Effects of a novel method of acute tryptophan depletion on plasma tryptophan and cognitive performance in healthy volunteers.

Abstract

Disorders associated with low levels of serotonin (5-HT) are characterized by mood and cognitive disturbances. Acute tryptophan depletion (ATD) is an established method for lowering 5-HT levels and an important tool to study the effects of reduced 5-HT on mood and cognition in human subjects. The traditional ATD method, i.e., administration of separate amino acids (AAs), has several disadvantages. The AA mixture is costly, unpalatable and associated with gastrointestinal discomfort.
The University of Maastricht developed a new and inexpensive method for ATD: a natural collagen protein (CP) mixture with low tryptophan (TRP) content. The reductions in plasma TRP after taking this CP mixture were compared with the reductions achieved taking the traditional AA mixture, and effects on memory and reversal learning were studied.
Fifteen healthy young volunteers participated in a double-blind, counterbalanced within-subject study. Reversal learning, verbal memory and pattern recognition were assessed at baseline and 3-4 h after taking the CP mixture.
The new ATD method significantly reduced plasma TRP by 74% and the ratio between TRP and the other large AAs (TRP/LNAA) by 82%. The placebo mixture did not change these measures. Delayed recognition reaction time on the verbal learning task was increased following ATD. No other cognitive effects were found.
The CP mixture was shown to be an efficient tool for lowering plasma TRP in humans. The validity of this method with regard to behavioral changes remains to be established in healthy, vulnerable and clinical populations.

Full text

Page 1

Psychopharmacology (2005) 178: 92–99
DOI 10.1007/s00213-004-2141-y
ORIGINAL INVESTIGATION
E. A. T. Evers.D. E. Tillie.F. M. van der Veen.
C. K. Lieben.J. Jolles.N. E. P. Deutz.J. A. J. Schmitt
Effects of a novel method of acute tryptophan depletion
on plasma tryptophan and cognitive performance in healthy
volunteers
Received: 11 March 2004 / Accepted: 5 May 2004 / Published online: 23 December 2004
# Springer-Verlag 2004
Abstract Rationale: Disorders associated with low levels
of serotonin (5-HT) are characterized by mood and cog-
nitive disturbances. Acute tryptophan depletion (ATD) is
an established method for lowering 5-HT levels and an
important tool to study the effects of reduced 5-HT on
mood and cognition in human subjects. The traditional
ATD method, i.e., administration of separate amino acids
(AAs), has several disadvantages. The AA mixture is
costly, unpalatable and associated with gastrointestinal
discomfort.Objectives: The University of Maastricht
developed a new and inexpensive method for ATD: a
natural collagen protein (CP) mixture with low trypto-
phan (TRP) content. The reductions in plasma TRP after
taking this CP mixture were compared with the reduc-
tions achieved taking the traditional AA mixture, and
effects on memory and reversal learning were studied.
Methods: Fifteen healthy young volunteers participated
in a double-blind, counterbalanced within-subject study.
Reversal learning, verbal memory and pattern recognition
were assessed at baseline and 3–4 h after taking the CP
mixture.Results: The new ATD method significantly
reduced plasma TRP by 74% and the ratio between TRP
and the other large AAs (TRP/LNAA) by 82%. The pla-
cebo mixture did not change these measures. Delayed
recognition reaction time on the verbal learning task was
increased following ATD. No other cognitive effects were
found. Conclusions: The CP mixture was shown to be an
efficient tool for lowering plasma TRP in humans. The
validity of this method with regard to behavioral changes
remains to be established in healthy, vulnerable and clin-
ical populations.
Keywords Serotonin.Acute tryptophan depletion.
Memory.Reversal learning.Mood
Introduction
Serotonergic (5-HT) dysfunction is associated with dis-
rupted cognitive function, emotional processing and social
functioning, which are characteristic for depressive pa-
tients (Murphy et al. 2002). A pharmacological model to
study the role of 5-HT in human behavior is acute tryp-
tophan depletion (ATD), in which central 5-HT synthesis
is reduced lowering the brain availability of the 5-HT
precursor L-tryptophan (TRP) (reviewed by Reilly et al.
1997). Animal (Biggio et al. 1974; Gartside et al. 1992)
and human (Nishizawa et al. 1997, Carpenter et al. 1998;
Williams et al. 1999) studies have demonstrated that the
acute reduction of TRP to the brain is sufficient to produce
a rapid decrease in the synthesis and release of brain 5-HT.
Previous ATD studies showed impaired performance on
tasks involving memory consolidation (see Riedel et al.
2002 for an overview), decision making (Rogers et al.
1999, 2003), reversal learning (Murphy et al. 2002;
Rogers et al. 1999) and affective processing (Murphy et al.
2002; Rubinzstein et al. 2001) in healthy volunteers. The
aim of the present study was to investigate the effects of a
novel ATD method on the plasma TRP levels, on the ratio
between TRP and the other large neutral amino acids
(ΣLNAAs), and to validate the method on a behavioral
level. To this end, we assessed several cognitive functions
that are known to be sensitive to ATD, i.e., memory and
reversal learning.
This article was originally published under the DOI 10.1007/
s00213-004-1933-4. Unfortunately an unrelated paper appeared in
print and in the PDF version online. For this reason, all versions of
the correct article are now published here under the new DOI,
10.1007/s00213-004-2141-y.
E. A. T. Evers (*).D. E. Tillie.F. M. van der Veen.
C. K. Lieben.J. Jolles.J. A. J. Schmitt
Department of Psychiatry and Neuropsychology (DRT10),
Brain and Behavior Institute, Maastricht University,
P.O. Box 616 Maastricht, 6200, The Netherlands
e-mail: l.evers@np.unimaas.nl
Tel.: +31-43-3884086
Fax: +31-43-3884092
N. E. P. Deutz
Department of Surgery, Maastricht University,
Maastricht, 6200, The Netherlands

