- Email: [email protected]
Blunted prolactin response to fentanyl in depression. Normalizing effect of partial sleep deprivation
Psychiatry Research 118 (2003) 155–164
Blunted prolactin response to fentanyl in depression. Normalizing effect of partial sleep deprivation Ede Frecskaa,*, Andras Perenyib, Mihaly Aratoc a
Department of Psychiatry, University of Florida, Psychiatry Service 116A, VA Medical Center, Gainesville, FL 32608, USA b Clayton Community Mental Health Service, Clayton, Vic., Australia c Department of Psychiatry, University of Calgary, Calgary, AB, Canada Received 6 March 2002; received in revised form 17 September 2002; accepted 13 November 2002
Abstract There is some evidence that sleep deprivation (SD) might exert its antidepressant properties by involving endogenous opioid mechanisms. The authors investigated the effects of mu-receptor agonist administration on prolactin release in depressed patients before and after partial SD. Medication-free female depressed inpatients (Ns18) were participating in two fentanyl challenge tests after partial SD and undisturbed sleep, 3 days apart in random order. Healthy volunteer women (Ns10) were enrolled after full night sleep as comparison subjects. Five of them had placebo trials. Participants were given an intravenous injection of 0.1 mgy70 kg fentanyl at 9:00 AM. The prolactin secretory response to the opiate agonist was investigated for 1 h with serial blood sampling. After a night of undisturbed sleep, fentanyl administration prompted increases in plasma prolactin concentrations with blunted responses found in the depressed group. Following partial SD, the stimulated prolactin secretion of depressed patients increased significantly and was comparable to the response of comparison subjects. These findings suggest that SD acts via an opioidydopamine-related mechanism. An alternative explanation, based on serotonin involvement is addressed in the discussion. 䊚 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Antidepressants; Affective disorders; Dopamine; Opiates; Opioids; Monoamines; Serotonin
1. Introduction Sleep deprivation (SD) studies revealed fast and substantial improvement of mood in depressed ¨ patients. In their original study, Pflug and Tolle (1971) ascertained a remarkable efficacy response rate (69.5%) in depressed patients in the morning *Corresponding author. Tel.: q1-352-374-6014; fax: q1352-379-4170. E-mail address: [email protected] (E. Frecska).
hours after the first SD. Similar results were ¨ published by Rudolf and Tolle (1978) and reviewed by Wu and Bunney (1990). Partial SD is less stressful for the patient than total SD, and it has been shown to be as effective (Schilgen and ¨ Tolle, 1980). It is well documented that if SD is applied as a single treatment modality, its positive effect is usually transitory in nature: subsequent sleep tends to reverse the improvement (Kuhs and ¨ Tolle, 1991). The original interest in clinical research and
0165-1781/03/$ - see front matter 䊚 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0165-1781(03)00072-6
156
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
applications of SD has lessened, possibly due to the focus on pharmacotherapy and the dominance of pharmacological approaches in research on the etiology of psychiatric disorders (Wirz-Justice and Van den Hoofdakker, 1999). Lately, there is a resurgence of interest emphasizing the clinical usefulness of SD. First of all, SD is the only antidepressant therapy that works within 24 h. Second, the symptoms favorably influenced by SD include those that warrant the most intensive care due to their life-threatening potential: suicidal tendencies and severe psychomotor retardation. Third, it is a valuable augmentation strategy in treatmentrefractory depression (Gelenberg and Chesen, 2000). Apart from its clinical value, another aspect of SD is its scientific use. It provides a valuable research model for studying recovery processes in major depression. Patients can be investigated in depressed and non-depressed states within a short interval and without the need for pharmacological intervention (Gillin et al., 2001). Even a limited insight into the neurochemical changes induced by SD can provide important information about the recuperation process in successful treatments of depressive illness. Several abnormalities in neuroendocrine function have been reported in patients with major depression, including dysregulation of prolactin secretion, often in response to a specific pharmacologic stimulus (for review: Nicholas et al., 1998). Prolactin challenge tests have been considered to provide us with a window to the brain’s neurochemical processes. This approach, though indirect, may allow testing hypotheses regarding the biochemical basis of depressive illness. In our previous studies, using the mu-receptor agonist fentanyl as the stimulant of prolactin secretion, we found it a useful challenging agent for estimating the functional state of the central opioid system (Frecska and Arato, 2002). We also described a robust diurnal variation of opioid sensitivity (Frecska et al., 1988). It was suggested that this diurnal variation could be superimposed on the fundamental mood disturbance of major depression (Frecska et al., 1989). The role of endogenous opioids in the mood change following SD might be pertinent. In animal models of SD the administration of beta-endorphin markedly pro-
longed the insomnia with the connected psychomotor activation following SD. On the contrary, naloxone reduced the latency to sleep in a dosedependent manner (Gessa et al., 1995). To our knowledge, no human data are available on changes in endogenous opioid function after SD. Based on their animal studies, Fratta et al. (1987) proposed that SD induces a hyperactivation of the endogenous opioid system and supposed its involvement in the mood change. While their data did not exclude the role of other neurotransmitters, a strong myd opioid-D1 dopamine interaction was suggested. In the first part of the present study we intended to repeat our previous investigation (Frecska et al., 1989) when the prolactin secretory response to the micro-opioid agonist fentanyl was measured in patients with major depression. The objective of the second part of the study was to determine the effect of partial SD on fentanyl-induced prolactin responses in the same patients. 2. Methods Thirty-nine premenopausal female psychiatric inpatients with diagnoses of major depression were evaluated as candidates for the study. Twenty-one of them were not found eligible for the protocol due to high suicide risk, lack of a psychotropic medication-free period lasting for a minimum of 7 days prior to the study, improper timing of the menstrual cycle for the fentanyl challenge test, body mass index -20 or )25 kgym2, smoking, and having history of cardiac, endocrine illness andyor allergic reaction to opiates. Subjects with history of illicit drug use andyor alcohol consumption exceeding 50 gyweek were also excluded. Enrolled individuals were 18 consenting female inpatients who met DSM-IV criteria (American Psychiatric Association, 1994) for major depression (with the ‘severe’ modifier) and 10 healthy volunteer women as comparison subjects (mean age"S.D.: 37.8"11.3 and 35.2"10.2, respectively). Screening entailed physical examination with laboratory work-up (including urine toxicology, total blood cell count, basic metabolic panel and liver function test), and a semistructured psychiatric interview (World Health Organization, 1994)
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
for current and past history. Informed consents were obtained conforming to the guidelines set forth by the International Committee of Medical Journal Editors (1997). Fentanyl challenge tests were performed between days 4 and 11 of the participants’ menstrual cycle. Subjects were free of psychotropics for at least 7 days prior to testing. The maximum duration of the lead-in period was 10 days, which at times was necessary for optimization of the timing of the test with the menstrual cycle. Participants had a minimum of 6 h sleep with onset not later than 12 midnight and awakening not earlier than 4 AM on the night prior to the baseline fentanyl challenge test. In the morning hours of the test day, subjects rested supine and had an indwelling catheter inserted at 8:30 AM. Two basal blood samples were taken at 8:45 and 9:00 AM. Following the second blood sampling, 0.1 mgy70 kg fentanyl was administered in a 1min infusion. Blood samples were obtained for prolactin measurements at baseline (two), 15, 30, 45 and 60 min. During the challenge test, the IV access was maintained by using slow saline infusion. Depressive patients were tested with this procedure repeatedly on a separate day after partial, late-night SD. The lights-off time was 10 PM. Sleep-deprived patients were allowed to sleep in their bedrooms until 2 AM where they were monitored with 15-min checks. SD was started at 2 AM. Between 2 AM and the start of the test, SD patients were staying in an area next to the nursing station and were under constant observation. For the rest of the day and other nights, they were continued on 15-min checks. Patients were randomized regarding the order of the fentanyl challenge test with undisturbed sleep (SL trials) or with partial SD (SD trials). One half of them had the SD trial as the first procedure; the other half had the SL trial first. SD and SL trials were timed 72 h apart for the same individual. One of the patients withdrew her participation before the second challenge test. Since she had fentanyl challenge without partial SD, her results were included in the first part of this study. Placebo was also given in a single-blind fashion to five of the comparison subjects following an otherwise identical protocol. Serum prolactin was measured in duplicate by commercial radioimmunoassay kit.
