Context
The pathophysiology of major depressive disorder (MDD) includes disturbances in several neuroanatomical substrates and neurotransmitter systems. The challenge is to elucidate the brain mechanisms of MDD behavioral symptoms, chiefly those of anhedonia.
Objectives
To visualize the neuroanatomical substrates implicated in altered reward processing in MDD, using functional magnetic resonance imaging in combination with a dopaminergic probe (a 30-mg dose of oral dextroamphetamine sulfate) to stimulate the brain reward system; and to test the hypothesis that a hypersensitive response to dextroamphetamine in MDD involves the prefrontal cortex and the striatum.
Design and Interventions
Among subjects with MDD and healthy control subjects, functional magnetic resonance imaging data were collected before and after single-blind administration of dextroamphetamine.
Setting
Subjects were recruited through local newspaper advertisements and by word of mouth.
Participants
Twelve depressed subjects (mean age, 34.83 years; male-female ratio, 6:6) met criteria for MDD according to the DSM-IV, were not taking antidepressants, and had no comorbid Axis I disorders. Twelve control subjects (mean age, 29.33 years; male-female ratio, 5:7) were healthy volunteers without a history of Axis I disorders.
Main Outcome Measures
Functional magnetic resonance imaging blood oxygen level–dependent activation was measured during a controlled task, and dextroamphetamine-induced subjective effects were assessed using the Addiction Research Center Inventory.
Results
Subjects with MDD had a hypersensitive response to the rewarding effects of dextroamphetamine (2-fold increase; t21 = 2.74, P = .01), with altered brain activation in the ventrolateral prefrontal cortex and the orbitofrontal cortex and the caudate and putamen (F1,44 = 11.93, P = .001).
Conclusion
Dopamine-related neuroanatomical substrates are involved in altered reward processing in MDD, shedding light on the neurobiology of the anhedonic symptoms in MDD and suggesting these substrates as future therapeutic targets.
Depressed mood and markedly diminished interest in previously enjoyed activities (ie, anhedonia) are key characteristics of major depressive disorder (MDD). Accompanying symptoms required for the diagnosis, including guilt, anxiety, sleep disturbances, and appetite changes, and suicidal ideation render it challenging to elucidate the central neurobiological disturbances in this costly and prevalent mood disorder.1,2 Previous studies3-9 focused on characterizing brain mechanisms of the disorder and found that MDD is associated with altered brain activity in a wide network of regions, including the prefrontal cortex, limbic regions, and basal ganglia, and with abnormalities in several neurotransmitter and neuroendocrine systems, including serotonin, dopamine, norepinephrine, and the hypothalamic-pituitary-adrenal axis. The relationship between these neurobiological disturbances and the specific behavioral symptoms of depression remains unclear.
The brain reward system (BRS) can be defined as a neurobiological system that consists of extensive neural pathways that mediate reward behaviors such as pleasure and motivation.10-12 Rewards serve to elicit approach and consummatory behaviors, increase the frequency and intensity of the behaviors, maintain the behaviors, prevent extinction, and induce subjective feelings of pleasure or positive emotional states.13,14 In light of our previous findings,15 we proposed that the BRS may represent a key neuroanatomical substrate of anhedonia, a core symptom of depression. The BRS of patients diagnosed as having MDD was examined in that study by measuring the behavioral response to dextroamphetamine sulfate. As cocaine, this psychostimulant induces dopamine release in the mesocorticolimbic dopamine system, a major pathway of the BRS,16-18 and elicits well-detected and measurable rewarding effects (eg, stimulation and euphoria).19 Depressed patients with severe symptoms reported significantly enhanced dextroamphetamine-induced rewarding effects, indicating altered reward processing in MDD.15
Scientists have recently begun to visualize the human neuroanatomical substrates associated with positive subjective experiences from reinforcing drugs such as nicotine and dextroamphetamine, as well as from financial and food rewards, using techniques such as functional magnetic resonance imaging (fMRI) and positron emission tomography.14,20-27 The importance of mesocorticolimbic dopamine in the BRS has been observed in multiple studies with human subjects.28-31 The primary objective of the present study was to use dextroamphetamine as a dopaminergic BRS probe in association with fMRI to isolate the neuroanatomical substrates of the altered BRS response in MDD. We chose this pharmacological fMRI32 approach because of its superior temporal and spatial resolution for visualizing brain activity and because of its avoidance of use of radioactive tracers. We also tested the hypothesis that altered BRS activity is associated with a disruption in a spatially distributed network of dopamine-rich brain regions such as the striatum and the limbic and prefrontal cortical regions.
The inclusion criteria for the MDD and control groups were age between 18 and 65 years, willingness to participate, and ability to provide signed informed consent. The protocol was approved by the research ethics board of the Sunnybrook and Women’s College Health Sciences Centre. Subjects were excluded if they had a current or past cardiovascular disorder, had a medical condition requiring immediate investigation or treatment, were pregnant or lactating, were taking a medication known to interact with the study drug (confirmed by urinalysis), had a neurological disorder (eg, Parkinson disease), or had conditions for which an MRI would be contraindicated (eg, metallic implants).
Depressed subjects were recruited through local newspaper advertisements. After obtaining informed consent, a trained researcher (L.K.T. or U.E.B.) assessed potential subjects with the Structured Clinical Interview for Axis I DSM-IV Disorders (SCID-IV).33 This was followed by a medical consultation with a staff psychiatrist (N.H.). A urine sample was collected for drug toxicologic screening. Subjects were excluded if they were taking antidepressant medication, had a Hamilton Depression Rating Scale34 (HDRS) score of less than 24 (subjects with moderate MDD were excluded because these patients responded similarly to control subjects in a previous study15), had suicidal ideation posing an immediate threat to life, or had a concurrent Axis I disorder such as anxiety disorders and schizophrenia. Of 265 advertisement responders, 49 were scheduled for the SCID-IV, 168 were ineligible (mostly because of current antidepressant use) or refused to participate, and 48 could not be contacted. Of the 49 scheduled for the SCID-IV, 12 completed the present study, 14 completed another study, and 23 were ineligible or dropped out.