Page 2

A low-TRP collagen-protein (CP) mixture was used to
induce an ATD. This protein is derived from the selective
hydrolysis of CP and comprises the entire range of amino
acids (AAs) in the form of peptides. After administration,
these peptides are decomposed into AAs, and the mech-
anism of depletion is identical to that of the AA mixture.
In rats, the CP mixture significantly lowered plasma TRP
and TRP/LNAA ratios (−78%) and brain TRP and 5-HT
(−50%) concentrations (Lieben et al. 2004). Based on the
previous behavioral ATD studies and the demonstration
that the CP mixture was an efficient TRP-depletion
method in rats, it was hypothesized that this new method
of ATD results in reduced plasma TRP, a decreased TRP/
LNAA ratio, impaired delayed recall and recognition in
word and pattern-learning tasks, slower responding and
more errors in the reversal-learning task when compared
with the placebo in healthy volunteers.
Materials and methods
Subjects
Fifteen young healthy male (n=3) and female (n=12)
volunteers (mean age 21.8 years; SD=1.8) gave informed
consent and participated in this study, which was approved
by the medical ethics committee of the University Hos-
pital, Maastricht. The participants were free from signif-
icant past or present physical or psychiatric illness and
did not use medication other than oral anti-conceptives.
Female participants were not tested in the late luteal
phase of the menstrual cycle (days 21–28). All subjects
completed the study.
CP mixture
The CP mixture was purchased from PB Gelatins
(Tessenderlo, Belgium); see Table 1 for the AA composi-
tion. To obtain a drinkable mixture, 100 g of the CP
mixture was mixed with 200 g water. The placebo mixture
was identical in composition, but 1.2 g L-TRP (Sigma,
Zwijndrecht, The Netherlands) was added.
Design
This study was conducted according to a double-blind,
placebo-controlled crossover design. The participants
received a TRP-free CP mixture (TRP−) and a CP mixture
with 1.21 g TRP added (placebo) on separate occasions.
Treatment order was balanced over the two test days,
which were separated by at least 3 days.
Procedure
In a separate session, approximately 1 week before the
actual test days, the cognitive tasks were practiced to
minimize learning effects. The participants were instructed
not to drink alcohol on the days prior to the test days, not
to eat or drink (except water) after 2200 hours that evening
and to arrive at the laboratory well rested. A test day
started with a cognitive test battery, subjective assessments
of mood and adverse effects, and baseline blood sampling.
Subsequently, subjects received the TRP−or the placebo
mixture. A 3-h break followed to maximize TRP deple-
tion. The participants had free access to low-TRP food,
such as apples, tomatoes and protein-free candy, and
caffeine-free tea during the pause. These food items were
generally consumed in small quantities, and the intake of
carbohydrates does not meaningfully affect brain TRP
availability in the presence of large amounts of protein
(Teff et al. 1989). After the 3-h interval, cognition, mood
and adverse effects were assessed again and a blood
sample was taken. The duration of a test day was 5 h in
total.
Cognitive assessment
The cognitive test battery, mood and adverse effects
assessments took approximately 1 h to complete. On each
session parallel test versions were used, version order was
distributed among the participants using a 4×4 Latin
square.
Probabilistic reversal-learning task
In the reversal-learning task (described in detail by O’
Doherty et al. 2001), two abstract stimuli, composed of
two bars of different color, were randomly presented to the
left or right side of a computer screen. One stimuli was
advantageous (S+) and was usually (70%) associated with
large reward (addition of 80–250 points) and occasionally
(30%) with a small punishment (subtraction of 10–60
points), based on a pseudo-random sequence. The other
stimulus was disadvantageous (S−) and usually (60%)
associated with a large punishment (250–600 points) and
occasionally (40%) a small reward (30–60 points). Volun-
teers had to determine which stimulus was advantageous
based on the feedback—the number of points won or lost.
Once this was learned, i.e., the advantageous stimulus was
Table 1 Composition (grams) of the natural collagen protein
(tryptophan−) in 100 ml tap water
Aspartic acid + asparagines
Glutamic acid + glutamine
Hydroxyproline
Serine
Glycine
Histidine
Arginine
Threonine
Alanine
Proline
5.2
9.3
12.1
3.1
22.5
0.5
8.8
1.1
9.3
13.3
Tyrosine
Valine
Methionine
Cysteine
Isoleucine
Leucine
Hydroxylysine
Phenylalanine
Tryptophan
Lysine
0.4
2.1
0.6
0.2
1.4
3
1.4
1.9
0.1
3.6
93