157
Intra- and inter-assay coefficients of variation were less than 5 and 7%, respectively. Assay sensitivity was 1.0 ngyml. Data were examined for normality of distribution by using a normality plot and derived correlation coefficient. This analysis suggested that plasma prolactin concentrations were not normally distributed. Therefore, prolactin measurements were examined by non-parametric methods. Friedman’s repeated-measure ANOVA was used to confirm the prolactin-releasing effect of the test. Illness- and SD-related group differences in prolactin levels were analyzed by means of a twostage method using a summary measures procedure (Matthews et al., 1990). In the first stage, the area under the curve (AUC) was calculated with correction for the baseline as a summary measure of the net response for each individual from hormone concentrations at time points following the fentanyl challenge. In the second stage, these summary measures were analyzed by means of non-parametric tests. The Mann–Whitney U-test was applied to independent samples and the Wilcoxon matched pairs test was used on the dependent ones. 3. Result 3.1. Part 1, effect of illness (healthy vs. depressed subjects) In both subject groups, the serum concentration of prolactin increased for the first 30 min after fentanyl administration and decreased for the remainder of the test procedure. A discernible prolactin peak was observed at approximately 30 min (Fig. 1). Friedman’s repeated-measures ANOVA applied to the prolactin results of both the comparison and illness groups demonstrated a change in the mean concentration with time (xs 29.3, P-0.0001 and xs47.8, P-0.00001, respectively). There was no effect in the placebo group (xs5.7, NS). The mean ("S.E.) size of the summed response (AUC) was 187.8 ("56.5) in comparison subjects compared with 82.0 ("22.4) in the depressed group. Mann-Whitney test identified the significance of the difference at P-0.05 with zs2.11.
158
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
Fig. 1. Plasma prolactin responses (mean"S.E.) to 0.1 mgy70 kg IV fentanyl after undisturbed sleep. ØØØsØØØ, placebo controls (Ns5); –d–, comparison subjects (Ns10), --j--, depressive patients (Ns18).
3.2. Part 2, effect of SD (SL vs. SD trials in depressed patients) Prolactin baseline values did not differ between the two trials (Wilcoxon’s zs1.42, NS). Following both sleep and partial SD, there was a substantial prolactin response to fentanyl (Fig. 2). When Wilcoxon’s paired comparison test was applied to the summed responses (mean"SE of AUC after SL: 86.0"23.4; mean"SE of AUC after SD: 154.7"47.2), it revealed increased hormone secretion to fentanyl after partial SD (zs2.15, P0.05). That increase had a normalizing effect, since the prolactin response of depressed patients in their SD trials did not differ significantly from the results of comparison subjects in SL trials (MannWhitney zs0.84, NS). 4. Discussion In this study we demonstrated that fentanyl, administered as an intravenous bolus after a night of normal sleep, induced a robust and substantial
increase in prolactin plasma levels in comparison subjects whereas a blunted response was observed in depressed patients. A night of partial SD elicited increased prolactin secretion to fentanyl, bringing levels in patients close to those in controls after a full night of sleep. While the study design was lacking an arm in which comparison subjects had fentanyl challenge after SD, the normalizing effect of partial SD is suggested by the direction of change. There is no consensus in the literature about opiate-induced prolactin responses in depression. The blunted response found in our study is in agreement with reports of Extein et al. (1980) and Judd et al. (1982), but does not correspond to studies of Zis et al. (1985) and Banki and Arato (1987). Our earlier report (Frecska et al., 1989) represents a middle stance, since we found decreased opiate sensitivity in those patients who later committed suicide. Replicating the study with a larger sample size has certainly improved the power. Empirical human studies in regard to the effect of SD on opiate-induced prolactin secretion
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
159
Fig. 2. Plasma prolactin responses (mean"S.E.) to 0.1 mgy70 kg IV fentanyl in depressive patients. –d–, after undisturbed sleep; ---s--, after sleep deprivation.
are missing. The observed augmentation is a unique finding and supports the suggestion of Gehde and Emrich (1996), who postulate that SD prevents a nightly breakdown of the endogenous opioid tone. In accordance with this concept, we reported earlier that both healthy and depressed subjects have decreased opioid sensitivity in the AM as compared to the PM hours (Frecska et al., 1988, 1989). The mechanisms involved in the SD-induced restoration of the prolactin response to fentanyl in depressed patients warrant a detailed discussion. The way in which SD may influence fentanylstimulated prolactin release should be discussed in light of the neurotransmitter pathways regulating the lactotrophic system. 4.1. Regulation of prolactin secretion The hormone release of pituitary lactotrophic cells is modulated by the hypothalamic paraventricular nucleus (Minamitani et al., 1987). The regulatory inputs to the paraventricular neurons are
complex and only partially understood. Multiple neurotransmitters and neuroactive substances are implicated in the control of prolactin release including serotonin, opioid peptides, dopamine, histamine, cholecystokinin and estradiol (for review: Ben-Jonathan et al., 1989). The regulation of the lactotrophic system involves the tonic inhibition exerted by tuberoinfundibular dopaminergic (TIDA) neurons. These TIDA neurons represent a common final pathway in regulation of prolactin release. Serotonin exerts its prolactin-releasing properties by inhibiting TIDA neurons (Preziosi et al., 1983). In mammals, the influence of serotonin on prolactin appears to be the result of a synergistic action of 5-HT1A and 5-HT2Ay2C receptors (Seibyl et al., 1991; Meltzer and Maes, 1994; Franklin et al., 1999). TIDA nerve terminals are subjected to a similar inhibitory control by endogenous opioid peptides. Inhibition of TIDA neurons is responsible for the increased circulating levels of prolactin produced by micro-opioid agonists (for review: Moore and Lookingland, 1995). The inhibitory action of micro-opioids on TIDA neurons has
160
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
a dual mechanism (Koenig, 1989). One is a direct action due to their ability to hyperpolarize these neurons by increasing potassium conductance (Loose et al., 1990). The other involves the activation of serotonin pathways (Demarest and Moore, 1981). Based on the complexity of the prolactin regulation outlined above, a series of questions arises. What kind of opioid mechanism is responsible for the effect of SD on fentanyl-induced prolactin secretion? Are opioids involved at all? What if opioid sensitivity is unchanged and the shift described in our study reflects an enhanced serotonin effect downstream of the opioid receptors? Due to the limitations of our study (lack of data on treatment response, no comparison subjects after SD), we can address these questions only in the context of other SD studies and will answer them in the framework of the psychostimulant theory of SD. 4.2. Monoamines and SD Unlike what has been observed after chronic treatment with antidepressants, changes in adrenergic receptors after SD were modest and contradictory (Tsai et al., 1993). The evidence for involvement of serotonin mechanisms in the mood change after SD is conflicting. On one hand, in laboratory animals, total SD enhances the turnover of serotonin (Asikainen et al., 1997), increases the firing rate of serotonin neurons in the dorsal raphe nuclei (Gardner et al., 1997) and decreases the sensitivity of 5-HT1A autoreceptors (Prevot et al., 1996). In bipolar patients, the therapeutic effect of SD has been associated with homozygotic long variants of the 5-HT transporter-linked polymorphic region (Benedetti et al., 1999). On the other hand, brain 5-HT levels do not differ from control ´ et al., 1980) or 72 h levels after 24 h (Borbely (Wesemann et al., 1983) of SD, and extracellular 5-HT concentrations are not affected (Bjorvatn et al., 2002). SD does not act exactly as serotonergic antidepressants do, since its effect on the firing rate of septal neurons is different (Contreras et al., 1993), and tryptophan-depletion studies in humans (Neumeister et al., 1998) suggest that SD does not exert its antidepressant effects by involving
brain serotonin systems. A slightly different picture emerges about the role of serotonin in SD: it is the prolongation of the SD-induced therapeutic changes for which serotonin mechanisms are responsible and not the induction of clinical response per se (Neumeister et al., 1999). The findings of a recent study by Smeraldi et al. (1999) are consistent with this notion: pindolol, which augments serotonin release by blocking the 5HT1A autoreceptor, significantly improved the antidepressant effect of total SD, by preventing the short-term relapse after treatment. These results are markedly different from the interaction observed between SD and the dopaminergic amineptine, which has not been shown to lead to sustained mood amelioration but augmented the acute effects of total SD (Benedetti et al., 2001). 4.3. Prolactin and SD When citalopram-stimulated prolactin concentrations were studied in healthy subjects following sleep and SD (Seifritz et al., 1997), prolactin responses were blunted following SD indicating a down-regulation of 5-HT1A or 5-HT2 receptors. This finding goes against the assumption that our SD-related findings were serotonin-mediated. There is other literature data that fail to support serotonin involvement and point toward alternative hypotheses. For example, after fenfluramine challenge, blunted prolactin levels were found in patients who subsequently responded to SD (Kasper et al., 1988a). The same authors (Kasper et al., 1988b) indicated that SD decreases the nocturnal prolactin release in depressed patients and control subjects. Similarly, a blunted nocturnal increase of prolactin levels under SD was reported by Baumgartner et al. (1990). This physiological effect was attributed to a decrease in dopamine release during slow wave sleep (review: Ebert and Kaschka, 1995). Consistent with the dopamine theory, there is an increased prolactin response to sulpiride after SD, possibly attributable to prevention of the physiological decrease of the dopaminergic tone (Ebert et al., 1993). The present study is far from being conclusive about the exact mechanisms involved in the hormonal and therapeutic effect of SD. Combining
