Low serotonin (5-HT) transmission is thought to increase susceptibility to a wide range of substance use disorders and impulsive behaviours. Reference Virkkunen, Rawlings, Tokola, Poland, Guidotti and Nemeroff1-Reference Seo, Patrick and Kennealy3 In part, this might reflect an influence of 5-HT on ascending midbrain dopamine pathways. For example, in laboratory animals 5-HT can alter dopamine cell firing and release. Reference Kapur and Remington4-Reference Minabe, Emori and Ashby6 Moreover, increasing and decreasing 5-HT transmission can diminish and augment, respectively, the ability of drugs such as cocaine to induce a dopamine response. Reference Carta, Collu, Fadda and Stancampiano7-Reference Molina, Ahmed, Gatley, Volkow and Abumrad9 Decreasing 5-HT transmission can also augment cocaine's behavioural effects, enhancing self-administration responses and increasing cocaine-induced locomotor activity, conditioned place preferences and conditioned approach. Reference Carta, Collu, Fadda and Stancampiano7,Reference Fletcher, Korth and Chambers10,Reference Walsh and Cunningham11 Increasing 5-HT transmission, in comparison, diminishes these effects. Reference Czoty, Ginsburg and Howell8,Reference Molina, Ahmed, Gatley, Volkow and Abumrad9,Reference Peltier and Schenk12 In humans, the effect of serotonergic manipulations on drug-induced dopamine responses has not been investigated. To examine these hypothesised interactions, the present study tested the effect of an experimentally induced low 5-HT state on the rewarding and dopaminergic effects of cocaine. Central 5-HT transmission was reduced using the acute tryptophan depletion method. Dopamine responses were measured using positron emission tomography (PET) plus the labelled tracer [11C]raclopride. It was predicted that the induction of a low 5-HT state would increase both dopamine and appetitive responses to cocaine.
Method
The studies were carried out in accordance with the Declaration of Helsinki and were approved by the Research Ethics Board of the Montreal Neurological Institute. All participants gave written informed consent.
Participants
Seventeen non-dependent cocaine users (2 females, mean age 24.8 years, s.d. = 3.7) (Table 1) were recruited through advertisements in local newspapers. Ten of these participants are the same individuals as those in our previous report. Reference Cox, Benkelfat, Dagher, Delaney, Durand and McKenzie13 For all participants, the primary and preferred route of self-administration was intranasal. All were free of current or past substance dependence, as determined by a semi-structured clinical interview for DSM-IV diagnoses. Reference First, Spitzer, Gibbon and Williams14 In total 9 of the 17 participants were current smokers with a score of four or less on the Fagerström Test for Nicotine Dependence (FTND) Reference Heatherton, Kozlowski, Frecker and Fagerstrom15 (mean 0.6, s.d. = 1.3, n = 9). All were free of current substance misuse. Women were tested during the follicular phase.
Participants were free of any other current Axis I psychopathology and were physically healthy as determined by physical examination, electrocardiography and standard laboratory tests. Prior to each test session, participants abstained from alcohol for at least 24 h and from nicotine for at least 12 h, with the exception of one participant in experiment two (with an FTND score of 4) who smoked one cigarette on the morning of each test day (about 6 h prior to each PET scan) in order to minimise nicotine withdrawal effects. On the morning of each test day, all tested negative on a urine drug screen sensitive to cocaine, opiates, phencyclidine, barbiturates, Δ9-tetrahydrocannabinol, benzodiazepines and amphetamines (Triage Panel for Drugs of Abuse, Biosite Diagnostics, San Diego, California, USA, www.biosite.com).