Page 3

chosen four times out of five responses, the stimulus-
reward contingencies were reversed (S+ became S−, and
S− became S+). The participant’s task was to keep track
of the most profitable stimulus and to collect as many
points as possible. Two successive 8-min blocks were
performed per session. The stimuli used in the two re-
versal blocks were four differently colored abstract pat-
terns, and blocks were randomly assigned. The task
stopped after the 19th reversal or after 121 trials. Before
each assessment, subjects practiced an acquisition (eight
correct responses in a row) and a short reversal-learning
task (two reversals within 50 trials: 8 correct out of 9
trials). Dependent variables were the total number of re-
versals, the total number of preservations (number of
errors directly after the rule has changed), the mean re-
action time per trial, the mean reaction time of the first
response following a reversal switch and the total number
of points collected, in the two successive reversal blocks.
The visual verbal learning test
The visual verbal learning test (VVLT), a modified version
of the Rey Auditory Verbal Learning Test (Lezak 1995),
consisted of a list of 30 monosyllabic words in Dutch,
which were presented in three trials on a computer screen
(Riedel et al. 1999). Of the words, 12 were positively
loaded, 12 were negatively loaded and 6 were neutrally
loaded. Immediately after each presentation (immediate
recall) and 30 min later (delayed recall), the subjects were
asked what words they could remember. After that, a
delayed recognition test was presented; 30 words were
shown (15 new and 15 words from the original word list),
and the subject was asked to press “yes” for “familiar”
and “no” for “new”, as quickly as possible. According to
the theory of signal detection (Pollack and Norman
1964), the proportion of correctly recognized words (cr)
and the proportion of those falsely recognized (fr)
constituted the non-parametric sensitivity measure: A0¼
1 ? 1=4 fr=cr þ 1 ? cr
proportion of correctly recognized words corrected for
the participant’s response tendency. A was arc sin
transformed before statistical analyses.
The outcome variables were the number of correct
words recalled during the three immediate recall trials as a
measure of short-term memory, the number of correct
words produced on delayed free recall as a measure of
retrieval from long-term memory, A’ as a measure of
storage in long-term memory, and the medial reaction
times of correctly recognized words as a measure of speed
of long-term memory.
ðÞ= 1 ? fr
ðÞðÞ where A is in fact the
Abstract pattern-learning task
In the abstract pattern-learning task, 15 abstract stimuli
were shown one by one on a computer screen, followed by
the recognition task in which the participants had to press
the left button of a response box if they recognized the left
stimulus and the right button if they recognized the right
stimulus. Recognition was tested immediately after the
presentation of the patterns (immediate recognition) and
30 min later (delayed recognition). The dependent mea-
sure of interest was the proportion correctly recognized
patterns and the reaction times of the correct responses.
Questionnaires
Mood
A visual analogue version of the profile of mood states
(POMS) was used to assess mood (McNair et al. 1988).
This questionnaire consists of 32 bipolar sets of adjectives,
which measure five mood dimensions: anger, depression,
fatigue, tension and vigor. The items were scored on a
0- to 100-mm scale. When the participants felt as they
Fig. 1 The means and the standard errors (SE) of the blood plasma
levels of tryptophan (μmol/l) and the ratio between tryptophan and
the other large amino acids (TRP/LNAA) at baseline and 4 h after
start of the TRP depletion and placebo treatment. *P<0.001
94