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
our findings with previously published results, we can only raise doubt about serotonergic mediation of the observed shift in the prolactin response. At this point, we have no more than indirect evidence supporting an opioid-dopamine mechanism in the behavioral effect of SD. Animal studies revealed an active role of opioids, in interaction with dopamine, in the behavioral syndrome subsequent to SD and hyperactivation of the endogenous opioid system was suggested (Fratta et al., 1987; Fadda et al., 1993). Single photon-emission computed tomography before and after total SD showed significantly different dopamine receptor occupancy in responders, vs. non-responders, suggesting an enhanced dopamine release in responders (Ebert et al., 1994). Dopaminergic augmentation of SD using amineptine was recently reported in humans (Benedetti et al., 2001). Depending on receptor type and brain location, opioids produce diverging response patterns on different dopaminergic neurons. They increase the activity of the major meso-telencephalic dopamine neurons terminating in the striatum and limbic forebrain regions, but inhibit TIDA neurons (for review: Moore and Lookingland, 1995). The behavioral and physiological effects of SD (e.g. euphoria, hyperactivity and hormonal changes) can be attributed to such a opioid-dopamine interaction and are consistent with the psychostimulant theory of antidepressant action (for review: Ebert and Berger, 1998). The cingulate gyrus is a likely site of the Sinduced behavioral changes. Brain-imaging studies reported that clinical improvement during SD was associated with reduced metabolic activity in the area of the anterior cingulate cortex (Volk et al., 1997; Wu et al., 1999). Positron emission tomography studies have revealed high levels of opioid receptor binding in the human anterior cingulate region (Jones et al., 1991), which also receives generous dopaminergic projections from the ventral tegmental area. Anterior cingulate blood flow is increased by anaesthetic doses of fentanyl (Adler et al., 1997; Casey et al., 2000), a fact that does not seem to favor the opioid-dopamine model since the direction of change is opposite to the findings within responders after SD. It may help with—but will not resolve—the controversy that typically
161
subanaesthetic doses cause behavioral effects similar to those induced by SD. At present time, one can only speculate about the neurotransmitter mechanisms involved in the effects of SD. One way to explain the apparent divergence in the literature is to suppose that the acute effects of SD are based on opioid-dopamine mediation and serotonergic mechanisms play a more significant role in prolongation of the effect. This hypothesis needs further investigation. A direct crossover comparison of serotonergic and opiate effects in the same subjects would be a powerful experiment in the debate. In conclusion, SD has potentials toward better understanding of the biological processes involved in antidepressant action and should be considered in future studies of antidepressant therapy. When this article was near completion, the theater siege in Moscow on October 26, 2002 gave gruesome actuality to fentanyl research in humans. The results presented here and previous findings on robust diurnal change in fentanyl effects (Frecska et al., 1988) indicate that individual responses to fentanyl may vary significantly according to the subjects’ sleep pattern, emotional state and the time of the day. Similar findings may reveal some other factors involved in the tragic outcome of the hostage crisis, and show how difficult the dose calculation can be for narcotic opiate use as a nonlethal weapon. Russian authorities stated the fentanyl-derivative they used would not have normally caused death. They argued that hostages had died because they were in a compromised medical condition, such as dehydration, immobilization, food deprivation, and lack of oxygen, and were under severe psychological stress due to the conditions under which they had been kept as hostages. Indeed, some of the conditions they were referring to are known from animal studies to potentiate opiate effects (Calcagnetti and Holtzman, 1992; Hodgson and Bond, 1991; Sutton et al., 1997) and might have contributed to the fatalities. What happened in Moscow on that sad morning underlines the importance of collecting more information on intra- and inter-individual variability of fentanyl effects.
162
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
Acknowledgments The authors thank Eva Wiesenmayer, Maurice G. Mnich and Jose R. Gutierrez for their help in the preparation of the manuscript. This study was supported by funds from the University of Calgary. The authors have no affiliation with or financial interest in any profit-oriented organization that might potentially bias this work. References Adler, L.J., Gyulai, F.E., Diehl, D.J., Mintun, M.A., Winter, P.M., Firestone, L.L., 1997. Regional brain activity changes associated with fentanyl analgesia elucidated by positron emission tomography. Anesthesia and Analgesia 84, 120–126. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders. fourth ed. American Psychiatric Press, Washington, DC. Asikainen, M., Toppila, J., Alanko, L., Ward, D.J., Stenberg, D., Porkka-Heiskanen, T., 1997. Sleep deprivation increases brain serotonin turnover in the rat. NeuroReport 8, 1577–1582. Banki, C.M., Arato, M., 1987. Multiple hormonal responses to morphine: relationship to diagnosis and dexamethasone suppression. Psychoneuroendocrinology 12, 3–11. Baumgartner, A., Riemann, D., Berger, M., 1990. Neuroendocrinological investigations during sleep deprivation in depression. II. Longitudinal measurement of thyrotropin, TH, cortisol, prolactin, GH, and LH during sleep and sleep deprivation. Biological Psychiatry 28, 569–587. Ben-Jonathan, N., Arbogast, L.A., Hyde, J.F., 1989. Neuroendocrine regulation of prolactin release. Progress in Neurobiology 33, 399–447. Benedetti, F., Serretti, A., Colombo, C., Campori, E., Barbini, B., di Bella, D., Smeraldi, E., 1999. Influence of a functional polymorphism within the promoter of the serotonin transporter gene on the effects of total sleep deprivation in bipolar depression. American Journal of Psychiatry 156, 1450–1452. Benedetti, F., Campori, E., Barbini, B., Fulgosi, M.C., Colombo, C., 2001. Dopaminergic augmentation of sleep deprivation effects in bipolar depression. Psychiatry Research 104, 239–246. Bjorvatn, B., Gronli, J., Hamre, F., Sorensen, E., Fiske, E., Bjorkum, A.A., Portas, C.M., Ursin, R., 2002. Effects of sleep deprivation on extracellular serotonin in hippocampus and frontal cortex of the rat. Neuroscience 113, 323–330. ´ A.A., Steigrad, P., Tobler, I., 1980. Effect of sleep Borbely, deprivation on brain serotonin in the rat. Behavioural Brain Research 1, 205–210. Calcagnetti, D.J., Holtzman, S.G., 1992. Potentiation of morphine analgesia in rats given a single exposure to restraint
stress immobilization. Pharmacology, Biochemistry and Behaviour 41, 449–453. Casey, K.L., Svensson, P., Morrow, T.J., Raz, J., Jone, C., Minoshima, S., 2000. Selective opiate modulation of nociceptive processing in the human brain. Journal of Neurophysiology 84, 525–533. Contreras, C.M., Marvan, M.L., Alcala-Herrera, V., 1993. Sleep deprivation is a less potent agent than clomipramine in increasing firing rate in lateral septal neurons in the rat. Neuropsychobiology 27, 83–85. Demarest, K.T., Moore, K.E., 1981. Disruption of 5-hydroxytryptaminergic neuronal function blocks the action of morphine on tuberoinfundibular dopaminergic neurons. Life Sciences 28, 1345–1351. Ebert, D., Berger, M., 1998. Neurobiological similarities in antidepressant sleep deprivation and psychostimulant use: a psychostimulant theory of antidepressant sleep deprivation. Psychopharmacology 140, 1–10. Ebert, D, Kaschka, W, 1995. Humoral aspects of sleep deprivation wGermanx. Fortschritte der Neurologie-Psychiatrie 63, 441–450. Ebert, D., Kaschka, W., Stegbauer, P., Schrell, U., 1993. Prolactin response to sulpiride before and after sleep deprivation in depression. Biological Psychiatry 33, 666–669. Ebert, D., Feistel, H., Kaschka, W., Barocka, A., Pirner, A., 1994. Single photon emission computerized tomography assessment of cerebral dopamine D2 receptor blockade in depression before and after sleep deprivation—preliminary results. Biological Psychiatry 35, 880–885. Extein, I., Pottash, A.L., Gold, M.S., Sweeney, D.R., Martin, D.M., Goodwin, F.K., 1980. Deficient prolactin response to morphine in depressed patients. American Journal of Psychiatry 137, 845–846. Fadda, P., Martellotta, M.C., Gessa, G.L., Fratta, W., 1993. Dopamine and opioid interactions in sleep deprivation. Progress in Neuro-Pychopharmacology and Biological Psychiatry 17, 269–278. Franklin, M., Craven, R.D., Dowling, B., Campling, G., Elliott, J.M., Cowen, P.J., 1999. Effect of a long-term low tryptophan diet on the prolactin responses to the 5-HT1A and 5HT2C agonists, 8-OH-DPAT and mCPP, in the male rat. Journal of Psychopharmacology 13, 58–63. Fratta, W., Collu, M., Martellotta, M.C., Pichiri, M., Muntoni, F., Gessa, G.L., 1987. Stress-induced insomnia: opioiddopamine interactions. European Journal of Pharmacology 142, 437–440. Frecska, E., Arato, M., Banki, C.M., Bagdy, G., Perenyi, A., Mohari, K., Fekete, M.I., 1988. Diurnal variation in fentanyl-induced hormone responses and side effects. Neuropsychopharmacology 1, 235–238. Frecska, E., Arato, M., Banki, C.M., Mohari, K., Perenyi, A., Bagdy, G., Fekete, M.I., 1989. Prolactin response to fentanyl in depression. Biological Psychiatry 25, 692–696. Frecska, E., Arato, M., 2002. Opiate sensitivity test in patients with stereotypic movement disorder and trichotillomania. Progress in Neuro-Pychopharmacology and Biological Psychiatry 26, 909–912.
