insertion potentially by affecting SERT’s interaction with syntaxin 1.³¹ Additional investigations into the role of p38 MAPK regulation of SERT surface expression are necessary before the effects of antidepressants on these interactions can be investigated.

In addition to interactions with numerous kinases that control SERT availability at the plasma membrane, SERT forms a complex with protein phosphatase 2A (PP2A) in HEK-hSERT cells.³⁵ Previously, inhibitors targeting PP2A and protein phosphatase 1 were found to down-regulate SERT activity.²⁵ Furthermore, p38 MAPK appears to regulate the interaction between SERT and PP2A.³¹ Additional research must be performed to elucidate the complex network of kinase and phosphatase activity involved in controlling SERT expression. Moreover, the extent of SERT-regulating phosphorylation that occurs in vivo remains unresolved. Further studies are necessary both in vivo and in vitro to establish the role of kinase phosphorylation on SERT expression under antidepressant-treated conditions both acutely and after chronic administration. Interestingly, an evaluation of kinase mRNA levels in rat brain following 21 days of fluoxetine or citalopram treatment showed a decrease in numerous kinases including PKC and PKA.³⁶ The effect this down-regulation confers on kinase protein levels and subsequent SERT regulation requires further analysis.

Effects of Chronic Antidepressant Exposure on SERT

Although antidepressants function to rapidly increase synaptic 5-HT concentrations, this initial effect does not instantly alleviate the symptoms of depression. In fact, a maximal clinical response to antidepressants will not become apparent until several weeks of continuous treatment (for review see Gelenberg and Chesen³⁷). Thus, the acute blockade of SERTs to increase 5-HT signaling does not account for the typical delay in symptom improvement. Current thought proposes that the neuroadaptive changes necessary for the therapeutic benefits of some antidepressants results in part from down-regulation of postsynaptic β-adrenergic receptors.³⁸ Similar changes, however, are not characteristic for SSRIs and other antidepressants that act on SERT.³⁸ Therefore, neuroadaptive changes in SERT expression could be necessary for antidepressants to be effective. Reports of changes to SERT expression after treatment with antidepressants vary.

The effects of chronic antidepressant treatment on the expression level of SERT have been explored in few in vitro systems. Horschitz et al³⁹ treated HEK cells heterologously expressing rSERT with increasing concentrations of citalopram before determining the effects of treatment on [³H]5-HT uptake and [³H]-citalopram binding. Following a 3-day incubation with 500 nM citalopram, the maximal transport activity (V_max value) decreased to approximately 40% without a subsequent change in the K_m value for 5-HT. A similar reduction (~50%) in [³H]-citalopram binding to rSERT at the plasma membrane occurred after this treatment; although a reduction in binding of approximately 25% was evident with as little as 20 nM citalopram. Moreover, a similar decrease in rSERT expression was not evident in cells chronically treated with the antidepressant, desipramine, a NET inhibitor.³⁹ Unfortunately, the precise mechanism for citalopram-induced down-regulation of SERT in HEK cells remains unclear as well as how these results will translate to neuronal cell lines.

Although the above results were obtained in a heterologous expression system, similar patterns of SERT down-regulation have been documented after chronic treatment of rats and humans with antidepressants. Specifically, Benmansour and colleagues^40,41 reported that chronic treatment (15 or 21 days) with paroxetine or sertraline in rats reduced the number of SERT binding sites by as much as 80% in the CA3 region of the hippocampus without long-term changes to SERT mRNA levels. Furthermore, chronoamperometry measurements of 5-HT levels revealed an increase in the time necessary to clear 5-HT in chronic sertraline-treated rats compared with the time required following an acute fluvoxamine block of SERT in saline-treated rats.⁴¹ These data show that an overall decrease in the expression level of SERT induced by chronic antidepressant exposure results in 5-HT remaining in the synapse longer than observed with acute blockade of SERT alone. Additionally, Gould et al⁴² reported a 75% to 80% decrease in the lateral nucleus of the amygdala, dentate gyrus, and dorsal raphe SERT binding sites after 6 weeks of paroxetine administration to rats. Similar treatment with the selective NET inhibitor, reboxetine, did not result in a change to SERT binding.⁴² A study of 17 healthy humans administered 40 mg/day of citalopram found a reduction in diencephalon and brainstem SERT binding following 8 days of treatment.⁴³ Surprisingly, no further reduction in SERT binding occurred after an additional 8 days of treatment.⁴³ Changes in platelet 5-HT and SERT have been suggested as possible biomarkers for psychiatric illness (for review see Plein and Berk⁴⁴). Interestingly, Alvarez et al⁴⁵ reported a decrease in platelet SERT binding sites in depressed versus healthy individuals that was further reduced following 12-week administration of clomipramine or fluoxetine. Furthermore, a greater drop in platelet SERT binding sites was evident in antidepressant treatment responders versus nonresponders.⁴⁵ These data support a possible link between SERT expression and responsiveness to antidepressant treatment. Interestingly, the responders showed an initial increase in platelet inositol triphosphate levels that returned to normal following chronic antidepressant treatment. In nonresponders, the inositol triphosphate levels remained elevated. An increase in inositol triphosphate indicates an increase in the phosphoinositide signaling pathway that would also lead to an increase in 1,2-diacylglycerol (DAG). Activation of PKC by DAG could then regulate SERT surface expression.⁴⁵ A more complex system for regulation of SERT expression is suggested by these data, however, since a decrease in DAG, in addition to inositol triphosphate, should result in a decrease in PKC activity. Therefore, an increase in SERT expression would be expected if PKC alone were responsible for SERT regulation.

Not all studies of chronic antidepressant treatments report a decrease in SERT expression levels. Hébert et al⁴⁶ described no significant changes to brain SERT binding sites in rats treated 21 days with numerous antidepressants with the exception of decreased SERT levels in the entorhinal cortex induced by both fluoxetine and venlafaxine. Interestingly, treatment with the tricyclic antidepressants trimipramine or desipramine appeared to increase SERT binding in several cortical brain regions compared with the transporter inhibitors fluoxetine and venlafaxine.⁴⁶ Moreover, an increase in hippocampal SERT binding sites occurred after a 10-week treatment with amitriptyline in 24-month-old rats. The treatment did not change the SERT levels in 10-month-old rats.⁴⁷ An increase in SERT following antidepressant treatment could contribute to age-related differences in therapeutic effectiveness of antidepressants.⁴⁷ Finally, a study of depressed versus healthy individuals showed an initial decrease in lymphocyte SERT binding sites in depressed individuals compared with controls that increased after chronic fluoxetine treatment concurrent with a clinical improvement in depressed symptoms.⁴⁸ Further studies are necessary to confirm the link between clinical response and changes in SERT expression in areas outside of the brain as regulatory mechanisms could differ substantially between neuronal and non-neuronal cells.