Page 4

normally do, they were asked to mark the middle of the
line (score 50).
Adverse effects
Adverse effects, 31 items, were registered and scored on a
five-point scale from “no complaint at all” (0) to “severe
complaint” (4).
Plasma AAs
In total, four blood samples (4 ml) were drawn from each
volunteer and immediately centrifuged at 4°C (10 min,
4500 rpm). A 100-μl aliquot of plasma was mixed with
8 mg sulfasalicyl acid and frozen at −80°C until AA
analysis using high-performance liquid chromatography
(van Eijk 1993).
Statistical analysis
The outcome variables of the cognitive and mood as-
sessments, and the total number of adverse effects were
analyzed using a repeated-measures analysis of variance
(ANOVA) of the difference scores (performance scores
3–4 h after the drink administration minus baseline scores)
using treatment (TRP−, placebo) as within-subject factor
and treatment order (TRP−or placebo mixture first) as
between-subject factors.
Results
Levels of plasma TRP
Blood samples of two participants were missing. Figure 1
shows that plasma TRP concentrations and TRP/ΣLNAA
ratios at baseline did not differ between groups. Three
hours after drinking the TRP−mixture, plasma TRP was
significantly reduced by 74% (F1,11=60.8, P<0.001) and
the TRP/ΣLNAA ratios by 82% (F1,11=50.4, P<0.001),
compared with baseline. Small, non-significant changes in
Table 2 Mean scores (standard deviations) for the outcome variables of the visual verbal learning task, the probabilistic reversal-learning
task and the abstract visual learning task at baseline and 3–4 h after administration of the tryptophan (TRP)−and placebo mixture
TRP−condition
Baseline
Placebo condition
BaselineT3–4T3–4
Visual verbal learning task
Number of words correctly recalled on the three immediate recall trials
Trial 1
Trial 2
Trial 3
Number of words correctly recalled on delayed recall
Delayed recognition sensitivity measure A (percentage)
Median reaction time of correctly recognized words (ms)
Probabilistic reversal-learning task
Total number of reversals
Total number of preservations
Mean RT
Mean RT after a reversal
Total number of points
Abstract pattern recognition task
Proportion correctly recognized patterns
Immediate recognition
Delayed recognition
Reaction times for the correct responses
Immediate recognition
Delayed recognition
12.07 (3.8)
17.5 (4.4)
21.5 (3.9)
18.7 (6.2)
96 (4)
677 (92)
9.4 (3.3)
15.0 (4.9)
18.9 (7.7)
13.7 (7.0)
94 (6)
703 (98)
11.9 (3.1)
18.1 (3.6)
21.7 (3.7)
19.7 (6.0)
97 (2)
669 (69)
10.3 (2.7)
16.7 (3.8)
20.4 (4.2)
15.0 (5.5)
95 (6)
654 (74)
18.0 (1.5)
26.0 (7.9)
496 (88)
487 (90)
1381 (1217)
18.3 (1.3)
25.1 (6.9)
479 (109)
474 (114)
1451 (1674)
18.1 (1.1)
25.5 (6.6)
464 (89)
461 (94)
1400 (1776)
18.6 (1.1)
23.1 (6.7)
426 (64)
436 (76)
1732 (2173)
90.6 (11.0)
91.0 (8.3)
89.6 (14.0)
82.8 (15.6)
92.6 (6.2)
88.3 (8.2)
93.1 (8.1)
85.6 (11.7)
1604 (451)
1593 (411)
1613 (498)
1758 (486)
1580 (434)
1503 (372)
1593 (436)
1597 (530)
Table 3 Mean (standard deviation) scores on the subscales of the
profile of mood questionnaire at baseline and 4 h after administra-
tion of the tryptophan (TRP)-depleted mixture (TRP−) and the
placebo mixture
TRP−condition
BaselineT4
Placebo
Baseline T4
Depression
Anger
Fatigue
Vigor
Tension
5.6 (0.9)
5.7 (1.0)
5.1 (0.9)
4.7 (0.8)
5.4 (1.0)
5.5 (1.0)
5.6 (1.2)
4.9 (1.0)
5.3 (1.0)
5.4 (0.8)
5.5 (1.3)
5.6 (1.4)
4.9 (1.2)
6.0 (3.4)
5.6 (1.4)
5.4 (1.3)
5.5 (1.4)
4.8 (1.1)
5.2 (0.9)
5.6 (1.3)
95