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164 Gardner, J.P., Fornal, C.A., Jacobs, B.L., 1997. Effects of sleep deprivation on serotonergic neuronal activity in the dorsal raphe nucleus of the freely moving cat. Neuropsychopharmacology 17, 72–81. Gehde, E., Emrich, H.M., 1996. Systemteoretische konzepte ¨ zur wirkung des schlafentzuges. In: Kasper, S., Moller, H.J. (Eds.), Therapeutischer Schlafentzug. Springer, Wien, pp. 229–263. Gelenberg, A.J., Chesen, C.L., 2000. How fast are antidepressants? Journal of Clinical Psychiatry 61, 712–721. Gessa, G.L., Pani, L., Serra, G., Fratta, W., 1995. Animal models of mania. Advances in Biochemical Psychopharmacology 49, 43–66. Gillin, J.C., Buchsbaum, M., Wu, J., Clark, C., Bunney Jr., W., 2001. Sleep deprivation as a model experimental antidepressant treatment: findings from functional brain imaging. Depression and Anxiety 14, 37–49. Hodgson, D.M., Bond, N.W., 1991. Control of food delivery in food-deprived rats mediates analgesia. Behavioural Brain Research 44, 205–209. International Committee of Medical Journal Editors, 1997. Uniform requirements for manuscripts submitted to biomedical journals. The Journal of the American Medical Association 277, 927–934. Jones, A.K., Qi, L.Y., Fujirawa, T., Luthra, S.K., Ashburner, J., Bloomfield, P., Cunningham, V.J., Itoh, M., Fukuda, H., Jones, T., 1991. In vivo distribution of opioid receptors in man in relation to the cortical projections of the medial and lateral pain systems measured with positron emission tomography. Neuroscience Letters 126, 25–28. Judd, L.L., Risch, S.C., Parker, D.C., Janowsky, D.S., Segal, D.S., Huey, L.Y., 1982. Blunted prolactin response: a neuroendocrine abnormality manifested by depressed patients. Archives of General Psychiatry 39, 1413–1416. Kasper, S., Vieira, A., Wehr, T.A., Schmidt, R., Kick, H., Voll, G., Murphy, D.L., 1988. Serotonergically induced hormonal responses and the antidepressant effect of total sleep deprivation in patients with major depression. Psychopharmacology Bulletin 24, 450–453. Kasper, S., Sack, D.A., Wehr, T.A., Kick, H., Voll, G., Vieira, A., 1988. Nocturnal TSH and prolactin secretion during sleep deprivation and prediction of antidepressant response in patients with major depression. Biological Psychiatry 24, 631–641. Koenig, J.I., 1989. Pituitary gland: neuropeptides, neurotransmitters and growth factors. Toxicologic Pathology 17, 256–265. ¨ Kuhs, H., Tolle, R., 1991. Sleep deprivation therapy. Biological Psychiatry 29, 1129–1148 Review. Loose, M.D., Ronnekleiv, O.K., Kelly, M.J., 1990. Membrane properties and response to opioids of identified dopamine neurons in the guinea pig hypothalamus. Journal of Neuroscience 10, 3627–3634. Matthews, J.N., Altman, D.G., Campbell, M.J., Royston, P., 1990. Analysis of serial measurements in medical research. British Medical Journal 300, 230–235.
163
Meltzer, H.Y., Maes, M., 1994. Effect of pindolol on the L-5HTP-induced increase in plasma prolactin and cortisol concentrations in man. Psychopharmacology 114, 635–643. Minamitani, N., Minamitani, T., Lechan, R.M., BollingerGruber, J., Reichlin, S., 1987. Paraventricular nucleus mediates prolactin secretory responses to restraint stress, ether stress, and 5-hydroxy-L-tryptophan injection in the rat. Endocrinology 120, 860–867. Moore, K.E., Lookingland, K.J., 1995. Dopaminergic neuronal systems in the hypothalamus. In: Bloom, F.E., Kupfer, D.J. (Eds.), Psychopharmacology: The Fourth Generation of Progress. Raven Press, New York, pp. 245–256. Neumeister, A., Praschak-Rieder, N., Hesselmann, B., Vitouch, O., Rauh, M., Barocka, A., Tauscher, J., Kasper, S., 1998. Effects of tryptophan depletion in drug-free depressed patients who responded to total sleep deprivation. Archives of General Psychiatry 55, 167–172. Neumeister, A., Praschak-Rieder, N., Willeit, M., Stastny, J., Kasper, S., 1999. Monoamine depletion in non-pharmacological treatments for depression. Advances in Experimental Medicine and Biology 467, 29–33. Nicholas, L., Dawkins, K., Golden, R.N., 1998. Psychoneuroendocrinology of depression. Prolactin. Psychiatric Clinics of North America 21, 341–358. ¨ Pflug, B., Tolle, R., 1971. Therapy of endogenous depressions using sleep deprivation. Practical and theoretical consequences wGermanx. Nervenarzt 42, 117–124. Prevot, E., Maudhuit, C., Le Poul, E., Hamon, M., Adrien, J., 1996. Sleep deprivation reduces the citalopram-induced inhibition of serotoninergic neuronal firing in the nucleus raphe dorsalis of the rat. Journal of Sleep Research 5, 238–245. Preziosi, P., Vacca, M., Cerrito, F., 1983. Serotonergic and endogenous opiates interplay in the control of ACTH and prolactin release. Archives Internationales de Pharmacodyn´ amie et de Therapie 263, 325–327. ¨ Rudolf, G.A., Tolle, R., 1978. Sleep deprivation and circadian rhythm in depression. Psychiatria Clinica 11, 198–212. ¨ Schilgen, B., Tolle, R., 1980. Partial sleep deprivation as therapy for depression. Archives of General Psychiatry 37, 267–271. Seibyl, J.P., Krystal, J.H., Price, L.H., Woods, S.W., D’Amico, C., Heninger, G.R., Charney, D.S., 1991. Effects of ritanserin on the behavioral, neuroendocrine, and cardiovascular responses to meta-chlorophenylpiperazine in healthy human subjects. Psychiatry Research 38, 227–236. ¨ Seifritz, E., Muller, M.J., Annen, O., Nil, R., Hatzinger, M., Hemmeter, U., Moore, P., Holsboer-Trachsler, E., 1997. Effect of sleep deprivation on neuroendocrine response to a serotonergic probe in healthy male subjects. Journal of Psychiatric Research 31, 543–554. Smeraldi, E., Benedetti, F., Barbini, B., Campori, E., Colombo, C., 1999. Sustained antidepressant effect of sleep deprivation combined with pindolol in bipolar depression. A placebocontrolled trial. Neuropsychopharmacology 20, 380–385. Sutton, L.C., Lea, S.E., Will, M.J., Schwartz, B.A., Hartley, C.E., Poole, J.C., Watkins, L.R., Maier, S.F., 1997. Inescap-
164
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
able shock-induced potentiation of morphine analgesia. Behavioral Neuroscience 111, 1105–1113. Tsai, L.L., Bergmann, B.M., Perry, B.D., Rechtschaffen, A., 1993. Effects of chronic total sleep deprivation on central noradrenergic receptors in rat brain. Brain Research 602, 221–227. Volk, S.A., Kaendler, S.H., Hertel, A., Maul, F.D., Manoocheri, R., Weber, R., Georgi, K., Pflug, B., Hor, G., 1997. Can response to partial sleep deprivation in depressed patients be predicted by regional changes of cerebral blood flow? Psychiatry Research 75, 67–74. Wesemann, W., Weiner, N., Rotsch, M., Schulz, E., 1983. Serotonin binding in rat brain: circadian rhythm and effect of sleep deprivation. Journal of Neural Transmission Supplementum 18, 287–294. World Health Organization, 1994. Schedules for Clinical Assessment in Neuropsychiatry. American Psychiatric Press, Washington, DC.
Wirz-Justice, A., Van den Hoofdakker, R.H., 1999. Sleep deprivation in depression: what do we know, where do we go? Biological Psychiatry 46, 445–453. Wu, J.C., Bunney, W.E., 1990. The biological basis of an antidepressant response to sleep deprivation and relapse: review and hypothesis. American Journal of Psychiatry 147, 14–21. Wu, J., Buchsbaum, M.S., Gillin, J.C., Tang, C., Cadwell, S., Wiegand, M., Najafi, A., Klein, E., Hazen, K., Bunney Jr., W.E., Fallon, J.H., Keator, D., 1999. Prediction of antidepressant effects of sleep deprivation by metabolic rates in the ventral anterior cingulate and medial prefrontal cortex. American Journal of Psychiatry 156, 1149–1158. Zis, A.P., Haskett, R.F., Albala, A.A., Carroll, B.J., Lohr, N.E., 1985. Prolactin response to morphine in depression. Biological Psychiatry 20, 287–292.
