- Email: [email protected]
Effects of fluoxetine administration on mood response to tryptophan depletion in healthy subjects
Effects of Fluoxetine Administration on Mood Response to Tryptophan Depletion in Healthy Subjects Linda C. Barr, George R. Heninger, Wayne Goodman, Dennis S. Charney, and Lawrence H. Price
Short-term reduction in plasma tryptophan (tryptophan depletion) produces a relapse of depressive symptoms in 60% of previously depressed patients recently recovered with serotonin reuptake inhibitor treatment. Tryptophan depletion does not consistently increase depressive symptoms in unmedicated depressed patients or in depressed patients whose symptoms are remitted with a norepinephrine reuptake inhibitor. These data suggest that serotonin reuptake inhibitor treatment itself may confer vulnerability to the development of depressive symptoms during tryptophan depletion. In order to further investigate this possibility, six healthy individuals underwent double-blind placebo-controlled tryptophan depletion before and following six weeks of treatment with fluoxetine 20 mg/day. No increased vulnerability to the mood-lowering effects of tryptophan depletion occurred as a result of fluoxetine treatment. Additionally, fluoxetine treatment itself was not associated with changes in mood or quality of life in these healthy volunteers. These data indicate that serotonin reuptake inhibitor treatment alone does not produce the depressive effects of tryptophan depletion that are observed in serotonin reuptake inhibitor-treated depressed and obsessive compulsive disorder patients. © 1997 Society of Biological Psychiatry
Key Words: Tryptophan, serotonin, mood, depression, fluoxetine BIOL PSYCHIATRY 1997;41;949--954
Introduction The tryptophan (5-HT) depletion (TD) paradigm reliably produces transient decreases in plasma free and total tryptophan (TRP) (Moja et al 1989). Preclinical data From the Clinical Neuroscience Research Unit, Abraham Ribicoff Research Facilities, Connecticut Mental Health Center, Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut. Address reprint requests to Linda C. Barr, MD, #116 A Veterans An'airs Medical Center, 950 Cambell Avenue, West Haven, CT 06516. Received October 13, 1995; revised April 8, 1996.
© 1997 Society of Biological Psychiatry
suggest these reductions in plasma TRP may be accompanied by parallel reductions in brain TRP and brain 5-HT (Moja et al 1989; Curzon 1979; Femstrom 1979). A 15-amino acid (AA) drink (Young et al 1985, 1989) deficient in TRP induces protein synthesis in the liver, which in turn rapidly depletes plasma TRP stores. Maximal TD occurs about 5 hours after drink administration and the peak behavioral effects are observed about 2 hours later. Administration of a low TRP diet for 24 hours prior to the AA drink, together with the depleting effects of the 0006-3223/96/$17.00 PII S0006-3223(96)00224-7
950
BIOLPSYCHIATRY
L.C. Barr et al
1996;41:949-954
drink, decreases plasma TRP to approximately 85%-90% of baseline (Delgado et al 1990). This in turn is estimated to reduce central 5-HT levels by approximately 60%-70% (Moja et al 1989). In healthy subjects, TD both significantly decreases nocturnal secretion of melatonin (Zimmerman et al 1993) (which is synthesized from 5-HT) and significantly increases yohimbine-induced anxiety (Goddard et al 1994). These findings provide indirect evidence that TD in humans is associated with alterations in central 5-HT function. TD is associated with the development of depressive symptoms in remitted depressed (Delgado et al 1990) and obsessive compulsive disorder (Barr et al 1994) patients. These effects appear most prominent in those patients who received treatment with serotonin reuptake inhibitors (SRIs). TD has inconsistent effects on mood in unmedicated depressed (Delgado et al 1994) and obsessive compulsive patients, suggesting that serotonin reuptake inhibitor treatment itself may contribute to vulnerability to the mood-lowering effects of TD. In healthy men not taking medication, TD induces a mild lowering of mood without development of clinically significant depressive symptoms (Young et al 1985; Smith et al 1987). Healthy men with mildly elevated Beck depression scale scores or those with a family history of affective disorder (Benkelfat et al 1994) appear to be particularly vulnerable to TD. These data suggest that the development of clinically significant depressive symptoms may rest on an interaction between underlying vulnerability to the development of depression, the presence of acute illness, and SRI treatment. The current investigation examined the behavioral consequences of TD prior to and following fluoxetine treatment in healthy subjects without a family history of depression. The specific aim of this study was to assess whether 6 weeks of the SRI fluoxetine, given in antidepressant doses, would confer on healthy subjects the vulnerability to the depressant effects of TD similar to that which has been observed in psychiatric patients receiving SRI treatment.
Methods Healthy Subjects The study was conducted at the Clinical Neuroscience Research Unit of the Connecticut Mental Health Center. Subjects were recruited by posting requests for "healthy subjects." Six subjects (4 men and 2 women, mean ___ SD age = 32.2 __+ 7.5 years) participated. Subjects were included if they were physically healthy and had no personal history of psychiatric disorder as determined by the Structured Clinical Interview for DSM-III-R (SCIDNP) (Spitzer et al 1992; Williams et al 1992) and had no
reported history of psychiatric illness in first-degree relatives. The initial mean 24-item modified Hamilton Depression Scale (HAM-D) (Hamilton 1960; Mazure et al 1986) score for the six subjects was 0.25 _+ 0.62 and the mean + SD Beck Depression Inventory (BDI) score was 1.5 ___ 1.31. Subjects were excluded from the study by the presence of any serious medical or neurologic illness as assessed by complete medical history, physical exam, and screening laboratory testing and serum pregnancy test (for women). Written informed consent was obtained prior to participation.