Page 5

plasma TRP (+8%) and TRP/ΣLNAA ratio (−2%) were
observed following administration of the placebo mixture.
Cognitive assessments
The results of the cognitive assessments are summarized in
Tables 3, 4. Delayed recognition reaction time of the
VVLT was increased following ATD (F1,13=9.2, P=0.01).
No treatment effects were seen on delayed recognition ac-
curacy, delayed recall or overall immediate recall.
No effects of ATD were seen on the outcome variables
of the probabilistic reversal-learning task and the abstract
pattern-learning task. None of the analyses showed an
interaction between treatment order and the effects of the
treatment (Table 2).
Adverse effects and mood assessment
None of the subscales of the POMS—depression, anger,
tension, fatigue and vigor—showed an effect of treatment
(Table 3). As is shown in Table 4, neither the TRP−nor the
placebo mixture was associated with robust adverse effects
at T4. None of the participants scored above “1” (bothered
a bit) on any of the adverse effect items. The total adverse-
effects score was identical in the ATD versus the placebo
condition. Treatment order did not modify the outcome of
the analyses.
Discussion
The TRP−CP mixture appears to be an efficient tool to
lower plasma TRP in humans. Four hours after adminis-
tration, plasma TRP concentrations were reduced by 74%
and the ratio TRP/ΣLNAA by 82%. These levels of TRP
depletion are comparable to those obtained by adminis-
tration of the “classic” AA mixture (Table 5). Furthermore,
the present results show that our placebo mixture is es-
Table 4 The number of participants experiencing the physical
complaint mentioned
Adverse effectsTRP−condition
Baseline
Placebo
Baseline T4 T4
Headache
Sleepiness
Dizziness
Nausea
Restlessness
Heart palpitations
Stomach ache
Bloating
Heartburn
Loss of appetite
Hunger
Diarrhea
Feeling cold
Feeling warm
Dry mouth
Trembling
Feeling tired
A hazy view
Sweating
Sedation
Feeling feeble
Tightness in the chest
Decreased concentration
Tingling
Nervousness
Irritation
Listless
Bothered by bright light
Bothered by hard sounds
A warm head
The feeling that you could faint
3
8
0
1
2
0
1
2
0
0
2
9
0
4
3
0
1
5
2
5
5
1
5
1
2
0
46
7
3
4
3
0
2
5
1
5
6
0
5
0
6
2
10
3
1
5
0
1
0
0
1
13
0
7
0
7
2
9
2
0
7
4
0
3
0
3
0
1
0
0
1
0
12
0
7
1
6
0
10
1
0
7
5
1
3
0
2
1
3
0
1
1
1
10
0
0
8
8
0
8
0
0
1
2
0
0
2
1
10
2
1
8
4
0
11
1
0
0
4
0
0
2
0
Table 5 The decrease in total plasma tryptophan (TRP), the ratio
between TRP and the other large amino acids (ratio) in the
tryptophan depletion (TRP−) and the placebo conditions as a result
of acute TRP depletion (ATD) by amino acid administration and
the collagen protein (CP) method. A selection of recent studies
using the traditional ATD method is shown in which the level of
depletion was reported or could be calculated. For each study, the
amount of amino acids (AA) and tryptophan (TRP) in grams (g),
and the number of hours (t) between consuming the mixture and
the providing blood samples are given. The last study mentioned is
the present study in which a CP mixture was used
References Plasma TRP
TRP−(%)
Ratio
TRP−(%)
AA (g)TRP (g)t
Placebo (%)Placebo (%)
Riedel et al. (1999)
Schmitt et al. (2000)
Rubinsztein et al. (2001)
Murphy et al. (2002)
Anderson et al. (2003)
CP-mixture
−67
−63
−80
−70
−84
−74
+22
+80
+47
−78
−79
−20
+20
100
100
53
86
100/86a
100
3.0
4.6
2.0
1.9
2.3/1.8a
1.2
6
5
5
7
6
4
+118
+8
−82
−2
a100 g of AAs with 2.3 g TRP added for males and 86 g with 1.8 g TRP added for females.
96

