Gyorgy Bagdy, Marton Graf, Zsuzsanna E. Anheuer, Edit A. Modos and Sandor Kantor

International Journal of Neuropsychopharmacology (2001), 4, 399 408. Copyright 2001 CINP DOI: 10.1017 S1461145701002632 Anxiety-like effects induced by acute fluoxetine, sertraline or m-cpp treatment are reversed by pretreatment with the 5-HT 2C antagonist SB-242084 but not the 5-HT 1A antagonist WAY-100635 BRIEF REPORT Gyorgy Bagdy, Marton Graf, Zsuzsanna E. Anheuer, Edit A. Modos and Sandor Kantor Laboratory of Neurochemistry and Experimental Medicine, and Department of Neurology, Faculty of Health Sciences, Semmelweis University, National Institute of Psychiatry and Neurology, Budapest, Huvosvolgyi ut 116, H-1021, Hungary Abstract The possible role of 5-HT and 5-HT s in the anxiety induced by fear, acute treatment with SSRI A C antidepressants or the 5-HT agonist m-cpp were tested in the social interaction anxiety test in male Sprague Dawley rats. Fluoxetine (2 5 10 mg kg, i.p.), sertraline (15 mg kg, i.p.) and m-cpp (0 5 2 0 mg kg, i.p.) all had an anxiogenic-like profile (decrease in time of total social interaction and increase in self-grooming compared to vehicle) under low-light, familiar arena test conditions. All these effects were reversed by pretreatment with the highly subtype-selective 5-HT antagonist, SB-242084 at doses of either 0 05 C or 0 2 mg kg, i.p. In contrast, the selective 5-HT antagonist WAY-100635 (0 05 and 0 2 mg kg, A s.c.) failed to reverse SSRI-induced decrease in time of total social interaction, further, it augmented selfgrooming response. SB-242084 (0 2 mg kg) and WAY-100635 (0 05 and 0 2 mg kg) reversed hypolocomotion caused by the SSRI antidepressants. SB-242084, tested alone against vehicle under high-light, unfamiliar arena test conditions associated with fear, caused significant anxiolysis at 0 2 mg kg and higher doses. These results suggest that increased anxiety in rodents, and possibly, also in humans (e.g. agitation or jitteriness after SSRIs and panic after m-cpp), caused by acute administration of SSRI antidepressants or m-cpp, are mediated by activation of 5-HT s. Blockade of 5-HT autos may exacerbate certain C A acute adverse effects of SSRI antidepressants. Both 5-HT and 5-HT s are involved in the SSRIinduced decrease in locomotor activity. In addition, our studies confirm data that subtype-selective 5-HT C A C antagonists have strong anxiolytic actions. Received 1 April 2001; Reviewed 31 May 2001; Revised 5 August 2001; Accepted 19 August 2001 Key words: Anxiety, 5-HT, 5-HT, 5-HT, locomotor activity, m-cpp, SSRI C A antidepressant. Introduction SSRI antidepressants are therapeutically active in several anxiety disorders (Nutt and Glue, 1989; Papp, 1999; Westenberg and Den Boer, 1993). Clinically, the effect of these agents are characterized by delayed onset with the likelihood of initial exacerbation, e.g. agitation or jitteriness, especially in panic disorder patients, and that Address for correspondence: Dr G. Bagdy, Laboratory of Neurochemistry and Experimental Medicine, and Department of Neurology, Faculty of Health Sciences, Semmelweis University, National Institute of Psychiatry and Neurology, Budapest, Huvosvolgyi ut 116, H-1021, Hungary. Tel.: 36-1-3915407 Fax: 36-1-3915305 E-mail: bag13638 helka.iif.hu may cause a decrease in treatment acceptance (Nutt and Glue, 1989; Papp, 1999; Pohl et al., 1988; Westenberg and Den Boer, 1993). In animal models of anxiety SSRI antidepressants are, in general, characterized by an anxiogenic-like profile after acute administration (Griebel, 1995; Handley et al., 1993). A transient anxiogenic-like effect of fluoxetine in the social interaction test that disappeared after chronic treatment was described by our group in earlier studies (To and Bagdy 1999; To et al., 1999). Pharmacological characterization of the effects of the 5-HT agonist m-chlorophenylpiperazine (m-cpp) led to the hypothesis that activation of 5-HT C s may mediate anxiety in humans and rodents although pharmacological tools used in these studies were not able to differentiate between 5-HT B and 5-HT C s