Procedure Subjects participated in a TD and sham test prior to and during fluoxetine administration. The overall design of the study is shown in Figure 1.
TD Test TD and sham testing were preceded by a 24-hour low-TRP diet (10500 kJ (4000 kcal), 7% protein, 27% fat, 64% carbohydrate) supplemented with either placebo (depletion diet) or 1.5 g of TRP (sham diet). After consuming the 24-hour diet, on the morning of the study, patients received a 15-AA drink that was supplemented with either placebo (depletion test) or 2.3 g of TRP (sham test). A detailed description of the diet and AA drink appears elsewhere (Delgado et al 1990). The sequence of depletion and sham testing was randomly assigned in a double-blind fashion with tests separated by 7-10 days. Normal TRP intake resumed 8 hours after the AA drink was given. Behavioral ratings were obtained by a research psychiatrist immediately before starting the diet (day 1: 9AM), and 15 minutes prior to and 5 and 7 hours following administration of the AA drink (day 2 : 8 : 4 5 AM, 2PM, 4PM). Behavioral ratings included the following: a 13-item 100 mm self-rated visual analog scale (VAS) (talkative, happy, drowsy, nervous, sad, calm, depressed, anxious, energetic, fearful, mellow, high, angry), the 65-item Profile of Mood States (POMS), modified to examine current mood (McNair 1992), and the HAM-D. Free and total plasma TRP levels were obtained prior to starting the 24-hour diet, 15 minutes prior to administration of the AA drink, and 5 hours following it.
Fluoxetine Administration Following completion of the initial pair of TD and sham tests, each subject was given fluoxetine 20 mg/day for 6 weeks. Subjects were evaluated weekly during fluoxetine treatment in order to assess compliance and to obtain behavioral ratings. Weekly behavioral ratings during fluoxetine administration consisted of the HAM-D, the
Mood Response to Tryptophan Depletion
BIOLPSYCHIATRY
951
1996;41:949 -954
RATINGS OF MOOD AND "QUALITY OF LIFE"
Week Number -2
0
EOW 1
EOW 2
EOW 3
EOW 4
EOW 5
EOW 6
-1
A ., t,,.,-r,o
Day #
n ~-
D.. >.
ra
12.
+1
+1
@1
÷1
C3
C3 13_
O..
3 q
q
1
Day #
TRYP Depletion or Sham
2
2
TRYPorDepletionsham
I
Figure 1. Design of study.
POMS, the Beck Depression Inventory (BDI) (Beck et al 1961) and a quality of life scale adapted from those employed by Testa et al (1989).
Biochemical Methods Total plasma TRP was assayed by high-performance liquid chromatography with fluorometric detection. Free plasma TRP was assayed by obtaining the ultrafiltrate of plasma from cellulose-based filters (30,000-molecular weight cutoff (1000g) at room temperature and subjecting the ultrafiltrate to the high-performance liquid chromatography with fluorometric detection method (Anderson et al 1981).
Statistical Analysis Change in TRP levels and behavioral ratings were analyzed by means of an analysis of variance (ANOVA) with repeated measures. Main effects of test type (TD vs sham) and time (changes over the time points sampled) were determined, as was the interaction of challenge X time (indicating the overall effect of the TRP depletion during each of the 2 phases of the study). The reported p values for all ANOVAs incorporate the Huynh-Feldt correction factor. Significant effects identified by ANOVA were investigated using post h o c t tests. The Wilcoxon matchedpairs signed rank test was employed to assess the effect of fluoxetine treatment on mood and quality of life.
Results Changes in Total and Free Plasma TRP TRP depletion prior to fiuoxetine treatment decreased mean ± SD total plasma TRP from 64.7 ___ 14.2 ~zmol/L at baseline to 7.5 ± 1.8 ~zmol/L 5 hours after the TRP-free AA drink was administered (88% decrease) (t = 8.7; df = 1,5; p < 0.0001). Mean free plasma TRP decreased from 9.8 ± 2.7 ~mol/L at baseline to 1.12 ± 39 txmol/L following the AA drink (89% decrease) (t = 9.6; df = 1,5; p < 0.0001). With sham testing, mean total plasma TRP increased from 59.8 _ 9.9 txmol/L at baseline to 115 ± 25.9 ~mol/L following the sham drink (t = 5.79; df = 1,5; p = 0.002). Mean free plasma TRP increased from 8.9 ± 3.4 txmol/L at baseline to 17.1 ± 7.8 p~mol/L 5 hours after the AA drink (t = - 4 . 0 ; df = 1,5; p = 0.002). Following fluoxetine treatment, mean plasma TRP levels showed similar changes with the TRP depletion and sham procedures. Total plasma TRP during TRP depletion decreased from 60.2 - 9.8 Ixmol/L to 6.6 --- 2.7 ~mol/L following the TRP-free AA drink (89% decrease) (t = 13.4; df = 1,5; p < 0.001). Free plasma TRP during TRP depletion decreased from 7.7 ± 1.8 p~mol/L to 1.0 ± .1 p~mol/L following the TRP-free AA drink (87% decrease) (t = 9.43; df = 1,5; p < 0.001). With sham testing mean total plasma TRP increased from 59.3 --- 13.7 Ixmol/L to 102.9 --- 29.3 p~mol/L (t = 4.41; df = 1,5; p = 0.01). Mean free plasma TRP increased from 7.7 ± .8 txmol/L at baseline to 16.6 ___ 9.3 rLmol/L following the sham drink (t = -2.58; df = 1,5; p = 0.06).