Healthy control subjects were recruited by electronic mail advertising locally in the hospital surroundings and by word of mouth. After providing informed consent, they were screened using the SCID-IV and the HDRS. A urine sample was collected for drug toxicologic screening. The exclusion criteria for the control subjects were a HDRS score of greater than 6, a history of mood disorders, or any other DSM-IV, Axis I mental illness.
Potential subjects were screened by telephone using a standard screening form based on the criteria described in the previous subsection, the HDRS, and an MRI clinical screening form. They were then invited for an assessment session that included informed consent, the SCID-IV, a medical consultation, and urine sampling. Eligible subjects were recruited for a study session day.
At the beginning of the study session day, a battery of symptom assessment scales was administered, including the Beck Depression Inventory,35 the Snaith-Hamilton Pleasure Scale,36 and a modified version of the Sunnybrook Psychomotor Agitation and Retardation Questionnaire,37 to evaluate depression, anhedonia, and psychomotor symptom severity, respectively. The severity of the depressive episodes experienced by subjects during the 2 to 3 weeks before the study session day was evaluated using the HDRS.34 The Apathy Evaluation Scale38 was used to better characterize anhedonic symptoms. The syndrome of apathy is defined as a lack of motivation without evidence of diminished level of consciousness, emotional distress, or cognitive deficits. This syndrome can occur in association with stroke, dementia, and depression. Items in this scale include negative symptoms experienced in depression such as diminished work or interest, psychomotor retardation, anergy, and lack of insight.39 In a depressed patient, the HDRS–Apathy Evaluation Scale ratio must be greater than 3.5 to diagnose apathy.
Thirty-milligram oral dextroamphetamine doses, prepared in identical 10-mg capsules filled with drug or dextrose powder, were obtained from the hospital pharmacy and administered in a single-blind fashion to all subjects. The Addiction Research Center Inventory (ARCI),40-42 a drug effect measurement tool, was administered at baseline and was repeated 40, 80, and 140 minutes after drug administration. The ARCI, the main outcome measure of the rewarding effects of dextroamphetamine in this study, is a well-validated questionnaire designed to measure positive effects of drugs that are reinforcing (ie, can promote drug self-administration) and negative effects (ie, increased anxiety and agitation). Specific sets of these questions (eg, “I feel now as I have felt after a very exciting experience; I feel so good I know people can tell it”) comprise empirically derived scale classes (eg, Amphetamine and Stimulation Euphoria) validated to measure characteristic effects of drugs or drug.
Blood pressure and heart rate were measured at baseline and 140 minutes after drug administration for physiological assessment of the effects of dextroamphetamine and for safety monitoring. A blood sample was drawn 140 minutes after drug administration, and concentrations were determined using gas chromatography–mass spectrometry.43
Functional magnetic resonance imaging was performed using a research-dedicated whole-body MRI scanner operating at 1.5 T (software version LX 8.4; General Electric Medical Systems, Milwaukee, Wis). Sagittal multisection “scout” images to localize brain anatomy (256 × 192-pixel matrix, 1-cm thick) and axial, high spatial resolution, 3-dimensional images (1.2-mm thick) were acquired using fast spoiled-gradient echo imaging. Functional magnetic resonance images were obtained using T2-weighted single-shot spiral k-space readout and off-line reconstruction44 (echo time, 40 milliseconds; repetition time, 2000 milliseconds; 80º angle; 20-cm field of view; and 90 × 90-pixel matrix), a technique that is less sensitive to patient head motion than the more commonly used echo planar imaging.45
A set of fMRI experiments was performed twice for all subjects, before drug administration and then at peak drug effect (90 minutes after drug administration). In each set, fMRI was performed with the International Affective Picture System (IAPS)46, in a block design. The IAPS task (described in the next paragraph) was used for 3 reasons. First, the IAPS task engages several brain regions47 that are hypothesized to be involved in the BRS dysfunction in MDD. Second, the IAPS task provides a set of validated stimuli46 that help evoke well-characterized and controlled behaviors in healthy adults, while minimizing unwanted interindividual effects (eg, somnolence and restlessness) that might occur in a resting state (without a task). Third, the IAPS task acts as a probe of relevant behaviors, making it possible to observe the action of drugs during a much longer period than has been possible in past pharmacological fMRI experiments.21,23 Among patients with present or past drug addiction, these studies measured fluctuations in blood oxygen level–dependent (BOLD) signals corresponding immediately to intravenous drug administration (pharmacokinetic approach). In the present study, using an intravenous psychostimulant in “inexperienced” subjects would have compromised subject safety and recruitment.
The IAPS task consisted of rating a set of validated pictures containing neutral, positive, and negative images of humans, animals, and objects. The IAPS task was programmed using specialized software (E-Prime, version 1, beta 4; Psychology Software Tools, Inc, Pittsburgh, Pa) to ensure precise timing of stimulus presentation and to measure accuracy, picture ratings, and reaction times. An fMRI-compatible visual stimulus presentation system was used, which included a set of goggles connected to an LCD projector (Silent Vision, model SV022; Avotec, Inc, Jensen Beach, Fla). The goggles contained adjustments for visual acuity. Subject responses were recorded using fiberoptic-coupled keypads containing 2 buttons for each hand (Lumitouch; Light Wave Technology Inc, Surrey, British Columbia), providing measurements of accuracy and reaction times. Subjects were instructed to interpret or capture the images, presented in blocks of neutral, positive, and negative emotional content (4.5 seconds per picture and 4 pictures per block), and to rate the level of pleasantness for each picture by pressing 1 of the 4 available buttons (unpleasant, somewhat unpleasant, somewhat pleasant, or pleasant). Each picture block (3 neutral, 3 positive, and 3 negative) alternated with blocks of control stimuli. Control stimuli consisted of fixation crosses instead of pictures. The total duration of the task was approximately 8 minutes. The selected pictures and the order of block presentation had been shown to activate brain regions important to the study, including subregions of the prefrontal cortex (eg, dorsolateral), cingulate, amygdala, and nucleus accumbens.47 The order of block presentation (neutral, positive, and negative) was used to allow time for subjects to process the emotions and to minimize overlap. The IAPS stimuli have been validated in terms of the affect and the degree of arousal elicited in healthy control subjects.48 Pictures used in the task before drug administration were different from those used in the task at peak drug effect. However, each set of pictures contained similar themes (eg, 2 different pictures of children). Pictures were not repeated within tasks for the same subject.
