170
BIOL PSYCHIATRY 1992;32:170-- 178
Platelet Imipramine Binding and Plasma Cortisol Levels in Israeli Civilians During the Gulf War Ronit Weizman, Irit Gil-Ad, Leon Karp, Nathaniel Laor, Yerachmiel Barber, and Moshe Rehavi
[31"1] imipramine binding to platelets and plasma cortisol levels were measured in nine Israeli civilians before, during, and after repeated missile attacks. Hamilton Anxiety Rating Scale (HARS), and Beck Depression Inventory (BDI) were similar before and during the war and decreased significantly after the war. A trend toward increase in platelet imipramine binding values was observed during the war when compared with the postwar values (19%; p = 0.056), and/or when compared with prewar values (26%; p = 0.063). However, one-way analysis of variance with repeated measures did not reveal a significant statistical difference IF (2,6) = p = 2.07; NS.I among the three time points. A significant correlation was found between HARS score, but not BDi, and imipramine binding values in the prewar and postwar time points (r = 0.87; 0.71, respectively). Plasma cortisol levels did not alter significantly during the study period.
Introduction During the Gulf War (which lasted 40 days), the city of Tel-Aviv was under repeated (almost nightly) missile attacks. This uncontrollable environmental stress was used in the present study as a paradigm for situational stress-related anxiety. Such condition may lead to the development of learned helplessness because of the repeated experience of failing to attain control over the threat to life (Seligman et al 1980; Abramson et al 1978). Both behavioral and pharmacological studies suggest that learned helplessness may be a useful model for depression (Abramson et al 1978; Drugan et al 1989). Similarly, an association between stressful life events and depression was reported (Weiss and Simson 1985; Breier et al 1987; Ormel and Sanderman 1989). Treatment with antidepressants can prevent the behavioral deficits in animals exposed to uncontrollable stress (Sherman et al 1982) as well as the neurochemical changes induced by chronic stress (Adell et al 1989). The serotonin "transporter", as measmed by [3Hlimipramine binding (Raisman et al 1979; Langer et al 1980; Rehavi et al 1980; Paul et al 1981) is decreased in platelets of major depressed patients (Briley et al 1980; Paul et al 1981, 1984; Raisman et al 1982;
From the Tel Aviv Community Mental Health Center ~,RW, YB, NL), and Sadder Faculty of Medicine (RW, LK. MR), TelAviv University, Tel Aviv; the Institute of Pediatric and Adolescent Endocrinology, Beilinson Medical Center, Petah Tikva (IG-A); and the Geha Psychiatric Hospital, Petah Tikva (LK), Tel-Aviv University, Tel-Aviv, Israel. Yale Child Study C.enter Address reprint requests to Moshe Rehavi, Ph.D., Department of Physiology and Pharmacology, Sackler Faculty of Medicine, Tel Aviv University, TeI-Aviv 69978, Israel. Received December 19, 1991: revised April 9, 1992. © 1992 Society of Biological Psychiatry
0006-3223/921505.00
Imipramine Binding during the Gulf War
BIOLPSYCHIATRY 1992;32:170-178
171
Rehavi et al 1984a) although some investigators demonstrated unaltered imipramine binding in major depression (Rehavi et al 1984b; Whitaker et al 1984). Reduction of the maximal binding capacity of [3Hlimipramme binding was reported in rats demonstrating learned helplessness (Sherman and Petty 1984). Yet, a recent study (Edwards et al 1991), using [3H]paroxetine to label the serotonin uptake site, has exhibited increased density in the hippocampus and hypothalamus and no alteration in the cerebral cortex in learned helpless rats. The possible role of the serotonergic system in mediating the impact of repeated stress and of learned helplessness in humans has not yet been clearly elucidated. Platelet [3H]imipramine binding and plasma cortisol levels in normal volunteers were measured before, during, and after the Gulf War. These parameters were correlated to ratings of depression and anxiety. To the best of our knowledge, the present study is the first one to test the hypothesis concerning the involvement of the serotonergic system in stress-exposed humans under natural disastrous conditions. Moreover, this study is unique in that it ventures to capture the development of the clinical events and their biological underpinnings from the very first moments of stress: the time of alert, the peak of attack, and the relaxation period. Naturally, the participation of subjects was hard to come by and hard to control. Twelve subjects participated, but because of technical problems, three could not complete the study. However, because of the relevance of the serotonergic system in stress, learned helplessness and depression, and the unique opportunity that had presented itself to us, we thought it desirable to report our findings.
Methods
Subjects The study population consisted of nine healthy, drug-free volunteers (five men, four women), 28-37 years old (mean -+ SD 30.9 +- 3.4 years). The participants were medical personnel from two mental health centers in Tel-Aviv who gave informed consent to participate in the study. Current and past history of psychiatric disorders was excluded using a clinical interview. None of the participants reported any traumatic event (e.g., road accident, loss of a spouse or child, or divorce) in the past. Four of the five male subjects had an active combat history in a previous war.
Procedure Blood samples for determination of platelet imipramine binding and plasma cortisol levels were obtained three times, between 8:00 AM and 9:00 AM. The first time point was 1 day before the outbreak of the war (prewar) followed by a second blood sample after 4 weeks of almost daily missile attacks, that is, 10 days before the end of the war (during war). A third blood sample was collected 4 weeks after the end of the war (postwar), 40 days after the second timepoint. Beck depression inventory (BDI) (Beck et al 1961) and Hamilton Anxiety Rating Scale (HARS) (Hamilton 1959) were completed at the same timepoints.