400 G. Bagdy et al. (a) (b) Figure 1. Effects of different doses of (a) fluoxetine (2 5 10 0 mg kg, i.p.) and (b) m-cpp (0 5 2 0 mg kg, i.p.) in the social interaction anxiety test under low-light, familiar arena conditions. Significant effects (p 0 05) compared to vehicle are denoted by *. (Bagdy, 1996, 1998; Bagdy et al., 1992; Bristow et al., 2000; Kennett et al., 1989; Murphy et al., 1991). Furthermore, the role of either 5-HT, namely 5-HT A, 5-HT A,5-HT B, and 5-HT C s, or other mechanisms, e.g. noradrenergic and neuropeptide systems have been suggested in acute and or chronic effects of SSRI antidepressants (Bristow et al., 2000; Dekeyne et al., 2000; Handley et al., 1993; Kennett et al., 1994; Lerer et al., 1999; Matto et al., 1996; Quested et al., 1997; Szabo et al., 2000; To and Bagdy 1999; To et al., 1999), but only one study used a 5-HT C subtype-selective antagonist (Dekeyne et al., 2000). To study the possible role of 5-HT A and 5-HT C s in acute SSRI administration-induced anxiety, WAY-100635, a subtypeselective 5-HT A antagonist (Fletcher et al., 1996; Griebel, 1999) and SB-242084, a subtype-selective 5-HT C antagonist (Kennett et al., 1997) were used as a pretreatment before administration of SSRI antidepressants. In addition to social interaction another measure of anxiety, self-grooming, and the anxiolytic-like actions of SB-242084 per se were also tested. Method All procedures used in this study were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals. The protocol of the study was approved by the local Ethical Committee. Male Sprague

5-HT C s mediate SSRI-induced anxiety 401 (a) (b) Figure 2. Effects of fluoxetine (5 0 mg kg, i.p.) after pretreatment with (a) SB-242084 (0 05 and 0 2 mg kg, i.p.), (b) WAY- 100635 (0 05 and 0 2 mg kg s.c.) or vehicle in the social interaction anxiety test under low-light, familiar arena conditions. Significant effects (p 0 05) compared to vehicle vehicle are denoted by *; significant effects (p 0 05) of fluoxetine compared to the antagonist vehicle are denoted by. Dawley rats (240 330 g, Crl: CDRBR, Charles River, Hungary), were used in the studies. The animals (4 per cage) were kept under standard conditions, with standard food (CRLT AM, Charles River, Hungary) and water freely available. The temperature was 21 1 C, the 12 h light dark cycle started at 06:00 hours. The social interaction test was performed as described previously (To and Bagdy, 1999). Rats were housed in a room adjacent to the testing room at least for 2 wk before the tests. The animals were randomly assigned to treatments. In the low-light (5 lx), familiar arena conditions animals were individually pre-exposed to the test box and the injections three times on days preceding the test. In the high-light (300 lx), unfamiliar arena conditions animals were not habituated to the test box. Rats were tested for social interaction with an unknown test partner that did not differ by more than 15 g in weight. At the end of each test the box was thoroughly wiped with detergent and dried to remove odours. The animals were tested in a random order between 09:00 and 13:00 hours in an

402 G. Bagdy et al. adjacent room for 7 5 min. The evenly illuminated box (60 60 40 cm), was marked in 10 cm compartments by lines on the floor. The behaviour of animals was recorded on videotape. It was scored later by a person unaware of the drug treatment as described earlier (To and Bagdy, 1999). The following behaviours were included in total social interaction: sniffing partner, anogenital sniffing, peaceful following, grooming partner, crawling under, climbing over, chasing, aggressive grooming, dominant posture, submissive posture, biting, boxing, kicking, pushing, wrestling. Rearing, crossing of lines and self-grooming were also scored. SB-242084 (6-chloro-5-methyl-1-[(2-[2-methylpyrid- 3-yloxy]pyrid-5-yl)carbamoyl]indoline dihydrochloride, Sigma-Aldrich, Budapest, Hungary), sertraline hydrochloride and fluoxetine hydrochloride (both kindly donated by EGIS Pharmaceuticals Ltd, Budapest, Hungary) were dissolved in 10% solution of 2-hydroxypropylβ-cyclodextrin (Research Biochemicals International, Natick, MA, USA). WAY-100635 maleate (N-[2- [4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-N-2-pyridinyl cyclohexanecarboxamide maleate, Research Biochemicals International) and m-cpp (meta-chlorophenylpiperazine hydrochloride, Research Biochemicals International), were dissolved in physiological saline. WAY-100635 was administered s.c., all other drugs i.p., in a volume of 1 ml kg. SB-242084 was administered 5 min before the SSRI antidepressants, WAY-100635 20 min before the SSRI antidepressants or m-cpp, which was administered 20 min before the test. The data were analysed using one- or two-way ANOVA followed by Tukey Kramer and Kruskal Wallis tests followed by the Mann Whitney rank sum test (Super ANOVA and StatViewSE Graph, Abacus Concepts, Berkeley, CA, USA). Each treatment group consisted of 10 14 animals that is, 5 7 pairs of rats. Each rat was tested only once. Data on the figures and in the text are expressed as mean S.E.M. Results Fluoxetine (Figure 1a; 2 5 10 mg kg, i.p.) and m-cpp (Figure 1b; 0 5 2 0 mg kg, i.p.) caused dose-dependent decrease in time of total social interaction [F(3,42) 7 68, p 0 001 and F(3,36) 11 01, p 0 001 for fluoxetine and m-cpp, respectively], increase in self-grooming and decrease in line crossings (Figure 1a, b). Minimal effective doses for time of total social interaction were 5 0 and 0 5 mg kg, for