Blunted prolactin response to fentanyl in depression. Normalizing effect of partial sleep deprivation Ede Frecskaa,*, Andras Perenyib, Mihaly Aratoc a
Department of Psychiatry, University of Florida, Psychiatry Service 116A, VA Medical Center, Gainesville, FL 32608, USA b Clayton Community Mental Health Service, Clayton, Vic., Australia c Department of Psychiatry, University of Calgary, Calgary, AB, Canada Received 6 March 2002; received in revised form 17 September 2002; accepted 13 November 2002
Abstract There is some evidence that sleep deprivation (SD) might exert its antidepressant properties by involving endogenous opioid mechanisms. The authors investigated the effects of mu-receptor agonist administration on prolactin release in depressed patients before and after partial SD. Medication-free female depressed inpatients (Ns18) were participating in two fentanyl challenge tests after partial SD and undisturbed sleep, 3 days apart in random order. Healthy volunteer women (Ns10) were enrolled after full night sleep as comparison subjects. Five of them had placebo trials. Participants were given an intravenous injection of 0.1 mgy70 kg fentanyl at 9:00 AM. The prolactin secretory response to the opiate agonist was investigated for 1 h with serial blood sampling. After a night of undisturbed sleep, fentanyl administration prompted increases in plasma prolactin concentrations with blunted responses found in the depressed group. Following partial SD, the stimulated prolactin secretion of depressed patients increased significantly and was comparable to the response of comparison subjects. These findings suggest that SD acts via an opioidydopamine-related mechanism. An alternative explanation, based on serotonin involvement is addressed in the discussion. 䊚 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Antidepressants; Affective disorders; Dopamine; Opiates; Opioids; Monoamines; Serotonin
1. Introduction Sleep deprivation (SD) studies revealed fast and substantial improvement of mood in depressed ¨ patients. In their original study, Pflug and Tolle (1971) ascertained a remarkable efficacy response rate (69.5%) in depressed patients in the morning *Corresponding author. Tel.: q1-352-374-6014; fax: q1352-379-4170. E-mail address: [email protected] (E. Frecska).
hours after the first SD. Similar results were ¨ published by Rudolf and Tolle (1978) and reviewed by Wu and Bunney (1990). Partial SD is less stressful for the patient than total SD, and it has been shown to be as effective (Schilgen and ¨ Tolle, 1980). It is well documented that if SD is applied as a single treatment modality, its positive effect is usually transitory in nature: subsequent sleep tends to reverse the improvement (Kuhs and ¨ Tolle, 1991). The original interest in clinical research and
0165-1781/03/$ - see front matter 䊚 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0165-1781(03)00072-6
156
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
applications of SD has lessened, possibly due to the focus on pharmacotherapy and the dominance of pharmacological approaches in research on the etiology of psychiatric disorders (Wirz-Justice and Van den Hoofdakker, 1999). Lately, there is a resurgence of interest emphasizing the clinical usefulness of SD. First of all, SD is the only antidepressant therapy that works within 24 h. Second, the symptoms favorably influenced by SD include those that warrant the most intensive care due to their life-threatening potential: suicidal tendencies and severe psychomotor retardation. Third, it is a valuable augmentation strategy in treatmentrefractory depression (Gelenberg and Chesen, 2000). Apart from its clinical value, another aspect of SD is its scientific use. It provides a valuable research model for studying recovery processes in major depression. Patients can be investigated in depressed and non-depressed states within a short interval and without the need for pharmacological intervention (Gillin et al., 2001). Even a limited insight into the neurochemical changes induced by SD can provide important information about the recuperation process in successful treatments of depressive illness. Several abnormalities in neuroendocrine function have been reported in patients with major depression, including dysregulation of prolactin secretion, often in response to a specific pharmacologic stimulus (for review: Nicholas et al., 1998). Prolactin challenge tests have been considered to provide us with a window to the brain’s neurochemical processes. This approach, though indirect, may allow testing hypotheses regarding the biochemical basis of depressive illness. In our previous studies, using the mu-receptor agonist fentanyl as the stimulant of prolactin secretion, we found it a useful challenging agent for estimating the functional state of the central opioid system (Frecska and Arato, 2002). We also described a robust diurnal variation of opioid sensitivity (Frecska et al., 1988). It was suggested that this diurnal variation could be superimposed on the fundamental mood disturbance of major depression (Frecska et al., 1989). The role of endogenous opioids in the mood change following SD might be pertinent. In animal models of SD the administration of beta-endorphin markedly pro-
longed the insomnia with the connected psychomotor activation following SD. On the contrary, naloxone reduced the latency to sleep in a dosedependent manner (Gessa et al., 1995). To our knowledge, no human data are available on changes in endogenous opioid function after SD. Based on their animal studies, Fratta et al. (1987) proposed that SD induces a hyperactivation of the endogenous opioid system and supposed its involvement in the mood change. While their data did not exclude the role of other neurotransmitters, a strong myd opioid-D1 dopamine interaction was suggested. In the first part of the present study we intended to repeat our previous investigation (Frecska et al., 1989) when the prolactin secretory response to the micro-opioid agonist fentanyl was measured in patients with major depression. The objective of the second part of the study was to determine the effect of partial SD on fentanyl-induced prolactin responses in the same patients. 2. Methods Thirty-nine premenopausal female psychiatric inpatients with diagnoses of major depression were evaluated as candidates for the study. Twenty-one of them were not found eligible for the protocol due to high suicide risk, lack of a psychotropic medication-free period lasting for a minimum of 7 days prior to the study, improper timing of the menstrual cycle for the fentanyl challenge test, body mass index -20 or )25 kgym2, smoking, and having history of cardiac, endocrine illness andyor allergic reaction to opiates. Subjects with history of illicit drug use andyor alcohol consumption exceeding 50 gyweek were also excluded. Enrolled individuals were 18 consenting female inpatients who met DSM-IV criteria (American Psychiatric Association, 1994) for major depression (with the ‘severe’ modifier) and 10 healthy volunteer women as comparison subjects (mean age"S.D.: 37.8"11.3 and 35.2"10.2, respectively). Screening entailed physical examination with laboratory work-up (including urine toxicology, total blood cell count, basic metabolic panel and liver function test), and a semistructured psychiatric interview (World Health Organization, 1994)
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
for current and past history. Informed consents were obtained conforming to the guidelines set forth by the International Committee of Medical Journal Editors (1997). Fentanyl challenge tests were performed between days 4 and 11 of the participants’ menstrual cycle. Subjects were free of psychotropics for at least 7 days prior to testing. The maximum duration of the lead-in period was 10 days, which at times was necessary for optimization of the timing of the test with the menstrual cycle. Participants had a minimum of 6 h sleep with onset not later than 12 midnight and awakening not earlier than 4 AM on the night prior to the baseline fentanyl challenge test. In the morning hours of the test day, subjects rested supine and had an indwelling catheter inserted at 8:30 AM. Two basal blood samples were taken at 8:45 and 9:00 AM. Following the second blood sampling, 0.1 mgy70 kg fentanyl was administered in a 1min infusion. Blood samples were obtained for prolactin measurements at baseline (two), 15, 30, 45 and 60 min. During the challenge test, the IV access was maintained by using slow saline infusion. Depressive patients were tested with this procedure repeatedly on a separate day after partial, late-night SD. The lights-off time was 10 PM. Sleep-deprived patients were allowed to sleep in their bedrooms until 2 AM where they were monitored with 15-min checks. SD was started at 2 AM. Between 2 AM and the start of the test, SD patients were staying in an area next to the nursing station and were under constant observation. For the rest of the day and other nights, they were continued on 15-min checks. Patients were randomized regarding the order of the fentanyl challenge test with undisturbed sleep (SL trials) or with partial SD (SD trials). One half of them had the SD trial as the first procedure; the other half had the SL trial first. SD and SL trials were timed 72 h apart for the same individual. One of the patients withdrew her participation before the second challenge test. Since she had fentanyl challenge without partial SD, her results were included in the first part of this study. Placebo was also given in a single-blind fashion to five of the comparison subjects following an otherwise identical protocol. Serum prolactin was measured in duplicate by commercial radioimmunoassay kit.
157
Intra- and inter-assay coefficients of variation were less than 5 and 7%, respectively. Assay sensitivity was 1.0 ngyml. Data were examined for normality of distribution by using a normality plot and derived correlation coefficient. This analysis suggested that plasma prolactin concentrations were not normally distributed. Therefore, prolactin measurements were examined by non-parametric methods. Friedman’s repeated-measure ANOVA was used to confirm the prolactin-releasing effect of the test. Illness- and SD-related group differences in prolactin levels were analyzed by means of a twostage method using a summary measures procedure (Matthews et al., 1990). In the first stage, the area under the curve (AUC) was calculated with correction for the baseline as a summary measure of the net response for each individual from hormone concentrations at time points following the fentanyl challenge. In the second stage, these summary measures were analyzed by means of non-parametric tests. The Mann–Whitney U-test was applied to independent samples and the Wilcoxon matched pairs test was used on the dependent ones. 3. Result 3.1. Part 1, effect of illness (healthy vs. depressed subjects) In both subject groups, the serum concentration of prolactin increased for the first 30 min after fentanyl administration and decreased for the remainder of the test procedure. A discernible prolactin peak was observed at approximately 30 min (Fig. 1). Friedman’s repeated-measures ANOVA applied to the prolactin results of both the comparison and illness groups demonstrated a change in the mean concentration with time (xs 29.3, P-0.0001 and xs47.8, P-0.00001, respectively). There was no effect in the placebo group (xs5.7, NS). The mean ("S.E.) size of the summed response (AUC) was 187.8 ("56.5) in comparison subjects compared with 82.0 ("22.4) in the depressed group. Mann-Whitney test identified the significance of the difference at P-0.05 with zs2.11.