Procedure
Experiment 1
All participants underwent PET [11C]raclopride scans following the self-administration of:
-
(a) condition 1: cocaine hydrochloride (1.0 mg/kg) intranasally plus a tryptophan-deficient amino acid mixture (acute tryptophan depletion) orally;
-
(b) condition 2: cocaine (1.0 mg/kg) plus a nutritionally balanced amino acid mixture; and
Table 1Drug-use histories
Experiment 1 (n = 10) Experiment 2 (n = 7) Participants exposed to drug, n Mean (s.d.) total range Participants exposed to drug, n Mean (s.d.) total range Cocaine Occasions used, past 12 months 10 18.4 (6.0) 12–28 7 12.3 (8.7) 6–30 Occasions used, lifetime 10 43.9 (25.9) 18–100 7 72.1 (66.4) 18–200 Quantity used per occasion in past 12 months, g 10 0.38 (0.25) 0.125–1 7 0.29 (0.12) 0.2–0.5 Amphetamine, occasions used, lifetime 6 18.9 (16.4) 0–45 4 26.8 (28.1) 0–60 MDMA, occasions used, lifetime 9 20.8 (22.6) 0–62 7 18.7 (13.4) 2–40 Ephedrine, occasions used, lifetime 3 21.7 (17.6) 0–40 0 Opiates, occasions used, lifetime 1 60 2 1 (0) 0–1 Cannabis, occasions used, lifetime 10 388 (610) 10–1550 a 7 565.7 (588.6) 20–1500 Psilocybin, occasion used, lifetime 8 14 (12.5) 0–35 6 3.5 (2.1) 0–6 LSD, occasions used, lifetime 7 14.1 (16.2) 0–50 3 5.3 (2.5) 0–8 Ketamine, occasions used, lifetime 5 5.8 (6.4) 0–15 0 Alcohol, occasions used, lifetime 10 394 (279) 100–816 b 7 632.1 (355.4) 175–1270 MDMA, 3,4-Methylenedioxymethamphetamine; LSD, lysergic acid diethylamide.
a Missing data on four participants for lifetime use of cannabis.
b Missing data on five participants for lifetime use of alcohol.
-
• condition 3: placebo powder (100 mg lactose) plus nutritionally balanced amino acid mixture.
The effect of cocaine alone (condition 2 v. 3) has recently been described elsewhere. Reference Cox, Benkelfat, Dagher, Delaney, Durand and McKenzie13 Here we focus on the effect of acute tryptophan depletion on cocaine-induced dopamine release. In order to keep the physical amount of cocaine and placebo powder constant, the active drug was mixed with lactose to yield a total of 100 mg. All test sessions were carried out under double-masked conditions, in a randomised, counterbalanced order.
On each test day, participants came to the laboratory at 09.00 h, were weighed and ingested an amino acid mixture, as described previously. Reference Young, Smith, Pihl and Ervin16 Then 4.5 h post-amino acid mixture, participants self-administered cocaine or placebo and were scanned immediately afterward. Self-report questionnaires, cardiovascular measures and blood tests to assess plasma amino acid levels were administered at five time points:
-
(a) baseline, prior to amino acid drink;
-
(b) 4.5 h post-amino acid drink, immediately prior to cocaine self-administration;
-
(c) immediately after tracer injection, about 15 min post-cocaine self-administration;
-
(d) mid-scan, 45 min post-cocaine self-administration; and
-
(e) the end of the scan, 75 min post-cocaine self-administration.
Cocaine plasma concentrations were obtained at times (b) to (e). Throughout testing, a project-dedicated nurse and emergency medicine physician were onsite to monitor participant's safety and a psychiatrist was on-call. Following each test session participants remained overnight for observation. Prior to being discharged, participants were evaluated by an emergency medicine physician. There were no serious adverse events.
Experiment 2
A second study was conducted to assess the effect of acute tryptophan depletion on [11C]raclopride binding values in the absence of cocaine self-administration. Participants underwent PET scans with the labelled tracer [11C]raclopride following the ingestion of:
-
(a) a tryptophan-deficient amino acid mixture (acute tryptophan depletion); and
-
(b) a nutritionally balanced amino acid mixture.
Both sessions were carried out under double-blind conditions, in a randomised, counterbalanced order. On each test day, participants ingested an amino acid mixture and were scanned 4.5 h later. Blood draws to assess plasma amino acid levels were collected at three different time points:
-
(a) baseline, prior to amino acid drink;
-
(b) 4.5 h post-amino acid drink; and
-
(c) at the end of the scan.
Image acquisition and processing
All participants were scanned on a Siemens ECAT HR + PET scanner (CTI/Siemens, Knoxville, Tenn) with lead septa removed (63-slice coverage, with a maximum resolution of 4.2 mm full-width at half maximum (FWHM) in the centre of the field of view (FOV)). Immediately after the transmission scan 7 mCi of [11C]raclopride was injected as a bolus into the antecubital vein. Emission data were collected over 60 min in 26 time frames of progressively longer duration.