Although the above-mentioned studies only represent a small portion of the studies examining the effects of chronic antidepressant treatment on SERT binding sites (for further review see Schloss and Silliams²² and Schloss and Henn⁴⁹), the disparities described could be reflective of the systems used to measure SERT levels. For instance, differences in the specificity and cell permeability of the radiolabels used for these studies could account for variability of the results. Moreover, different antidepressants may elicit unique effects on SERT availability due to differences in recognition of the antidepressants by SERT. Additionally, the dose of antidepressant administered as well as the route of antidepressant administration could affect the local concentration of drug in the brain. Benmansour et al^40,41 and Gould et al⁴² used osmotic minipumps to continuously dispense the drugs to the rats, whereas Hébert et al⁴⁶ administered the drugs to the rats through a single daily ip injection. In addition, the region of the brain under examination could account for differences in expression levels.⁴⁶ Finally, the population size, ethnicity, and age of subjects in human studies must be considered before generalizing the results.

Receptor Regulation of SERT

As described earlier, SERT is subject to regulation through kinase and phosphatase activity. These signaling proteins themselves are regulated by the activity of neurotransmitter receptors. Thus, the effect of chronic antidepressant treatment on various receptors and the subsequent effect the receptors confer on SERT is an area of active investigation. Unfortunately, specific receptor-mediated regulation of SERT remains poorly understood. One area of exploration involves the serotonin 5-HT_1A receptor. The 5-HT_1A receptor is believed to be important in the therapeutic effects observed upon chronic antidepressant treatment because this receptor is responsible for regulating firing of serotonergic neurons.⁵⁰ The increase in extracellular 5-HT caused by blockade of SERT by antidepressants results in a decrease in vesicular 5-HT release due to activation of the negative feedback mechanism facilitated by presynaptic 5-HT_1A receptors. Desensitization of the presynaptic 5-HT_1A receptor must then occur to return serotonergic neuronal firing to normal.^50,51 In the median and dorsal raphe, chronic treatment with fluoxetine leads to a decrease in presynaptic 5-HT_1A receptor G-protein coupling without a decrease in receptor binding sites. These findings suggest that receptor desensitization occurs at the level of effector coupling.⁵² Currently, no data support a direct role for 5-HT_1A receptor regulation of SERT, although SERT expression is decreased in 5-HT_1A receptor knockout mice,⁵³ an effect that could be a result of disrupting serotonergic signaling.

Recent exploration of the 5-HT_1B receptor suggests a role for this receptor in regulating SERT activity in vivo.⁵⁴ Specifically, activating presynaptic 5-HT_1B receptors results in increased clearance of 5-HT by SERT independent of 5-HT_1A receptor activity.⁵⁵ In agreement with these findings, studies in 5-HT_1B receptor knockout mice report a larger increase in extracellular ventral hippocampus 5-HT concentration following a single ip injection of fluoxetine compared with wild-type mice, although an alteration in SERT levels in the knockout mice could explain these results.⁵⁶ Interestingly, no significant change to extracellular 5-HT concentrations was evident in the frontal cortex suggesting differences in SERT regulation between the 2 brain areas.⁵⁶ Additionally, treatment of rats for 21 days with fluoxetine decreased 5-HT_1B receptor mRNA levels in the dorsal raphe nucleus.⁵⁷ No concurrent measurement of changes to presynaptic 5-HT_1B receptor protein levels was assessed in this study. If a decrease in 5-HT_1B receptors were to occur, however, a decrease in SERT function would be considered likely since presynaptic 5-HT_1B receptors regulate 5-HT uptake.⁵⁵

The therapeutic benefits of antidepressants may rely in part on desensitization of 5-HT_1A and 5-HT_1B receptors.⁵⁰ Interestingly, co-administration of the 5-HT_1A receptor antagonist pindolol with an SSRI increased the SSRI efficacy and decreased the lag between the start of drug treatment and a clinical response.⁵⁸ Studies by Shalom et al⁵⁹ showed that repeated application of a 5-HT_1B receptor antagonist in the rat frontal cortex prevents fluoxetine-induced 5-HT_1B receptor desensitization. Because of the importance of receptor desensitization in mediating the therapeutic benefits of chronic SSRI treatment, co-administration of an SSRI and a 5-HT_1B receptor antagonist may not be more effective over an extended period of time than an SSRI alone.⁵⁹ Further investigation into the mechanism behind 5-HT_1B receptor control of SERT activity could help with designing drug regimens for depressed individuals.

The role of the α₂-adrenergic receptor in the noradrenaline system, and the subsequent changes to the receptor’s expression and function during depressed episodes are currently under intensive investigation. Specifically, changes to α₂-adrenergic receptors are known to occur when examining antidepressants that specifically inhibit NET with changes being dependent on the type of antidepressant.⁴⁹ Interestingly, Ansah and colleagues⁶⁰ recently reported that α₂-adrenergic receptor activation could inhibit SERT activity both in vitro and in vivo, and the presence of Ca²⁺ was essential for this inhibition. Application of an α₂-adrenergic receptor agonist at a concentration that resulted in maximal SERT inhibition in synaptosomal preparations did not decrease SERT activity in transfected cell lines suggesting activation of a pathway in vivo that was unavailable in transfected cells rather than a direct interaction of the receptor or drug with SERT. Furthermore, when analyzing the effects of the α₂-adrenergic receptor agonist UK14304 in vivo an increase in the amount of time necessary to clear 5-HT was evident. UK14304 did not change SERT phosphorylation or the V_max value, but did decrease the apparent K_m value for 5-HT.⁶⁰ Thus, regulation of SERT caused by the activation of α₂-adrenergic receptors does not appear to be changing the surface expression of SERT. Activation of α₂-adrenergic receptors, however, could influence the interaction of an ancillary protein with SERT that affects recognition of 5-HT. The influence that α₂-adrenergic receptors have on SERT expression and activity requires further investigation to determine what, if any, role these receptors play on changing SERT levels during antidepressant treatment.