158
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
Fig. 1. Plasma prolactin responses (mean"S.E.) to 0.1 mgy70 kg IV fentanyl after undisturbed sleep. ØØØsØØØ, placebo controls (Ns5); –d–, comparison subjects (Ns10), --j--, depressive patients (Ns18).
3.2. Part 2, effect of SD (SL vs. SD trials in depressed patients) Prolactin baseline values did not differ between the two trials (Wilcoxon’s zs1.42, NS). Following both sleep and partial SD, there was a substantial prolactin response to fentanyl (Fig. 2). When Wilcoxon’s paired comparison test was applied to the summed responses (mean"SE of AUC after SL: 86.0"23.4; mean"SE of AUC after SD: 154.7"47.2), it revealed increased hormone secretion to fentanyl after partial SD (zs2.15, P0.05). That increase had a normalizing effect, since the prolactin response of depressed patients in their SD trials did not differ significantly from the results of comparison subjects in SL trials (MannWhitney zs0.84, NS). 4. Discussion In this study we demonstrated that fentanyl, administered as an intravenous bolus after a night of normal sleep, induced a robust and substantial
increase in prolactin plasma levels in comparison subjects whereas a blunted response was observed in depressed patients. A night of partial SD elicited increased prolactin secretion to fentanyl, bringing levels in patients close to those in controls after a full night of sleep. While the study design was lacking an arm in which comparison subjects had fentanyl challenge after SD, the normalizing effect of partial SD is suggested by the direction of change. There is no consensus in the literature about opiate-induced prolactin responses in depression. The blunted response found in our study is in agreement with reports of Extein et al. (1980) and Judd et al. (1982), but does not correspond to studies of Zis et al. (1985) and Banki and Arato (1987). Our earlier report (Frecska et al., 1989) represents a middle stance, since we found decreased opiate sensitivity in those patients who later committed suicide. Replicating the study with a larger sample size has certainly improved the power. Empirical human studies in regard to the effect of SD on opiate-induced prolactin secretion
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
159
Fig. 2. Plasma prolactin responses (mean"S.E.) to 0.1 mgy70 kg IV fentanyl in depressive patients. –d–, after undisturbed sleep; ---s--, after sleep deprivation.
are missing. The observed augmentation is a unique finding and supports the suggestion of Gehde and Emrich (1996), who postulate that SD prevents a nightly breakdown of the endogenous opioid tone. In accordance with this concept, we reported earlier that both healthy and depressed subjects have decreased opioid sensitivity in the AM as compared to the PM hours (Frecska et al., 1988, 1989). The mechanisms involved in the SD-induced restoration of the prolactin response to fentanyl in depressed patients warrant a detailed discussion. The way in which SD may influence fentanylstimulated prolactin release should be discussed in light of the neurotransmitter pathways regulating the lactotrophic system. 4.1. Regulation of prolactin secretion The hormone release of pituitary lactotrophic cells is modulated by the hypothalamic paraventricular nucleus (Minamitani et al., 1987). The regulatory inputs to the paraventricular neurons are
complex and only partially understood. Multiple neurotransmitters and neuroactive substances are implicated in the control of prolactin release including serotonin, opioid peptides, dopamine, histamine, cholecystokinin and estradiol (for review: Ben-Jonathan et al., 1989). The regulation of the lactotrophic system involves the tonic inhibition exerted by tuberoinfundibular dopaminergic (TIDA) neurons. These TIDA neurons represent a common final pathway in regulation of prolactin release. Serotonin exerts its prolactin-releasing properties by inhibiting TIDA neurons (Preziosi et al., 1983). In mammals, the influence of serotonin on prolactin appears to be the result of a synergistic action of 5-HT1A and 5-HT2Ay2C receptors (Seibyl et al., 1991; Meltzer and Maes, 1994; Franklin et al., 1999). TIDA nerve terminals are subjected to a similar inhibitory control by endogenous opioid peptides. Inhibition of TIDA neurons is responsible for the increased circulating levels of prolactin produced by micro-opioid agonists (for review: Moore and Lookingland, 1995). The inhibitory action of micro-opioids on TIDA neurons has
160
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
a dual mechanism (Koenig, 1989). One is a direct action due to their ability to hyperpolarize these neurons by increasing potassium conductance (Loose et al., 1990). The other involves the activation of serotonin pathways (Demarest and Moore, 1981). Based on the complexity of the prolactin regulation outlined above, a series of questions arises. What kind of opioid mechanism is responsible for the effect of SD on fentanyl-induced prolactin secretion? Are opioids involved at all? What if opioid sensitivity is unchanged and the shift described in our study reflects an enhanced serotonin effect downstream of the opioid receptors? Due to the limitations of our study (lack of data on treatment response, no comparison subjects after SD), we can address these questions only in the context of other SD studies and will answer them in the framework of the psychostimulant theory of SD. 4.2. Monoamines and SD Unlike what has been observed after chronic treatment with antidepressants, changes in adrenergic receptors after SD were modest and contradictory (Tsai et al., 1993). The evidence for involvement of serotonin mechanisms in the mood change after SD is conflicting. On one hand, in laboratory animals, total SD enhances the turnover of serotonin (Asikainen et al., 1997), increases the firing rate of serotonin neurons in the dorsal raphe nuclei (Gardner et al., 1997) and decreases the sensitivity of 5-HT1A autoreceptors (Prevot et al., 1996). In bipolar patients, the therapeutic effect of SD has been associated with homozygotic long variants of the 5-HT transporter-linked polymorphic region (Benedetti et al., 1999). On the other hand, brain 5-HT levels do not differ from control ´ et al., 1980) or 72 h levels after 24 h (Borbely (Wesemann et al., 1983) of SD, and extracellular 5-HT concentrations are not affected (Bjorvatn et al., 2002). SD does not act exactly as serotonergic antidepressants do, since its effect on the firing rate of septal neurons is different (Contreras et al., 1993), and tryptophan-depletion studies in humans (Neumeister et al., 1998) suggest that SD does not exert its antidepressant effects by involving
brain serotonin systems. A slightly different picture emerges about the role of serotonin in SD: it is the prolongation of the SD-induced therapeutic changes for which serotonin mechanisms are responsible and not the induction of clinical response per se (Neumeister et al., 1999). The findings of a recent study by Smeraldi et al. (1999) are consistent with this notion: pindolol, which augments serotonin release by blocking the 5HT1A autoreceptor, significantly improved the antidepressant effect of total SD, by preventing the short-term relapse after treatment. These results are markedly different from the interaction observed between SD and the dopaminergic amineptine, which has not been shown to lead to sustained mood amelioration but augmented the acute effects of total SD (Benedetti et al., 2001). 4.3. Prolactin and SD When citalopram-stimulated prolactin concentrations were studied in healthy subjects following sleep and SD (Seifritz et al., 1997), prolactin responses were blunted following SD indicating a down-regulation of 5-HT1A or 5-HT2 receptors. This finding goes against the assumption that our SD-related findings were serotonin-mediated. There is other literature data that fail to support serotonin involvement and point toward alternative hypotheses. For example, after fenfluramine challenge, blunted prolactin levels were found in patients who subsequently responded to SD (Kasper et al., 1988a). The same authors (Kasper et al., 1988b) indicated that SD decreases the nocturnal prolactin release in depressed patients and control subjects. Similarly, a blunted nocturnal increase of prolactin levels under SD was reported by Baumgartner et al. (1990). This physiological effect was attributed to a decrease in dopamine release during slow wave sleep (review: Ebert and Kaschka, 1995). Consistent with the dopamine theory, there is an increased prolactin response to sulpiride after SD, possibly attributable to prevention of the physiological decrease of the dopaminergic tone (Ebert et al., 1993). The present study is far from being conclusive about the exact mechanisms involved in the hormonal and therapeutic effect of SD. Combining
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