fluoxetine and m-cpp, respectively. At these doses fluoxetine, but not m-cpp, caused a significant decrease in the number of line crossings. The dose response curves for self-grooming were clearly different, Figure 3. Effects of sertraline (15 mg kg, i.p.) after pretreatment with SB-242084 (0 05 and 0 2 mg kg, i.p.) or vehicle in the social interaction anxiety test under low-light, familiar arena conditions. Significant effects (p 0 05) compared to vehicle vehicle are denoted by *; significant effects (p 0 05) of sertraline compared to the antagonist vehicle are denoted by. namely, a biphasic response was found for m-cpp but not for fluoxetine. Pretreatment with SB-242084 (0 05 and 0 2 mg kg, i.p.) significantly reversed effects of fluoxetine (Figure 2a) and sertraline (Figure 3) on time of total social interaction [F(2,68) 6 14, p 0 01 and F(2,60) 7 19, p 0 01 for pretreatment treatment interaction for fluoxetine and sertraline, respectively]. Effect of m-cpp (0 5 mg kg,

5-HT C s mediate SSRI-induced anxiety 403 Figure 4. Effects of m-cpp (0 5 mg kg, i.p.) after pretreatment with SB-242084 (0 05 mg kg, i.p.) or vehicle in the social interaction anxiety test under low-light, familiar arena conditions. Significant effects (p 0 05) compared to vehicle vehicle are denoted by *. i.p.) on time of total social interaction was also significantly reversed by pretreatment with SB-242084 [Figure 4; pretreatment treatment interaction, F(1,36) 12 64, p 0 01]. SB-242084 also reversed other effects, thus, self-grooming and decrease in line crossings induced by fluoxetine (Figure 2a) and sertraline (Figure 3) and selfgrooming induced by m-cpp (Figure 4), but the dose response effects of SB-242084 on the various parameters were different. Anxiety-related effects, like social in- Figure 5. Effects of different doses of SB-242084 (0 05 0 5 mg kg, i.p.) in the social interaction anxiety test under high-light, unfamiliar arena conditions. Significant effects (p 0 05) compared to vehicle are denoted by *. teraction and self-grooming responses, were reversed by 0 05 mg kg of SB-242084, but a higher dose was needed for the reversal of hypolocomotion. Pretreatment with WAY-100635 (Figure 2b; 0 05 and 0 2 mg kg, s.c.) failed to reverse the effects of fluoxetine on time of total social interaction or self grooming [pretreatment treatment interactions, F(2,68) 0 71, p 0 49 and F(2,68) 0 37, p 0 69, for the two parameters, respectively]. Even more, WAY-100635 alone slightly increased self-grooming, and, as a pretreatment, it augmented self-grooming response caused by fluoxetine [Figure 2b; pretreatment effect, F(2,68) 3 30, p 0 05]. In contrast, it reversed the effect of fluoxetine on line crossings very efficiently [Figure 2b; pretreatment treatment interaction, F(2,68) 5 81, p 0 005]. WAY-

404 G. Bagdy et al. 100635 alone did not have any effect on time of total social interaction or line crossings (Figure 2b). Possible anxiolytic effects of SB-242084 alone (Figure 5; 0 05 0 5 mg kg, i.p.) were tested in separate experiments under high-light, unfamiliar arena conditions. SB- 242084 had significant effects on time of total social interaction and self-grooming. The minimal effective dose for both social interaction and self-grooming was 0 2 mg kg. Locomotor activity was not significantly altered, although a strong trend, a 41% increase was found in the number of line crossings at 0 5 mg kg SB- 242084 compared to vehicle (Figure 5). Discussion Anxiogenic-like effect of fluoxetine (Bristow et al., 2000; Griebel, 1995; To and Bagdy 1999; To et al., 1999) and citalopram (Dekeyne et al., 2000) in the social interaction test was described in earlier studies. Acute administration of these compounds caused significant decrease in the time of total social interaction and fluoxetine caused also a decrease in the number of line crossings, and an increase in time of self-grooming (G. Bagdy, Z. E. Anheuer and S. Kantor, personal communication; Bristow et al., 2000; Dekeyne et al., 2000; To and Bagdy 1999; To et al., 1999). Under lowlight, familiar arena conditions the surrounding is comfortable for the animals, and thus, basal anxiety is low, so anxiogenic compounds, including, e.g. corticotropinreleasing hormone (CRH) or m-cpp efficiently alter parameters like social interaction (File and Hyde, 1978; Guy and Gardner, 1985; Kantor et al., 2000; Kennett et al., 1989; To et al., 1999) or self-grooming (Dunn and File, 1987; Kantor et al., 2000; To et al., 1999). The number of line crossings, a measure of locomotor activity, is decreased by some, e.g. cholecystokinin (CCK) or higher doses of m-cpp, but not all compounds that increase anxiety (File and Hyde, 1978; Guy and Gardner, 1985; Kennett et al., 1989; To and Bagdy 1999). Anxiogenic action of SSRI antidepressants is also