SERT Polymorphisms and Antidepressant Response

Although the interplay between SERT, receptors, and signaling proteins must be considered for the current antidepressants, development of future drugs for the treatment of depression should not only consider these interactions, but also genetic polymorphisms of the SERT gene. Ramamoorthy and colleagues⁶¹ isolated a single human SERT gene to chromosome 17q11.1-q12. Later studies revealed a polymorphism in the promoter region of the human SERT gene that resulted in a deletion of a 44-basepair repeat.⁶² Further investigation into the functional implications of the promoter deletion showed a 3-fold decrease in activity of the short (s) versus the long (l) promoter in JAR cells as measured by luciferase-promoter fusion constructs. Moreover, PKC- and cAMP-activated transcription was significantly greater at the l SERT gene promoter versus the s promoter.⁶² Prevalence of the s versus the l alleles in 505 subjects revealed that 19% of the individuals were homozygous for the s/s genotype whereas 49% were l/s and the remaining subjects were l/l.⁶³ Basal transcriptional activity of the promoters were ascertained from lymphoblasts cultured from individuals with l/l, l/s, and s/s genotypes. Similar transcriptional activity to that shown in JAR cells was reported; a concurrent decrease in SERT protein expression was evident in lymphoblasts containing the s/s and l/s genotypes compared with the l/l genotype.⁶³

Because these initial studies confirm that differences exist in the activities of the promoter as well as the distribution of the polymorphism in humans, the consequences of these allelic variations on antidepressant efficacy and response has become a subject of active inquiry. Rausch and colleagues⁶⁴ determined the SERT promoter polymorphism in 51 individuals diagnosed with major depression and examined the effects of fluoxetine on depression symptoms. Interestingly, patients with an l allele (either as l/l or l/s) exhibited an increase in the response to both fluoxetine and placebo as measured by a decrease in the Hamilton depression score compared with individuals with the s/s genotype.⁶⁴ In agreement with these findings, Chinese patients with the l/l genotype also showed a greater decrease in Hamilton scores after fluoxetine treatment compared with their s/s and l/s genotype counterparts.⁶⁵ Moreover, a more rapid onset of responsiveness for elderly people with the l/l genotype has been reported. Pollock et al⁶⁶ showed that after a week of treatment with paroxetine, l/l elderly patients had a decrease in the Hamilton depression score compared with both the l/s and s/s genotypes suggesting a faster rate of onset for the clinical effect. There was no change in onset of response in patients treated with nortriptyline suggesting a response unique to SERT-selective antidepressants.⁶⁶ Similar results indicating a faster rate of onset for response in l/l genotypes were obtained in a study of 176 elderly subjects treated with sertraline with no change in response to placebo-treated subjects.⁶⁷ Interestingly, neither study reported a further change in response subsequent to the initial response at the end of either a 12-week or 8-week study.^66,67 Thus, these data suggest a link between response to SSRIs and the age of the depressed subject.

Another polymorphism of the SERT gene results in a mutation in the eighth TMD of the SERT protein. This rare polymorphism was initially identified in 2 families with multiple neuropsychiatric illnesses.⁶⁸ The hSERT I425V mutation results in increased 5-HT transport in human cervical epitheloid carcinoma and COS-7 cells transfected with the I425V hSERT cDNA. The change in transport was accompanied with a decrease in the K_m value and an increase in the V_max value without a change in the surface expression suggesting a mutation-induced alteration to the transporter’s activity.⁶⁹ Additionally, as previously shown with activation of SERT through the A₃ adenosine receptor,³⁰ a nitric oxide donor increased wild-type SERT activity. The increase in SERT activity following nitric oxide stimulation was equivalent to the activity of the I425V hSERT mutant in the absence of nitric oxide. hSERT I425V activity, however, was insensitive to nitric oxide. Therefore, hSERT I425V functions at a level equivalent to wild-type SERT after activating cGMP-signaling pathways.⁶⁹ Furthermore, the hSERT I425V polymorphism exists on the l allele.⁶⁸ Thus, individuals with this polymorphism would have increased expression of a catalytically enhanced transporter. Interestingly, individuals with the hSERT I425V polymorphism were resistant to antidepressant treatment.⁶⁸

Although the present research strongly suggests genetic links to differences in response rates as well as overall response to antidepressants, specifically SSRIs, further investigation is necessary to determine how the changes in the SERT gene promoter alter the expression, regulation, and activity of the SERT protein. Rausch et al⁶⁴ reported a difference in the initial 5-HT K_m value by platelet SERT between subjects who responded to fluoxetine treatment versus nonresponders. This change in 5-HT interactions did not appear to be dependent on the allelic variant, although subjects with the l allele and an initial increase in the 5-HT K_m value did exhibit the most favorable response to treatment.⁶⁴ Furthermore, investigation into how the I425V polymorphism disrupts SERT regulation by PKG is necessary to further the understanding of how antidepressants affect SERT activity and expression. Such information could lead to the development of therapies for individuals who do not respond to common antidepressants.

Cytokines and Depression

In addition to antidepressants acting on the biogenic amine neurotransmitters, other signaling pathways may be involved with antidepressant action as well as the pathophysiology of affective disorders. One alternative to the monoamine hypothesis of depression is the macrophage hypothesis of depression developed by RS Smith in 1991.⁷⁰ Smith hypothesized that gross secretion of macrophage cytokines such as interleukin-1, tumor necrosis factor-α, and interferon-α (IFN-α) are responsible for some cases of major depression. Increases in these substances may elicit a hypersecretion of corticotrophin-releasing factor, resulting in hypercortisolemia, which may play a role in depressive diseases as a result of defects in negative feedback mechanisms controlling these substances.⁷¹ In fact, a recent study has indicated that patients suffering from depressive syndromes with a history of suicidal attempts were associated with lowered adrenocorticotropin and cortisol levels, particularly after a recent suicide attempt.⁷² Additionally, cancer patients receiving IFN-α or interleukin-2 therapy had significant decreases in serum tryptophan, the amino acid precursor to 5-HT.⁷³ Decreased serum tryptophan upon cytokine treatment may be due to induced tryptophan catabolism leading to increases in kynurenine and quinolinic acid. Patients who received IFN-α treatment for malignant melanoma showed significant increases in kynurenine as well as decreases in tryptophan and approximately half of these patients developed major depression during treatment.⁷⁴ However, patients who received paroxetine 2 weeks prior to IFN-α treatment did not develop major depression despite increases in kynurenine and quinolinic acid.⁷⁴ Although major depression frequently occurs in patients receiving cytokine therapy, these depressive symptoms can be attenuated with antidepressant treatment. Although a detailed discussion of cytokines and their purported link to depression via the hypothalamic-pituitary-adrenal axis is outside the scope of this review, it has been reviewed in detail elsewhere.^71,75