[3H]Imipramine Binding Assay Preparation of platelet membranes and [3H]imipramine (specific activity 55.4 Ci/mmol, New England Nuclear, Boston, MA) binding assay were performed as previously described (Weizman et al 1986). Eight concentrations of [3H]imipramine (0.2-8 nM) were
172
BiOLPSYCHIATRY
R. Weizman et al
1992;32:170-178
n
6 5
[] [] []
Pro war During w a r Post war
-7
4 3 2 1 0
Figure 1. HARS scores of the subjects are prewar (1 day before the outbreak of the war), during war (after 4 weeks of missile attacks), and postwar (4 weeks after the end of the war). **p < 0.005 versus prewar and during war scores.
used to examine the binding to platelets of each subject. Scatchard plots were constructed and both the maximal binding (Bm~,) of [aH]imipramine and its affinity to its binding sites (Kd) were determined by linear regression analysis. The reliability for the binding assay is 95%. [3H]Imipramine was chosen to label the serotonin transporter as this ligand has been used extensively in our previous clinical studies and we have a reliable database concerning its binding parameters in humans.
Hormonal Determination Plasma cortisol levels were measured by solid-phase radioimmunoassay (RIA) using materials provided by Diagnostic Products Corporation (Los Angeles, CA). The antibody was covalently bound to the inner surface of a polypropylene tube. The sensitivity of the assay was 0.5 I~g/dl, and the mean intraassay coefficient of variation was 7%. All the hormonal assays were run in the same batch.
Statistical Evaluation Data were analyzed using three-way analysis of variance (ANOVA) with repeated measures with gender, time, and combat experience as independent variants for intragroup variation. The Pearson correlation test was used to evaluate the correlation between the clinical information (depression and anxiety ratings) and the laboratory data. All results are expressed as mean _+ SEM.
Results The clinical and laboratory data of the study population are illustrates in Figures I-4. The mean HARS and BDI scores were significantly elevated before and during the war as compared with postwar values [HARS: F (2,6) = 12.77; p < 0.005, BDI: F(2,6) = ! 1.04, p < 0.005], (Figures 1,2). A 26% and 19% increase in platelet [3H]imipramine binding were obser:ed during the war compared with prewar and postwar values, respectively. The individual Bmax values are given in Table 1. As shown in Table 1, the during war Bmax values were elevated compared with the prewar and post war values in six and seven out of nine
Imipramine Binding during the Gulf War
4.0
Pre war
BIOLPSYCHIATRY 1992;32:170-178
173
m
[ ] Duringwar IHI Post w a r
3.0 m
2.0
1.0
0.0
Figure 2. BDI scores of the subjects before, during, and after the repeated missile attacks. **p < 0.005 versus prewar and during war scores.
subjects, respectively. Although the difference in Bmax did not reach a significant statistical level [F(2,6) - 2.07; NS], a trend toward increase was observed during the exposure to missile attacks (during war versus prewar: 552 --. 58 versus 439 __. 57 fmol/mg protein, paired t = 2.16, df = 8, p = 0.063; during war versus postwar: 552 _.+ 58 versus 465 +-. 42 fmol/mg protein, paired t = 2.24, df = 8, p = 0.056) (Figure 3). The dissociation constant (Kd) for [aH]imiprarnine binding in platelet remained unaltered during the entire study period [F(2,6) = 1.09; NS]. A significant (p < 0.05) positive correlation was observed between the Bmax values and HARS scores at the prewar and postwar time points (r = 0.87 and 0.71, respectively), but not during the war (Table 2). No correlation was found between BDI scores and Bmx values. Basal morning plasma cortisol levels were nonsignificantly decreased before and during the war when compared with postwar values (Fig. 4). Cortisol levels did not correlate either with [3Hlimipramine binding values or with anxiety and depression scores. Imipramine binding values and plasma cortisol levels as well as HARS and BDI scores did not differ significantly in men and women, but sample sizes were too small to rule this out. Four men had a past combat experience (served as soldiers in one of the previous wars in Israel). Their HARS, BDI, and laboratory measures did not differ from the values of the noncombat subjects. Table 1. The Individual Bmax (fmol/mg protein) Values of the Study Population Subject no
Prewar
During war
Postwar
1 2 3 4 5 6 7 8 9
112 297 367 389 505 583 499 690 510
390 460 696 374 445 469 531 874 733
306 320 608 502 355 393 569 635 499
439 57
552 58
465 42
Mean SEM
174
BIOLPSYCHIATRY
R. Weizman et al
1992;32:170-178
Table 2. The Correlation Between HARS Scores and Bmax(fmol/mg protein) Values Time Point
HARS
B,m~
r
p
Prewar During war Postwar
5.8 _-+ 0.8 5.0 _ 1.0 1.4 -4- 0.4
439 ± 57 552 ± 58 465 ± 42
0.87 0.19 0.71
0.024 0.62 0.034
Discussion The present study began 1 day before the outbreak of the Gulf War. At that time, we expected the crisis to resolve in a matter of days. Somehow, perhaps in the service of denial, we did not expect active attacks on Tel-Aviv. We did expect, however, a short period of stress. Hence, the initial aim of our study was to evaluate the involvement of the serotonergic system and adrenocortical response to short-term environmental stress. As we completed 1 day of sampling (12 subjects), our area was hit by a Scud attack which capped the size of our study sample and turned it into the "pre-Scud" sample. Furthermore, it called for a change in design. We could either increase the sample regardless of the dramatic environmental change (which would have been methodologically flawed), or discard our initial sample yet stay with our initial aim and collect, a new study sample under the new conditions (taking the chance that conditions may vary again). The third possibility, the one we actually endorsed, was to take the unique advantage of having collected a pre-Scud sample and continue to follow our subjects in order to assess biological parameters at model points of change in stress. As expected, a significant decrease in anxiety and depression scores was observed after the war when compared with prewar and during war scores. The normal range of plasma cortisol levels, despite the exposure to the stressful situation, may be an artifact of design imposed by situational constrains: all missile attacks occurred at night, whereas hormonal mcasui'cmeni~ were perlormed in the mornh~g. It is likely that the peak release of cortisol was during and immediately after the missile attack event at night, and thus was undetectable the following morning (>6 hr after that night missile attack). Alter-
600 -g
500
[ ] Pr¢ war [] During war [ ] Post war
k..
