فرضیه سروتونین افسردگی زمستان: اثرات رفتاری و غددی عصبی گیرنده آگونیست نسبی 5 HT1A در بیماران مبتلا به اختلال عاطفی فصلی و افراد سالم

Winter depressions in seasonal affective disorder (SAD) are associated with central serotonergic (5-HT) dysfunction. SAD patients demonstrate rather specific, state-dependent, abnormal increases in `activation-euphoria' ratings following intravenous infusion of the 5-HT receptor agonist meta-chlorophenylpiperazine (m-CPP). Several studies are also consistent with abnormal serotonergic regulation of the hypothalamic–pituitary–adrenal (HPA) axis in SAD. Here, we investigated the effects of the 5-HT1A receptor partial agonist ipsapirone, which produces behavioral effects and HPA-axis activation, to further characterize the 5-HT receptor subtype-specificity of these disturbances in SAD. Eighteen SAD patients and 18 control subjects completed two drug challenges (ipsapirone 0.3 mg/kg and placebo) separated by 3–5 days in randomized order. We measured behavioral responses with the NIMH self-rating scale, and plasma ACTH, cortisol, and prolactin concentrations. Compared with placebo, ipsapirone was associated with significant increases in self-rated `functional deficit' and `altered self-reality', and in each of the hormones. There were no differences between groups on any measures. The level of depression in SAD patients was inversely correlated with their ipsapirone-induced cortisol responses. There were significant drug×order effects on baseline `anxiety' scores, ACTH and cortisol concentrations, such that subjects were significantly more stressed (higher `anxiety', ACTH and cortisol) prior to their first challenge compared with their second. In conclusion, post-synaptic 5-HT1A receptors appear to function normally in SAD. The previously observed m-CPP-induced behavioral abnormality may be mediated by either 5-HT2C or 5-HT7 receptors.

In its original formulation, the serotonin (5-HT) hypothesis of seasonal affective disorder (SAD) (Rosenthal et al., 1984) stated that in winter, the decreased amount of sunlight leads to a deficiency of central serotonergic transmission and the development of clinical depression in patients with SAD. This hypothesis was based, in part, on the following observations. First, many of the symptoms of winter depression (e.g. hypersomnia and hyperphagia) were known to be regulated by hypothalamic 5-HT (Jouvet, 1969; Fernstrom and Wurtman, 1971). Second, hypothalamic 5-HT content had been shown to vary seasonally in post-mortem analyses of brains of healthy volunteers, with troughs in the winter (Carlsson et al., 1980). Third, seasonal rhythms had been reported in several peripheral indices of 5-HT metabolism [e.g. cerebrospinal fluid 5-hydroxyindoleacetic acid and platelet 5-HT uptake (Wirz-Justice and Richter, 1979; Swade and Coppen, 1980)] which were thought to reflect similar seasonal rhythms in central 5-HT metabolism. Thus, patients with SAD were thought to be abnormally vulnerable to the decreases in hypothalamic 5-HT that normally occur during the winter (Rosenthal et al., 1995; Kasper et al., 1996) — a hypothesis that attracted considerable experimental support from investigations in both humans and animals (Rosenthal and Blehar, 1989). Consistent with the original 5-HT hypothesis of SAD, a number of treatment efficacy studies have since demonstrated that agents that enhance brain 5-HT transmission, such as d-fenfluramine ( O'Rourke et al., 1989), tranylcypromine ( Dilsaver and Jaeckle, 1990), fluoxetine ( Ruhrman et al., 1993; Childs et al., 1995; Lam et al., 1995), moclobemide ( Partonen and Lönnqvist, 1996), and tryptophan ( McGrath et al., 1990; Lam et al., 1997) alleviate the symptoms of winter depression. Similarly, high carbohydrate meals, which increase brain 5-HT content ( Fernstrom and Wurtman, 1971), are `craved' by depressed patients with SAD ( Rosenthal et al., 1989; Krauchi