Hypothalamic-pituitary-thyroid Axis

The hypothalamus is directly responsible for regulating the pituitary, which in turn secretes thyroid-stimulating hormone (TSH). Dysregulation of this tightly controlled system can result in thyroid dysfunction. Interestingly, hypothyroidism and depression share many of the same psychological and physical symptoms including anhedonia, fatigue, depressed mood, and cognitive disturbances, as well as sleep disturbances and weight change. Depression, in addition to hypothyroidism, may be a result of an imbalance of the regulatory functions of the hypothalamic-pituitary-thyroid (HPT) axis. Recently, Bschor and colleagues⁷⁶ reported a relationship between thyroid function and lithium pharmacotherapy. Although lithium therapy is one of the oldest treatments for mood disorders and remains a useful strategy for the treatment of refractory depression, the mode of lithium action has been poorly understood. Following lithium treatment, Bschor et al⁷⁶ reported significantly higher serum TSH levels, as well as significantly decreased triiodothyronine (T₃) and thyroxine (T₄) levels, resulting in a significant reduction of HPT system activity. Although the study did not provide evidence that thyroid status would predict responses to lithium therapy, it does provide support for a relationship between the HPT and depression.

Direct effects of psychotropic therapy on the HPT axis have been difficult to assess. Many studies are inconclusive and have considerable methodological differences making the results difficult to interpret. At the molecular level, in vitro studies have shown that incubation of thyroid cells with chlorpromazine causes a decrease of ¹³¹I uptake, an important step in thyroid hormone synthesis.⁷⁷ Additionally, many antipsychotics and antidepressants with an alkylamino side chain (alimemazine, chlorpromazine, clomipramine, imipramine, and desipramine) can have a strong donor/acceptor interaction with iodine that could lead to iatrogenic hypothyroidism or thyroiditis.⁷⁷

The possible interactions between antidepressants and thyroid metabolites are less clearly interpreted in clinical studies. Shelton and colleagues⁷⁸ reported that T₃ levels decreased transiently for both desipramine and fluoxetine treatment groups whereas desipramine treatment caused a significant increase in total T₄. Also, no significant treatment effects on free T₄, TSH, or TSH response to a thryrotropin-releasing hormone test were observed.⁷⁸ A similar study determined the antidepressant effects on thyroid function while controlling for the circadian activity of the thyrotrophs. Depressed patients had a significantly lower TSH response to protirelin, free T₄, and free T₃ than control patients.⁷⁹ The authors concluded that significant changes in thyroid function tests were associated with clinical recovery and not due to a direct effect of the antidepressant drug treatment.

Conversely, thyroid hormone levels could exert a positive influence on antidepressant treatment of depressive disorders. Sauvage et al⁷⁷ postulated that T₃ may enhance the effects of antidepressants through a potentiation of the β-adrenergic receptor system. Co-administration of T₃ during an antidepressant regimen may accelerate a response to antidepressant treatment to improve the overall treatment outcome.⁷⁷ Overall, depression and the HPT axis seem to be intimately linked, although this link remains to be clearly defined.

G-proteins, cAMP Response Element Binding Protein

Several hypotheses regarding the mechanism of action of antidepressants have been developed, including the monoamine, cytokine, and HPT hypotheses. However, all of these hypotheses fail to address the significant lag time (approximately 2 to 4 weeks) between initiating a treatment regimen and detecting a clinical response. Molecular interactions of antidepressants at targets such as transporters and receptors are well studied and these interactions are known to take place following the first dose of drug treatment. Additional mechanisms must account for the lengthy response time. One such mechanism proposed is an increase in G-protein coupled signaling.²⁷ An increase in the coupling of the heterotrimeric G-protein subunit Gs_α to adenylyl cyclase results in increased cAMP, cAMP-dependent activation of protein kinase A, and subsequent phosphorylation and activation of a multitude of downstream targets. The up- or down-regulation of these targets is dependent on transcriptional and translational events that require days or even weeks for full molecular and cellular effects to be achieved and may contribute to the lag time between initiation of pharmacotherapy and evidence of a clinical response. These targets may include the cAMP response element binding protein (CREB), a nuclear transcription factor that up-regulates gene expression. Chronic antidepressant treatment has been linked to the up-regulation of CREB in the hippocampus, cortex, and amygdala.^80,81 Wallace and coworkers⁸¹ demonstrated that overexpression of CREB in the basolateral amygdala after training in the learned helpless model of depression caused a decrease in escape failures, an antidepressant response. In contrast, induction of CREB before training caused an increase in escape failures, a converse effect, as well as increased immobility in the forced swim test, indicating a prodepressive effect. Although the temporal expression of CREB was critical in determining antidepressant outcomes, it nonetheless implicates CREB as a key player in depression, learning, and memory, and signifies the need for controlled regulation.

Brain-derived Neurotrophic Factor

Brain-derived neurotrophic factor (BDNF) is a neurotrophin (ie, a peptide that regulates growth and survival of neurons) that functions as a neurotransmitter modulator and a regulator of plasticity mechanisms.⁸² Human studies have shown that serum BDNF levels in antidepressant-naïve depressed patients are lower than those in antidepressant-treated patients or healthy controls. These findings indicate that decreased BDNF may be important in the pathophysiology of depression.⁸² Shimizu et al⁸² hypothesized that a decrease in BDNF may reduce the neuroprotective effect and cause stress-induced neuronal damage, leading to biological vulnerability and susceptibility to depressive diseases.⁸² Several studies have reported robust increases in BDNF mRNA levels in several parts of the brain including the cortex and hippocampus after various (SSRI, TCA, MAOI) antidepressant treatments.⁸³ Coppell and coworkers⁸⁴ replicated this finding in rats with repeated administration of tranylcypromine, fluoxetine, paroxetine, or sertraline. Their data show BDNF mRNA levels were up-regulated in the hippocampus after long-term drug treatment. However, 4 hours after the last injection BDNF mRNA levels were down-regulated indicating that BDNF expression is differentially regulated in acute and chronic drug treatments. Interestingly, desipramine, maprotiline, and mianserin did not increase BDNF levels 24 hours post-injection.⁸⁴ Therefore, the regulation of BDNF may be drug specific as well as temporally regulated. Finally, CREB-deficient mice, unlike wild-type controls, do not exhibit increases in BDNF mRNA levels upon chronic desipramine administration.⁸⁵ Conti and colleagues⁸⁵ were unable to detect increased levels of BDNF mRNA after chronic fluoxetine administration in either control or CREB-deficient mice providing further evidence that this regulatory effect may be antidepressant specific. In summary, the data demonstrate that CREB may act upstream of BDNF and is necessary for antidepressant-induced alterations in BDNF mRNA expression.