For anatomical coregistration, high-resolution (1 mm) T 1-weighted magnetic resonance images (MRI) were obtained for all participants on a 1.5 T Siemens scanner, using gradient echo pulse sequence (repetition time (TR) = 9.7 ms, echo time (TE) = 4 ms, flip angle 12°, FOV = 250 and matrix 256×256). These volumes were corrected for image intensity non-uniformity, Reference Sled, Zijdenbos and Evans17 and linearly and non-linearly transformed into standardised stereotaxic Talairach-like space Reference Talairach and Tournoux18 using automated feature matching to the MNI305 template. Reference Collins, Neelin, Peters and Evans19 Each individual's MRI was then coregistered to their summed radioactivity PET images. Reference Evans, Marrett, Neelin, Collins, Worsley and Dai20
The PET images were reconstructed using a 6 mm FWHM Hanning filter and corrected for movement. Reference Costes, Dagher, Larcher, Evans, Collins and Reilhac21 Parametric images were generated by computing [11C]raclopride binding potential (BPND) at each voxel using a simplified kinetic model that uses the cerebellum as a reference tissue devoid of dopamine D2/3 receptors to describe the kinetics of the free and specifically bound ligand. Reference Gunn, Lammertsma, Hume and Cunningham22 The BPND expresses the relationship between the estimated concentration of available dopamine D2/D3 receptors (Bavail), the dissociation constant of the radiotracer from D2/D3 receptors (Kd) and the free fraction of non-specifically bound tracer in the brain (FND):
In a resting condition BPND is proportional to the concentration of available D2/D3 receptors; BPND decreases when dopamine release is elicited, and the magnitude of the change in BPND has been shown to be proportional to the increase in dopamine transmission. Reference Laruelle23 Next, t-maps were generated, representing t-tests of voxel-by-voxel changes in [11C]raclopride BPND Reference Aston, Gunn, Worsley, Ma, Evans and Dagher24 between the following conditions.
Experiment 1
-
(a) Nutritionally balanced amino acid mixture + cocaine v. nutritionally balanced amino acid mixture + placebo;
-
(b) acute tryptophan depletion + cocaine v. nutritionally balanced amino acid mixture + cocaine;
-
(c) acute tryptophan depletion + cocaine v. nutritionally balanced amino acid mixture + placebo.
Experiment 2
-
(a) Acute tryptophan depletion v. nutritionally balanced amino acid mixture.
For both experiments, voxels of statistically significant changes were identified by thresholding the t-maps at a value of t≥4.1 (adjusted for sample size, Experiment 1: t = 4.05, Experiment 2: t≥4.12) which corresponds to P≤0.05 Bonferroni corrected for multiple comparisons, based on a search volume of the entire striatum and a spatial resolution of 8 mm at FWHM. Reference Worsley, Marrett, Neelin, Vandal, Friston and Evans25 To exclude potential noise, clusters of less than 5 voxels were ignored.
Self-report scales
Drug craving, mood and other subjective states were assessed with 14 visual analogue scale (VAS) items (happy, rush, high, euphoria, excited, anxious, energetic, mind-racing, alert, like drug effect, want cocaine, desire cocaine, urge for cocaine and crave cocaine) anchored at 1 (least) and 10 (most). A total craving score was calculated by combining the scores on all four craving-related VAS items: ‘want cocaine’, ‘urge for cocaine’, ‘desire cocaine’ and ‘crave cocaine’.
Plasma amino acid and cocaine levels
Tryptophan and tyrosine levels were measured using gradient reverse-phase high-performance liquid chromatography with fluorometric detection. In Experiment 1, tryptophan levels were available from six participants at five time points whereas tyrosine levels were available for eight participants at three time points:
Plasma concentrations of tryptophan at five (Experiment 1) and three (Experiment 2) different time points
| Pre-amino acid | 4.5 h | Start scan | Mid-scan | Post-scan | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Treatment | n | μg/ml, mean (s.d.) | n | μg/ml, mean (s.d.) | n | μg/ml, mean (s.d.) | n | μg/ml, mean (s.d.) | n | μg/ml, mean (s.d.) |
| Nutritionally balanced amino acid mixture + placebo | 10 | 11.2 (1.3) | 10 | 19.8 (6.1) | 9 | 17.1 (5.3) | 10 | 14.7 (4.5) | 10 | 12.7 (4.0) |
| Nutritionally balanced amino acid mixture + cocaine | 10 | 11.0 (1.5) | 8 | 23.7 (5.9) | 10 | 19.4 (5.2) | 9 | 17.1 (5.0) | 10 | 15.1 (5.1) |
| Acute tryptophan depletion + cocaine | 10 | 11.9 (2.2) | 10 | 1.5 (0.6) | 10 | 1.2 (0.6) | 9 | 1.2 (0.7) | 10 | 1.4 (0.8) |
| Nutritionally balanced amino acid mixture | 6 | 10.3 (0.6) | 7 | 21.5 (5.3) | 7 | 14.7 (5.0) | ||||
| Acute tryptophan depletion | 7 | 10.5 (1.6) | 7 | 0.9 (0.3) | 7 | 0.9 (0.3) | ||||
baseline, prior to amino acid drink; 4.5 h post-amino acid drink; and at the end of the scan. In Experiment 1, plasma levels of cocaine were analysed with gas chromatography mass spectroscopy, using solid phase extraction. Data were missing from two participants (n = 8). For Experiment 2, tryptophan levels were missing from one participant (n = 6); tyrosine levels were missing from two participants (n = 5).