952
BIOLPSYCHIATRY
L.C. Barr et al
1996;4 ! : 9 4 9 - 9 5 4
,0o] (N=4)
[]
Sham
•
Depletion
Work
8O C
E
g
e
,
0
60
60
0
55
0 O
50
(~ 40 0
<
e
a5'
>,
40
2O O 35
0
Day 1
9AM
Day 2
2PM
Day 2
4PM Day 2
O,
Figure 2. Mean ± SD for self ratings of "happy" for healthy subjects following tryptophan and sham depletion prior to fluoxetine treatment. *p = 0.07 compared to 9AM day 2 by paired t test; **p = 0.013 compared to 9AM day 2 by paired t test.
Baseline
Week 6
Daily Living 50-
Behavioral Effects of TD Prior to Fluoxetine Treatment Self-report VAS ratings revealed a trend toward a significant main effect of challenge (F = 4.08; df = 1,10; p = 0.07) and a significant challenge X time interaction (F = 6.15; df = 2,20; p = 0.008) for self-ratings of happy. As shown in Figure 2, depletion decreased the mean _+ SD self-rating of happy from 51.2 ___ 9.1 before the A A drink to 42.7 ___ 13.8 at 300 min (t = 2.2, df = 1,5, p = 0.08) and to 35.3 _ 16.3 at 420 min (t = 3.8, df = 1,5, p = 0.013). In contrast, during sham testing self-ratings of happy were unchanged from the baseline of 51.7 ___ 25.9 (300 min = 62.7 --- 12.2 (t = 1.5, df = 1,5,p = 0.19) and 4 2 0 m i n = 6 3 . 0 - l l . 0 ( t = - 1 . 3 , d f = 1,5, p = 0 . 2 3 ) ) (Figure 2). A N O V A of H A M - D ratings also revealed a significant main effect of challenge (F = 8.37; df = 1,10; p = 0.02) and a challenge X time interaction (F = 3.72; df = 3,30; p = 0.02) Subsequent examination of the means demonstrated that these effects were based partly on small differences in the baseline mean H A M - D scores. A N O V A revealed no significant effects of challenge, time or challenge X time interactions for the BDI, POMS subscales, or other VAS self-ratings.
O)
45.
.e ~
-
~.
ao-
'5
9
,00 '-I
35.
•1~
30-
o
25.
"6
"
0
Baseline
Week
6
Leisure 8580. 75-
O
7065-
60. 55-
g
50. 45. 0
Behavioral Effects of TD During Fluoxetine Treatment A N O V A revealed no significant effects of challenge, time, or challenge X time interactions for the HAM-D, BDI, POMS subscales, or V A S self-ratings, except for selfreport ratings of hunger which showed a main effect of time. Mean ratings of hunger increased during both TD
Baseline
Week 6
Figure 3. Ratings of mood and quality of life during fluoxetine treatment. and sham testing. No subject reported symptoms of depression or sadness with TD or sham testing either before or following fluoxetine treatment.
Mood Response to Tryptophan Depletion
Weekly Ratings of Mood and Quality of Life During FLX Treatment Comparison of the POMS subscales (depression, tension, anger, vigor, fatigue, confusion) and the quality of life subscales (healthy, cognitive, work, daily living, leisure) at baseline and following 6 weeks of fluoxetine treatment revealed no significant changes (Figure 3).
Discussion In this small sample of healthy volunteers, fluoxetine treatment was not associated with the development of clinically significant changes in mood following TD. Rather, fluoxetine treatment appeared to have a protective effect against the very mild mood-lowering effect of TD seen when subjects were drug-free. It is possible that TD loses some potency in reducing mood when the treatment is applied repeatedly and that this might explain the absence of a mood-lowering effect when subjects underwent a second tryptophan depletion while receiving fluoxetine. The six-week time interval between depletions appears to reduce the likelihood that this would explain the absence of an effect. These findings suggest that the clinically significant mood-lowering effects of TD which are seen in depressed and obsessive-compulsive patients treated with SRIs are not produced solely by an interaction between SRI treatment and TD. Rather, the current finding further establishes the particular vulnerability of patients with depression and obsessive-compulsive disorder to the moodlowering effects of TD. The exact nature of this vulnerability remains undefined as yet. It is perhaps self-evident that those patients who have responded to serotonin uptake inhibitor treatment might be particularly vulnerable to the effects of tryptophan depletion. Similarly, Miller et al (1996) have reported that patients who have responded to tricyclic antidepressants are particularly vulnerable to the depressant effects of treatment with alpha-methylparatyrosine, an inhibitor of norepinephrine and dopamine synthesis. Taken together, these findings suggest that these two classes of antidepressants differ in their mechanism of action.
BIOLPSYCHIATRY 1996;41:949-954
953
Fluoxetine treatment did not appear to improve mood or quality of life in these healthy subjects. This finding contradicts recent speculation that fluoxetine treatment may be associated with substantial psychotropic effects in individuals free of psychiatric disease (Fox, 1994). The number of subjects in the study was small, however, and thus prohibits any firm conclusion regarding the effects of fluoxetine in healthy subjects. Nonetheless, we are not aware of any other published report on this issue that has systematically and objectively assessed changes in mood and quality of life. The small sample size also limits the capacity of the study to detect possibly important differences between men and women in their vulnerability to the moodlowering effects of TD. Preliminary data gathered by our group have suggested that some healthy women may experience prominent mood worsening in association with TD. In a study of the effects of TD on nocturnal melatonin secretion, Zimmerman et al (1993) studied five women in the early follicular phase of their menstrual cycle. TD was associated with the development of mild depressive symptoms in four of the five women studied. The finding of a mild mood-lowering effect of TD in healthy subjects is consistent with a study by Weltzin et al (1994) who also observed mild increases in depressed mood in healthy subjects undergoing TD; however, Oldman et al (1994) did not observe a mood-lowering effect of TD in a group of healthy women. Differences in the composition of the amino acid mixture used to accomplish the TD may explain this discrepancy. This study does not address whether individuals who are at increased risk for the future development of depression (by virtue of family or past personal history) might be especially vulnerable to the mood-lowering effects of TD, as has been suggested by Benkelfat et al (1994) in their study of at-risk men. Future studies will need to examine the interaction between sex and at-risk states with respect to the depressant effects of TD. Supported in part by National Institute of Mental Health Grants MH36229, MH14276, MH30929, and MH45802 and by the State of Connecticut.