Future Horizons of Antidepressant Development

Future treatments for depression will likely diverge further from the single target specificity of the SSRIs or the tricyclic antidepressants. Currently, novel 5-HT_1D and 5-HT_1B receptor antagonists that block presynaptic autoreceptors are under development. These agents would reverse the inhibitory role of the autoreceptors resulting in the subsequent release of 5-HT.⁸⁶ In addition, selective, reversible inhibitors of MAOs (RIMAs) such as befloxatone, teloxantrone, moclobemide, and brofaromine are being redeveloped.¹¹ If taken at low to moderate doses, RIMAs can be displaced from the enzyme and do not require the dietary restrictions of conventional MAOIs.¹¹ The emergency contraceptive mifepristone (RU-486) in addition to antagonizing progesterone receptors also antagonizes glucocorticoid receptors. Blocking the toxic effects of cortisol on glucocorticoid receptors may have therapeutic benefits in treating psychotic depression and bipolar disorder by enhancing neurocognitive function.^11,87 Furthermore, novel peptides that act as antagonists for the corticotropin-releasing factor are under development.¹¹ In fact, a recent study found that patients with mood disorders had a statistically significant increased number of corticotrophin-releasing hormone immunoreactive (CRH-IR) neurons in the paraventricular nucleus of the hypothalamus as well as a significantly higher number of CHR-IR neurons expressing the estrogen receptor alpha.⁸⁸ Finally, antagonists for the neurokinin receptors are being developed as inhibitors of substance P and other neurokinins. Antagonizing the effects of these peptides may have efficacy in treating mood disorders, psychosis, and anxiety.¹¹ Indeed, several neurokinin receptor antagonists have demonstrated the same antidepressant effect as amitriptyline and desipramine in a rodent forced-swim test model.⁸⁹

Conclusion

Clinical depression is thought to result in part from disruptions in the serotonergic system. Numerous studies have identified SERT as a pharmacological target for treating these symptoms. Not surprisingly, an interaction of antidepressants with SERT alone is not sufficient for immediate therapeutic benefit. Although a preponderance of research has focused on determining the regulation of SERT expression, the exact mechanisms responsible remain undefined. Thus, until the normal processes of SERT regulation are delineated, the implications of discrepancies in SERT levels and function due to chronic antidepressant treatment cannot be ascertained. Furthermore, development of new antidepressants with greater therapeutic benefits will be hindered without knowing the contribution of the various receptors and signaling proteins to this regulation as well as the contribution of non-amine systems to the etiology of depression. Finally, recent evidence shows that allelic differences exist in the responsiveness to current antidepressants. Due to the prevalence of depression in society, further investigation into why the l/l genotype of SERT allows for faster onset of therapeutic response to antidepressants and how drugs can be designed to increase the responsiveness of the s/s SERT genotype must be addressed.

References

1. Heninger GR, Delgado PL, Charney DS.  The revised monoamine theory of depression: a modulatory role for monoamines, based on new findings from monoamine depletion experiments in humans. Pharmacopsychiatry. 1996;29:2-11.
PubMed 

2. Nutt DJ.  The neuropharmacology of serotonin and noradrenaline in depression. Int Clin Psychopharmacol. 2002;17:S1-12.
PubMed 

3. Rush A, Ryan N. Current and emerging therapeutics for depression. In: Davis K, Charney D, Coyle J, Nemeroff C,  eds. Neuropsychopharmacology: The Fifth Generation of Progress. New York, NY: Raven Press; 2002:1081-1095. 

4. Owens MJ, Morgan WN, Plott SJ, Nemeroff CB.  Neurotransmitter receptor and transporter binding profile of antidepressants and their metabolites. J Pharmacol Exp Ther. 1997;283:1305-1322.
PubMed 

5. Ban TA.  Pharmacotherapy of depression: a historical analysis. J Neural Transm. 2001;108:707-716.
PubMed  DOI: 10.1007/s007020170047

6. Richelson E.  The clinical relevance of antidepressant interaction with neurotransmitter transporters and receptors. Psychopharmacol Bull. 2002;36:133-150.
PubMed 

7. Steffens DC, Krishnan KR, Helms MJ.  Are SSRIs better than TCAs? Comparison of SSRIs and TCAs: a meta-analysis. Depress Anxiety. 1997;6:10-18.
PubMed  DOI: 10.1002/(SICI)1520-6394(1997)6:1<10::AID-DA2>3.0.CO_2-9

8. Song F, Freemantle N, Sheldon TA, et al.  Selective serotonin reuptake inhibitors: meta-analysis of efficacy and acceptability. BMJ. 1993;306:683-687.
PubMed 

9. Fava M.  New approaches to the treatment of refractory depression. J Clin Psychiatry. 2000;61:26-32.
PubMed 

10. Gutierrez MA, Stimmel GL, Aiso JY.  Venlafaxine: a 2003 update. Clin Ther. 2003;25:2138-2154.
PubMed  DOI: 10.1016/S0149-2918(03)80210-2

11. Stahl SM, Grady MM.  Differences in mechanism of action between current and future antidepressants. J Clin Psychiatry. 2003;64:13-17.
PubMed 

12. Shelton RC.  The dual-action hypothesis: does pharmacology matter? J Clin Psychiatry. 2004;65:5-10.
PubMed 

13. Barker EL, Perlman MA, Adkins EM, et al.  High affinity recognition of serotonin transporter antagonists defined by species-scanning mutagenesis. An aromatic residue in transmembrane domain I dictates species-selective recognition of citalopram and mazindol. J Biol Chem. 1998;273:19459-19468.
PubMed  DOI: 10.1074/jbc.273.31.19459

14. Ravna AW, Sylte I, Dahl SG.  Molecular mechanism of citalopram and cocaine interactions with neurotransmitter transporters. J Pharmacol Exp Ther. 2003;307:34-41.
PubMed  DOI: 10.1124/jpet.103.054593

15. Kitayama S, Shimada S, Xu H, Markham L, Donovan DM, Uhl GR.  Dopamine transporter site-directed mutations differentially alter substrate transport and cocaine binding. Proc Natl Acad Sci USA. 1992;89:7782-7785.
PubMed 

16. Barker EL, Moore KR, Rakhshan F, Blakely RD.  Transmembrane domain I contributes to the permeation pathway for serotonin and ions in the serotonin transporter. J Neurosci. 1999;19:4705-4717.
PubMed 