Statistics
All data were analysed using repeated measures ANOVAs, followed by Greenhouse-Geisser corrections and planned comparisons as appropriate (SPSS version 12).
Results
Plasma levels of tryptophan, tyrosine and cocaine
As expected, tryptophan values increased and decreased on the nutritionally balanced amino acid and acute tryptophan depletion tests, respectively, as reflected by test×time interactions (Experiment 1: F(8, 40) = 25.3, P≤0.001; Experiment 2: F(2, 10) = 71.4, P≤0.0001). Importantly, cocaine administration did not alter the magnitude of the tryptophan increase nor its time course on the sessions with the nutritionally balanced amino acid mixtures (P = 0.4). Ingestion of the acute tryptophan depletion mixture led to significant decreases in plasma tryptophan levels that were sustained throughout the scanning procedure (Experiment 1: P≤0.0001; Experiment 2: P≤0.0001) (Table 2). In comparison, acute tryptophan depletion did not affect plasma levels of tyrosine as indicated by the absence of test×time interactions (Experiment 1: F(4, 28) = 1.17, P = 0.346; Experiment 2: F(2, 8) = 3.05, P = 0.104).
In Experiment 1, plasma concentrations of cocaine differed significantly between the three test sessions as reflected by a test× time interaction (F(6, 42) = 24.4, P=0.0001). Plasma cocaine levels increased significantly in the cocaine conditions compared with nutritionally balanced amino acid mixture + placebo (P = 0.0001 for all time points). Plasma cocaine levels peaked at the mid-scan time point, 45 min post-cocaine self-administration (mean 149 ng/ml (s.d. = 44) and 155 ng/ml (s.d. = 32) for nutritionally balanced amino acid mixture + cocaine and acute tryptophan depletion + cocaine respectively); the plasma cocaine levels did not differ on the two drug test sessions (P≥0.4 for all time points). Cocaine was not detected in plasma on any of the drug-free test days or on any baseline measure prior to cocaine self-administration.
PET data: analyses of t-maps
Experiment 1
As previously reported, Reference Cox, Benkelfat, Dagher, Delaney, Durand and McKenzie13 cocaine plus nutritionally balanced amino acid mixture v. placebo powder plus nutritionally balanced amino acid mixture produced a significant decrease in [11C]raclopride BPND values (440 voxels, peak t = 6.2), consistent with an effect of intranasal cocaine on extracellular dopamine levels with peak effects in the ventral striatum and post-commissural putamen (Fig. 1(a) and online Table DS1(a)). Compared with this effect of cocaine alone (nutritionally balanced amino acid mixture + cocaine), cocaine in the low 5-HT session (acute tryptophan depletion + cocaine) yielded significantly greater changes in [11C]raclopride BPND values, particularly in dorsal aspects of the anterior and posterior putamen, both left and right, as well as bilateral caudate (208 voxels, peak t = 5.4) (Fig. 1(b) and online Table DS1(b)). A comparison of the effect of cocaine plus acute tryptophan depletion to the placebo test session (nutritionally balanced amino acid mixture + placebo) yielded decreased [11C]raclopride BPND values throughout the entire striatum (2867 voxels, peak t = 9.2) (Fig. 1(c) and online Table DS1(c)).