et al., 1990), while rapid tryptophan depletion, which lowers brain 5-HT ( Moja et al., 1989), reverses the antidepressant effects of both light therapy ( Lam et al., 1996; Neumeister et al., 1997; Neumeister et al., 1998b) and summer ( Neumeister et al., 1998a) in remitted patients with SAD. Neuroendocrine challenge studies have also provided support for the 5-HT hypothesis of SAD. Probes administered have included the 5-HT precursor 5-hydroxytryptophan (Jacobsen et al., 1987), the 5-HT2C partial agonist meta-chlorophenylpiperazine (m-CPP) ( Joseph-Vanderpool et al., 1993; Jacobsen et al., 1994; Garcia-Borreguero et al., 1995; Schwartz et al., 1997a; Levitan et al., 1998), the non-specific 5-HT releasing agent dl-fenfluramine ( Coiro et al., 1993; Yatham and Michalon, 1995), the 5-HT1D antagonist sumitriptan ( Yatham et al., 1997), and ovine-corticotropin releasing hormone ( Joseph-Vanderpool et al., 1991). In two of these studies, diminished growth hormone responsivity has been found in patients with SAD ( Schwartz et al., 1997a; Yatham et al., 1997). Mixed results have been found with regard to prolactin ( Coiro et al., 1993; Garcia-Borreguero et al., 1995; Schwartz et al., 1997a; Levitan et al., 1998). The most frequently ( Joseph-Vanderpool et al., 1991; Coiro et al., 1993; Schwartz et al., 1997a), albeit inconsistently ( Jacobsen et al., 1987; Garcia-Borreguero et al., 1995; Yatham and Michalon, 1995; Levitan et al., 1998) replicated neuroendocrine abnormality has been a blunted responsivity of the hypothalamic–pituitary–adrenal (HPA) axis to various probes in patients with SAD. Perhaps the most conspicuous abnormality to emerge from pharmacological challenge studies involving patients with SAD has been the finding that intravenous administration of m-CPP produces rather specific, state-dependent, abnormal increases in `activation-euphoria' ratings [one of six subscales on the NIMH self-rating scale ( van Kammen and Murphy, 1975; Murphy et al., 1989)] in patients with SAD during winter depression ( Jacobsen et al., 1994; Schwartz et al., 1997a), but not during remission either following light treatment ( Jacobsen et al., 1994; Schwartz et al., 1997a) or in the summer ( Joseph-Vanderpool et al., 1993). As such, this abnormal behavioral response to m-CPP may be considered a replicated state-marker of the depressed condition in SAD ( Schwartz et al., 1997a). The `activation-euphoria' subscale is comprised of four items: `I feel elated', `I feel talkative', `I feel energetic', and `I have racing thoughts'. As such, the induction by m-CPP of significant and abnormal increases in `activation-euphoria' ratings in patients with winter depression bears some resemblance to a switch into hypomania. The 5-HT receptor subtype-specificity of this abnormal m-CPP-induced `activation-euphoria' response in patients with SAD is unknown. m-CPP binds to a variety 5-HT receptor subtypes in the following descending order of affinities (pKds); 5-HT2C (7.7), 5-HT3 (7.0), 5-HT2A (6.7), 5-HT1B (6.6), 5-HT1A (6.5), 5-HT7 (6.4), 5-HT1D (5.8), 5-HT6 (5.6) ( Monsma et al., 1993; Hoyer et al., 1994; Sleight et al., 1995). m-CPP also has moderate affinity for α2-adrenergic receptors (6.2) and considerably lower affinity for other adrenergic and dopamine receptors (4.6–5.6) and for the human serotonin transporter (<4.0). Studies in which m-CPP has been administered to humans following pre-treatment with a 5-HT receptor antagonist do not yet permit a conclusion regarding the 5-HT receptor subtype(s) responsible for m-CPP-induced `activation-euphoria' ( Kahn and Wetzler, 1991; Pigott et al., 1991 and Pigott et al., 1993; Seibyl et al., 1991; Broocks et al., 1997). Hence, our working hypothesis has been that the state-dependent, abnormal increases in m-CPP-induced `activation-euphoria' scores in patients with SAD are mediated by the 5-HT receptor subtype at which m-CPP has its