supported by data that fluoxetine caused significant c-fos-like immunoreactivity in the central nucleus of amygdala, and paraventricular nucleus and increased ACTH and corticosterone secretion (Gibbs and Vale, 1983; Overstreet et al., 2000). The 5-HT family consists of three subtypes, namely, the 5-HT, the 5-HT and the 5-HT (formerly A B C 5-HT ), and several antagonists, like ritanserin C has similar affinity for all three subtypes (Bonhaus et al., 1997). SB-242084, the first subtype-selective 5-HT C antagonist, has very high affinity (pk i 9 0) for the cloned human 5-HT and has over 100- and C 158-fold selectivity over the cloned human 5-HT and 5- A HT s (Kennett et al., 1997). Furthermore, it has B at least 100-fold selectivity for this over all other s tested (Kennett et al., 1997). The affinities of m- CPP for these subtypes are 5-HT 5-HT 5-HT B C A (Bonhaus et al., 1997; Murphy et al., 1991), but functional characterization on recombinant human s showed that, compared to 5-HT, m-cpp has 65, 24 and 22% relative efficacies on 5-HT, 5-HT and 5-HT C B A s, respectively (Porter et al., 1999). SB-242084 reversed the actions of SSRI antidepressants and m-cpp very efficiently in our study; in the dose of 0 05 mg kg it reversed decrease of social interaction and increase in selfgrooming caused by all compounds tested. These data, together with a recently published study of Dekeyne et al. (2000), where SB-242084 effectively reversed anxiogenic action of citalopram, strongly suggest that 5-HT C s mediate the anxiogenic actions of SSRI antidepressants caused by increased synaptic 5-HT concentration. SB-242084 alone failed to alter anxiety-related behaviours, namely, time of total social interaction and self-grooming, but pretreatment with this compound significantly reversed these effects of SSRI antidepressants and m-cpp, and in addition, hypolocomotion caused by fluoxetine and sertraline under low-light, familiar arena conditions. Because of the -binding profile of this compound, these data suggest that SSRIs and m-cpp mediate anxiogenic responses by activation of 5-HT C s. However, other explanations might also be possible, thus first, the effect of the compounds on the locomotion may confound the interpretation of the data, secondly, the anxiolytic effect of SB-242084 per se may lead to a false conclusion. To discuss the first question, our data show that decreases in locomotion caused by fluoxetine are, in some cases parallel to its effect on total social interaction, that might suggest that the decrease in social interaction may be caused by the effects on locomotion. However, these effects showed clear dissociation in the case of m-cpp and citalopram, thus, m-cpp failed to decrease locomotion at the doses of 0 5 mg kg, but it caused significant decrease on time of total social interaction. A similar dissociation was found in the case of citalopram (G. Bagdy, Z. E. Anheuer and S. Kantor, personal communication; Dekeyne et al., 2000). The efficacies of SB-242084 on SSRI-induced social interaction and locomotion were also dissociated, namely, anxiety-like effects were reversed by 0 05 mg kg SB- 242084, while higher doses of the compound were needed for the reversal of hypolocomotion (see Results). In addition, WAY-100635 completely reversed hypolocomotion caused by fluoxetine but it failed to reverse the decrease in social interaction. Thus, the decrease and reversal in social interaction cannot be considered as a secondary

5-HT C s mediate SSRI-induced anxiety 405 effect of locomotion. Furthermore, in addition to social interaction, we used another measure of anxiety, namely self-grooming. This stereotype behaviour is increased by aversive stimuli and anxiogenic compounds like CRH or m-cpp measured either in the social interaction test or in single cages (Bagdy et al. 1992; Dunn and File, 1987; Kantor et al., 2000; To et al., 1999). Conversely, it is attenuated by anxiolytic compounds under high anxiety conditions. Under certain conditions, this parameter is a more sensitive and less variable measure than time of total social interaction (Dunn and File, 1987; Kantor et al., 2000; To et al., 1999). For example, WAY-100635 pretreatment enhanced self-grooming response but not social interaction caused by fluoxetine in this study, and the minimal effective dose for fluoxetine was lower for self-grooming than for social interaction (see Results). It is likely that augmentation of this stereotypical behaviour by WAY-100635 is caused by the enhancement of the effect of fluoxetine on extracellular 5-HT, as was shown in microdialysis studies (Malagie et al., 1996). It has been suggested that m-cpp-induced self-grooming is mediated by 5-HT C (formerly 5-HT C ) s, although subtype-selective 5-HT C antagonists have not been used, and neuropeptides