Experiment 2
Acute tryptophan depletion alone did not decrease [11C]raclopride BPND values anywhere. In comparison, the voxel-wise analysis indicated that acute tryptophan depletion alone, as compared with the nutritionally balanced amino acid mixture, increased [11C]raclopride BPND in ventrolateral aspects of the right
Three t-maps.
(a) Decreased [11C]raclopride BPND in nutritionally balanced amino acid mixture + cocaine compared with nutritionally balanced amino acid mixture + placebo. (b) Decreased [11C]raclopride BPND in acute tryptophan depletion + cocaine compared with nutritionally balanced amino acid mixture + cocaine. (c) Decreased [11C]raclopride BPND in acute tryptophan depletion + cocaine compared with nutritionally balanced amino acid mixture + placebo.
posterior putamen and bilateral anterior putamen (141 voxels, peak t = 5.4) (Fig. 2 and online Table DS1(d)).
Behavioural data
Cocaine produced its prototypical subjective effects, as reflected by time×condition interactions for the total craving score (F(3.2, 28.8) = 3.66, P≤0.022) as well as drug wanting (F(3.79, 34.1) = 4.68, P≤0.005), drug desire (F(3.53, 31.8) = 3.35, P≤0.025), liking (F(2.60, 23.4) = 5.38, P≤0.008), high (F(2.53, 22.8) = 5.59, P≤0.007), euphoria (F(2.47, 22.2) = 4.14, P≤0.023) and rush (F(2.97, 26.7) = 3.58, P≤0.027) (Table 3). On the low 5-HT test session the cocaine-induced craving response was augmented as reflected by greater increases in cocaine-induced drug wanting (t(9) = –2.67, P≤0.025, Fig. 3) and the total craving score (t(9) = 2.28, P≤0.049, Table 3). Cocaine's effects on drug liking, high, euphoria and rush, in comparison, were unaltered following acute tryptophan depletion (P≥0.05, Table 3). In Experiment 2, acute tryptophan depletion alone did not produce any significant changes in subjective measures as indicated by the absence of significant time×condition interactions on the VAS scales (F(4, 12)≤2.1, P≥0.14).
Discussion
Main findings
The present findings provide the first demonstration in humans that lowered 5-HT transmission increases cocaine-induced craving states and dopamine responses. These observations may suggest a mechanism by which individuals with disorders associated with
A t-map illustrating increased [11C]raclopride BPND in acute tryptophan depletion compared with nutritionally balanced amino acid mixture, indicating decreases in dopamine release.
Cocaine-increased self-reports of ‘want cocaine’.
*Significantly different from nutritionally balanced amino acid (AA) mixture + placebo; †significantly different from nutritionally balanced amino acid mixture + cocaine. ATD, acute tryptophan depletion; BAL, nutritionally balanced amino acid mixture.
Subjective effects in Experiment 1
| Subjective effects, mean (s.e.m.) | |||||
|---|---|---|---|---|---|
| Pre-amino acid | 4.5 h | Start scan | Mid-scan | Post-scan | |
| High** | |||||
| Nutritionally balanced amino acid mixture + placebo | 1.4 (0.2) | 1.2 (0.1) | 1.4 (0.2) | 1.1 (0.1) | 1.1 (0.1) |
| Nutritionally balanced amino acid mixture + cocaine | 1.8 (0.6) | 1.1 (0.1) | 3.8 (0.6)†† | 2.3 (0.4)† | 1.4 (0.2) |
| Acute tryptophan depletion + cocaine | 1.4 (0.2) | 1.1 (0.1) | 4.7 (0.9)†† | 3.1 (0.5)†† | 1.3 (0.2) |
| Like drug** | |||||
| Nutritionally balanced amino acid mixture + placebo | 1.9 (0.5) | 1.4 (0.2) | 1.4 (0.2) | 1.2 (0.1) | 1.1 (0.1) |
| Nutritionally balanced amino acid mixture + cocaine | 1.4 (0.2) | 1 (0)† | 4.1 (0.5)†† | 2.5 (0.4) | 1.8 (0.3) |
| Acute tryptophan depletion + cocaine | 2 (0.6) | 1.6 (0.4) | 5.2 (1.0)†† | 3.7 (0.6)†† | 2 (0.4)† |
| Rush* | |||||
| Nutritionally balanced amino acid mixture + placebo | 2 (0.4) | 2 (0.5) | 1.5 (0.2) | 1.2 (0.1) | 1.3 (0.2) |
| Nutritionally balanced amino acid mixture + cocaine | 2.2 (0.5) | 2.1 (0.8) | 4 (0.4)††† | 2.1 (0.3)† | 1.9 (0.4) |
| Acute tryptophan depletion + cocaine | 1.8 (0.3) | 1.6 (0.2) | 4.4 (0.9)† | 3.1 (0.5)†† | 1.8 (0.3) |
| Euphoria* | |||||
| Nutritionally balanced amino acid mixture + placebo | 1.4 (0.2) | 1.1 (0.1) | 1.1 (0.1) | 1 (0) | 1.1 (0.1) |