In addition, 2 simple control tasks were administered at baseline and at peak drug effect to investigate potential drug effects on simple visual sensory and motor output. Finger tapping was performed to visualize the primary sensorimotor cortex and supplementary motor area, and a mondrianlike colored pattern was flickered at a rate of 8 Hz to visualize the primary visual cortex and other visual association areas. Both tasks were implemented in a block design (15-second “on condition,” 15-second “off condition,” and 8 repetitions), with approximately 4 minutes for each task. The off condition consisted of a blank screen. Because the brain regions activated with these tasks were not expected to be sensitive to dextroamphetamine administration, any observed changes would provide insight regarding whether dextroamphetamine significantly affected the neurovascular coupling that underlies BOLD signals.
Statistical analysis of the behavioral data was conducted using a standard software package (SPSS for Windows, version 10.0.0; SPSS Inc, Chicago, Ill). For the dextroamphetamine behavioral effects, the main dependent outcome variable consisted of a composite of change scores from previously described subscales of the ARCI that measure positive reinforcing effects, namely, abuse-potential, amphetamine, benzedrine, morphine-benzedrine, and stimulation euphoria.15 The peak dextroamphetamine effect was defined as the highest scale score among the 80- and 140-minute recordings. The corresponding baseline score was subtracted from this value to measure the change. Because of the different score ranges within the various scales, baseline and peak scores were converted to a score on a 100% scale before being added to the composite score. These ARCI data were analyzed using an independent t test contrasting the MDD subjects with the control subjects. Similarly to the ARCI scores, the baseline (predrug) picture ratings and reaction times were subtracted from the corresponding postdrug scores, which were then assessed in an independent t test contrasting the MDD subjects with the control subjects. Pearson product moment correlation was used for bivariate tests. Finally, independent t tests were performed to compare demographic and baseline measurements, including age and symptom scale scores, between the MDD subjects and the controls.
The imaging data were assessed with the Analysis of Functional Neuroimages statistical program.49 Data sets were corrected for motion, detrended to eliminate slow global changes in the fMRI signal, transformed into Talairach coordinates, and spatially filtered (using a gaussian filter with a full width at half maximum of 6 mm). Brain activation maps were generated for each data set (at baseline and at peak drug effect for the MDD subjects and the controls) using the conventional block design correlation method, which involves contrasting blocks of task performance (on condition vs off condition).50 For the IAPS task, all picture blocks (on condition) were contrasted with the blocks of fixation crosses (off condition). Therefore, signal intensity values represented the percentage change in BOLD signal between the on and off conditions. The brain activation maps were submitted to a 2-way analysis of variance in the Analysis of Functional Neuroimages program to assess the effects of mood (contrasting the MDD subjects vs the controls) and drug (contrasting before administration vs after administration), as well as the interaction between these 2 independent factors. We report the results at an uncorrected threshold of P = .001 in table format but use P = .01 for illustration purposes. In each region of interaction, there were the following 3 possible scenarios: (1) There was no predrug difference between the MDD subjects and the controls, whereas there was a statistically significant postdrug difference. (2) Predrug and postdrug differences occurred, but the magnitude of the difference was statistically different between the MDD subjects and the controls. (3) There was a statistically significant predrug difference between the MDD subjects and the controls but no postdrug difference. Therefore, to best answer the question concerning the direction of change in the regions of interaction, post hoc t tests comparing the MDD group with the control group before drug administration and after drug administration were performed with the Analysis of Functional Neuroimages program (at an uncorrected threshold of P = .001 and illustrated at P = .01).
Twelve subjects with severe MDD and 12 healthy control subjects completed the study. Table 1 gives the characteristics of the study population. Between-group t tests showed that the differences between the groups were statistically significant (P<.05) for all symptom assessment measures. In the MDD group, the duration of episodes reported during the SCID-IV interviews ranged from 2 to 24 months (mean, 13.75 months). Seven of 12 patients were experiencing their first MDD episode, and 2 of the 12 had been prescribed an antidepressant (2 days previously in one subject and 1 month previously in the other subject). No MDD subjects had been hospitalized for mental illness, and most reported good physical health except for migraine headaches (n = 2) and stomach ulcer (n = 1).
Independent t tests comparing the ARCI composite scores of the MDD subjects vs the controls revealed that the MDD subjects had a hypersensitive response to the rewarding effects of dextroamphetamine (2-fold increase; t22 = 1.92, P = .07), replicating previous findings.19 The difference became statistically significant (t21 = 2.74, P = .01) with the removal of an outlier in the MDD group, who reported dextroamphetamine effects at least 2 SDs below the mean. The mean ARCI composite scores of the MDD group and the control group in the present study were statistically equivalent to previous findings.15 Except for age (our study group was younger), no statistical differences between the studies were found in predrug or postdrug measurements. Pearson product moment correlation tests revealed positive correlations between ARCI scores and measures of anhedonia, including the Snaith-Hamilton Pleasure Scale (n = 23; r = 0.58, P = .005) and the Apathy Evaluation Scale (n = 23; r = 0.43, P = .049). Therefore, the degree of dextroamphetamine reward was associated with the severity of anhedonia.
Table 2 gives the IAPS task results and the physiological measures. The t tests comparing baseline picture ratings and reaction times (postdrug scores minus predrug scores) in the MDD subjects vs the controls showed no differences during the IAPS task. The ratings of the pleasantness of the IAPS pictures were equivalent between the MDD subjects and the controls before and after drug administration. Therefore, neither mood nor drug influenced these behavioral measures. Heart rate and blood pressure were increased by dextroamphetamine (F = 9.35, P = .006), but no differences were found between the MDD subjects and the controls.