40O 300 v
E re~
2O0 100 0
Figure 3. [3H]lmipraminebinding values (Bmax)in platelets of the subjects exposed to the Gulf War. Tile during-war binding value was higher compared with prewar (p = 0.063) and postwar value (p = 0.056).
Imipramine Binding ,during the Gulf War
BIOLPSYCHIATRY 1992;32:170--178
! 75
12 10
[] [] []
Pre w~r During war Post ~,~r
8 v
6 L_
o
¢.!
4
Figure 4. Basal morning plasma cortisol levels in the study population.
natively, it is possible that the repeated exposure to the stress led to habituation and adaptation of the hypothalamic-pituitary-adrenai axis. It is noteworthy that unaltered cortisol levels were previously reported by us in examination stress (Karl) et al 1989) and in soldiers exposed to repeated parachute jumps (Dar et al 1991). Owing to the spasmodic nature of secretion, it is not surprising that single estimations of plasma cortisol level were not different at the three time points. Unfortunately, due to the war situation, we were unable to measure 24 hr urinary-free cortisol excretion. A line of evidence suggests possible involvement of the serotonergic system in stress (Adell et al 1989). Previous studies have demonstrated that acute stress is associated with increased plasma serotonin (Evron et al 1982) and whole blood serotonin levels (Davis et al 1985), as well as increased utilization of brain serotonin (Adell et al 1988) and marked stimulation of serotonin turnover (Joseph and Kennett 1983). Chronic stress was demonstrated to induce increased synthesis of serotonin (Adell et ai 1988) in rat brain. Pa~'tial agonists at the serotonin-lA receptor have anxiolytic effects (Feighner et al 1982; Traber and Glaser 1987). Drugs that activate the serotonergic system are anxiogenic, whereas drugs that interfere with serotonergic neurotransmission possess anxiolytic properties (Ceulemans et al 1985; Chopin and Briley 1987; Soderpalm and Engel 1989; Stutzmann et at 1991). The involvement of the serotonin uptake site in chronic stress is yet unclear. Sherman and Petty (1984) showed diminished brain [3H]imipramine binding in learned helpless rats. However, two recent studies (File et al 1990; Edwards et al 1991) demonstrated increased serotonin uptake capacity in zat brain fc!lowing repeated stress. Repeated exposure to the situational stress during the war tended to increase the platelet imipramine maximal binding capacity. However, the elevated binding capacity during the war was not statistically different from the prewar and postwar values. The small number of participants can account for the lack of statistical significance. Though not statistically significant, the trend toward increase in platele~ "serotonin transporter" in humans in stressful conditions is intriguing. The associettion between stress-induced anxiety and increased density of the serotonin transporter is supported by the observation of a significant positive correlation between HARS scores and Bmaxvalues in the prewar and postwar time points. However, it is not clear why such a correlation did not exist during the war. It is possible that the increase in depression (as measured by BDI) during the war and the habituation to the repeated stress blunted the full expression of the upregulatory effect of the anxiety on the imipramine maximal binding capacity. Selective
176
BIOLPSYCHIATRY
R. Weizman et al
1992;32:170-178
inhibition of serotonin reuptake by fluoxetine is associated with insomnia, nervousness, and anxiety (Cooper 1988). Because chronic stress inducez enhanced serotonergic sensitivity (Mogilnicka 1981; Kennett et al 1985), it is possible that the increased serotonin uptake capacity is a compensatory mechanism to oppose the hyperfunction of the serotonergic system. One is tempted to speculate that hypofunction of serotonergic transmission in major depression leads to an adaptive down-regulation of the presynaptic serotonin transporter in order to diminish the reuptake and to maintain adequate neurotransmitter concentration in the synaptic cleft. In contrast, enhanced serotonergic transmission following repeated stress leads to upregulation of the transporter to amplify the reuptake process from the synaptic cleft to the presynaptic neuron. This hypothesis, however, calls for further testing of the role of the serotonergic system in the neurobiological response of stress. It seems that prolonged stress of different kinds (anticipatory~ before war and uncontrolled lethal threat--during war) causes habituation or adaptation of both the hypothalamic-pituitary-adrenal ax;.s and the serotonergic system.
References Abramson LY, Seligman MEP, Teasdale JD (1978): Learned helplessness in humans: Critique and reformulation. J. Abnorm Psychol 78:49-74. Adell A, Garcia-Marquez C, Armafio A, Gelpi E (1988): Chronic stress increases serotonin and noradrenaline in rat brain and sensitizes their responses to a further acute stress. J Neurochem 50:1678-1681. Adell A, Garcia-Marquez C, Armario A, Gelpi E (1989): Chronic administration of ¢lomipramine prevents the increase in serotonin and noradrenaline induced by chronic stress. Psychopharmacology 99:22-26. Beck AT, Wand CH, Mendelsohn M, Mock J, Ernbaugh J (1961): An inventory for measuring depression. Arch Gen Psychiatry 453-59. Brcier A, Albus M, Pickar D, Zahn TP, Wolkowitz OM, Paul SM (1987): Controllable and uncontrollable stress ill humans: Alterations in mood, neuroendocrine ano psychophysiologi¢ function. Am J Psychiat~. 144:14i9-1425. Briley MS, Langer SZ, Reisman R, Seehter, D, Zarifian E (19'.~0): Tritiated imipramine binding sites are decreased in platelets of untreated depressed patient:;. Science 2~'~9:303-305. Ceulemans DL, Hoppenbrouwers HL, Gelders ¥G, Reyntjens AJ (1985): The inl?ueneeof ritanserin, a serotonin antagonist, in anxiety disorders: A double-blind, placebo contmllod study versus Iorazepam, Pharmacopsyehiatry 18:303-305. Chopin P, Briley M (1987): Animal models of anxiety: The effect ot compouads that modify 5]-iT neurotransmission. Trends Pharmacol Sci 8:383-388. Cooper GL (1988): The safety of fluoxetine---~n update. Br J Psychiatry 153(Suppl 3):77-86. Dar DE, Weizman A, Karl) L, et al (1991): Platelet peripheral bcnzodiazepine receptor in repeated stress. Life Sci 48:341-346. Davis DD, Dunlop SR, Shea P, Brittain H, Hendrie HC (1985): Biological stress responses in high and low trait anxious students. Biol Psychiatr), 20:860-868. Drugan RC, Morrow LA, Weizman R, et al (1989): Stress-induced behavior depression in the rat is associated with a decrease in GABA receptor-mediated chloride ion flux and brain benzodiazepine receptor occupancy. Brain Res 487:45-51. Edwards E, Harkins K, Wright G, Henn F (1991): Modulation of [3H]paroxetine binding to the 5hydroxytryptamine uptake site in an animal model of depression. J Neurochem 56:1581-1586. Evron S, Pfeifer Y, Sadovsky E, Sulman G (1982): Plasma levels of serotonin compared with corfisol as stress indicators in gynecological operations, isr J Med Sci 18:806--808.