| Nutritionally balanced amino acid mixture + cocaine | 1.5 (0.2) | 1.1 (0.1) | 2.4 (0.4)† | 1.5 (0.2)† | 1.2 (0.1) |
| Acute tryptophan depletion + cocaine | 1.7 (0.2) | 1.2 (0.1) | 3.6 (0.8) | 2.1 (0.4)†† | 1.2 (0.1) |
| Want cocaine** | |||||
| Nutritionally balanced amino acid mixture + placebo | 1.9 (0.5) | 1.3 (0.1) | 1.1 (0.1) | 1.1 (0.1) | 1.1 (0.1) |
| Nutritionally balanced amino acid mixture + cocaine | 1.5 (0.3) | 1.5 (0.4) | 3.6 (0.6)†† | 2.3 (0.8) | 1.3 (0.2) |
| Acute tryptophan depletion + cocaine | 1.9 (0.7) | 1.3 (0.2) | 4.1 (0.8)†† | 3.7 (0.8)††‡ | 1.8 (0.3) |
| Total craving score* | |||||
| Nutritionally balanced amino acid mixture + placebo | 1.3 (0.1) | 1.1 (0.1) | 1.1 (0.1) | 1.1 (0.1) | 1.1 (0.1) |
| Nutritionally balanced amino acid mixture + cocaine | 1.2 (0.1) | 1.3 (0.2) | 2.8 (0.5)†† | 2.2 (0.8) | 1.2 (0.1) |
| Acute tryptophan depletion + cocaine | 1.5 (0.4) | 1.2 (0.1) | 3.3 (0.6)†† | 2.9 (0.7) † ‡ | 1.5 (0.2) |
* Treatment × time interaction, significant at P ≤ 0.05
** P ≤ 0.01
† significantly different from nutritionally balanced amino acid mixture + placebo, at P=0.05, P ≤ 0.01, P ≤ 0.001
‡ significantly different from nutritionally balanced amino acid mixture + cocaine, significant at P ≤ 0.05.
low 5-HT Reference Virkkunen, Rawlings, Tokola, Poland, Guidotti and Nemeroff1-Reference Seo, Patrick and Kennealy3 are at increased risk for drug-seeking behaviour and the progression to substance misuse.
The present findings are supported by considerable preclinical evidence for 5-HT/dopamine interactions in the nigrostriatal and mesolimbic dopamine systems. Serotonin neurons in the raphe nuclei project to dopamine cell bodies and terminal regions including the ventral tegmental area, substantia nigra and ventral plus dorsal striatum. Reference Hervé, Pickel, Joh and Beaudet26,Reference Fibiger and Miller27 Inhibitory effects of 5-HT on both the behavioural and dopaminergic effects of cocaine have been reported. For example, administration of a tryptophan-deficient diet to rats significantly increases locomotor activity, conditioned place preference and dopamine responses to amphetamine and cocaine. Reference Carta, Collu, Fadda and Stancampiano7,Reference Carta, Fadda and Stancampiano28 In comparison, increasing 5-HT neurotransmission has the opposite effect, diminishing cocaine-induced striatal dopamine responses and self-administration behaviour. Reference Czoty, Ginsburg and Howell8,Reference Molina, Ahmed, Gatley, Volkow and Abumrad9
In the present study, acute tryptophan depletion alone did not increase dopamine release, consistent with the hypothesis that the lowered 5-HT was augmenting the cocaine's effects. In comparison, acute tryptophan depletion alone did elicit evidence of decreased dopamine release, particularly within the ventrolateral putamen. In animal models, the predominant effect of 5-HT is to inhibit dopamine cell firing and release, although this occurs most consistently within the substantia nigra and dorsal striatum. Reference Kapur and Remington4,Reference Gervais and Rouillard29 In comparison, some evidence suggests that 5-HT can increase resting dopamine transmission in more ventral regions. For example electrical stimulation of the dorsal raphe increases the cell firing of most dopamine neurons in the ventral tegmental area, Reference Gervais and Rouillard29 whereas 5-HT depletion leads to decreased ventral tegmental area dopamine cell firing. Reference Minabe, Emori and Ashby6