synthesized by the hypothalamic paraventricular nucleus (Bagdy and Makara, 1995; Bagdy et al., 1992). m-cpp causes significant increase of this response at low doses where absolutely no locomotion effect can be observed (Bagdy et al., 1992). Dose-dependence of self-grooming and locomotion response of m-cpp were also dissociated in this study. Thus, anxiety-like responses of SSRIs and m-cpp, and anxiolytic effects of SB-242084 cannot be explained on the basis of activity changes. A similar conclusion was drawn also by Bristow et al. (2000) who measured anxiety and locomotion after fluoxetine treatment. To address the second question, namely, that the anxiolytic effect of SB-242084 per se may lead to a false conclusion, we performed an experiment with SB-242084 at high-light, unfamiliar arena conditions. Although at low-light, familiar arena conditions, where SB-242084 reversed the effects of m-cpp and the SSRIs, the effects of SB-242084 alone were not significant, additional experiments were necessary, because under low-light, familiar arena conditions anxiolytic compounds per se show weak effects, but their anxiolytic-like effects are enhanced under conditions where the animals sense more fear (File and Hyde, 1978; Guy and Gardner, 1985; Kantor et al., 2000). Under high light, unfamiliar arena conditions SB-242084 alone caused dose-dependent anxiolytic-like activity. Significant effect was found, however, only at 0 2 and 0 5 but not at 0 05 mg kg. This finding is similar to that described by Kennett et al. (1997), despite the fact that we used a 2-fold shorter observation period, similar to that also used by other groups (File and Hyde, 1978; Guy and Gardner, 1985). In contrast, SB-242084 completely reversed social interaction and self-grooming effects of the SSRIs and m-cpp at the dose of 0 05 mg kg, thus, this latter effect can not be explained by the anxiolytic action of this compound. The anxiogenic-like effects of fluoxetine were abolished after chronic treatment (Bristow et al., 2000; To and Bagdy, 1999; To et al., 1999). This finding is in parallel with the clinical and experimental data obtained with chronic SSRI treatment (Handley et al., 1993; Lightowler et al., 1994; Nutt and Glue, 1989; Westenberg and Den Boer, 1993). There is a higher activation of postsynaptic s by the excess extracellular 5-HT produced by chronic, compared to acute, SSRI treatment (Artigas, 1993; Bel and Artigas 1993; Invernizzi et al., 1994). Furthermore, disappearance of fluoxetine-induced acute anxiety is accelerated by additional subchronic administration of WAY-100635 (Bristow et al., 2000). The antagonists used for characterization of s that mediate the effect of fluoxetine used in the work of Bristow et al. (2000) could not differentiate between 5- HT B and 5-HT C s. By the use of the subtypeselective SB-242084, our data show that fluoxetineinduced anxiety is mediated by 5-HT C s. Thus, the mechanism of action of delayed attenuation of fluoxetine-induced anxiety possibly includes desensitization of 5-HT C s. This is supported also by the clinical data that m-cpp exacerbates symptoms before, but not after chronic treatment with clomipramine (Zohar et al., 1988), and m-cpp-induced neuroendocrine and hyperthermic responses are attenuated after chronic citalopram treatment (Quested et al., 1997). In rats, m-cpp-induced hypolocomotion was attenuated after chronic SSRI treatment (Kennett et al., 1994). The role of 5-HT B or 5-HT C s have been suggested in these responses of m-cpp at least in rats (Aulakh et al., 1992; Bagdy, 1996, 1998; Bagdy et al., 1989, 1992; Bristow et al., 2000; Lucki et al., 1989), and, in addition, effectiveness of SB-242084 in self-grooming and social interaction responses in this study provide further evidence for the involvement of 5-HT C s in m-cpp-induced anxiety, thus, these data support the view that desensitization of 5-HT C s is caused by chronic SSRI treatment. Fluoxetine has modest affinity for 5-HT C s (Bonhaus et al., 1997; Jenck et al., 1994; Palvimaki et al., 1996). In an in vivo functional model for 5-HT C s, however, it failed to have any significant 5-HT C antagonist action (Jenck et al., 1994). This might be explained by its 5-HT-mimetic, reuptake inhibiting effect (Jenck et al., 1994). Indeed, its affinity for the 5-HT uptake site is much higher than its affinity for the 5-HT C