Imipramine Binding during the Gulf War
BIOLPSYCHIATRY 1992;32:170-178
177
Feighner JP, Meredith CH, Hendfickson GA (1982): A double-blind comparison of buspirone and diazepam in outpatients with generalized anxiety disorder. J Clin Psychiatry 43:103-107. File SE, Andrews N, Zharkovsky A (1990): Handling habituation and chlordiazepoxide have different effects on GABA and 5-HT function in the frontal cortex and hippocampus. Eur J Pharmacol 190:229-234. Hamilton M (1959): The assessment of anxiety states by rating. Br J Med Psychol 32:50-55. Joseph MH, Kennett GA (1983): Stress-induced release of 5-HT in the hippoeampus and its dependence on increased tryptophan availability: An in vivo electrochemical study. Brain Res 270:251-257. Karp L, Weizman A, Tyano S, Garish M (1989): Examination stress, platelet peripheral benzodiazepine binding sites and plasma hormone levels. Life Sci 44:1077-1082. Kennett GA, Dickinson SL, Curzon G (1985): Enhancement of some 5-HT-dependent behavioral responses following repeated immobilization in rats. Brain Res 330:253-264. Langer SZ, Moret C, Raisman R, Dubocovich ML, Briley M (1980): High affinity [3H]imipramine binding in rat hypothalamus: Association with uptake of serotonin but not of norepinephrine. Science 210:! 133-1135. Mogilnieka E (1981): REM sleep deprivation changes behavioral response to catecholaminergic and serotonergic receptor activation in rats. Ptmrmacol Biochem Behav 15:149- ! 5 !. Ormel J, Sanderman R 0989): Life events, porsonal control and depression. In Steptoe A, Appels AD (eds), Strauss, Personal Control and Health. Chichester, UK" John Wiley & Sons, pp 193213. Paul SM, Rehavi M, Rice KC, Ittah Y, Skolnick P (1981): Does high affinity [3H]imiFra~ine binding label serotonin reuptake sites in brain and platelet? Life Sci 28:2753-2760. Paul SM, Rehavi M, Skolnick P, Goodwin FK (1984): High affinity binding of antidepressants to biogenic amine transporter sites in human brain and platelet: Studies in depression. In Post RM, Ballenger JC (eds), Neurobioiogy of Mood Disorders. Baltimore: Williams & Wilkins, pp 846853. Raisman R, Briley M, Langer SZ (1979): Specific tricyclic antidepressant binding sites in rat brain. Nature 281:148-150. Raisman R, Briley MS, Bouchamy F, Sechter D, Zarifian E, Langer SZ (1982): 3H-Imipramine binding and serotonin uptake in platelets from untreated depressed patients and control volunteers. Psychopharmacology 77:332-335. Rehavi M, Paul SM, Skolnick P, Goodwin FK (1980): Demonstration of specific high affinity binding sites for [3H]imipramine in human brain. Life Sci 26:2273-2279. Rehavi M, Paul SM, Skolnick P (1984a): High-affinity binding sites for tricyclic antidepressants in brain and platelets. In Marangos PJ, Campbell IC, Cohen RM (eds), Brain Recepwr Methodologies, Part B. Orlando, FL: Academic Press, pp 279-295. Rehavi M, Weizman R, Carel C, Apter A, Tyano S (1984b): High-affinity 3H-imipramine binding in platelets of children and adolescents with major affective disorders. Psychiatry Res 13:3139. Seligman MEP, Weiss JM, Weinranb M, Schulman A (1980): Coping behavior, learned helplessness, physiological change and learned inactivity. Behav Res Ther 18:459-512. Sherman AD, Petty F (1984): ]Learnedhelplessness decreases [3H]-imipraminebinding in rat cortex. J Affective Disord 6:25-32. Sherman AD, Sacquitne JL, Petty F (1982): Specificity of the learned helplessness model of depression. Pharmacol Biochem Behav 16:449-454. Soderpalm B, Engel JA (1989): Does the pCPA-induced anticonflict effect involve activation of the GABAA/benzodiazepine chloride ionophore receptor complex? J Neural Transm 76:145153. Stutzmann JM, Eon B, Darche F, et al (1991): Are 5-HT2 antagonists endowed with anxiolytic properties in rodents? Neurosci Lett 128:4-8.
178
BIOLPSYCHIATRY 1992:32:170-178
R. Weizman et al
Traber J, Glaser T (1987): 5-HTIA receptor-related anxiolytics. Trends Pharmacoi Sci 8:432-437. Wei~s JM, Simson PG (I 985): Neurochemical basis of stress-induced depression. Psychopharmacol 1lull 21:447-457. Weitman R, Carmi M, Tyano S, Apter A, Rehavi M (1986): High affinity [3H]imipramine binding and serotonin uptake in platelets of adolescent females suffering from anorexia nervosa. Life Sci 38:1235-1242. Whitaker PM, Warsh JJ, Stancer HC, Persad E, Vint CK (1984): Seasonal variation in platelet 3H-imipramine binding: Comparable values in control and depressed populations. Psychiatry Res ! 1:127-135.