The dopaminergic effects of cocaine and acute tryptophan depletion demonstrated some regional specificity, and the relevant neuroanatomy has been studied in depth. Reference Haber, Kim, Mailly and Calzavara30 The ventral striatum receives dense input from the amygdala, hippocampus and limbic cortex, and dopamine transmission within this area is thought to influence the salience of, and sustained interest in, motivationally relevant stimuli such as rewards. Reference Robinson and Berridge31-Reference Phillips, Vacca and Ahn36 The dorsal striatum, in comparison, receives less input from limbic regions and more from the associative and sensorimotor cortical areas. Dopamine transmission within the dorsal striatum has been implicated in the regulation of stimulus-response habit learning. Reference McDonald and White37,Reference Everitt and Robbins38 In primates, these subdivisions are not sharply delineated, raising the possibility that limbic system-mediated incentive motivational processes extend in a gradation from ventromedial striatum through to more dorsal aspects. Reference Haber, Kim, Mailly and Calzavara30 If low 5-HT states augment drug-induced dopamine responses within the dorsal striatum, the result might be enhanced motivation to obtain drug reward and susceptibility to compulsive, habit-like drug-seeking behaviour.
Limitations
The conclusions suggested by the present study should be interpreted in light of the following considerations. First, the challenge dose of cocaine was modest, 1.0 mg/kg taken intranasally. However, using this dose protected against ceiling effects, increasing our ability to identify augmentations of the drug-induced dopamine response as a result of acute tryptophan depletion. Second, the sample sizes are modest, but within the commonly accepted range for assessing pharmacological challenges within participants. Third, cocaine and 5-HT can be vasoconstrictors. Altered cerebral blood flow following acute tryptophan depletion or cocaine could produce confounding effects on the observed changes in BPND. However, the effects of cocaine plus acute tryptophan depletion were synergistic rather than antagonistic. Moreover, simulation studies indicate that even large changes in blood flow have negligible effects on receptor ligand binding, as measured by the method used here. Reference Aston, Gunn, Worsley, Ma, Evans and Dagher24 Fourth, in a previous study conducted in individuals who were cocaine dependent, acute tryptophan depletion was reported to leave cocaine-induced craving unaltered and to decrease the drug's euphorigenic effects. Reference Aronson, Black, McDougle, Scanley, Jatlow and Kosten39 The apparent discrepancy with the current results may be explained by differences in participant population (dependent v. non-dependent) or dose of cocaine administered (2.0 v. 1.0 mg/kg). For example, as the authors noted, the diminished high may have been secondary to an acute tryptophan depletion-induced increase in cocaine-elicited anxiety. In the present study with a lower dose of cocaine, significant anxiogenic responses were not produced. The ability of other serotonergic manipulations to influence cocaine's behavioural effects has also been dependent on the cocaine dose, both in humans and in laboratory animals. Reference Walsh and Cunningham11,Reference Peltier and Schenk12 Finally, 15 of the 17 participants were male, precluding our ability to test for potential gender differences.
Implications
In conclusion, the present study's primary findings were that, in humans, a low 5-HT state diminished resting limbic dopamine levels and augmented cocaine-induced striatal dopamine responses plus the desire to use the drug. Although this is the first study in humans of the effects of lowered 5-HT transmission on dopamine release, and the results will require replication, the observed combination of effects might delineate monoamine interactions within human brain and identify a mechanism by which individuals with low serotonergic tone are at elevated risk for substance misuse and various comorbid disorders.
Funding
This work was supported by an operating grant from the Canadian Institutes of Health Research to M.L. (). M.L. and C.B. are both recipients of research chairs from McGill University. S.M.L.C. received a fellowship from the McGill University Health Centre.
Acknowledgements
We thank Franceen Lenoff, Kathleen Auclair, Sam McKenzie and the PET technicians at the McConnell Brain Imaging Centre for excellent technical assistance, and the Emergency Medicine Services physicians for their valuable medical support.
eLetters
No eLetters have been published for this article.