406 G. Bagdy et al. (Palvimaki et al., 1996). The dose of fluoxetine used in this study was based on the dose response curves (Figure 1a), the dose of sertraline on the ratio of clinical doses and affinities for the 5-HT uptake sites of these compounds (Kasper et al., 1994; Palvimaki et al., 1996). An increase in serotonergic function produced by SSRI antidepressants underlies their clinical effectiveness in anxiety disorders and major depression (Artigas, 1993; Mongeau et al., 1997). Interestingly, acute increase in 5- HT neurotransmission usually increases fear and anxietylike behaviour (Griebel, 1995), although it may have also an opposite effect depending on the type of the test (Hetem et al., 1996). Increased anxiety-like behaviour was described in Fawn-Hooded rats, a strain with inherited impairment of 5-HT storage and uptake (Kantor et al., 2001). Graeff et al. (1996) suggested that at least three serotonergic pathways are involved in the regulation of anxiety, thus, separate pathways facilitate conditioned fear, inhibit inborn fight flight reactions and promote resistance to chronic, unavoidable stress. SSRIs and m- CPP increase anxiety in the social interaction test or in the elevated plus maze in rats (Griebel, 1995; Handley et al., 1993; Kantor et al., 2000; Kennett et al., 1989; To and Bagdy, 1999). Anxiogenic effects of m-cpp in clinical studies are also well known (Arato and Bagdy, 1998; Kahn and Wetzler, 1991; Murphy et al., 1991), and anxiety related transient effects of SSRI antidepressants, like agitation or jitteriness have been described (Nutt and Glue, 1989; Papp, 1999; Pohl et al., 1988; Westenberg and Den Boer, 1993). Stress hormones, like ACTH and cortisol or corticosterone are also released after acute administration of these compounds in both humans and rodents (Bagdy et al., 1989; Murphy et al., 1991). However, ritanserin, a non-subtype-selective 5-HT antagonist that reverses some effects of m-cpp, may also increase or decrease anxiety (Bagdy 1998; Guimares et al., 1997; Kahn and Wetzler, 1991). Furthermore, activation of the 5-HT C efficiently alters release of other neurotransmitters like noradrenaline, adrenaline, dopamine or glutamate (Bagdy et al., 1989; Di Matteo et al., 1999; Marcoli et al., 1998; Millan et al., 1998), and neuropeptides like CRH, oxytocin and vasopressin (Bagdy et al., 1989, 1992; Calogero et al., 1989). Some of these neurotransmitters systems may mediate anxiolytic and antidepressant activity of SSRIs. The role of CCK in SSRI-induced anxiety-like effects is supported by the data that citalopram-induced decrease in exploratory behaviour is reversed by a CCK B antagonist (Matto et al., 1996). A progressive attenuation of the firing activity of locus coeruleus noradrenergic neurons by sustained SSRI treatment has been shown in electrophysiological studies (Szabo et al., 2000). Furthermore, CCK or CRH-induced anxiety is also decreased after chronic SSRI treatment (To and Bagdy, 1999; To et al., 1999; Van Megen et al., 1997). Thus, different anatomical pathways and other neurotransmitter neuropeptide systems and s may mediate or modulate 5-HT C -induced changes. In conclusion, our studies provide evidence that acute anxiogenic effects of SSRI antidepressants are reversed by pretreatment with low doses of the subtype-selective 5-HT C antagonist SB-242084, and thus, they suggest that anxiety-related side-effects of acute SSRI treatment, like agitation or jitteriness, may be mediated by 5-HT C s. Acknowledgements These studies were supported by the Hungarian Research Fund Grant T020500, Ministry of Health Research Grant T023 2000, and Postdoctoral Ph.D. Fellowship Program of the Semmelweis University, Ministry of Culture and Education, Hungary. References Arato M, Bagdy G (1998). Gender difference in m-cpp challenge test in healthy volunteers. International Journal of Neuropsychopharmacology 1, 121 124. Artigas F (1993). 5-HT and antidepressants: new views from microdialysis studies. Trends in Pharmacological Sciences 14, 262. Aulakh CS, Hill JL, Murphy DL (1992). Effect of various serotonin subtype selective antagonists alone and on m-cpp-induced neuroendocrine changes in rats. Journal of Pharmacology and Experimental Therapeutics 263, 588 595. Bagdy G (1996). Role of the hypothalamic paraventricular nucleus in 5-HT A,5-HT A and 5-HT C -mediated oxytocin, prolactin and ACTH corticosterone responses. Behavioural Brain Research 73, 277 280. Bagdy G (1998). Serotonin, anxiety, and stress hormones. Focus on 5-HT subtypes, species and gender differences. Annals of the New York Academy of Sciences 851, 357 363. Bagdy G, Calogero AE, Murphy DL, Szemeredi K (1989). Serotonin agonists cause parallel activation of the sympathoadrenomedullary system and the hypothalamopituitary adrenocortical axis in conscious rats. Endocrinology 125, 2664 2669. Bagdy G, Kalogeras KT, Szemeredi K (1992). Effect of 5-HT C and 5-HT stimulation on excessive grooming, penile erection, and plasma oxytocin concentrations. European Journal of Pharmacology 229, 9 14. Bagdy G, Makara GB (1995). Paraventricular nucleus controls 5-HT -mediated corticosterone and prolactin but not oxytocin and penile erection responses. European Journal of Pharmacology 275, 301 305.