Page 6

sentially neutral with regard to effects on plasma TRP as
well as the TRP/LNAA ratio. This constitutes an important
advantage over placebo AA mixtures, in which neutral
TRP/LNAA ratios can only be maintained at the cost of
marked increases of TRP levels (van der Does 2001;
Weltzin et al. 1994). Both TRP/LNAA (van der Does
2002; Moore et al. 2000) and free TRP levels (Biggio et al.
1974; Moja et al. 1989; Fadda 2000) have been suggested
as being the predominant mediator of central TRP
availability, meaning that changes in either parameter
may result in an active control that may either under-
estimate or overestimate the effects of the depletion drinks
(Reilly et al. 1997).
Although poorly documented, consumption of an ATD
and placebo AA drink can be associated with quite un-
pleasant adverse effects, particularly gastrointestinal com-
plaints with occasional vomiting, in some individuals.
These adverse events occur predominantly in the first 1 h
or 2 h after drink administration and subside in the ensuing
hours, having little or no impact on assessments done 5 h
or 6 h after the start of the treatment. However, these
transient adverse effects can be quite discomforting for the
subjects and may ultimately lead to subjects withdrawing
from the study (Schmitt et al. 2000; Riedel et al. 1999;
Danjou et al. 1990; Klaassen et al. 1999), while vomiting
may diminish the level of depletion. It is clear that avoid-
ing or minimizing adverse side effects would be highly
desirable. In the present study, four subjects (all female)
reported significant nausea, three of whom vomited,
approximately 1 h after drink administration. Vomiting or
nausea was not specifically related to either the TRP−(two
occasions) or placebo (two occasions) mixtures. Vomiting
did not substantially affect the level of TRP depletion:
plasma TRP level was reduced by 80% and 94%, TRP/
LNAA ratios by 95% and 95% in the two individuals
mentioned. Further, excluding these subjects from analyses
of the cognitive data did not affect the outcome. Although
the current study was not specifically designed to in-
vestigate the time course, magnitude and prevalence of
drink-related side effects—this would require multiple as-
sessments over time and a non-drink group to ascertain
time of day effects—it was speculated that the CP mixture
would elicit minimal gastrointestinal effects because of its
increased palatability and lack of separate AAs. Our cur-
rent observations do not support this notion, but more
systematic investigation is warranted. However, our data
do clearly show that at the time of testing, i.e., 3–4 h after
drink administration, the CP mixtures are not associated
with significant adverse effects, which could interfere with
performance and mood assessments.
Administration of TRP−CP protein induced a mild
reduction of long-term memory function, which was
apparent only in terms of reduced speed of delayed word
recognition, but not accuracy. It is important to note that
slowing of responses in the word recognition task cannot
be attributed to a general reduction of psychomotor speed
as no effects of ATD were seen on reaction time measures
of other tasks. Overall, the memory effect is rather modest
compared with previous findings showing impaired accu-
racy of delayed recall and/or recognition (Harrison et al.
2004; Sobczak et al. 2002; Riedel et al. 1999; Schmitt et
al. 2000), as well as reduced speed of recognition (Riedel
et al. 1999). Nevertheless, our findings are in line with
accumulating evidence implicating 5-HT in long-term
memory functioning (Riedel 2004; Buhot et al. 2000;
Meneses 1999). It is therefore rather unexpected that long-
term memory for abstract patterns was unaffected by ATD
in the current study. Our results conflict with those re-
ported by Rubinzstein et al. (2001), who found that ATD
impaired delayed recognition of previously presented
abstract patterns. However, in the latter study, feedback
was given on the correctness of each response, and task
performance may have been modulated by altered feed-
back processing following ATD. Elliott et al. (1996)
showed that depressed patients [associated with reduced
levels of serotonin (Maes and Meltzer 1995)] were over-
sensitive to negative feedback and in general showed a
bias for negative stimuli (Murphy et al. 1999). The ab-
sence of ATD effects on this task, however, should also
be viewed in the context of the overall picture of rather
modest ATD effects, which may be related to more gen-
eral methodological factors.
An overall explanation for the relatively mild long-term
memory effects may be the timing of the post-treatment
assessments. Generally, a 4-h to 5-h interval is maintained
between ATD drink administration and subsequent testing,
allowing for a maximal depletion of peripheral and pre-
sumably central TRP levels. A limited set of data from a
series of pilot studies (unpublished data) suggested that
maximal TRP depletion following TRP−CP administra-
tion was achieved after a 3-h to 4-h interval (Fig. 2).
However, it is possible that, due to a delay between
peripheral depletion and central 5-HT deficiency (Biggio
et al. 1974), this time may be too short to produce robust
Fig. 2 The ratio between tryptophan and the other large amino acids
(TRP/LNAA in percentage change) for four young healthy
volunteers (p1–4) tested in a pilot experiment. Blood levels of
TRP and the other large amino acids were measured at baseline and
at hourly intervals until 6 h after consumption of the TRP−collagen
protein mixture (100 g)
97