References Anderson GM, Young JG, Cohen DJ, Schlicht KR, Patel N (1981): Liquid-chromatographic determination of serotonin and tryptophan in whole blood and plasma: Clin Chem 27:775-776. (Editorial) (1994): "Artificial paradise encapsulated (The controversy surrounding Prozac)(fluoxetine)". Fox, Robin 343:865. Barr LC, Goodman WK, McDougle CJ, Delgado PL, Heninger GR, Charney DS, Price LH (1994): Tryptophan depletion in
patients with obsessive-compulsive disorder who respond to serotonin reuptake inhibitors. Arch Gen Psychiatry 51:309317. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J (1961): An inventory for measuring depression. Arch Gen Psychiatry 4:561-571. Benkelfat C, Ellenbogen MA, Dean P, Palmour RM, Young SN (1994): Mood-lowering effect of tryptophan depletion. En-
954
BIOLPSYCHIATRY
L.C. Barr et al
1996;41:949-954
hanced susceptibility in young men at genetic risk for major affective disorder. Arch Gen Psychiatry 51(9):687-97. Curzon G (1979): Relationships between plasma, CSF and brain tryptophan. J Neural Transm 15(suppl):81-92. Delgado PL, Price LH, Aghajanian GK, Landis H, Heninger GR (1990): Serotonin function and the mechanism of antidepressant action. Reversal of antidepressant-induced remission by rapid depletion of plasma tryptophan. Arch Gen Psychiatry 47:411-418. Delgado PL, Price LH, Miller HL, Salomon RM, Aghajanian GK, Heninger GR, Charney DS (1994): Serotonin and the neurobiology of depression: Effects of tryptophan depletion in drug-free depressed patients. Arch Gen Psychiatry 51:86574. Fernstrom JD (1979): Effects of the diet on brain neurotransmitters. Metabolism 26:207-223. Goddard AW, Charney DS, Germine M, Woods SW, Heninger GR, Krystal JH, Goodman WK, Price LH (1994): Effects of tryptophan depletion on responses to yohimbine in healthy human subjects. Biol Psychiatry 38:74-85. Hamilton M (1960): A rating scale for depression. J Neurol Neurosurg Psychiatry 23:56-62. Mazure CM, Nelson JC, Price LH (1986): Reliability and validity of the symptoms of major depressive illness. Arch Gen Psychiatry 43:451-456. McNair DM, Lorr M, Droppleman LF (1992): EDITS Manual for the Profile of Mood States. San Diego, CA: EdiTS/Educational and Industrial Testing Service. Miller HI, Delgado PL, Salomon RM, Berman R, Krystal JH, Heninger GR, Chamey DS (1996): Clinical and biochemical effects of catecholamine depletion on antidepressant-induced remission of depression. Arch Gen Psychiatry 53:117-128. Moja EA, Cipollo P, Castoldi D, Tofanetti O (1989): Dose response decrease in plasma tryptophan and in brain trypto-
phan and serotonin after tryptophan-free amino acid mistures in rats. Life Sci 44:971-976. Oldman AD, Walsh AES, Salkovskis P, Laver DA, Cowen PJ (1994): Effect of acute tryptophan depletion on mood and appetite in healthy female volunteers. J Psychopharmacology 8(1):8-13. Smith SE, Pihl RO, Young SN, Ervin FR (1987): A test of possible cognitive and environmental influences on the mood lowering effect of tryptophan depletion in normal males. Psychopharmacology 91:451-457. Spitzer RL, Williams JBW, Gibbon M (1992): The structured clinical interview for DSM-III-R (SCID). I: History, rationale, and description. Arch Gen Psychiatry 49:624-629. Testa MA, Sudilovsky A, Rippey RM, Williams GH (1989): A short form for clinical assessment of quality of life among hypertensive patients. Am J Prev Med 5:82-89. Weltzin TE, Fernstrom JD, Mc Conaha C, Kaye WH (1994): Acute tryptophan depletion in bulemia: Effects on large neutral amino acids. Biol Psychiatry 35:388-397. Williams JBW, Gibbon M, First MB, Spitzer RL, Davies M, Borus J, Howes MJ, Kane J, Pope HG Jr, Rounsaville B, Wittchen H-U (1992): The structured clinical interview for DSM-III-R (SCID). II. Multisite test-retest reliability. Arch Gen Psychiatry 49:630-636. Young SN, Smith SE, Pihl R, Ervin FR (1985): Tryptophan depletion causes a rapid lowering of mood in normal males. Psychopharmacology (Berlin) 87:173-177. Young SN, Ervin FR, Pihl RO, Finn P (1989): Biochemical aspects of tryptophan depletion in primates. Psychopharmacology (Berlin) 98:508-511. Zimmerman RC, McDougle C J, Schumacher M, Olcese J, Mason JW, Heninger GR, Price LH (1993a): Effects of acute tryptophan depletion on nocturnal melatonin secretion in humans. J Clin Endocr Metab 76:1106-1164.