5-HT C s mediate SSRI-induced anxiety 407 Bel N, Artigas F (1993). Chronic treatment with fluvoxamine increases extracellular serotonin in frontal cortex but not in raphe nuclei. Synapse 15, 243 245. Bonhaus DW, Weinhardt KK, Taylor M, Desouza A, McNeeley PM, Szczepanski K, Fontana DJ, Trinh J, Rocha CL, Dawson MW, Flippin LA, Eglen RM (1997). RS- 102221: a novel high affinity and selective, 5-HT C antagonist. Neuropharmacology 36, 621 629. Bristow LJ, O Connor D, Watts R, Duxon MS, Hutson PH (2000). Evidence for accelerated desensitisation of 5-HT C s following combined treatment with fluoxetine and the 5-HT A antagonist, WAY 100,635, in the rat. Neuropharmacology 39, 1222 1236. Calogero AE, Bernardini R, Margioris AN, Bagdy G, Gallucci WT, Munson PJ, Tamarkin L, Tomai TP, Brady L, Gold PW, Chrousos PW (1989). Effects of serotonergic agonists and antagonists on corticotropin-releasing hormone secretion by explanted rat hypothalami. Peptides 10, 189 200. Dekeyne A, Denorme B, Monneyron S, Millan MJ (2000). Citalopram reduces social interaction in rats by activation of serotonin (5-HT) C s. Neuropharmacology 39, 1114 1117. Di Matteo V, Di Giovanni G, Di Mascio M, Esposito E (1999). SB 242084, a selective serotonin C antagonist, increases dopaminergic transmission in the mesolimbic system. Neuropharmacology 38, 1195 1205. Dunn AJ, File SE (1987). Corticotropin-releasing factor has an anxiogenic action in the social interaction test. Hormones and Behavior 21, 193 202. File SE, Hyde JRG (1978). Can social interaction be used to measure anxiety? British Journal of Pharmacology 62, 19 24. Fletcher A, Froster EA, Bill DJ, Brown G, Cliffe IA, Hartley JE, Jones DE, McLenachan A, Stanhope KJ, Critchley DJP, Childs KJ, Middlefell VC, Lanfumay L, Corradetti R, Laporte A-M, Gozlan H, Hamon M, Dourish CT (1996). Electrophysiological, biochemical, neurohormonal and behavioural studies with WAY-100635, a potent, selective and silent 5-HT A antagonist. Behavioural Brain Research 73, 337 353. Gibbs DM, Vale W (1983). Effect of the serotonin reuptake inhibitor fluoxetine on corticotropin-releasing factor and vasopressin secretion into the hypophyseal portal blood. Brain Research 280, 176 179. Graeff FG, Guimeras TS, De Andrade TG (1996). Role of 5-HT in stress, anxiety and depression. Pharmacology, Biochemistry and Behavior 54, 129 141. Griebel G (1995). 5-Hydroxytryptamine-interacting drugs in animal models of anxiety disorders: more than 30 years of research. Pharmacology and Therapeutics 65, 319 395. Griebel G (1999). 5-HT A blockers as potential drug candidates for the treatment of anxiety disorders. Drug News and Perspectives 12, 484 490. Guimaraes FS, Mbaya PS, Deakin JF (1997). Ritanserin facilitates anxiety in a stimulated public-speaking paradigm. Journal of Psychopharmacology 11, 225 231. Guy AP, Gardner CR (1985). Pharmacological characterization of a modified social interaction model of anxiety in the rat. Neuropsychobiology 13, 194 200. Handley SL, McBlane JW, Critchley MAE, Njung e K (1993). Multiple serotonin mechanisms in animal models of anxiety: environmental, emotional and cognitive factors. Behavioural Brain Research 58, 203 210. Hetem LA, de Souza CJ, Guimaraes ES, Zuardi AW, Graeff FG (1996). Effect of d-fenfluramine on human experimental anxiety. Psychopharmacology 127, 276 282. Invernizzi R, Bramante M, Samanin R (1994). Chronic treatment with citalopram facilitates the effect of a challenge dose on cortical serotonin output: role of presynaptic 5-HT A s. European Journal of Pharmacology 260, 243 246. Jenck F, Moreau J, Mutel V, Martin JR (1994). Brain 5-HT C s and antidepressants. Progress in Neuropsychopharmacology and Biological Psychiatry 18, 563 574. Kahn RS, Wetzler S (1991). m-chlorophenylpiperazine as a probe of serotonin function. Biological Psychiatry 30, 1139 1166. Kantor S, Anheuer ZE, Bagdy G (2000). High social anxiety and low aggression in Fawn-Hooded rats. Physiology and Behavior 71, 551 557. Kantor S, Graf M, Anheuer ZE, Bagdy G (2001). Rapid desensitization of 5-HT A s in Fawn-Hooded rats after chronic fluoxetine treatment. European Neuropsychopharmacology 11, 15 24. Kasper S, Ho flich G, Scholl HP, Mo ller HJ (1994). Safety and antidepressant efficacy of selective serotonin re-uptake inhibitors. Human Psychopharmacology 9, 1 12. Kennett GA, Lightowler S, De Biasi V, Stevens NC, Wood MD, Tulloch IF, Blackburn TP (1994). Effect of chronic administration of selective 5-hydroxytryptamine and noradrenaline uptake inhibitors on a putative index of 5-HT C/ B function. Neuropharmacology 33, 1581 1588. Kennett GA, Whitton P, Shah K, Curzon G (1989). Anxiogenic-like effects of mccp and TFMPP in animal models are opposed by 5-HT C agonists. European Journal of Pharmacology 164, 445 454. Kennett GA, Wood MD, Bright F, Trail B, Riley G, Holland V, Avenell KY, Stean T, Upton N, Bromidge S, Forbes IT, Brown AM, Middlemiss DN, Blackburn TP (1997). SB242084, a selective and brain penetrant 5-HT C antagonist. Neuropharmacology 36, 609 620. Lightowler S, Kennet GA, Williamson IJR, Blackburn TP, Tulloch IF (1994). Anxiolytic-like effect of paroxetine in a rat social interaction test. Pharmacology, Biochemistry and Behavior 49, 281 285. Lerer B, Gelfin Y, Gorfine M, Allolio B, Lesch KP, Newman ME (1999). 5-HT A function in normal subjects on clinical doses of fluoxetine: blunted temperature and hormone responses to ipsapirone challenge. Neuropsychopharmacology 20, 628 639. Lucki I, Ward HR, Frazer A (1989). Effect of 1-(m-