Page 7

effects on memory function. In addition, it is possible that
the use of an active control mixture that increases TRP
levels leads to an overestimation of the cognitive effects of
ATD in previous studies, since TRP depletion may have
been compared with a state of enhanced TRP availability
(Reilly et al. 1997). It can be argued that these factors may
also underlie the absence of effect on the probabilistic
reversal-learning task. However, other factors relating to
the procedure and task characteristics may have also
influenced the results. Task familiarity has been shown to
modulate the effects of ATD on a number of tests of
executive functioning (Gallagher et al. 2003), including
probabilistic reversal learning (Murphy et al. 2002). It
appears that ATD may diminish performance predomi-
nantly when the task is novel to the participants, e.g., on a
single reversal switch on the first test day (Murphy et al.
2002). In this respect, it is important to note that the
inclusion of separate practice sessions in the present study
excludes any novelty effects. Furthermore, in the pre-
sented study, multiple reversal switches were assessed at
each session as the reversal-learning task was designed to
allow blocked fluorescence magnetic resonance imaging
analyses in future research. It is always dangerous to draw
conclusions based on negative results, especially when
other factors may be involved, but the lack of ATD effects
on reversal learning when novelty effects are excluded
appears to be in line with the notion that the effects of
ATD on probabilistic reversal learning may depend on task
familiarity.
In conclusion, we have demonstrated that administra-
tion of a low-TRP collagen-based protein depletes peri-
pheral TRP, decreases the peripheral TRP/LNAA ratio and
presumably diminishes TRP availability in the brain. The
biochemical effects are very similar to those observed after
a traditional ATD that is based on AA drinks, but with a
more neutral placebo with regard to both TRP and TRP/
LNAA levels. Our data suggest that the CP method is a
suitable alternative for the AA mixtures in TRP depletion
research. Further studies will need to focus on the side-
effect profile, optimal lag time for maximum biochemical
and behavioral effects, and the validation of the method
with regard to mood and cognitive changes in healthy,
vulnerable and clinical populations.
Acknowledgment
are funded by the ZonMW grant 912-02-050.
E. A. T. Evers, F. M. van der Veen and J. Jolles
References
Anderson IM, Richell RA, Bradshaw CM (2003) The effects of
acutetryptophan depletion on probabilistic choice. J Psycho-
pharmacol 17:3–7
Biggio G, Fadda F, Fanni P, Tagliamonte A, Gessa G (1974) Rapid
depletion of serum tryptophan, brain tryptophan, serotonin and
5 hydroxyindolacetic by a tryptophan-free diet. Life Sci
14:1321–1329
Buhot M, Martin S, Segu L (2000) Role of serotonin in memory
impairment. Ann Med 32:210–221
Carpenter LL, Anderson GM, Pelton GH, Gudin JA, Kirwin PD,
Price LH, Heninger GR, McDougle CJ (1998) Tryptophan
depletion during continuous CSF sampling in healthy human
volunteers. Neuropsychopharmacology 19:26–35
Danjou P, Hamon M, Lacomblez L, Warot D (1990) Psychomotor,
subjective and neuroendocrine effects of acute tryptophan
depletion in the healthy volunteer. Psychiatry Psychobiol 5:31–
38
van der Does AJ (2001) The effects of tryptophan depletion on
mood and psychiatric symptoms (review). J Affect Disord
64:107–119
van Eijk HM, Rooyakkers DR, Deutz NE (1993) Rapid routine in
amino acids in plasma by high-performance liquid chromatog-
raphy with a 2–3 microns spherisorb ODS ll column. J
Chromatogr 620:143–148
Elliott R, Sahakian BJ, Herrod JJ, Robbins TW, Paykel ES (1996)
Neuropsychological impairments in unipolar depression: the
influence of perceived failure on subsequent performance.
Psychol Med 26:975–989
Fadda F (2000) Tryptophan-free diets: a physiological tool to study
brain serotonin function. News Physiol Sci 15:260–264
Gallagher P, Massey AE, Young AH, McAllister-Williams RH
(2003) Effects of acute tryptophan depletion on executive
function in healthy volunteers. BMC Psychiatry 3:10
Gartside SE, Cowen PJ, Sharp T (1992) Effect of amino-acid loads
on hippocampal 5-HT release in vivo evoked by electrical
stimulation of the dorsal raphe nucleus and D-fenfluramine
administration. Br J Pharmacol 107:448P
Harrison BJ, Olver JS, Norman TR, Burrows GD, Wesnes KA,
Nathan PJ (2004) Selective effects of acute serotonin and
catecholamine depletion on memory in healthy women. J
Psychopharmacol 18:32–40
Klaassen T, Riedel WJ, van Someren A, Deutz NE, Honig A, van
Praag HM (1999) Mood effects of 24-hour tryptophan deple-
tion in healthy first-degree relatives of patients with affective
disorders. Biol Psychiatry 15:489–497
Lezak MD (1995) Neuropsychological assessment. Oxford Uni-
versity, New York
Lieben CK, Blokland A, Westerink B, Deutz NE (2004) Acute
tryptophan and serotonin depletion using an optimized trypto-
phan-free protein-carbohydrate mixture in the adult rat.
Neurochem Int 44:9–16
Maes M, Meltzer HY (1995) Psychopharmacology: the fourth
generation of progress. Psychopharmacology: the serotonin
hypothesis of major depression. Raven, New York
McNair DM, Lorr DM, Droppelman LF (1988) Manual for the
profile of mood states. San Diego, California
Meneses A (1999) 5-HT system and cognition. Neurosci Biobehav
Rev 23:1111–1125
Moja EA, Cipolla P, Castoldi D, Tofanetti O (1989) Dose-response
decrease in plasma tryptophan and in brain tryptophan and
serotonin after tryptophan-free amino acid mixtures in rats. Life
Sci 44:971–976
Moore P, Landolt HP, Seifritz E, Clark C, Bhatti T, Kelsoe J,
Rapaport M, Gillin JC (2000) Clinical and physiological
consequences of rapid tryptophan depletion. Neuropsychophar-
macology 23:601–622
Murphy FC, Sahakian BJ, Rubinsztein JS, Michael A, Rogers RD,
Robbins TW, Paykel ES (1999) Emotional bias and inhibitory
control processes in mania and depression. Psychol Med
29:1307–1321
Murphy FC, Smith KA, Cowen PJ, Robbins TW, Sahakian BJ
(2002) The effects of tryptophan depletion on cognitive and
affective processing in healthy volunteers. Psychopharmacolo-
gy 163:42–53
Nishizawa S, Benkelfat C, Young SN, Leyton M, Mzengeza S, De
Montigny C, Blier P, Diksic M (1997) Differences between
males and females in rates of serotonin synthesis in human
brain. Proc Natl Acad Sci U S A 94:5308–5313
O’ Doherty J, Kringelbach ML, Rolls ET, Hornak J, Andrews C
(2001) Abstract reward and punishment representations in the
human orbitofrontal cortex. Nature 4:95–102
98