To compare differences in brain activation between the MDD subjects and the controls, we examined the interaction between mood and drug (using the 2-way analysis of variance model described in the last paragraph of the “Methods” section). The resulting regions of interaction with corresponding Talairach coordinates (x, y, and z) are given in Table 3. The patterns of change in these regions of interaction were explored by means of post hoc t tests, revealing no differences between the MDD subjects and the controls before drug administration, except for the right posterior cingulate, but revealing significant differences after drug administration. Therefore, the interaction corresponded to the first of the 3 possible scenarios described at the end of the “Methods” section. The last column of Table 3 gives the decreases in brain activation in the MDD subjects vs the controls, which were seen in all regions of interaction except those near Brodmann area (BA) 38, which showed increases compared with the controls. The decreases were caused by mean deactivations (negative BOLD signals) in the MDD subjects vs mean activations (positive BOLD signals) in the controls, whereas the increases were caused by the opposite, that is, activations in the MDD subjects and deactivations in the controls.
Results of the Pearson product moment correlation tests in the regions of interaction showed statistically significant correlations between brain activity (peak BOLD signal changes minus baseline values) and degree of dextroamphetamine reward (peak ARCI scores minus baseline values) in the ventrolateral prefrontal cortex (BA 10) (n = 23; r = −0.62, P = .002) and the premotor cortex (BA 6) (n = 23; r = −0.43, P = .04). In addition, a trend was observed in the caudate and putamen (n = 23; r = 0.36, P = .09). The effects of the mood and drug independent factors are given in Table 4. Neither of the simple control tasks (finger tapping or visual stimuli) showed statistically significant mood, drug, or interaction effects.
The MDD group demonstrated enhanced dextroamphetamine-induced rewarding effects compared with the control group, reproducing previous findings.15 This hypersensitive response correlated with the severity of anhedonic symptoms, providing further support for the involvement of the BRS in the pathophysiology of MDD. Combining the dopaminergic probe (dextroamphetamine) with fMRI enabled us to determine the neuroanatomical substrates of altered reward processing in the MDD subjects vs the controls. The link between dextroamphetamine-induced dopamine release and rewarding effects in humans is well established.16,17,51,52 Therefore, results of this study provide support for the involvement of dopaminergic mechanisms in the altered reward processing in MDD.
The neuroanatomical substrates shown in the Figure, particularly the caudate and putamen and the prefrontal region (BA 10), are regions that mediate reward and contain mesocorticolimbic dopaminergic neurons.53-58 In our study reward scores (eg, euphoria and increased energy) of depressed patients correlated with underlying changes in brain activity in certain dopamine-rich regions, including BAs 10 and 6 and the caudate and putamen. The present study showed that this hypersensitive dextroamphetamine response was associated with negative BOLD signals rather than increased brain activation in most of the regions of interaction compared with the controls (Table 3, last column). The physiological mechanisms of negative BOLD signals are thought to be induced by reduced blood flow (ie, active neuronal inhibition and decreased cortical excitability).59-61 The relative decrease in brain activity among depressed subjects could reflect exaggerated deactivation of glutamate–mediated transmission by amphetamine. According to Paladini et al,62 amphetamine selectively inhibits metabotropic glutamate receptor–mediated inhibitory postsynaptic potentials in dopamine neurons, which subsequently induces burst firing of dopamine neurons. Therefore, depressed subjects in our study may have experienced enhanced behavioral effects due to abnormal disinhibition of dopaminergic neurons. Evidence of abnormal glutamate transmission in MDD has been reported,63 and disturbances in glutamate-mono amine interactions have been proposed as pathophysiological mechanisms in mental illnesses such as schizophrenia and depression.64,65 Furthermore, N-methyl-D-aspartate receptor antagonists (metatropic glutamate receptor subtypes 1 and 5) have antidepressant properties that may be exploited for future pharmacotherapeutic strategies.66 As previously proposed,15 an alternative to the glutamatergic inhibition concept is that the hypersensitive dextroamphetamine behavioral response in depressed patients is mediated by stimulation of a higher density of D2 dopamine receptors67-69 or by stimulation in synapses containing a lower density of dopamine transporters,70,71 which are thought to be neurobiological responses secondary to a chronic hypodopaminergic state in depression. Negative BOLD signals in the prefrontal regions and striatum may be due to increased D2 receptor–mediated inhibition.64 However, the notion of a dopamine storage deficit was refuted by findings that experimental depletion of dopamine in unmedicated MDD patients does not produce exacerbation of mood symptoms.72 Furthermore, Parsey et al73 found that depressed patients at baseline showed striatal D2 dopamine receptor availability that was similar to that in controls and that amphetamine-induced displacement of the D2 ligand (indicating dopamine release) was not different in the striatum between patients and controls.
Unlike the previously discussed regions of interaction such as the caudate and putamen, a different pattern occurred in the remaining regions of interaction, that is, increased brain activity with dextroamphetamine in the MDD subjects compared with the controls in the temporal pole (BA 38) and increased activity before dextroamphetamine administration followed by normalization (ie, a response equal to that of controls) with dextroamphetamine in the posterior cingulate. One report indicates that cortical gray matter volumes in temporal lobes correlate with degrees of cocaine euphoria.74 D2/D3 dopamine receptor densities in the temporal cortex represent a source of extrastriatal dopamine and are an important therapeutic mechanism of atypical antipsychotic medications.75,76 Decreased availability of these extrastriatal dopamine receptors has been linked to negative symptoms of schizophrenia such as anhedonia.77 Antipsychotic medications are also used to augment the efficacy of antidepressants78 and have in a few cases possibly induced manic or hypomanic symptoms.79 A few reports link the posterior cingulate to reward mediation26 via dopaminergic mechanisms.80,81 Similar to the pattern in this study, normalization of activity has been induced with antidepressant treatment,82 whereas tryptophan depletion (ie, induction of depressive symptoms) has been linked to increased posterior cingulate activation.83
The overall differences in brain activation found between the MDD subjects and the controls (Table 4) were consistent with previous neuroimaging findings in depressed patients at rest (ie, without pharmacological probing), including the prefrontal cortex (BAs 10, 8, 6, 47, and 45), cingulate gyrus (BA 24), basal ganglia (caudate, putamen, and globus pallidus), and subcortical structures such as the hypothalamus.4,9,84 Furthermore, the overall effect of dextroamphetamine involved regions that have been linked to short-term rewarding effects of psychostimulants, including the striatum (caudate and putamen and nucleus accumbens), cingulate, and frontal cortex.16,85,86 Therefore, the main effects given in Table 4 demonstrate that our pharmacological fMRI protocol provided consistent neuroimaging data and reliable interaction findings. The results of the control tasks involving finger tapping and visual stimuli also support the interpretation of the data. The associated activation maps, which primarily depicted primary sensory and motor cortices, showed no effect of dextroamphetamine on these tasks, as anticipated. This lends additional credence to the observation and interpretation of activity within regions associated with the BRS that are specific to dextroamphetamine administration. Compliance was excellent in the MDD and control groups, with the groups performing the tasks with similar picture ratings and reaction times (Table 2).