17. Paczkowski FA, Bryan-Lluka LJ.  Tyrosine residue 271 of the norepinephrine transporter is an important determinant of its pharmacology. Brain Res Mol Brain Res. 2001;97:32-42.
PubMed  DOI: 10.1016/S0169-328X(01)00295-9

18. Roubert C, Cox PJ, Bruss M, Hamon M, Bonisch H, Giros B.  Determination of residues in the norepinephrine transporter that are critical for tricyclic antidepressant affinity. J Biol Chem. 2001;276:8254-8260.
PubMed  DOI: 10.1074/jbc.M009798200

19. Paczkowski FA, Bonisch H, Bryan-Lluka LJ.  Pharmacological properties of the naturally occurring Ala(457)Pro variant of the human norepinephrine transporter. Pharmacogenetics. 2002;12:165-173.
PubMed  DOI: 10.1097/00008571-200203000-00010

20. Roman DL, Walline CC, Rodriguez GJ, Barker EL.  Interactions of antidepressants with the serotonin transporter: a contemporary molecular analysis. Eur J Pharmacol. 2003;479:53-63.
PubMed  DOI: 10.1016/j.ejphar.2003.08.056

21. Barker EL, Blakely RD.  Identification of a single amino acid, phenylalanine 586, that is responsible for high affinity interactions of tricyclic antidepressants with the human serotonin transporter. Mol Pharmacol. 1996;50:957-965.
PubMed 

22. Schloss P, Williams DC.  The serotonin transporter: a primary target for antidepressant drugs. J Psychopharmacol. 1998;12:115-121.
PubMed 

23. Vetulani J, Nalepa I.  Antidepressants: past, present and future. Eur J Pharmacol. 2000;405:351-363.
PubMed  DOI: 10.1016/S0014-2999(00)00565-3

24. Qian Y, Galli A, Ramamoorthy S, Risso S, DeFelice LJ, Blakely RD.  Protein kinase C activation regulates human serotonin transporters in HEK-293 cells via altered cell surface expression. J Neurosci. 1997;17:45-57.
PubMed 

25. Ramamoorthy S, Giovanetti E, Qian Y, Blakely RD.  Phosphorylation and regulation of antidepressant-sensitive serotonin transporters. J Biol Chem. 1998;273:2458-2466.
PubMed  DOI: 10.1074/jbc.273.4.2458

26. Ramamoorthy S, Blakely RD.  Phosphorylation and sequestration of serotonin transporters differentially modulated by psychostimulants. Science. 1999;285:763-766.
PubMed  DOI: 10.1126/science.285.5428.763

27. Donati RJ, Rasenick MM.  G protein signaling and the molecular basis of antidepressant action. Life Sci. 2003;73:1-17.
PubMed  DOI: 10.1016/S0024-3205(03)00249-2

28. Ramamoorthy S, Cool DR, Mahesh VB, et al.  Regulation of the human serotonin transporter. Cholera toxin-induced stimulation of serotonin uptake in human placental choriocarcinoma cells is accompanied by increased serotonin transporter mRNA levels and serotonin transporter-specific ligand binding. J Biol Chem. 1993;268:21626-21631.
PubMed 

29. Zhu CB, Hewlett WA, Feoktistov I, Biaggioni I, Blakely RD.  Adenosine receptor, protein kinase G, and p38 mitogen-activated protein kinase-dependent up-regulation of serotonin transporters involves both transporter trafficking and activation. Mol Pharmacol. 2004;65:1462-1474.
PubMed  DOI: 10.1124/mol.65.6.1462

30. Miller KJ, Hoffman BJ.  Adenosine A3 receptors regulate serotonin transport via nitric oxide and cGMP. J Biol Chem. 1994;269:27351-27356.
PubMed 

31. Samuvel DJ, Jayanthi LD, Bhat NR, Ramamoorthy S.  A role for p38 mitogen-activated protein kinase in the regulation of the serotonin transporter: evidence for distinct cellular mechanisms involved in transporter surface expression. J Neurosci. 2005;25:29-41.
PubMed  DOI: 10.1523/JNEUROSCI.3754-04.2005

32. Haase J, Killian AM, Magnani F, Williams C.  Regulation of the serotonin transporter by interacting proteins. Biochem Soc Trans. 2001;29:722-728.
PubMed  DOI: 10.1042/BST0290722

33. Quick MW.  Role of syntaxin 1A on serotonin transporter expression in developing thalamocortical neurons. Int J Dev Neurosci. 2002;20:219-224.
PubMed  DOI: 10.1016/S0736-5748(02)00021-7

34. Quick MW.  Regulating the conducting states of a mammalian serotonin transporter. Neuron. 2003;40:537-549.
PubMed  DOI: 10.1016/S0896-6273(03)00605-6

35. Bauman AL, Apparsundaram S, Ramamoorthy S, Wadzinski BE, Vaughan RA, Blakely RD.  Cocaine and antidepressant-sensitive biogenic amine transporters exist in regulated complexes with protein phosphatase 2A. J Neurosci. 2000;20:7571-7578.
PubMed 

36. Rausch JL, Gillespie CF, Fei Y, et al.  Antidepressant effects on kinase gene expression patterns in rat brain. Neurosci Lett. 2002;334:91-94.
PubMed  DOI: 10.1016/S0304-3940(02)01106-0

37. Gelenberg AJ, Chesen CL.  How fast are antidepressants? J Clin Psychiatry. 2000;61:712-721.
PubMed 

38. Potter WZ, Hollister LE. Antidepressant agents. In: Katzung BG,  ed. Basic & Clinical Pharmacology. New York, NY: Lange Medical Books/McGraw-Hill; 2004:482-496. 