our findings with previously published results, we can only raise doubt about serotonergic mediation of the observed shift in the prolactin response. At this point, we have no more than indirect evidence supporting an opioid-dopamine mechanism in the behavioral effect of SD. Animal studies revealed an active role of opioids, in interaction with dopamine, in the behavioral syndrome subsequent to SD and hyperactivation of the endogenous opioid system was suggested (Fratta et al., 1987; Fadda et al., 1993). Single photon-emission computed tomography before and after total SD showed significantly different dopamine receptor occupancy in responders, vs. non-responders, suggesting an enhanced dopamine release in responders (Ebert et al., 1994). Dopaminergic augmentation of SD using amineptine was recently reported in humans (Benedetti et al., 2001). Depending on receptor type and brain location, opioids produce diverging response patterns on different dopaminergic neurons. They increase the activity of the major meso-telencephalic dopamine neurons terminating in the striatum and limbic forebrain regions, but inhibit TIDA neurons (for review: Moore and Lookingland, 1995). The behavioral and physiological effects of SD (e.g. euphoria, hyperactivity and hormonal changes) can be attributed to such a opioid-dopamine interaction and are consistent with the psychostimulant theory of antidepressant action (for review: Ebert and Berger, 1998). The cingulate gyrus is a likely site of the Sinduced behavioral changes. Brain-imaging studies reported that clinical improvement during SD was associated with reduced metabolic activity in the area of the anterior cingulate cortex (Volk et al., 1997; Wu et al., 1999). Positron emission tomography studies have revealed high levels of opioid receptor binding in the human anterior cingulate region (Jones et al., 1991), which also receives generous dopaminergic projections from the ventral tegmental area. Anterior cingulate blood flow is increased by anaesthetic doses of fentanyl (Adler et al., 1997; Casey et al., 2000), a fact that does not seem to favor the opioid-dopamine model since the direction of change is opposite to the findings within responders after SD. It may help with—but will not resolve—the controversy that typically
161
subanaesthetic doses cause behavioral effects similar to those induced by SD. At present time, one can only speculate about the neurotransmitter mechanisms involved in the effects of SD. One way to explain the apparent divergence in the literature is to suppose that the acute effects of SD are based on opioid-dopamine mediation and serotonergic mechanisms play a more significant role in prolongation of the effect. This hypothesis needs further investigation. A direct crossover comparison of serotonergic and opiate effects in the same subjects would be a powerful experiment in the debate. In conclusion, SD has potentials toward better understanding of the biological processes involved in antidepressant action and should be considered in future studies of antidepressant therapy. When this article was near completion, the theater siege in Moscow on October 26, 2002 gave gruesome actuality to fentanyl research in humans. The results presented here and previous findings on robust diurnal change in fentanyl effects (Frecska et al., 1988) indicate that individual responses to fentanyl may vary significantly according to the subjects’ sleep pattern, emotional state and the time of the day. Similar findings may reveal some other factors involved in the tragic outcome of the hostage crisis, and show how difficult the dose calculation can be for narcotic opiate use as a nonlethal weapon. Russian authorities stated the fentanyl-derivative they used would not have normally caused death. They argued that hostages had died because they were in a compromised medical condition, such as dehydration, immobilization, food deprivation, and lack of oxygen, and were under severe psychological stress due to the conditions under which they had been kept as hostages. Indeed, some of the conditions they were referring to are known from animal studies to potentiate opiate effects (Calcagnetti and Holtzman, 1992; Hodgson and Bond, 1991; Sutton et al., 1997) and might have contributed to the fatalities. What happened in Moscow on that sad morning underlines the importance of collecting more information on intra- and inter-individual variability of fentanyl effects.
162
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
Acknowledgments The authors thank Eva Wiesenmayer, Maurice G. Mnich and Jose R. Gutierrez for their help in the preparation of the manuscript. This study was supported by funds from the University of Calgary. The authors have no affiliation with or financial interest in any profit-oriented organization that might potentially bias this work. References Adler, L.J., Gyulai, F.E., Diehl, D.J., Mintun, M.A., Winter, P.M., Firestone, L.L., 1997. Regional brain activity changes associated with fentanyl analgesia elucidated by positron emission tomography. Anesthesia and Analgesia 84, 120–126. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders. fourth ed. American Psychiatric Press, Washington, DC. Asikainen, M., Toppila, J., Alanko, L., Ward, D.J., Stenberg, D., Porkka-Heiskanen, T., 1997. Sleep deprivation increases brain serotonin turnover in the rat. NeuroReport 8, 1577–1582. Banki, C.M., Arato, M., 1987. Multiple hormonal responses to morphine: relationship to diagnosis and dexamethasone suppression. Psychoneuroendocrinology 12, 3–11. Baumgartner, A., Riemann, D., Berger, M., 1990. Neuroendocrinological investigations during sleep deprivation in depression. II. Longitudinal measurement of thyrotropin, TH, cortisol, prolactin, GH, and LH during sleep and sleep deprivation. Biological Psychiatry 28, 569–587. Ben-Jonathan, N., Arbogast, L.A., Hyde, J.F., 1989. Neuroendocrine regulation of prolactin release. Progress in Neurobiology 33, 399–447. Benedetti, F., Serretti, A., Colombo, C., Campori, E., Barbini, B., di Bella, D., Smeraldi, E., 1999. Influence of a functional polymorphism within the promoter of the serotonin transporter gene on the effects of total sleep deprivation in bipolar depression. American Journal of Psychiatry 156, 1450–1452. Benedetti, F., Campori, E., Barbini, B., Fulgosi, M.C., Colombo, C., 2001. Dopaminergic augmentation of sleep deprivation effects in bipolar depression. Psychiatry Research 104, 239–246. Bjorvatn, B., Gronli, J., Hamre, F., Sorensen, E., Fiske, E., Bjorkum, A.A., Portas, C.M., Ursin, R., 2002. Effects of sleep deprivation on extracellular serotonin in hippocampus and frontal cortex of the rat. Neuroscience 113, 323–330. ´ A.A., Steigrad, P., Tobler, I., 1980. Effect of sleep Borbely, deprivation on brain serotonin in the rat. Behavioural Brain Research 1, 205–210. Calcagnetti, D.J., Holtzman, S.G., 1992. Potentiation of morphine analgesia in rats given a single exposure to restraint
stress immobilization. Pharmacology, Biochemistry and Behaviour 41, 449–453. Casey, K.L., Svensson, P., Morrow, T.J., Raz, J., Jone, C., Minoshima, S., 2000. Selective opiate modulation of nociceptive processing in the human brain. Journal of Neurophysiology 84, 525–533. Contreras, C.M., Marvan, M.L., Alcala-Herrera, V., 1993. Sleep deprivation is a less potent agent than clomipramine in increasing firing rate in lateral septal neurons in the rat. Neuropsychobiology 27, 83–85. Demarest, K.T., Moore, K.E., 1981. Disruption of 5-hydroxytryptaminergic neuronal function blocks the action of morphine on tuberoinfundibular dopaminergic neurons. Life Sciences 28, 1345–1351. Ebert, D., Berger, M., 1998. Neurobiological similarities in antidepressant sleep deprivation and psychostimulant use: a psychostimulant theory of antidepressant sleep deprivation. Psychopharmacology 140, 1–10. Ebert, D, Kaschka, W, 1995. Humoral aspects of sleep deprivation wGermanx. Fortschritte der Neurologie-Psychiatrie 63, 441–450. Ebert, D., Kaschka, W., Stegbauer, P., Schrell, U., 1993. Prolactin response to sulpiride before and after sleep deprivation in depression. Biological Psychiatry 33, 666–669. Ebert, D., Feistel, H., Kaschka, W., Barocka, A., Pirner, A., 1994. Single photon emission computerized tomography assessment of cerebral dopamine D2 receptor blockade in depression before and after sleep deprivation—preliminary results. Biological Psychiatry 35, 880–885. Extein, I., Pottash, A.L., Gold, M.S., Sweeney, D.R., Martin, D.M., Goodwin, F.K., 1980. Deficient prolactin response to morphine in depressed patients. American Journal of Psychiatry 137, 845–846. Fadda, P., Martellotta, M.C., Gessa, G.L., Fratta, W., 1993. Dopamine and opioid interactions in sleep deprivation. Progress in Neuro-Pychopharmacology and Biological Psychiatry 17, 269–278. Franklin, M., Craven, R.D., Dowling, B., Campling, G., Elliott, J.M., Cowen, P.J., 1999. Effect of a long-term low tryptophan diet on the prolactin responses to the 5-HT1A and 5HT2C agonists, 8-OH-DPAT and mCPP, in the male rat. Journal of Psychopharmacology 13, 58–63. Fratta, W., Collu, M., Martellotta, M.C., Pichiri, M., Muntoni, F., Gessa, G.L., 1987. Stress-induced insomnia: opioiddopamine interactions. European Journal of Pharmacology 142, 437–440. Frecska, E., Arato, M., Banki, C.M., Bagdy, G., Perenyi, A., Mohari, K., Fekete, M.I., 1988. Diurnal variation in fentanyl-induced hormone responses and side effects. Neuropsychopharmacology 1, 235–238. Frecska, E., Arato, M., Banki, C.M., Mohari, K., Perenyi, A., Bagdy, G., Fekete, M.I., 1989. Prolactin response to fentanyl in depression. Biological Psychiatry 25, 692–696. Frecska, E., Arato, M., 2002. Opiate sensitivity test in patients with stereotypic movement disorder and trichotillomania. Progress in Neuro-Pychopharmacology and Biological Psychiatry 26, 909–912.