Short-term reduction in plasma tryptophan (tryptophan depletion) produces a relapse of depressive symptoms in 60% of previously depressed patients recently recovered with serotonin reuptake inhibitor treatment. Tryptophan depletion does not consistently increase depressive symptoms in unmedicated depressed patients or in depressed patients whose symptoms are remitted with a norepinephrine reuptake inhibitor. These data suggest that serotonin reuptake inhibitor treatment itself may confer vulnerability to the development of depressive symptoms during tryptophan depletion. In order to further investigate this possibility, six healthy individuals underwent double-blind placebo-controlled tryptophan depletion before and following six weeks of treatment with fluoxetine 20 mg/day. No increased vulnerability to the mood-lowering effects of tryptophan depletion occurred as a result of fluoxetine treatment. Additionally, fluoxetine treatment itself was not associated with changes in mood or quality of life in these healthy volunteers. These data indicate that serotonin reuptake inhibitor treatment alone does not produce the depressive effects of tryptophan depletion that are observed in serotonin reuptake inhibitor-treated depressed and obsessive compulsive disorder patients. © 1997 Society of Biological Psychiatry
Key Words: Tryptophan, serotonin, mood, depression, fluoxetine BIOL PSYCHIATRY 1997;41;949--954
Introduction The tryptophan (5-HT) depletion (TD) paradigm reliably produces transient decreases in plasma free and total tryptophan (TRP) (Moja et al 1989). Preclinical data From the Clinical Neuroscience Research Unit, Abraham Ribicoff Research Facilities, Connecticut Mental Health Center, Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut. Address reprint requests to Linda C. Barr, MD, #116 A Veterans An'airs Medical Center, 950 Cambell Avenue, West Haven, CT 06516. Received October 13, 1995; revised April 8, 1996.
© 1997 Society of Biological Psychiatry
suggest these reductions in plasma TRP may be accompanied by parallel reductions in brain TRP and brain 5-HT (Moja et al 1989; Curzon 1979; Femstrom 1979). A 15-amino acid (AA) drink (Young et al 1985, 1989) deficient in TRP induces protein synthesis in the liver, which in turn rapidly depletes plasma TRP stores. Maximal TD occurs about 5 hours after drink administration and the peak behavioral effects are observed about 2 hours later. Administration of a low TRP diet for 24 hours prior to the AA drink, together with the depleting effects of the 0006-3223/96/$17.00 PII S0006-3223(96)00224-7
950
BIOLPSYCHIATRY
L.C. Barr et al
1996;41:949-954
drink, decreases plasma TRP to approximately 85%-90% of baseline (Delgado et al 1990). This in turn is estimated to reduce central 5-HT levels by approximately 60%-70% (Moja et al 1989). In healthy subjects, TD both significantly decreases nocturnal secretion of melatonin (Zimmerman et al 1993) (which is synthesized from 5-HT) and significantly increases yohimbine-induced anxiety (Goddard et al 1994). These findings provide indirect evidence that TD in humans is associated with alterations in central 5-HT function. TD is associated with the development of depressive symptoms in remitted depressed (Delgado et al 1990) and obsessive compulsive disorder (Barr et al 1994) patients. These effects appear most prominent in those patients who received treatment with serotonin reuptake inhibitors (SRIs). TD has inconsistent effects on mood in unmedicated depressed (Delgado et al 1994) and obsessive compulsive patients, suggesting that serotonin reuptake inhibitor treatment itself may contribute to vulnerability to the mood-lowering effects of TD. In healthy men not taking medication, TD induces a mild lowering of mood without development of clinically significant depressive symptoms (Young et al 1985; Smith et al 1987). Healthy men with mildly elevated Beck depression scale scores or those with a family history of affective disorder (Benkelfat et al 1994) appear to be particularly vulnerable to TD. These data suggest that the development of clinically significant depressive symptoms may rest on an interaction between underlying vulnerability to the development of depression, the presence of acute illness, and SRI treatment. The current investigation examined the behavioral consequences of TD prior to and following fluoxetine treatment in healthy subjects without a family history of depression. The specific aim of this study was to assess whether 6 weeks of the SRI fluoxetine, given in antidepressant doses, would confer on healthy subjects the vulnerability to the depressant effects of TD similar to that which has been observed in psychiatric patients receiving SRI treatment.
Methods Healthy Subjects The study was conducted at the Clinical Neuroscience Research Unit of the Connecticut Mental Health Center. Subjects were recruited by posting requests for "healthy subjects." Six subjects (4 men and 2 women, mean ___ SD age = 32.2 __+ 7.5 years) participated. Subjects were included if they were physically healthy and had no personal history of psychiatric disorder as determined by the Structured Clinical Interview for DSM-III-R (SCIDNP) (Spitzer et al 1992; Williams et al 1992) and had no
reported history of psychiatric illness in first-degree relatives. The initial mean 24-item modified Hamilton Depression Scale (HAM-D) (Hamilton 1960; Mazure et al 1986) score for the six subjects was 0.25 _+ 0.62 and the mean + SD Beck Depression Inventory (BDI) score was 1.5 ___ 1.31. Subjects were excluded from the study by the presence of any serious medical or neurologic illness as assessed by complete medical history, physical exam, and screening laboratory testing and serum pregnancy test (for women). Written informed consent was obtained prior to participation.
Procedure Subjects participated in a TD and sham test prior to and during fluoxetine administration. The overall design of the study is shown in Figure 1.