Elucidating the neuroanatomical substrates of altered reward processing in MDD provides potential targets for the development of more effective antidepressants. Disturbances in dopaminergic and other BRS pathways (eg, opioids) have been found in animal models of depression (eg, chronic mild stress).87 Findings in animal models also suggest that the mesocorticolimbic dopamine pathway may be a potential target for more rapid onset of action,88 a significant problem in current antidepressant treatment. However, interpretations involving single neurotransmitters are not ideal, in light of the ability of dextroamphetamine to modulate the release of neurotransmitters other than dopamine, as well as the involvement of multiple neurotransmitter systems in the neurobiology of the BRS and MDD.19 Instead of the continued use of single or multiple neurotransmitter system targeting, perhaps a more promising treatment approach would be to consider the BRS, specifically the regions identified in this study, as a target for future “plasticity-enhancing” drugs such as N-methyl-D-aspartate receptor antagonists. This approach may be more in line with the complex pathophysiology of MDD.66,89 Findings in animal models suggest that antidepressants in general produce neurotrophic and neuroplastic effects (eg, increases in brain-derived neurotrophic factor), including neurogenesis, which could explain the 6 to 8 weeks required for a therapeutic response and the alleviation of depressive symptoms.90 The dopaminergic system and the BRS components (eg, the nucleus accumbens and the caudate and putamen) exhibit important plastic properties, as seen after repeated drug administration,91 with environmental changes,92 and during postnatal development.93 Therefore, the brain reward system may be an excellent target for novel treatment approaches in drug development research.
Correspondence: Usoa E. Busto, PharmD, Clinical Neuroscience Department, Centre for Addiction and Mental Health, 33 Russell St, Room 1053, Toronto, Ontario, Canada M5S 2S1 (usoa_busto@camh.net).
Submitted for Publication: March 31, 2004; final revision received March 29, 2005; accepted April 28, 2005.
Funding/Support: This study was supported in part by a grant from the Canadian Institutes of Health Research, Ottawa, Ontario. Dr Tremblay was supported by the Ontario Graduate Scholarship and the Ben Cohen Bursary Fund, University of Toronto.
Previous Presentation: This study was presented in part at the 106th Annual Meeting of the American Society for Clinical Pharmacology and Therapeutics; March 25, 2002; Atlanta, Ga; and at the 32nd Annual Meeting of the Society for Neuroscience; November 4, 2002; Orlando, Fla.
Acknowledgment: We thank Anthony J. Levitt, MD, FRCPC, and Stephen Sokolov, MD, FRCPC, for their help in the subject assessments.
1.American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington, DC American Psychiatric Association1994;161- 198
2.Greden
JF The burden of disease for treatment-resistant depression.
J Clin Psychiatry 2001;6226- 31
PubMedGoogle Scholar 4.Videbech
P PET measurements of brain glucose metabolism and blood flow in major depressive disorder: a critical review.
Acta Psychiatr Scand 2000;10111- 20
PubMedGoogle ScholarCrossref 5.Brody
ALBarsom
MWBota
RGSaxena
S Prefrontal-subcortical and limbic circuit mediation of major depressive disorder.
Semin Clin Neuropsychiatry 2001;6102- 112
PubMedGoogle ScholarCrossref 6.Leonard
BE Evidence for a biochemical lesion in depression.
J Clin Psychiatry 2000;6112- 17
PubMedGoogle Scholar 7.Galard
RCatalan
RCastellanos
JMHallart
JM Plasma corticotropin-releasing factor in depressed patients before and after the dexamethasone suppression test.
Biol Psychiatry 2002;51463- 468
PubMedGoogle ScholarCrossref 8.Ichimiya
TSuhara
TSudo
YOkubo
YNakayama
KNankai
MInoue
MYasuno
FTakano
AMaeda
JShibuya
H Serotonin transporter binding in patients with mood disorders: a PET study with [
11C](+)McN5652.
Biol Psychiatry 2002;51715- 722
PubMedGoogle ScholarCrossref 9.Mayberg
HS Modulating dysfunctional limbic-cortical circuits in depression: towards development of brain-based algorithms for diagnosis and optimised treatment.
Br Med Bull 2003;65193- 207
PubMedGoogle ScholarCrossref 10.Bozarth
MA Pleasure systems in the brain. Warburton
DMed.
Pleasure The Politics and the Reality New York, NY John Wiley & Sons1994;
Google Scholar 14.Elliott
RFriston
KJDolan
RJ Dissociable neural responses in human reward systems.
J Neurosci 2000;206159- 6165
PubMedGoogle Scholar 15.Tremblay
LKNaranjo
CACardenas
LHerrmann
NBusto
UE Probing the brain reward system in major depressive disorder: altered response to dextroamphetamine.
Arch Gen Psychiatry 2002;59409- 416
PubMedGoogle ScholarCrossref 16.Drevets
WCGautier
CPrice
JCKupfer
DJKinahan
PEGrace
AAPrice
JLMathis
CA Amphetamine-induced dopamine release in human ventral striatum correlates with euphoria.
Biol Psychiatry 2001;4981- 96
PubMedGoogle ScholarCrossref 17.Cardenas
LHoule
SKapur
SBusto
UE Oral
D-amphetamine causes prolonged displacement of [
11C]raclopride as measured by PET.
Synapse 2004;5127- 31
PubMedGoogle ScholarCrossref 19.Naranjo
CATremblay
LKBusto
UE The role of the brain reward system in depression.