BIOL PSYCHIATRY 1992;32:170-- 178
Platelet Imipramine Binding and Plasma Cortisol Levels in Israeli Civilians During the Gulf War Ronit Weizman, Irit Gil-Ad, Leon Karp, Nathaniel Laor, Yerachmiel Barber, and Moshe Rehavi
[31"1] imipramine binding to platelets and plasma cortisol levels were measured in nine Israeli civilians before, during, and after repeated missile attacks. Hamilton Anxiety Rating Scale (HARS), and Beck Depression Inventory (BDI) were similar before and during the war and decreased significantly after the war. A trend toward increase in platelet imipramine binding values was observed during the war when compared with the postwar values (19%; p = 0.056), and/or when compared with prewar values (26%; p = 0.063). However, one-way analysis of variance with repeated measures did not reveal a significant statistical difference IF (2,6) = p = 2.07; NS.I among the three time points. A significant correlation was found between HARS score, but not BDi, and imipramine binding values in the prewar and postwar time points (r = 0.87; 0.71, respectively). Plasma cortisol levels did not alter significantly during the study period.
Introduction During the Gulf War (which lasted 40 days), the city of Tel-Aviv was under repeated (almost nightly) missile attacks. This uncontrollable environmental stress was used in the present study as a paradigm for situational stress-related anxiety. Such condition may lead to the development of learned helplessness because of the repeated experience of failing to attain control over the threat to life (Seligman et al 1980; Abramson et al 1978). Both behavioral and pharmacological studies suggest that learned helplessness may be a useful model for depression (Abramson et al 1978; Drugan et al 1989). Similarly, an association between stressful life events and depression was reported (Weiss and Simson 1985; Breier et al 1987; Ormel and Sanderman 1989). Treatment with antidepressants can prevent the behavioral deficits in animals exposed to uncontrollable stress (Sherman et al 1982) as well as the neurochemical changes induced by chronic stress (Adell et al 1989). The serotonin "transporter", as measmed by [3Hlimipramine binding (Raisman et al 1979; Langer et al 1980; Rehavi et al 1980; Paul et al 1981) is decreased in platelets of major depressed patients (Briley et al 1980; Paul et al 1981, 1984; Raisman et al 1982;
From the Tel Aviv Community Mental Health Center ~,RW, YB, NL), and Sadder Faculty of Medicine (RW, LK. MR), TelAviv University, Tel Aviv; the Institute of Pediatric and Adolescent Endocrinology, Beilinson Medical Center, Petah Tikva (IG-A); and the Geha Psychiatric Hospital, Petah Tikva (LK), Tel-Aviv University, Tel-Aviv, Israel. Yale Child Study C.enter Address reprint requests to Moshe Rehavi, Ph.D., Department of Physiology and Pharmacology, Sackler Faculty of Medicine, Tel Aviv University, TeI-Aviv 69978, Israel. Received December 19, 1991: revised April 9, 1992. © 1992 Society of Biological Psychiatry
0006-3223/921505.00
Imipramine Binding during the Gulf War
BIOLPSYCHIATRY 1992;32:170-178
171
Rehavi et al 1984a) although some investigators demonstrated unaltered imipramine binding in major depression (Rehavi et al 1984b; Whitaker et al 1984). Reduction of the maximal binding capacity of [3Hlimipramme binding was reported in rats demonstrating learned helplessness (Sherman and Petty 1984). Yet, a recent study (Edwards et al 1991), using [3H]paroxetine to label the serotonin uptake site, has exhibited increased density in the hippocampus and hypothalamus and no alteration in the cerebral cortex in learned helpless rats. The possible role of the serotonergic system in mediating the impact of repeated stress and of learned helplessness in humans has not yet been clearly elucidated. Platelet [3H]imipramine binding and plasma cortisol levels in normal volunteers were measured before, during, and after the Gulf War. These parameters were correlated to ratings of depression and anxiety. To the best of our knowledge, the present study is the first one to test the hypothesis concerning the involvement of the serotonergic system in stress-exposed humans under natural disastrous conditions. Moreover, this study is unique in that it ventures to capture the development of the clinical events and their biological underpinnings from the very first moments of stress: the time of alert, the peak of attack, and the relaxation period. Naturally, the participation of subjects was hard to come by and hard to control. Twelve subjects participated, but because of technical problems, three could not complete the study. However, because of the relevance of the serotonergic system in stress, learned helplessness and depression, and the unique opportunity that had presented itself to us, we thought it desirable to report our findings.
Methods
Subjects The study population consisted of nine healthy, drug-free volunteers (five men, four women), 28-37 years old (mean -+ SD 30.9 +- 3.4 years). The participants were medical personnel from two mental health centers in Tel-Aviv who gave informed consent to participate in the study. Current and past history of psychiatric disorders was excluded using a clinical interview. None of the participants reported any traumatic event (e.g., road accident, loss of a spouse or child, or divorce) in the past. Four of the five male subjects had an active combat history in a previous war.
Procedure Blood samples for determination of platelet imipramine binding and plasma cortisol levels were obtained three times, between 8:00 AM and 9:00 AM. The first time point was 1 day before the outbreak of the war (prewar) followed by a second blood sample after 4 weeks of almost daily missile attacks, that is, 10 days before the end of the war (during war). A third blood sample was collected 4 weeks after the end of the war (postwar), 40 days after the second timepoint. Beck depression inventory (BDI) (Beck et al 1961) and Hamilton Anxiety Rating Scale (HARS) (Hamilton 1959) were completed at the same timepoints.