TD Test TD and sham testing were preceded by a 24-hour low-TRP diet (10500 kJ (4000 kcal), 7% protein, 27% fat, 64% carbohydrate) supplemented with either placebo (depletion diet) or 1.5 g of TRP (sham diet). After consuming the 24-hour diet, on the morning of the study, patients received a 15-AA drink that was supplemented with either placebo (depletion test) or 2.3 g of TRP (sham test). A detailed description of the diet and AA drink appears elsewhere (Delgado et al 1990). The sequence of depletion and sham testing was randomly assigned in a double-blind fashion with tests separated by 7-10 days. Normal TRP intake resumed 8 hours after the AA drink was given. Behavioral ratings were obtained by a research psychiatrist immediately before starting the diet (day 1: 9AM), and 15 minutes prior to and 5 and 7 hours following administration of the AA drink (day 2 : 8 : 4 5 AM, 2PM, 4PM). Behavioral ratings included the following: a 13-item 100 mm self-rated visual analog scale (VAS) (talkative, happy, drowsy, nervous, sad, calm, depressed, anxious, energetic, fearful, mellow, high, angry), the 65-item Profile of Mood States (POMS), modified to examine current mood (McNair 1992), and the HAM-D. Free and total plasma TRP levels were obtained prior to starting the 24-hour diet, 15 minutes prior to administration of the AA drink, and 5 hours following it.
Fluoxetine Administration Following completion of the initial pair of TD and sham tests, each subject was given fluoxetine 20 mg/day for 6 weeks. Subjects were evaluated weekly during fluoxetine treatment in order to assess compliance and to obtain behavioral ratings. Weekly behavioral ratings during fluoxetine administration consisted of the HAM-D, the
Mood Response to Tryptophan Depletion
BIOLPSYCHIATRY
951
1996;41:949 -954
RATINGS OF MOOD AND "QUALITY OF LIFE"
Week Number -2
0
EOW 1
EOW 2
EOW 3
EOW 4
EOW 5
EOW 6
-1
A ., t,,.,-r,o
Day #
n ~-
D.. >.
ra
12.
+1
+1
@1
÷1
C3
C3 13_
O..
3 q
q
1
Day #
TRYP Depletion or Sham
2
2
TRYPorDepletionsham
I
Figure 1. Design of study.
POMS, the Beck Depression Inventory (BDI) (Beck et al 1961) and a quality of life scale adapted from those employed by Testa et al (1989).
Biochemical Methods Total plasma TRP was assayed by high-performance liquid chromatography with fluorometric detection. Free plasma TRP was assayed by obtaining the ultrafiltrate of plasma from cellulose-based filters (30,000-molecular weight cutoff (1000g) at room temperature and subjecting the ultrafiltrate to the high-performance liquid chromatography with fluorometric detection method (Anderson et al 1981).
Statistical Analysis Change in TRP levels and behavioral ratings were analyzed by means of an analysis of variance (ANOVA) with repeated measures. Main effects of test type (TD vs sham) and time (changes over the time points sampled) were determined, as was the interaction of challenge X time (indicating the overall effect of the TRP depletion during each of the 2 phases of the study). The reported p values for all ANOVAs incorporate the Huynh-Feldt correction factor. Significant effects identified by ANOVA were investigated using post h o c t tests. The Wilcoxon matchedpairs signed rank test was employed to assess the effect of fluoxetine treatment on mood and quality of life.
Results Changes in Total and Free Plasma TRP TRP depletion prior to fiuoxetine treatment decreased mean ± SD total plasma TRP from 64.7 ___ 14.2 ~zmol/L at baseline to 7.5 ± 1.8 ~zmol/L 5 hours after the TRP-free AA drink was administered (88% decrease) (t = 8.7; df = 1,5; p < 0.0001). Mean free plasma TRP decreased from 9.8 ± 2.7 ~mol/L at baseline to 1.12 ± 39 txmol/L following the AA drink (89% decrease) (t = 9.6; df = 1,5; p < 0.0001). With sham testing, mean total plasma TRP increased from 59.8 _ 9.9 txmol/L at baseline to 115 ± 25.9 ~mol/L following the sham drink (t = 5.79; df = 1,5; p = 0.002). Mean free plasma TRP increased from 8.9 ± 3.4 txmol/L at baseline to 17.1 ± 7.8 p~mol/L 5 hours after the AA drink (t = - 4 . 0 ; df = 1,5; p = 0.002). Following fluoxetine treatment, mean plasma TRP levels showed similar changes with the TRP depletion and sham procedures. Total plasma TRP during TRP depletion decreased from 60.2 - 9.8 Ixmol/L to 6.6 --- 2.7 ~mol/L following the TRP-free AA drink (89% decrease) (t = 13.4; df = 1,5; p < 0.001). Free plasma TRP during TRP depletion decreased from 7.7 ± 1.8 p~mol/L to 1.0 ± .1 p~mol/L following the TRP-free AA drink (87% decrease) (t = 9.43; df = 1,5; p < 0.001). With sham testing mean total plasma TRP increased from 59.3 --- 13.7 Ixmol/L to 102.9 --- 29.3 p~mol/L (t = 4.41; df = 1,5; p = 0.01). Mean free plasma TRP increased from 7.7 ± .8 txmol/L at baseline to 16.6 ___ 9.3 rLmol/L following the sham drink (t = -2.58; df = 1,5; p = 0.06).