Page 8

Pollack I, Norman, DA (1964) A non-parametric analysis of re-
cognition experiments. Psychon Sci 1:125–126
Reilly JG, McTavish SF, Young AH (1997) Rapid depletion of
plasma tryptophan: a review of studies and experimental
methodology. J Psychopharmacol 11:381–392
Riedel W (2004) Cognitive changes after acute tryptophan deple-
tion: what can they tell us? Psychol Med 34:3–8
Riedel WJ, Klaassen T, Deutz NEP, Someren van A, Praag van HM
(1999) Tryptophan depletion in normal volunteers produces
selective impairment in memory consolidation. Psychopharma-
cology 141:362–369
Riedel WJ, Klaassen T, Schmitt JA (2002) Tryptophan, mood, and
cognitive function (review). Brain Behav Immun 16:581–589
Rogers RD, Blackhaw AJ, Middleton HC, Matthews K, Hawtin H,
Crowley C, Hopwood A, Wallace C, Deakin JFW, Sahakian BJ,
Robbins TW (1999) Tryptophan depletion impairs stimulus-
reward learning while methylphenidate disrupts attentional
control in healthy young adults: implications for the monoam-
inergic basis of impulsive behaviour. Psychopharmacology
146:482–491
Rogers RD, Tunbridge EM, Bhagwagar Z, Drevets WC, Sahakian
BJ, Carter CS (2003) Tryptophan depletion alters decision-
making of healthy volunteers through altered processing of
reward cues. Neuropsychopharmacology 28:153–162
Rubinsztein JS, Rogers RD, Riedel WJ, Mehta MA, Robbins TW,
Sahakian BJ (2001) Acute tryptophan depletion impairs
maintenance of “affective set” and delayed visual recognition
in healthy volunteers. Springer, Berlin Heidelberg New York
Schmitt JA, Jorissen BL, Sobzak S, van Boxtel MP, Hogervorst E,
Deutz NE, Riedel WJ (2000) Tryptophan depletion impairs
memory consolidation but improves focused attention in healthy
young volunteers. J Psychopharmacol 14:21–29
Sobczak S, Riedel WJ, Booij I, Aan Het Rot M, Deutz NE, Honig A
(2002) Cognition following acute tryptophan depletion: dif-
ference between first-degree relatives of bipolar disorder pa-
tients and matched healthy control volunteers. Psychol Med
32:503–515
Teff KL, Young SN, Blundell JE (1989) The effect of protein or
carbohydrate breakfasts on subsequent plasma amino acid
levels, satiety and nutrient selection in normal males. Pharma-
col Biochem Behav 34:829–837
Weltzin TE, Fernstrom MH, Kaye WH (1994) Serotonin and
bulimia nervosa. Nutr Rev 52:399–408
Williams WA, Shoaf SE, Hommer D, Rawlings R, Linnoila M
(1999) Effects of acute tryptophan depletion on plasma and
cerebrospinal fluid tryptophan and 5-hydroxyindoleacetic acid
in normal volunteers. J Neurochem 72:1641–1647
99