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164 Gardner, J.P., Fornal, C.A., Jacobs, B.L., 1997. Effects of sleep deprivation on serotonergic neuronal activity in the dorsal raphe nucleus of the freely moving cat. Neuropsychopharmacology 17, 72–81. Gehde, E., Emrich, H.M., 1996. Systemteoretische konzepte ¨ zur wirkung des schlafentzuges. In: Kasper, S., Moller, H.J. (Eds.), Therapeutischer Schlafentzug. Springer, Wien, pp. 229–263. Gelenberg, A.J., Chesen, C.L., 2000. How fast are antidepressants? Journal of Clinical Psychiatry 61, 712–721. Gessa, G.L., Pani, L., Serra, G., Fratta, W., 1995. Animal models of mania. Advances in Biochemical Psychopharmacology 49, 43–66. Gillin, J.C., Buchsbaum, M., Wu, J., Clark, C., Bunney Jr., W., 2001. Sleep deprivation as a model experimental antidepressant treatment: findings from functional brain imaging. Depression and Anxiety 14, 37–49. Hodgson, D.M., Bond, N.W., 1991. Control of food delivery in food-deprived rats mediates analgesia. Behavioural Brain Research 44, 205–209. International Committee of Medical Journal Editors, 1997. Uniform requirements for manuscripts submitted to biomedical journals. The Journal of the American Medical Association 277, 927–934. Jones, A.K., Qi, L.Y., Fujirawa, T., Luthra, S.K., Ashburner, J., Bloomfield, P., Cunningham, V.J., Itoh, M., Fukuda, H., Jones, T., 1991. In vivo distribution of opioid receptors in man in relation to the cortical projections of the medial and lateral pain systems measured with positron emission tomography. Neuroscience Letters 126, 25–28. Judd, L.L., Risch, S.C., Parker, D.C., Janowsky, D.S., Segal, D.S., Huey, L.Y., 1982. Blunted prolactin response: a neuroendocrine abnormality manifested by depressed patients. Archives of General Psychiatry 39, 1413–1416. Kasper, S., Vieira, A., Wehr, T.A., Schmidt, R., Kick, H., Voll, G., Murphy, D.L., 1988. Serotonergically induced hormonal responses and the antidepressant effect of total sleep deprivation in patients with major depression. Psychopharmacology Bulletin 24, 450–453. Kasper, S., Sack, D.A., Wehr, T.A., Kick, H., Voll, G., Vieira, A., 1988. Nocturnal TSH and prolactin secretion during sleep deprivation and prediction of antidepressant response in patients with major depression. Biological Psychiatry 24, 631–641. Koenig, J.I., 1989. Pituitary gland: neuropeptides, neurotransmitters and growth factors. Toxicologic Pathology 17, 256–265. ¨ Kuhs, H., Tolle, R., 1991. Sleep deprivation therapy. Biological Psychiatry 29, 1129–1148 Review. Loose, M.D., Ronnekleiv, O.K., Kelly, M.J., 1990. Membrane properties and response to opioids of identified dopamine neurons in the guinea pig hypothalamus. Journal of Neuroscience 10, 3627–3634. Matthews, J.N., Altman, D.G., Campbell, M.J., Royston, P., 1990. Analysis of serial measurements in medical research. British Medical Journal 300, 230–235.
163
Meltzer, H.Y., Maes, M., 1994. Effect of pindolol on the L-5HTP-induced increase in plasma prolactin and cortisol concentrations in man. Psychopharmacology 114, 635–643. Minamitani, N., Minamitani, T., Lechan, R.M., BollingerGruber, J., Reichlin, S., 1987. Paraventricular nucleus mediates prolactin secretory responses to restraint stress, ether stress, and 5-hydroxy-L-tryptophan injection in the rat. Endocrinology 120, 860–867. Moore, K.E., Lookingland, K.J., 1995. Dopaminergic neuronal systems in the hypothalamus. In: Bloom, F.E., Kupfer, D.J. (Eds.), Psychopharmacology: The Fourth Generation of Progress. Raven Press, New York, pp. 245–256. Neumeister, A., Praschak-Rieder, N., Hesselmann, B., Vitouch, O., Rauh, M., Barocka, A., Tauscher, J., Kasper, S., 1998. Effects of tryptophan depletion in drug-free depressed patients who responded to total sleep deprivation. Archives of General Psychiatry 55, 167–172. Neumeister, A., Praschak-Rieder, N., Willeit, M., Stastny, J., Kasper, S., 1999. Monoamine depletion in non-pharmacological treatments for depression. Advances in Experimental Medicine and Biology 467, 29–33. Nicholas, L., Dawkins, K., Golden, R.N., 1998. Psychoneuroendocrinology of depression. Prolactin. Psychiatric Clinics of North America 21, 341–358. ¨ Pflug, B., Tolle, R., 1971. Therapy of endogenous depressions using sleep deprivation. Practical and theoretical consequences wGermanx. Nervenarzt 42, 117–124. Prevot, E., Maudhuit, C., Le Poul, E., Hamon, M., Adrien, J., 1996. Sleep deprivation reduces the citalopram-induced inhibition of serotoninergic neuronal firing in the nucleus raphe dorsalis of the rat. Journal of Sleep Research 5, 238–245. Preziosi, P., Vacca, M., Cerrito, F., 1983. Serotonergic and endogenous opiates interplay in the control of ACTH and prolactin release. Archives Internationales de Pharmacodyn´ amie et de Therapie 263, 325–327. ¨ Rudolf, G.A., Tolle, R., 1978. Sleep deprivation and circadian rhythm in depression. Psychiatria Clinica 11, 198–212. ¨ Schilgen, B., Tolle, R., 1980. Partial sleep deprivation as therapy for depression. Archives of General Psychiatry 37, 267–271. Seibyl, J.P., Krystal, J.H., Price, L.H., Woods, S.W., D’Amico, C., Heninger, G.R., Charney, D.S., 1991. Effects of ritanserin on the behavioral, neuroendocrine, and cardiovascular responses to meta-chlorophenylpiperazine in healthy human subjects. Psychiatry Research 38, 227–236. ¨ Seifritz, E., Muller, M.J., Annen, O., Nil, R., Hatzinger, M., Hemmeter, U., Moore, P., Holsboer-Trachsler, E., 1997. Effect of sleep deprivation on neuroendocrine response to a serotonergic probe in healthy male subjects. Journal of Psychiatric Research 31, 543–554. Smeraldi, E., Benedetti, F., Barbini, B., Campori, E., Colombo, C., 1999. Sustained antidepressant effect of sleep deprivation combined with pindolol in bipolar depression. A placebocontrolled trial. Neuropsychopharmacology 20, 380–385. Sutton, L.C., Lea, S.E., Will, M.J., Schwartz, B.A., Hartley, C.E., Poole, J.C., Watkins, L.R., Maier, S.F., 1997. Inescap-
164
E. Frecska et al. / Psychiatry Research 118 (2003) 155–164
able shock-induced potentiation of morphine analgesia. Behavioral Neuroscience 111, 1105–1113. Tsai, L.L., Bergmann, B.M., Perry, B.D., Rechtschaffen, A., 1993. Effects of chronic total sleep deprivation on central noradrenergic receptors in rat brain. Brain Research 602, 221–227. Volk, S.A., Kaendler, S.H., Hertel, A., Maul, F.D., Manoocheri, R., Weber, R., Georgi, K., Pflug, B., Hor, G., 1997. Can response to partial sleep deprivation in depressed patients be predicted by regional changes of cerebral blood flow? Psychiatry Research 75, 67–74. Wesemann, W., Weiner, N., Rotsch, M., Schulz, E., 1983. Serotonin binding in rat brain: circadian rhythm and effect of sleep deprivation. Journal of Neural Transmission Supplementum 18, 287–294. World Health Organization, 1994. Schedules for Clinical Assessment in Neuropsychiatry. American Psychiatric Press, Washington, DC.
Wirz-Justice, A., Van den Hoofdakker, R.H., 1999. Sleep deprivation in depression: what do we know, where do we go? Biological Psychiatry 46, 445–453. Wu, J.C., Bunney, W.E., 1990. The biological basis of an antidepressant response to sleep deprivation and relapse: review and hypothesis. American Journal of Psychiatry 147, 14–21. Wu, J., Buchsbaum, M.S., Gillin, J.C., Tang, C., Cadwell, S., Wiegand, M., Najafi, A., Klein, E., Hazen, K., Bunney Jr., W.E., Fallon, J.H., Keator, D., 1999. Prediction of antidepressant effects of sleep deprivation by metabolic rates in the ventral anterior cingulate and medial prefrontal cortex. American Journal of Psychiatry 156, 1149–1158. Zis, A.P., Haskett, R.F., Albala, A.A., Carroll, B.J., Lohr, N.E., 1985. Prolactin response to morphine in depression. Biological Psychiatry 20, 287–292.