Prog Neuropsychopharmacol Biol Psychiatry 2001;25781- 823
PubMedGoogle ScholarCrossref 20.Breiter
HCGollub
RLWeisskoff
RMKennedy
DNMakris
NBerke
JDGoodman
JMKantor
HLGastfriend
DRRiorden
JPMathew
RTRosen
BRHyman
SE Acute effects of cocaine on human brain activity and emotion.
Neuron 1997;19591- 611
PubMedGoogle ScholarCrossref 21.Breiter
HCRosen
BR Functional magnetic resonance imaging of brain reward circuitry in the human.
Ann N Y Acad Sci 1999;877523- 547
PubMedGoogle ScholarCrossref 22.Childress
ARMozley
PDMcElgin
WFitzgerald
JReivich
MO’Brien
CP Limbic activation during cue-induced cocaine craving.
Am J Psychiatry 1999;15611- 18
PubMedGoogle Scholar 23.Stein
EAPankiewicz
JHarsch
HHCho
JKFuller
SAHoffmann
RGHawkins
MRao
SMBandettini
PABloom
AS Nicotine-induced limbic cortical activation in the human brain: a functional MRI study.
Am J Psychiatry 1998;1551009- 1015
PubMedGoogle Scholar 24.Schlaepfer
TEStrain
ECGreenberg
BDPreston
KLLancaster
EBigelow
GEBarta
PEPearlson
GD Site of opioid action in the human brain: mu and kappa agonists’ subjective and cerebral blood flow effects.
Am J Psychiatry 1998;155470- 473
PubMedGoogle Scholar 25.O’Doherty
JKringelbach
MLRolls
ETHornak
JAndrews
C Abstract reward and punishment representations in the human orbitofrontal cortex.
Nat Neurosci 2001;495- 102
PubMedGoogle ScholarCrossref 26.Small
DMZatorre
RJDagher
AEvans
ACJones-Gotman
M Changes in brain activity related to eating chocolate: from pleasure to aversion.
Brain 2001;1241720- 1733
PubMedGoogle ScholarCrossref 27.Knutson
BAdams
CMFong
GWHommer
D Anticipation of increasing monetary reward selectively recruits nucleus accumbens.
J Neurosci 2001;21 (RC159)1- 5
Google Scholar 28.Koepp
MJGunn
RNLawrence
ADCunningham
VJDagher
AJones
TBrooks
DJBench
CJGrasby
PM Evidence for striatal dopamine release during a video game.
Nature 1998;393266- 268
PubMedGoogle ScholarCrossref 29.Breiter
HCAharon
IKahneman
DDale
AShizgal
P Functional imaging of neural responses to expectancy and experience of monetary gains and losses.
Neuron 2001;30619- 639
PubMedGoogle ScholarCrossref 30.Laruelle
MAbi-Dargham
Avan Dyck
CHRosenblatt
WZea-Ponce
YZoghbi
SSBaldwin
RMCharney
DSHoffer
PBKung
HFInnis
RB SPECT imaging of striatal dopamine release after amphetamine challenge.
J Nucl Med 1995;361182- 1190
PubMedGoogle Scholar 31.Volkow
NDWang
GJFischman
MWFoltin
RWFowler
JSAbumrad
NNVitkun
SLogan
JGatley
SJPappas
NHitzemann
RShea
CE Relationship between subjective effects of cocaine and dopamine transporter occupancy.
Nature 1997;386827- 830
PubMedGoogle ScholarCrossref 33.Spitzer
MBGibbon
RLWilliams
MJanet
BW Structured Clinical Interview for Axis I DSM-IV Disorders: Patient Edition (SCID-I/P, Version 2.0). New York Biometrics Research Department, New York State Psychiatric Institute1994;
35.Kendall
PCHollon
SDBeck
ATHammen
CLIngram
RE Issues and recommendations regarding use of the Beck Depression Inventory.
Cogn Ther Res 1987;11289- 299
Google ScholarCrossref 36.Snaith
RPHamilton
MMorley
SHumayan
AHargreaves
DTrigwell
P A scale for the assessment of hedonic tone: the Snaith-Hamilton Pleasure Scale.
Br J Psychiatry 1995;16799- 103
PubMedGoogle ScholarCrossref 37.Glassman
SWestrich
NParker
KLevitt
A Psychomotor function in depression.
Proceedings of the 150th Annual Meeting of the American Psychiatric Association, San Diego, Calif, 17-22 May 1997 Washington, DC American Psychiatric Association1997;
Google Scholar 38.Marin
RSRuth
CBiedrzycki
CFirinciogullari
S Reliability and validity of the Apathy Evaluation Scale.
Psychiatry Res 1991;38143- 162
PubMedGoogle ScholarCrossref 39.Marin
RSFirinciogullari
SBiedzycki
RC The sources of convergence between measures of apathy and depression.
J Affect Disord 1993;28117- 124
PubMedGoogle ScholarCrossref 40.Martin
WRSloan
JWSapira
JDJasinski
DR Physiologic, subjective, and behavioral effects of amphetamine, methamphetamine, ephedrine, phenmetrazine, and methylphenidate in man.
Clin Pharmacol Ther 1971;12245- 258
PubMedGoogle Scholar 41.Cole
JOOrzack
MHBeake
BBird
MBar-Tal
Y Assessment of the abuse liability of buspirone in recreational sedative users.
J Clin Psychiatry 1982;4369- 75
PubMedGoogle Scholar 42.Hill
HEHaertzen
CAWolbach
JA
JrMiner
EJ The Addiction Research Center Inventory: standardization of scales which evaluate subjective effects of morphine, amphetamine, pentobarbital, alcohol, LSD-25, pirahexyl and chloropromazine.
Psychopharmacologia 1963;65167- 183
PubMedGoogle ScholarCrossref 43.Midha
KKMcGilveray
IJCooper
JK A GLC-EDC assay for simultaneous determination of fenfluramine and norfenfluramine in human plasma and urine.
Can J Pharm Sci 1979;1418- 21
Google Scholar 46.Center for the Study of Emotion and Attention (CSEA-MINH), International Affective Picture System: Digitized Photographs. Gainesville Center for Research in Psychophysiology, University of Florida1999;
47.Paradiso
SJohnson
DLAndreasen
NCO’Leary
DSWatkins
GLPonto
LLHichwa
RD Cerebral blood flow changes associated with attribution of emotional valence to pleasant, unpleasant, and neutral visual stimuli in a PET study of normal subjects.