highest affinity, namely, the 5-HT2C receptor. Acute administration of a selective 5-HT1A receptor agonist may help to clarify the 5-HT receptor subtype-specificity of this m-CPP-induced behavioral abnormality in patients with SAD. 5-HT1A receptors are located in a variety of brain regions that modulate emotional responses and that could potentially mediate m-CPP-induced `activation-euphoria' [e.g. the prefrontal cortex, hippocampus ( Pazos et al., 1987; De Vry, 1995)]. Several previous studies have measured the behavioral responses following acute administration of a selective 5-HT1A receptor agonist, and have reported feelings, such as `light-headedness' and `nervousness' ( Murphy et al., 1991). The relationship, if any, between these feelings and `activation-euphoria' is unknown. Hence in theory, acute administration of a selective 5-HT1A receptor agonist could establish whether 5-HT1A receptor stimulation alone is sufficient to induce either significant or abnormal increases in `activation-euphoria' scores. If so, then the m-CPP-induced behavioral abnormality in patients with SAD could well be mediated by 5-HT1A receptors, given m-CPPs moderate affinity for this receptor. In addition, acute administration of a selective 5-HT1A receptor agonist might also help to clarify the 5-HT receptor subtype-specificity of the putative HPA-axis abnormality in patients with SAD, since ACTH release is regulated by several 5-HT receptor subtypes including 5-HT1A receptors ( Calogero et al., 1990 and Calogero et al., 1993; Fuller, 1992; Saphier et al., 1994). Ipsapirone is a relatively selective 5-HT1A receptor partial agonist, having more than 100 times the affinity for 5-HT1A receptors as for any other 5-HT receptor subtype [pKis: 5-HT1A (7.7), 5-HT2A (5.1), 5-HT1D (4.9), 5-HT2C (4.5), and 5-HT3 (<5) ( Hoyer, 1988; Neijt et al., 1988; Van Wijngaarden et al., 1990)]. Ipsapirone also has less affinity for α1- and α2-adrenergic receptors (pKis=6.6 and 5.6, respectively) and for the dopamine-2 receptor (6.4) than for the 5-HT1A receptor, while its affinity for the 5-HT7 receptor is presently unknown ( Lovenberg et al., 1993). In the present study, we administered the selective 5-HT1A receptor partial agonist ipsapirone to 18 depressed patients with SAD and 18 healthy control subjects in the winter. We measured behavioral responses, and plasma ACTH and cortisol concentrations. In addition, because of several reports suggesting that the regulation of prolactin secretion is (1) abnormal in SAD ( Coiro et al., 1993; Garcia-Borreguero et al., 1995; Oren et al., 1996); and (2) stimulated by some 5-HT1A agonists ( Cowen et al., 1990), we also measured plasma prolactin concentrations in response to ipsapirone. We had two main hypotheses. First, we predicted that although ipsapirone administration would be associated with significant behavioral effects (i.e. ipsapirone would be psychoactive), ipsapirone would not be associated with significant increases in `activation-euphoria' scores in either group (i.e. 5-HT1A receptor stimulation is probably not a primary event in mediating our previously observed m-CPP-induced behavioral abnormality). Second, we predicted that there would be no differences between patients with SAD and control subjects on any of the six NIMH behavioral subscales following ipsapirone administration [i.e. in patients with SAD, the range of behavioral abnormalities observed in response to serotonergic, psychoactive drug administration is not only quite specific (only one of six NIMH subscales) and state-dependent (only during winter depression), but also 5-HT receptor subtype specific (only in response to 5-HT2C receptor stimulation)]. We had no particular hypothesis about the HPA-axis or prolactin responses to ipsapirone. We have previously reported the thermoregulatory responses to ipsapirone from the subjects in the present study ( Schwartz et al., 1998).