39. Horschitz S, Hummerich R, Schloss P.  Down-regulation of the rat serotonin transporter upon exposure to a selective serotonin reuptake inhibitor. Neuroreport. 2001;12:2181-2184.
PubMed  DOI: 10.1097/00001756-200107200-00027

40. Benmansour S, Cecchi M, Morilak DA, et al.  Effects of chronic antidepressant treatments on serotonin transporter function, density, and mRNA level. J Neurosci. 1999;19:10494-10501.
PubMed 

41. Benmansour S, Owens WA, Cecchi M, Morilak DA, Frazer A.  Serotonin clearance in vivo is altered to a greater extent by antidepressant-induced downregulation of the serotonin transporter than by acute blockade of this transporter. J Neurosci. 2002;22:6766-6772.
PubMed 

42. Gould GG, Pardon MC, Morilak DA, Frazer A.  Regulatory effects of reboxetine treatment alone, or following paroxetine treatment, on brain noradrenergic and serotonergic systems. Neuropsychopharmacology. 2003;28:1633-1641.
PubMed  DOI: 10.1038/sj.npp.1300236

43. Kugaya A, Seneca NM, Snyder PJ, et al.  Changes in human in vivo serotonin and dopamine transporter availabilities during chronic antidepressant administration. Neuropsychopharmacology. 2003;28:413-420.
PubMed  DOI: 10.1038/sj.npp.1300036

44. Plein H, Berk M.  The platelet as a peripheral marker in psychiatric illness. Hum Psychopharmacol. 2001;16:229-236.
PubMed  DOI: 10.1002/hup.251

45. Alvarez JC, Gluck N, Arnulf I, et al.  Decreased platelet serotonin transporter sites and increased platelet inositol triphosphate levels in patients with unipolar depression: effects of clomipramine and fluoxetine. Clin Pharmacol Ther. 1999;66:617-624.
PubMed 

46. Hébert C, Habimana A, Élie R, Reader TA.  Effects of chronic antidepressant treatments on 5-HT and NA transporters in rat brain: an autoradiographic study. Neurochem Int. 2001;38:63-74.
PubMed  DOI: 10.1016/S0197-0186(00)00043-7

47. Yau JL, Kelly PA, Olsson T, Noble J, Seckl JR.  Chronic amitriptyline administration increases serotonin transporter binding sites in the hippocampus of aged rats. Neurosci Lett. 1999;261:183-185.
PubMed  DOI: 10.1016/S0304-3940(98)00997-5

48. Lima L, Urbina M.  Serotonin transporter modulation in blood lymphocytes from patients with major depression. Cell Mol Neurobiol. 2002;22:797-804.
PubMed  DOI: 10.1023/A:1021869310702

49. Schloss P, Henn FA.  New insights into the mechanisms of antidepressant therapy. Pharmacol Ther. 2004;102:47-60.
PubMed  DOI: 10.1016/j.pharmthera.2004.02.001

50. Piñeyro G, Blier P.  Autoregulation of serotonin neurons: role in antidepressant drug action. Pharmacol Rev. 1999;51:533-591.
PubMed 

51. Celada P, Puig M, Amargos-Bosch M, Adell A, Artigas F.  The therapeutic role of 5-HT1A and 5-HT2A receptors in depression. J Psychiatry Neurosci. 2004;29:252-265.
PubMed 

52. Hensler JG.  Differential regulation of 5-HT1A receptor-G protein interactions in brain following chronic antidepressant administration. Neuropsychopharmacology. 2002;26:565-573.
PubMed  DOI: 10.1016/S0893-133X(01)00395-5

53. Ase AR, Reader TA, Hen R, Riad M, Descarries L.  Regional changes in density of serotonin transporter in the brain of 5-HT1A and 5-HT1B knockout mice, and of serotonin innervation in the 5-HT1B knockout. J Neurochem. 2001;78:619-630.
PubMed  DOI: 10.1046/j.1471-4159.2001.00437.x

54. Daws LC, Gerhardt GA, Frazer A.  5-HT1B antagonists modulate clearance of extracellular serotonin in rat hippocampus. Neurosci Lett. 1999;266:165-168.
PubMed  DOI: 10.1016/S0304-3940(99)00277-3

55. Daws LC, Gould GG, Teicher SD, Gerhardt GA, Frazer A.  5-HT(1B) receptor-mediated regulation of serotonin clearance in rat hippocampus in vivo. J Neurochem. 2000;75:2113-2122.
PubMed  DOI: 10.1046/j.1471-4159.2000.0752113.x

56. Malagié I, David DJ, Jolliet P, Hen R, Bourin M, Gardier AM.  Improved efficacy of fluoxetine in increasing hippocampal 5-hydroxytryptamine outflow in 5-HT1B receptor knock-out mice. Eur J Pharmacol. 2002;443:99-104.
PubMed  DOI: 10.1016/S0014-2999(02)01604-7

57. Neumaier JF, Root DC, Hamblin MW.  Chronic fluoxetine reduces serotonin transporter mRNA and 5-HT1B mRNA in a sequential manner in the rat dorsal raphe nucleus. Neuropsychopharmacology. 1996;15:515-522.
PubMed  DOI: 10.1016/S0893-133X(96)00095-4

58. Artigas F, Perez V, Alvarez E.  Pindolol induces a rapid improvement of depressed patients treated with serotonin reuptake inhibitors. Arch Gen Psychiatry. 1994;51:248-251.
PubMed 

59. Shalom G, Gur E, Van de Kar LD, Newman ME.  Repeated administration of the 5-HT(1B) receptor antagonist SB-224289 blocks the desensitisation of 5-HT(1B) autoreceptors induced by fluoxetine in rat frontal cortex. Naunyn Schmiedebergs Arch Pharmacol. 2004;370:84-90.
PubMed  DOI: 10.1007/s00210-004-0958-x

60. Ansah TA, Ramamoorthy S, Montanez S, Daws LC, Blakely RD.  Calcium-dependent inhibition of synaptosomal serotonin transport by the alpha 2-adrenoceptor agonist 5-bromo-N-[4,5-dihydro-1H-imidazol-2-yl]-6-quinoxalinamine (UK14304). J Pharmacol Exp Ther. 2003;305:956-965.
PubMed  DOI: 10.1124/jpet.102.047134

61. Ramamoorthy S, Bauman AL, Moore KR, et al.  Antidepressant- and cocaine-sensitive human serotonin transporter: molecular cloning, expression, and chromosomal localization. Proc Natl Acad Sci USA. 1993;90:2542-2546.
PubMed 

62. Heils A, Teufel A, Petri S, et al.  Allelic variation of human serotonin transporter gene expression. J Neurochem. 1996;66:2621-2624.
PubMed 

63. Lesch KP, Bengel D, Heils A, et al.  Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region. Science. 1996;274:1527-1531.
PubMed  DOI: 10.1126/science.274.5292.1527

64. Rausch JL, Johnson ME, Fei YJ, et al.  Initial conditions of serotonin transporter kinetics and genotype: influence on SSRI treatment trial outcome. Biol Psychiatry. 2002;51:723-732.
PubMed  DOI: 10.1016/S0006-3223(01)01283-5

65. Yu YW, Tsai SJ, Chen TJ, Lin CH, Hong CJ.  Association study of the serotonin transporter promoter polymorphism and symptomatology and antidepressant response in major depressive disorders. Mol Psychiatry. 2002;7:1115-1119.
PubMed  DOI: 10.1038/sj.mp.4001141