[3H]Imipramine Binding Assay Preparation of platelet membranes and [3H]imipramine (specific activity 55.4 Ci/mmol, New England Nuclear, Boston, MA) binding assay were performed as previously described (Weizman et al 1986). Eight concentrations of [3H]imipramine (0.2-8 nM) were
172
BiOLPSYCHIATRY
R. Weizman et al
1992;32:170-178
n
6 5
[] [] []
Pro war During w a r Post war
-7
4 3 2 1 0
Figure 1. HARS scores of the subjects are prewar (1 day before the outbreak of the war), during war (after 4 weeks of missile attacks), and postwar (4 weeks after the end of the war). **p < 0.005 versus prewar and during war scores.
used to examine the binding to platelets of each subject. Scatchard plots were constructed and both the maximal binding (Bm~,) of [aH]imipramine and its affinity to its binding sites (Kd) were determined by linear regression analysis. The reliability for the binding assay is 95%. [3H]Imipramine was chosen to label the serotonin transporter as this ligand has been used extensively in our previous clinical studies and we have a reliable database concerning its binding parameters in humans.
Hormonal Determination Plasma cortisol levels were measured by solid-phase radioimmunoassay (RIA) using materials provided by Diagnostic Products Corporation (Los Angeles, CA). The antibody was covalently bound to the inner surface of a polypropylene tube. The sensitivity of the assay was 0.5 I~g/dl, and the mean intraassay coefficient of variation was 7%. All the hormonal assays were run in the same batch.
Statistical Evaluation Data were analyzed using three-way analysis of variance (ANOVA) with repeated measures with gender, time, and combat experience as independent variants for intragroup variation. The Pearson correlation test was used to evaluate the correlation between the clinical information (depression and anxiety ratings) and the laboratory data. All results are expressed as mean _+ SEM.
Results The clinical and laboratory data of the study population are illustrates in Figures I-4. The mean HARS and BDI scores were significantly elevated before and during the war as compared with postwar values [HARS: F (2,6) = 12.77; p < 0.005, BDI: F(2,6) = ! 1.04, p < 0.005], (Figures 1,2). A 26% and 19% increase in platelet [3H]imipramine binding were obser:ed during the war compared with prewar and postwar values, respectively. The individual Bmax values are given in Table 1. As shown in Table 1, the during war Bmax values were elevated compared with the prewar and post war values in six and seven out of nine
Imipramine Binding during the Gulf War
4.0
Pre war
BIOLPSYCHIATRY 1992;32:170-178
173
m
[ ] Duringwar IHI Post w a r
3.0 m
2.0
1.0
0.0
Figure 2. BDI scores of the subjects before, during, and after the repeated missile attacks. **p < 0.005 versus prewar and during war scores.
subjects, respectively. Although the difference in Bmax did not reach a significant statistical level [F(2,6) - 2.07; NS], a trend toward increase was observed during the exposure to missile attacks (during war versus prewar: 552 --. 58 versus 439 __. 57 fmol/mg protein, paired t = 2.16, df = 8, p = 0.063; during war versus postwar: 552 _.+ 58 versus 465 +-. 42 fmol/mg protein, paired t = 2.24, df = 8, p = 0.056) (Figure 3). The dissociation constant (Kd) for [aH]imiprarnine binding in platelet remained unaltered during the entire study period [F(2,6) = 1.09; NS]. A significant (p < 0.05) positive correlation was observed between the Bmax values and HARS scores at the prewar and postwar time points (r = 0.87 and 0.71, respectively), but not during the war (Table 2). No correlation was found between BDI scores and Bmx values. Basal morning plasma cortisol levels were nonsignificantly decreased before and during the war when compared with postwar values (Fig. 4). Cortisol levels did not correlate either with [3Hlimipramine binding values or with anxiety and depression scores. Imipramine binding values and plasma cortisol levels as well as HARS and BDI scores did not differ significantly in men and women, but sample sizes were too small to rule this out. Four men had a past combat experience (served as soldiers in one of the previous wars in Israel). Their HARS, BDI, and laboratory measures did not differ from the values of the noncombat subjects. Table 1. The Individual Bmax (fmol/mg protein) Values of the Study Population Subject no
Prewar
During war
Postwar
1 2 3 4 5 6 7 8 9
112 297 367 389 505 583 499 690 510
390 460 696 374 445 469 531 874 733
306 320 608 502 355 393 569 635 499
439 57
552 58
465 42
Mean SEM
174
BIOLPSYCHIATRY
R. Weizman et al
1992;32:170-178
Table 2. The Correlation Between HARS Scores and Bmax(fmol/mg protein) Values Time Point
HARS
B,m~
r
p
Prewar During war Postwar
5.8 _-+ 0.8 5.0 _ 1.0 1.4 -4- 0.4
439 ± 57 552 ± 58 465 ± 42
0.87 0.19 0.71
0.024 0.62 0.034
Discussion The present study began 1 day before the outbreak of the Gulf War. At that time, we expected the crisis to resolve in a matter of days. Somehow, perhaps in the service of denial, we did not expect active attacks on Tel-Aviv. We did expect, however, a short period of stress. Hence, the initial aim of our study was to evaluate the involvement of the serotonergic system and adrenocortical response to short-term environmental stress. As we completed 1 day of sampling (12 subjects), our area was hit by a Scud attack which capped the size of our study sample and turned it into the "pre-Scud" sample. Furthermore, it called for a change in design. We could either increase the sample regardless of the dramatic environmental change (which would have been methodologically flawed), or discard our initial sample yet stay with our initial aim and collect, a new study sample under the new conditions (taking the chance that conditions may vary again). The third possibility, the one we actually endorsed, was to take the unique advantage of having collected a pre-Scud sample and continue to follow our subjects in order to assess biological parameters at model points of change in stress. As expected, a significant decrease in anxiety and depression scores was observed after the war when compared with prewar and during war scores. The normal range of plasma cortisol levels, despite the exposure to the stressful situation, may be an artifact of design imposed by situational constrains: all missile attacks occurred at night, whereas hormonal mcasui'cmeni~ were perlormed in the mornh~g. It is likely that the peak release of cortisol was during and immediately after the missile attack event at night, and thus was undetectable the following morning (>6 hr after that night missile attack). Alter-
600 -g
500
[ ] Pr¢ war [] During war [ ] Post war
k..