Am J Psychiatry 1999;1561618- 1629
PubMedGoogle Scholar 48.Lang
PJBradley
MMCurthbert
BN International Affective Picture System (IAPS): Instruction Manual and Affective Ratings. Gainesville Center for Research in Psychophysiology, University of Florida1999;
49.Cox
RW AFNI: software for analysis and visualization of functional magnetic resonance neuroimages.
Comput Biomed Res 1996;29162- 173
PubMedGoogle ScholarCrossref 50.Bandettini
PAJesmanowicz
AWong
ECHyde
JS Processing strategies for time-course data sets in functional MRI of the human brain.
Magn Reson Med 1993;30161- 173
PubMedGoogle ScholarCrossref 51.Martinez
DSlifstein
MBroft
AMawlawi
OHwang
DRHuang
YCooper
TKegeles
LZarahn
EAbi-Dargham
AHaber
SNLaruelle
M Imaging human mesolimbic dopamine transmission with positron emission tomography, II: amphetamine-induced dopamine release in the functional subdivisions of the striatum.
J Cereb Blood Flow Metab 2003;23285- 300
PubMedGoogle ScholarCrossref 52.Leyton
MDagher
ABoileau
ICasey
KBaker
GBDiksic
MGunn
RYoung
SNBenkelfat
C Decreasing amphetamine-induced dopamine release by acute phenylalanine/tyrosine depletion: a PET [
11C]raclopride study in healthy men.
Neuropsychopharmacology 2004;29427- 432
PubMedGoogle Scholar 53.Elliott
RNewman
JLLonge
OADeakin
JF Differential response patterns in the striatum and orbitofrontal cortex to financial reward in humans: a parametric functional magnetic resonance imaging study.
J Neurosci 2003;23303- 307
PubMedGoogle Scholar 54.Pochon
JBLevy
RFossati
PLehericy
SPoline
JBPillon
BLe Bihan
DDubois
B The neural system that bridges reward and cognition in humans: an fMRI study.
Proc Natl Acad Sci U S A 2002;995669- 5674
PubMedGoogle ScholarCrossref 55.Zink
CPagnoni
GMartin-Skurski
MChappelow
JBerns
G Human striatal responses to monetary reward depend on saliency.
Neuron 2004;42509- 517
PubMedGoogle ScholarCrossref 56.Haruno
MKuroda
TDoya
KToyama
KKimura
MSamejima
KImamizu
HKawato
M A neural correlate of reward-based behavioral learning in caudate nucleus: a functional magnetic resonance imaging study of a stochastic decision task.
J Neurosci 2004;241660- 1665
PubMedGoogle ScholarCrossref 57.Knutson
BFong
GBennett
SAdams
CHommer
D A region of mesial prefrontal cortex tracks monetarily rewarding outcomes: characterization with rapid event-related fMRI.
Neuroimage 2003;18263- 272
PubMedGoogle ScholarCrossref 58.Ramnani
NMiall
RC Instructed delay activity in the human prefrontal cortex is modulated by monetary reward expectation.
Cereb Cortex 2003;13318- 327
PubMedGoogle ScholarCrossref 59.Shmuel
AYacoub
EPfeuffer
JVan de Moortele
PFAdriany
GHu
XUgurbil
K Sustained negative BOLD, blood flow and oxygen consumption response and its coupling to the positive response in the human brain.
Neuron 2002;361195- 1210
PubMedGoogle ScholarCrossref 60.Hamzei
FDettmers
CRzanny
RLiepert
JBuchel
CWeiller
C Reduction of excitability (“inhibition”) in the ipsilateral primary motor cortex is mirrored by fMRI signal decreases.
Neuroimage 2002;17490- 496
PubMedGoogle ScholarCrossref 62.Paladini
CAFiorillo
CDMorikawa
HWilliams
JT Amphetamine selectively blocks inhibitory glutamate transmission in dopamine neurons.
Nat Neurosci 2001;4275- 281
PubMedGoogle ScholarCrossref 63.McCullumsmith
REMeador-Woodruff
JH Striatal excitatory amino acid transporter transcript expression in schizophrenia, bipolar disorder, and major depressive disorder.
Neuropsychopharmacology 2002;26368- 375
PubMedGoogle ScholarCrossref 64.West
ARFloresco
SBCharara
ARosenkranz
JAGrace
AA Electrophysiological interactions between striatal glutamatergic and dopaminergic systems.
Ann N Y Acad Sci 2003;100353- 74
PubMedGoogle ScholarCrossref 65.Pralong
EMagistretti
PStoop
R Cellular perspectives on the glutamate-monoamine interactions in limbic lobe structures and their relevance for some psychiatric disorders.
Prog Neurobiol 2002;67173- 202
PubMedGoogle ScholarCrossref 67.D’haenen
HABossuyt
A Dopamine D
2 receptors in depression measured with single photon emission computed tomography.
Biol Psychiatry 1994;35128- 132
PubMedGoogle ScholarCrossref 68.Ebert
DFeistel
HLoew
TPirner
A Dopamine and depression: striatal dopamine D
2 receptor SPECT before and after antidepressant therapy.
Psychopharmacology (Berl) 1996;12691- 94
PubMedGoogle ScholarCrossref 69.Klimke
ALarisch
RJanz
AVosberg
HMuller-Gartner
HWGaebel
W Dopamine D
2 receptor binding before and after treatment of major depression measured by [
123I]IBZM SPECT.
Psychiatry Res 1999;9091- 101
PubMedGoogle ScholarCrossref 70.Meyer
JHKruger
SWilson
AAChristensen
BKGoulding
VSSchaffer
AMinifie
CHoule
SHussey
DKennedy
SH Lower dopamine transporter binding potential in striatum during depression.
Neuroreport 2001;124121- 4125
PubMedGoogle ScholarCrossref 71.Klimek
VSchenck
JHan
HStockmeier
COrdway
G Dopaminergic abnormalities in amygdaloid nuclei in major depression: a postmortem study.