952
BIOLPSYCHIATRY
L.C. Barr et al
1996;4 ! : 9 4 9 - 9 5 4
,0o] (N=4)
[]
Sham
•
Depletion
Work
8O C
E
g
e
,
0
60
60
0
55
0 O
50
(~ 40 0
<
e
a5'
>,
40
2O O 35
0
Day 1
9AM
Day 2
2PM
Day 2
4PM Day 2
O,
Figure 2. Mean ± SD for self ratings of "happy" for healthy subjects following tryptophan and sham depletion prior to fluoxetine treatment. *p = 0.07 compared to 9AM day 2 by paired t test; **p = 0.013 compared to 9AM day 2 by paired t test.
Baseline
Week 6
Daily Living 50-
Behavioral Effects of TD Prior to Fluoxetine Treatment Self-report VAS ratings revealed a trend toward a significant main effect of challenge (F = 4.08; df = 1,10; p = 0.07) and a significant challenge X time interaction (F = 6.15; df = 2,20; p = 0.008) for self-ratings of happy. As shown in Figure 2, depletion decreased the mean _+ SD self-rating of happy from 51.2 ___ 9.1 before the A A drink to 42.7 ___ 13.8 at 300 min (t = 2.2, df = 1,5, p = 0.08) and to 35.3 _ 16.3 at 420 min (t = 3.8, df = 1,5, p = 0.013). In contrast, during sham testing self-ratings of happy were unchanged from the baseline of 51.7 ___ 25.9 (300 min = 62.7 --- 12.2 (t = 1.5, df = 1,5,p = 0.19) and 4 2 0 m i n = 6 3 . 0 - l l . 0 ( t = - 1 . 3 , d f = 1,5, p = 0 . 2 3 ) ) (Figure 2). A N O V A of H A M - D ratings also revealed a significant main effect of challenge (F = 8.37; df = 1,10; p = 0.02) and a challenge X time interaction (F = 3.72; df = 3,30; p = 0.02) Subsequent examination of the means demonstrated that these effects were based partly on small differences in the baseline mean H A M - D scores. A N O V A revealed no significant effects of challenge, time or challenge X time interactions for the BDI, POMS subscales, or other VAS self-ratings.
O)
45.
.e ~
-
~.
ao-
'5
9
,00 '-I
35.
•1~
30-
o
25.
"6
"
0
Baseline
Week
6
Leisure 8580. 75-
O
7065-
60. 55-
g
50. 45. 0
Behavioral Effects of TD During Fluoxetine Treatment A N O V A revealed no significant effects of challenge, time, or challenge X time interactions for the HAM-D, BDI, POMS subscales, or V A S self-ratings, except for selfreport ratings of hunger which showed a main effect of time. Mean ratings of hunger increased during both TD
Baseline
Week 6
Figure 3. Ratings of mood and quality of life during fluoxetine treatment. and sham testing. No subject reported symptoms of depression or sadness with TD or sham testing either before or following fluoxetine treatment.
Mood Response to Tryptophan Depletion
Weekly Ratings of Mood and Quality of Life During FLX Treatment Comparison of the POMS subscales (depression, tension, anger, vigor, fatigue, confusion) and the quality of life subscales (healthy, cognitive, work, daily living, leisure) at baseline and following 6 weeks of fluoxetine treatment revealed no significant changes (Figure 3).
Discussion In this small sample of healthy volunteers, fluoxetine treatment was not associated with the development of clinically significant changes in mood following TD. Rather, fluoxetine treatment appeared to have a protective effect against the very mild mood-lowering effect of TD seen when subjects were drug-free. It is possible that TD loses some potency in reducing mood when the treatment is applied repeatedly and that this might explain the absence of a mood-lowering effect when subjects underwent a second tryptophan depletion while receiving fluoxetine. The six-week time interval between depletions appears to reduce the likelihood that this would explain the absence of an effect. These findings suggest that the clinically significant mood-lowering effects of TD which are seen in depressed and obsessive-compulsive patients treated with SRIs are not produced solely by an interaction between SRI treatment and TD. Rather, the current finding further establishes the particular vulnerability of patients with depression and obsessive-compulsive disorder to the moodlowering effects of TD. The exact nature of this vulnerability remains undefined as yet. It is perhaps self-evident that those patients who have responded to serotonin uptake inhibitor treatment might be particularly vulnerable to the effects of tryptophan depletion. Similarly, Miller et al (1996) have reported that patients who have responded to tricyclic antidepressants are particularly vulnerable to the depressant effects of treatment with alpha-methylparatyrosine, an inhibitor of norepinephrine and dopamine synthesis. Taken together, these findings suggest that these two classes of antidepressants differ in their mechanism of action.
BIOLPSYCHIATRY 1996;41:949-954
953
Fluoxetine treatment did not appear to improve mood or quality of life in these healthy subjects. This finding contradicts recent speculation that fluoxetine treatment may be associated with substantial psychotropic effects in individuals free of psychiatric disease (Fox, 1994). The number of subjects in the study was small, however, and thus prohibits any firm conclusion regarding the effects of fluoxetine in healthy subjects. Nonetheless, we are not aware of any other published report on this issue that has systematically and objectively assessed changes in mood and quality of life. The small sample size also limits the capacity of the study to detect possibly important differences between men and women in their vulnerability to the moodlowering effects of TD. Preliminary data gathered by our group have suggested that some healthy women may experience prominent mood worsening in association with TD. In a study of the effects of TD on nocturnal melatonin secretion, Zimmerman et al (1993) studied five women in the early follicular phase of their menstrual cycle. TD was associated with the development of mild depressive symptoms in four of the five women studied. The finding of a mild mood-lowering effect of TD in healthy subjects is consistent with a study by Weltzin et al (1994) who also observed mild increases in depressed mood in healthy subjects undergoing TD; however, Oldman et al (1994) did not observe a mood-lowering effect of TD in a group of healthy women. Differences in the composition of the amino acid mixture used to accomplish the TD may explain this discrepancy. This study does not address whether individuals who are at increased risk for the future development of depression (by virtue of family or past personal history) might be especially vulnerable to the mood-lowering effects of TD, as has been suggested by Benkelfat et al (1994) in their study of at-risk men. Future studies will need to examine the interaction between sex and at-risk states with respect to the depressant effects of TD. Supported in part by National Institute of Mental Health Grants MH36229, MH14276, MH30929, and MH45802 and by the State of Connecticut.