66. Pollock BG, Ferrell RE, Mulsant BH, et al.  Allelic variation in the serotonin transporter promoter affects onset of paroxetine treatment response in late-life depression. Neuropsychopharmacology. 2000;23:587-590.
PubMed  DOI: 10.1016/S0893-133X(00)00132-9

67. Durham LK, Webb SM, Milos PM, Clary CM, Seymour AB.  The serotonin transporter polymorphism, 5HTTLPR, is associated with a faster response time to sertraline in an elderly population with major depressive disorder. Psychopharmacology (Berl). 2004;174:525-529.
PubMed 

68. Ozaki N, Goldman D, Kaye WH, et al.  Serotonin transporter missense mutation associated with a complex neuropsychiatric phenotype. Mol Psychiatry. 2003;8:933-936.
DOI: 10.1038/sj.mp.4001365

69. Kilic F, Murphy DL, Rudnick G.  A human serotonin transporter mutation causes constitutive activation of transport activity. Mol Pharmacol. 2003;64:440-446.
PubMed  DOI: 10.1124/mol.64.2.440

70. Smith RS.  The macrophage theory of depression. Med Hypotheses. 1991;35:298-306.
PubMed  DOI: 10.1016/0306-9877(91)90272-Z

71. Connor TJ, Leonard BE.  Depression, stress and immunological activation: the role of cytokines in depressive disorders. Life Sci. 1998;62:583-606.
PubMed  DOI: 10.1016/S0024-3205(97)00990-9

72. Pfennig A, Kunzel HE, Kern N, et al.  Hypothalamus-pituitary-adrenal system regulation and suicidal behavior in depression. Biol Psychiatry. 2005;57:336-342.
PubMed  DOI: 10.1016/j.biopsych.2004.11.017

73. Capuron L, Ravaud A, Neveu PJ, Miller AH, Maes M, Dantzer R.  Association between decreased serum tryptophan concentrations and depressive symptoms in cancer patients undergoing cytokine therapy. Mol Psychiatry. 2002;7:468-473.
PubMed  DOI: 10.1038/sj.mp.4000995

74. Capuron L, Neurauter G, Musselman DL, et al.  Interferon-alpha-induced changes in tryptophan metabolism: relationship to depression and paroxetine treatment. Biol Psychiatry. 2003;54:906-914.
PubMed  DOI: 10.1016/S0006-3223(03)00173-2

75. Schiepers OJ, Wichers MC, Maes M.  Cytokines and major depression. Prog Neuropsychopharmacol Biol Psychiatry. 2005;29:201-217.
PubMed  DOI: 10.1016/j.pnpbp.2004.11.003

76. Bschor T, Baethge C, Adli M, Lewitzka U, Eichmann U, Bauer M.  Hypothalamic-pituitary-thyroid system activity during lithium augmentation therapy in patients with unipolar major depression. J Psychiatry Neurosci. 2003;28:210-216.
PubMed 

77. Sauvage MF, Marquet P, Rousseau A, Raby C, Buxeraud J, Lachatre G.  Relationship between psychotropic drugs and thyroid function: a review. Toxicol Appl Pharmacol. 1998;149:127-135.
PubMed  DOI: 10.1006/taap.1998.8367

78. Shelton RC, Winn S, Ekhatore N, Loosen PT.  The effects of antidepressants on the thyroid axis in depression. Biol Psychiatry. 1993;33:120-126.
PubMed  DOI: 10.1016/0006-3223(93)90311-Z

79. Duval F, Mokrani MC, Crocq MA, et al.  Effect of antidepressant medication on morning and evening thyroid function tests during a major depressive episode. Arch Gen Psychiatry. 1996;53:833-840.
PubMed 

80. Nibuya M, Nestler EJ, Duman RS.  Chronic antidepressant administration increases the expression of cAMP response element binding protein (CREB) in rat hippocampus. J Neurosci. 1996;16:2365-2372.
PubMed 

81. Wallace TL, Stellitano KE, Neve RL, Duman RS.  Effects of cyclic adenosine monophosphate response element binding protein overexpression in the basolateral amygdala on behavioral models of depression and anxiety. Biol Psychiatry. 2004;56:151-160.
PubMed  DOI: 10.1016/j.biopsych.2004.04.010

82. Shimizu E, Hashimoto K, Okamura N, et al.  Alterations of serum levels of brain-derived neurotrophic factor (BDNF) in depressed patients with or without antidepressants. Biol Psychiatry. 2003;54:70-75.
PubMed  DOI: 10.1016/S0006-3223(03)00181-1

83. Duman RS.  Role of neurotrophic factors in the etiology and treatment of mood disorders. Neuromolecular Med. 2004;5:11-25.
PubMed  DOI: 10.1385/NMM:5:1:011

84. Coppell AL, Pei Q, Zetterstrom TS.  Bi-phasic change in BDNF gene expression following antidepressant drug treatment. Neuropharmacology. 2003;44:903-910.
PubMed  DOI: 10.1016/S0028-3908(03)00077-7

85. Conti AC, Cryan JF, Dalvi A, Lucki I, Blendy JA.  cAMP response element-binding protein is essential for the upregulation of brain-derived neurotrophic factor transcription, but not the behavioral or endocrine responses to antidepressant drugs. J Neurosci. 2002;22:3262-3268.
PubMed 

86. Pullar IA, Boot JR, Broadmore RJ, et al.  The role of the 5-HT1D receptor as a presynaptic autoreceptor in the guinea pig. Eur J Pharmacol. 2004;493:85-93.
PubMed  DOI: 10.1016/j.ejphar.2004.04.029

87. Young AH, Gallagher P, Watson S, Del-Estal D, Owen BM, Nicol Ferrier I.  Improvements in neurocognitive function and mood following adjunctive treatment with mifepristone (RU-486) in bipolar disorder. Neuropsychopharmacology. 2004;29:1538-1545.
PubMed  DOI: 10.1038/sj.npp.1300471

88. Bao AM, Hestiantoro A, Van Someren EJ, Swaab DF, Zhou JN.  Colocalization of corticotropin-releasing hormone and oestrogen receptor-{alpha} in the paraventricular nucleus of the hypothalamus in mood disorders. Brain. 2005;

89. Dableh LJ, Yashpal K, Rochford J, Henry JL.  Antidepressant-like effects of neurokinin receptor antagonists in the forced swim test in the rat. Eur J Pharmacol. 2005;507:99-105.
PubMed  DOI: 10.1016/j.ejphar.2004.11.024