Biol Psychiatry 2002;52740- 748
PubMedGoogle ScholarCrossref 72.Berman
RMSanacora
GAnand
ARoach
LMFasula
MKFinkelstein
COWachen
RMOren
DAHeninger
GRCharney
DS Monoamine depletion in unmedicated depressed patients.
Biol Psychiatry 2002;51469- 473
PubMedGoogle ScholarCrossref 73.Parsey
RVOquendo
MAZea-Ponce
YRodenhiser
JKegeles
LSPratap
MCooper
TBVan Heertum
RMann
JJLaruelle
M Dopamine D
2 receptor availability and amphetamine-induced dopamine release in unipolar depression.
Biol Psychiatry 2001;50313- 322
PubMedGoogle ScholarCrossref 74.Bartzokis
GBeckson
MLu
PHEdwards
NRapoport
RBridge
PMintz
J Cortical gray matter volumes are associated with subjective responses to cocaine infusion.
Am J Addict 2004;1364- 73
PubMedGoogle ScholarCrossref 75.Kemppainen
NLaine
MLaakso
MPKaasinen
VNagren
KVahlberg
TKurki
TRinne
JO Hippocampal dopamine D
2 receptors correlate with memory functions in Alzheimer’s disease.
Eur J Neurosci 2003;18149- 154
PubMedGoogle ScholarCrossref 76.Bressan
RAErlandsson
KJones
HMMulligan
RFlanagan
RJEll
PJPilowsky
LS Is regionally selective D
2/D
3 dopamine occupancy sufficient for atypical antipsychotic effect? an in vivo quantitative [
123I]epidepride SPET study of amisulpride-treated patients.
Am J Psychiatry 2003;1601413- 1420
PubMedGoogle ScholarCrossref 77.Tuppurainen
HKuikka
JViinamaki
HHusso-Saastamoinen
MBergstrom
KTiihonen
J Extrastriatal dopamine D
2/D
3 receptor density and distribution in drug-naive schizophrenic patients.
Mol Psychiatry 2003;8453- 455
PubMedGoogle ScholarCrossref 78.Barbee
JConrad
EJamhour
N The effectiveness of olanzapine, risperidone, quetiapine, and ziprasidone as augmentation agents in treatment-resistant major depressive disorder.
J Clin Psychiatry 2004;65975- 981
PubMedGoogle ScholarCrossref 79.Rachid
FBertschy
GBondolfi
GAubry
J Possible induction of mania or hypomania by atypical antipsychotics: an updated review of reported cases.
J Clin Psychiatry 2004;651537- 1545
PubMedGoogle ScholarCrossref 80.Elkashef
ADoudet
DBryant
TCohen
RLi
SWyatt
R 6-18F-DOPA PET study in patients with schizophrenia: positron emission tomography.
Psychiatry Res 2000;1001- 11
PubMedGoogle ScholarCrossref 81.Telang
FWVolkow
NDLevy
ALogan
JFowler
JSFelder
CWong
CWang
GJ Distribution of tracer levels of cocaine in the human brain as assessed with averaged [
11C]cocaine images.
Synapse 1999;31290- 296
PubMedGoogle ScholarCrossref 82.Drevets
WBogers
WRaichle
M Functional anatomical correlates of antidepressant drug treatment assessed using PET measures of regional glucose metabolism.
Eur Neuropsychopharmacol 2002;12527- 544
PubMedGoogle ScholarCrossref 83.Neumeister
ANugent
ACWaldeck
TGeraci
MSchwarz
MBonne
OBain
EELuckenbaugh
DAHerscovitch
PCharney
DSDrevets
WC Neural and behavioral responses to tryptophan depletion in unmedicated patients with remitted major depressive disorder and controls.
Arch Gen Psychiatry 2004;61765- 773
PubMedGoogle ScholarCrossref 84.Kimbrell
TAKetter
TAGeorge
MSLittle
JTBenson
BEWillis
MWHerscovitch
PPost
RM Regional cerebral glucose utilization in patients with a range of severities of unipolar depression.
Biol Psychiatry 2002;51237- 252
PubMedGoogle ScholarCrossref 85.Devous
MDTrivedi
MHRush
AJ Regional cerebral blood flow response to oral amphetamine challenge in healthy volunteers.
J Nucl Med 2001;42535- 542
PubMedGoogle Scholar 86.Vollm
BAde Araujo
IECowen
PJRolls
ETKringelbach
MLSmith
KAJezzard
PHeal
RJMatthews
PM Methamphetamine activates reward circuitry in drug naive human subjects.
Neuropsychopharmacology 2004;291715- 1722
PubMedGoogle ScholarCrossref 87.Gambarana
CMasi
FTagliamonte
AScheggi
SGhiglieri
ODe Montis
MG A chronic stress that impairs reactivity in rats also decreases dopaminergic transmission in the nucleus accumbens: a microdialysis study.
J Neurochem 1999;722039- 2046
PubMedGoogle ScholarCrossref 88.Willner
P The mesolimbic dopamine system as a target for rapid antidepressant action.
Int Clin Psychopharmacol 1997;12
((suppl 3))
S7- S14
PubMedGoogle ScholarCrossref 89.Manji
HKQuiroz
JASporn
JPayne
JLDenicoff
KGray
NAZarate
CA
JrCharney
DS Enhancing neuronal plasticity and cellular resilience to develop novel, improved therapeutics for difficult-to-treat depression.
Biol Psychiatry 2003;53707- 742
PubMedGoogle ScholarCrossref 90.Duman
RSNakagawa
SMalberg
J Regulation of adult neurogenesis by antidepressant treatment.
Neuropsychopharmacology 2001;25836- 844
PubMedGoogle ScholarCrossref 92.Uslaner
JBadiani
ADay
HEWatson
SJAkil
HRobinson
TE Environmental context modulates the ability of cocaine and amphetamine to induce c-
fos mRNA expression in the neocortex, caudate nucleus, and nucleus accumbens.
Brain Res 2001;920106- 116
PubMedGoogle ScholarCrossref 93.Benes
FMTaylor
JBCunningham
MC Convergence and plasticity of monoaminergic systems in the medial prefrontal cortex during the postnatal period: implications for the development of psychopathology.
Cereb Cortex 2000;101014- 1024
PubMedGoogle ScholarCrossref