References Anderson GM, Young JG, Cohen DJ, Schlicht KR, Patel N (1981): Liquid-chromatographic determination of serotonin and tryptophan in whole blood and plasma: Clin Chem 27:775-776. (Editorial) (1994): "Artificial paradise encapsulated (The controversy surrounding Prozac)(fluoxetine)". Fox, Robin 343:865. Barr LC, Goodman WK, McDougle CJ, Delgado PL, Heninger GR, Charney DS, Price LH (1994): Tryptophan depletion in
patients with obsessive-compulsive disorder who respond to serotonin reuptake inhibitors. Arch Gen Psychiatry 51:309317. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J (1961): An inventory for measuring depression. Arch Gen Psychiatry 4:561-571. Benkelfat C, Ellenbogen MA, Dean P, Palmour RM, Young SN (1994): Mood-lowering effect of tryptophan depletion. En-
954
BIOLPSYCHIATRY
L.C. Barr et al
1996;41:949-954
hanced susceptibility in young men at genetic risk for major affective disorder. Arch Gen Psychiatry 51(9):687-97. Curzon G (1979): Relationships between plasma, CSF and brain tryptophan. J Neural Transm 15(suppl):81-92. Delgado PL, Price LH, Aghajanian GK, Landis H, Heninger GR (1990): Serotonin function and the mechanism of antidepressant action. Reversal of antidepressant-induced remission by rapid depletion of plasma tryptophan. Arch Gen Psychiatry 47:411-418. Delgado PL, Price LH, Miller HL, Salomon RM, Aghajanian GK, Heninger GR, Charney DS (1994): Serotonin and the neurobiology of depression: Effects of tryptophan depletion in drug-free depressed patients. Arch Gen Psychiatry 51:86574. Fernstrom JD (1979): Effects of the diet on brain neurotransmitters. Metabolism 26:207-223. Goddard AW, Charney DS, Germine M, Woods SW, Heninger GR, Krystal JH, Goodman WK, Price LH (1994): Effects of tryptophan depletion on responses to yohimbine in healthy human subjects. Biol Psychiatry 38:74-85. Hamilton M (1960): A rating scale for depression. J Neurol Neurosurg Psychiatry 23:56-62. Mazure CM, Nelson JC, Price LH (1986): Reliability and validity of the symptoms of major depressive illness. Arch Gen Psychiatry 43:451-456. McNair DM, Lorr M, Droppleman LF (1992): EDITS Manual for the Profile of Mood States. San Diego, CA: EdiTS/Educational and Industrial Testing Service. Miller HI, Delgado PL, Salomon RM, Berman R, Krystal JH, Heninger GR, Chamey DS (1996): Clinical and biochemical effects of catecholamine depletion on antidepressant-induced remission of depression. Arch Gen Psychiatry 53:117-128. Moja EA, Cipollo P, Castoldi D, Tofanetti O (1989): Dose response decrease in plasma tryptophan and in brain trypto-
phan and serotonin after tryptophan-free amino acid mistures in rats. Life Sci 44:971-976. Oldman AD, Walsh AES, Salkovskis P, Laver DA, Cowen PJ (1994): Effect of acute tryptophan depletion on mood and appetite in healthy female volunteers. J Psychopharmacology 8(1):8-13. Smith SE, Pihl RO, Young SN, Ervin FR (1987): A test of possible cognitive and environmental influences on the mood lowering effect of tryptophan depletion in normal males. Psychopharmacology 91:451-457. Spitzer RL, Williams JBW, Gibbon M (1992): The structured clinical interview for DSM-III-R (SCID). I: History, rationale, and description. Arch Gen Psychiatry 49:624-629. Testa MA, Sudilovsky A, Rippey RM, Williams GH (1989): A short form for clinical assessment of quality of life among hypertensive patients. Am J Prev Med 5:82-89. Weltzin TE, Fernstrom JD, Mc Conaha C, Kaye WH (1994): Acute tryptophan depletion in bulemia: Effects on large neutral amino acids. Biol Psychiatry 35:388-397. Williams JBW, Gibbon M, First MB, Spitzer RL, Davies M, Borus J, Howes MJ, Kane J, Pope HG Jr, Rounsaville B, Wittchen H-U (1992): The structured clinical interview for DSM-III-R (SCID). II. Multisite test-retest reliability. Arch Gen Psychiatry 49:630-636. Young SN, Smith SE, Pihl R, Ervin FR (1985): Tryptophan depletion causes a rapid lowering of mood in normal males. Psychopharmacology (Berlin) 87:173-177. Young SN, Ervin FR, Pihl RO, Finn P (1989): Biochemical aspects of tryptophan depletion in primates. Psychopharmacology (Berlin) 98:508-511. Zimmerman RC, McDougle C J, Schumacher M, Olcese J, Mason JW, Heninger GR, Price LH (1993a): Effects of acute tryptophan depletion on nocturnal melatonin secretion in humans. J Clin Endocr Metab 76:1106-1164.