40O 300 v
E re~
2O0 100 0
Figure 3. [3H]lmipraminebinding values (Bmax)in platelets of the subjects exposed to the Gulf War. Tile during-war binding value was higher compared with prewar (p = 0.063) and postwar value (p = 0.056).
Imipramine Binding ,during the Gulf War
BIOLPSYCHIATRY 1992;32:170--178
! 75
12 10
[] [] []
Pre w~r During war Post ~,~r
8 v
6 L_
o
¢.!
4
Figure 4. Basal morning plasma cortisol levels in the study population.
natively, it is possible that the repeated exposure to the stress led to habituation and adaptation of the hypothalamic-pituitary-adrenai axis. It is noteworthy that unaltered cortisol levels were previously reported by us in examination stress (Karl) et al 1989) and in soldiers exposed to repeated parachute jumps (Dar et al 1991). Owing to the spasmodic nature of secretion, it is not surprising that single estimations of plasma cortisol level were not different at the three time points. Unfortunately, due to the war situation, we were unable to measure 24 hr urinary-free cortisol excretion. A line of evidence suggests possible involvement of the serotonergic system in stress (Adell et al 1989). Previous studies have demonstrated that acute stress is associated with increased plasma serotonin (Evron et al 1982) and whole blood serotonin levels (Davis et al 1985), as well as increased utilization of brain serotonin (Adell et al 1988) and marked stimulation of serotonin turnover (Joseph and Kennett 1983). Chronic stress was demonstrated to induce increased synthesis of serotonin (Adell et ai 1988) in rat brain. Pa~'tial agonists at the serotonin-lA receptor have anxiolytic effects (Feighner et al 1982; Traber and Glaser 1987). Drugs that activate the serotonergic system are anxiogenic, whereas drugs that interfere with serotonergic neurotransmission possess anxiolytic properties (Ceulemans et al 1985; Chopin and Briley 1987; Soderpalm and Engel 1989; Stutzmann et at 1991). The involvement of the serotonin uptake site in chronic stress is yet unclear. Sherman and Petty (1984) showed diminished brain [3H]imipramine binding in learned helpless rats. However, two recent studies (File et al 1990; Edwards et al 1991) demonstrated increased serotonin uptake capacity in zat brain fc!lowing repeated stress. Repeated exposure to the situational stress during the war tended to increase the platelet imipramine maximal binding capacity. However, the elevated binding capacity during the war was not statistically different from the prewar and postwar values. The small number of participants can account for the lack of statistical significance. Though not statistically significant, the trend toward increase in platele~ "serotonin transporter" in humans in stressful conditions is intriguing. The associettion between stress-induced anxiety and increased density of the serotonin transporter is supported by the observation of a significant positive correlation between HARS scores and Bmaxvalues in the prewar and postwar time points. However, it is not clear why such a correlation did not exist during the war. It is possible that the increase in depression (as measured by BDI) during the war and the habituation to the repeated stress blunted the full expression of the upregulatory effect of the anxiety on the imipramine maximal binding capacity. Selective
176
BIOLPSYCHIATRY
R. Weizman et al
1992;32:170-178
inhibition of serotonin reuptake by fluoxetine is associated with insomnia, nervousness, and anxiety (Cooper 1988). Because chronic stress inducez enhanced serotonergic sensitivity (Mogilnicka 1981; Kennett et al 1985), it is possible that the increased serotonin uptake capacity is a compensatory mechanism to oppose the hyperfunction of the serotonergic system. One is tempted to speculate that hypofunction of serotonergic transmission in major depression leads to an adaptive down-regulation of the presynaptic serotonin transporter in order to diminish the reuptake and to maintain adequate neurotransmitter concentration in the synaptic cleft. In contrast, enhanced serotonergic transmission following repeated stress leads to upregulation of the transporter to amplify the reuptake process from the synaptic cleft to the presynaptic neuron. This hypothesis, however, calls for further testing of the role of the serotonergic system in the neurobiological response of stress. It seems that prolonged stress of different kinds (anticipatory~ before war and uncontrolled lethal threat--during war) causes habituation or adaptation of both the hypothalamic-pituitary-adrenal ax;.s and the serotonergic system.
References Abramson LY, Seligman MEP, Teasdale JD (1978): Learned helplessness in humans: Critique and reformulation. J. Abnorm Psychol 78:49-74. Adell A, Garcia-Marquez C, Armafio A, Gelpi E (1988): Chronic stress increases serotonin and noradrenaline in rat brain and sensitizes their responses to a further acute stress. J Neurochem 50:1678-1681. Adell A, Garcia-Marquez C, Armario A, Gelpi E (1989): Chronic administration of ¢lomipramine prevents the increase in serotonin and noradrenaline induced by chronic stress. Psychopharmacology 99:22-26. Beck AT, Wand CH, Mendelsohn M, Mock J, Ernbaugh J (1961): An inventory for measuring depression. Arch Gen Psychiatry 453-59. Brcier A, Albus M, Pickar D, Zahn TP, Wolkowitz OM, Paul SM (1987): Controllable and uncontrollable stress ill humans: Alterations in mood, neuroendocrine ano psychophysiologi¢ function. Am J Psychiat~. 144:14i9-1425. Briley MS, Langer SZ, Reisman R, Seehter, D, Zarifian E (19'.~0): Tritiated imipramine binding sites are decreased in platelets of untreated depressed patient:;. Science 2~'~9:303-305. Ceulemans DL, Hoppenbrouwers HL, Gelders ¥G, Reyntjens AJ (1985): The inl?ueneeof ritanserin, a serotonin antagonist, in anxiety disorders: A double-blind, placebo contmllod study versus Iorazepam, Pharmacopsyehiatry 18:303-305. Chopin P, Briley M (1987): Animal models of anxiety: The effect ot compouads that modify 5]-iT neurotransmission. Trends Pharmacol Sci 8:383-388. Cooper GL (1988): The safety of fluoxetine---~n update. Br J Psychiatry 153(Suppl 3):77-86. Dar DE, Weizman A, Karl) L, et al (1991): Platelet peripheral bcnzodiazepine receptor in repeated stress. Life Sci 48:341-346. Davis DD, Dunlop SR, Shea P, Brittain H, Hendrie HC (1985): Biological stress responses in high and low trait anxious students. Biol Psychiatr), 20:860-868. Drugan RC, Morrow LA, Weizman R, et al (1989): Stress-induced behavior depression in the rat is associated with a decrease in GABA receptor-mediated chloride ion flux and brain benzodiazepine receptor occupancy. Brain Res 487:45-51. Edwards E, Harkins K, Wright G, Henn F (1991): Modulation of [3H]paroxetine binding to the 5hydroxytryptamine uptake site in an animal model of depression. J Neurochem 56:1581-1586. Evron S, Pfeifer Y, Sadovsky E, Sulman G (1982): Plasma levels of serotonin compared with corfisol as stress indicators in gynecological operations, isr J Med Sci 18:806--808.