408 G. Bagdy et al. chlorophenyl)piperazine and 1-(m-trifluoromethylphenyl)- piperazine on locomotor activity. Journal of Pharmacology and Experimental Therapeutics 249, 155 164. Malagie I, Trillat AC, Douvier E, Anmella MC, Dessalles MC, Jacquot C, Gardier AM (1996). Regional differences in the effect of the combined treatment of WAY 100,635 and fluoxetine: an in vivo microdialysis study. Naunyn- Schmiedeberg s Archives of Pharmacology 354, 785 790. Marcoli M, Maura G, Tortarolo M, Raiteri M (1998). Trazodone is a potent agonist at 5-HT C s mediating inhibition the N-methyl-D-aspartate nitric oxide cyclic GMP pathway in rat cerebellum. Journal of Pharmacology and Experimental Therapeutics 286, 983 986. Matto V, Harro J, Allikmets L (1996). The effects of cholecystokinin A and B antagonists, devazepide and L 365260, on citalopram-induced decrease of exploratory behaviour in rat. Journal of Physiology and Pharmacology 47, 661 669. Millan MJ, Dekeyne A, Gobert A (1998). Serotonin (5-HT) C s tonically inhibit dopamine (DA) and noradrenaline (NA), but not 5-HT release in the frontal cortex in vivo. Neuropharmacology 37, 953 955. Mongeau R, Blier P, De Montigny C (1997). The serotonergic and noradrenergic systems of the hippocampus: their interactions and the effects of antidepressant treatments. Brain Research Reviews 23, 145 195. Murphy DL, Lesch KP, Aulakh CS, Pigott TA (1991). Serotonin-selective arylpiperazines with neuroendocrine, behavioural, temperature, and cardiovascular effects in humans. Pharmacological Reviews 43, 528 552. Nutt DJ, Glue P (1989). Clinical pharmacology of anxiolytics and antidepressants: a psychopharmacological perspective. Pharmacology and Therapeutics 44, 309 334. Overstreet DH, Knapp DJ, Breese GR (2000). Mechanisms involved in the acute anxiogenic effects of fluoxetine. In: Serotonin: From the Molecule to the Clinic. A Serotonin Club Brain Research Bulletin Conference, Abstracts (p. 109). New Orleans, USA. Palvimaki EP, Roth BL, Majasuo H, Laakso A, Kuoppamaki M, Syvalahti E, Hietala (1996). Interactions of selective serotonin reuptake inhibitors with the serotonin 5-HT C. Psychopharmacology 126, 234 240. Papp LA (1999). Anxiety Disorders: Somatic Treatment. In: Sadock BJ, Sadock VA (Eds.), Kaplan and Sadock s Comprehensive Textbook of Psychiatry (7th edn) (pp. 1490 1498). New York: Lippincott Williams and Wilkins. Pohl R, Yeragani VK, Balon R, Lycaki H (1988). The jitteriness syndrome in panic disorder patients treated with antidepressants. Journal of Clinical Psychiatry 49, 100 104. Porter RHP, Benwell KR, Lamb H, Malcolm CS, Allen NH, Revell DF, Adams DR, Sheardown MJ (1999). Functional characterization of agonists at recombinant human 5-HT B and 5-HT C s in CHO-K1 cells. British Journal of Pharmacology 128, 13 20. Quested DJ, Sargent PA, Cowen PJ (1997). SSRI treatment decreases prolactin and hyperthermic responses to mcpp. Psychopharmacology 133, 305 308. Szabo ST, de Montigny C, Blier P (2000). Progressive attenuation of the firing activity of locus coeruleus noradrenergic neurons by sustained administration of selective serotonin reuptake inhibitors. International Journal of Neuropsychopharmacology 3, 1 11. To CT, Anheuer ZE, Bagdy G (1999). Effects of acute and chronic fluoxetine treatment on CRH induced anxiety. NeuroReport 10, 553 555. To CT, Bagdy G (1999). Anxiogenic effect of central CCK administration is attenuated by chronic fluoxetine or ipsapirone treatment. Neuropharmacology 38, 279 282. Van Megen HJGM, Westenberg HGM, Den Boer JA, Slaap B, Scheepmakers A (1997). Effect of the selective serotonin reuptake inhibitor fluvoxamine on CCK-4 induced panic attacks. Psychopharmacology 129, 357 364. Westenberg HGM, Den Boer JA (1993). Serotonin in anxiety related disorders. In: Vanhouette PM, Saxena PR, Paoletti R, Brunello N, Jackson AS (Eds.), Serotonin, from Cell Biology to Pharmacology and Therapeutics (pp. 249 254). Dordrecht: Kluwer Academic Publishers. Zohar J, Insel TR, Zohar-Kadouch RC, Hill JL, Murphy DL (1988). Serotonergic responsivity in obsessive compulsive disorder: effects of chronic clomipramine treatment. Archives of General Psychiatry 45, 167 172.