Imipramine Binding during the Gulf War
BIOLPSYCHIATRY 1992;32:170-178
177
Feighner JP, Meredith CH, Hendfickson GA (1982): A double-blind comparison of buspirone and diazepam in outpatients with generalized anxiety disorder. J Clin Psychiatry 43:103-107. File SE, Andrews N, Zharkovsky A (1990): Handling habituation and chlordiazepoxide have different effects on GABA and 5-HT function in the frontal cortex and hippocampus. Eur J Pharmacol 190:229-234. Hamilton M (1959): The assessment of anxiety states by rating. Br J Med Psychol 32:50-55. Joseph MH, Kennett GA (1983): Stress-induced release of 5-HT in the hippoeampus and its dependence on increased tryptophan availability: An in vivo electrochemical study. Brain Res 270:251-257. Karp L, Weizman A, Tyano S, Garish M (1989): Examination stress, platelet peripheral benzodiazepine binding sites and plasma hormone levels. Life Sci 44:1077-1082. Kennett GA, Dickinson SL, Curzon G (1985): Enhancement of some 5-HT-dependent behavioral responses following repeated immobilization in rats. Brain Res 330:253-264. Langer SZ, Moret C, Raisman R, Dubocovich ML, Briley M (1980): High affinity [3H]imipramine binding in rat hypothalamus: Association with uptake of serotonin but not of norepinephrine. Science 210:! 133-1135. Mogilnieka E (1981): REM sleep deprivation changes behavioral response to catecholaminergic and serotonergic receptor activation in rats. Ptmrmacol Biochem Behav 15:149- ! 5 !. Ormel J, Sanderman R 0989): Life events, porsonal control and depression. In Steptoe A, Appels AD (eds), Strauss, Personal Control and Health. Chichester, UK" John Wiley & Sons, pp 193213. Paul SM, Rehavi M, Rice KC, Ittah Y, Skolnick P (1981): Does high affinity [3H]imiFra~ine binding label serotonin reuptake sites in brain and platelet? Life Sci 28:2753-2760. Paul SM, Rehavi M, Skolnick P, Goodwin FK (1984): High affinity binding of antidepressants to biogenic amine transporter sites in human brain and platelet: Studies in depression. In Post RM, Ballenger JC (eds), Neurobioiogy of Mood Disorders. Baltimore: Williams & Wilkins, pp 846853. Raisman R, Briley M, Langer SZ (1979): Specific tricyclic antidepressant binding sites in rat brain. Nature 281:148-150. Raisman R, Briley MS, Bouchamy F, Sechter D, Zarifian E, Langer SZ (1982): 3H-Imipramine binding and serotonin uptake in platelets from untreated depressed patients and control volunteers. Psychopharmacology 77:332-335. Rehavi M, Paul SM, Skolnick P, Goodwin FK (1980): Demonstration of specific high affinity binding sites for [3H]imipramine in human brain. Life Sci 26:2273-2279. Rehavi M, Paul SM, Skolnick P (1984a): High-affinity binding sites for tricyclic antidepressants in brain and platelets. In Marangos PJ, Campbell IC, Cohen RM (eds), Brain Recepwr Methodologies, Part B. Orlando, FL: Academic Press, pp 279-295. Rehavi M, Weizman R, Carel C, Apter A, Tyano S (1984b): High-affinity 3H-imipramine binding in platelets of children and adolescents with major affective disorders. Psychiatry Res 13:3139. Seligman MEP, Weiss JM, Weinranb M, Schulman A (1980): Coping behavior, learned helplessness, physiological change and learned inactivity. Behav Res Ther 18:459-512. Sherman AD, Petty F (1984): ]Learnedhelplessness decreases [3H]-imipraminebinding in rat cortex. J Affective Disord 6:25-32. Sherman AD, Sacquitne JL, Petty F (1982): Specificity of the learned helplessness model of depression. Pharmacol Biochem Behav 16:449-454. Soderpalm B, Engel JA (1989): Does the pCPA-induced anticonflict effect involve activation of the GABAA/benzodiazepine chloride ionophore receptor complex? J Neural Transm 76:145153. Stutzmann JM, Eon B, Darche F, et al (1991): Are 5-HT2 antagonists endowed with anxiolytic properties in rodents? Neurosci Lett 128:4-8.
178
BIOLPSYCHIATRY 1992:32:170-178
R. Weizman et al
Traber J, Glaser T (1987): 5-HTIA receptor-related anxiolytics. Trends Pharmacoi Sci 8:432-437. Wei~s JM, Simson PG (I 985): Neurochemical basis of stress-induced depression. Psychopharmacol 1lull 21:447-457. Weitman R, Carmi M, Tyano S, Apter A, Rehavi M (1986): High affinity [3H]imipramine binding and serotonin uptake in platelets of adolescent females suffering from anorexia nervosa. Life Sci 38:1235-1242. Whitaker PM, Warsh JJ, Stancer HC, Persad E, Vint CK (1984): Seasonal variation in platelet 3H-imipramine binding: Comparable values in control and depressed populations. Psychiatry Res ! 1:127-135.
