Ondansetron and fluoxetine reduce sleep apnea in mice lacking monoamine oxidase A

Respiratory Physiology & Neurobiology 168 (2009) 230–238

Contents lists available at ScienceDirect

Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol

Ondansetron and fluoxetine reduce sleep apnea in mice lacking monoamine oxidase A C. Real a,∗ , I. Seif a , J. Adrien b,c , P. Escourrou a a b c

Univ Paris-Sud, EA 3544, Sérotonine et Neuropharmacologie, F-92296 Châtenay-Malabry cedex, France UPMC Univ Paris 06, UMR S677, F-75013 Paris, France INSERM U 975, CRicm-Hôpital Pitié Salpêtrière, F-75013 Paris, France

a r t i c l e Article history: Accepted 7 July 2009 Keywords: Sleep Apnea Breathing Serotonin MAOA Mice Ondansetron Fluoxetine

i n f o

a b s t r a c t Prospective clinical trials addressing the role of serotonin (5-HT) in sleep apnea have indicated that the 5-HT uptake inhibitor fluoxetine is beneficial to some patients with obstructive apnea, whereas the 5-HT3 receptor antagonist ondansetron seems of little value despite its efficacy in rat and dog models of sleep apnea (central and obstructive). Here, we examined the effect of these drugs in transgenic mice lacking monoamine oxidase A (Tg8), which exhibit ∼3-fold higher rates of central sleep apnea than their wild-type counterparts (C3H), linked to their enhanced 5-HT levels. Acute ondansetron (2 mg kg−1 , intraperitoneal), acute fluoxetine (16 mg kg−1 ) and 13-day chronic fluoxetine (1 or 16 mg kg−1 ) decreased by ∼80% the total (spontaneous and post-sigh) apnea index in Tg8 mice during non-rapid eye movement sleep, with no statistically significant effect on apnea in C3H mice. Our study shows that both drugs reduce the frequency of apneic episodes attributable to increased monoamine levels in this model of MAOA deficiency, and suggests that both may be effective in some patients with central sleep apneas. © 2009 Elsevier B.V. All rights reserved.

1. Introduction The prevalence and morbid consequences of sleep apnea have stimulated investigations of this breathing disorder over the past decades, but there is no widely effective pharmacotherapy for this condition (Smith et al., 2006). The pathogenesis of both the obstructive and central forms of sleep apnea is diverse, with a significant overlap between the two forms (White, 2005). Some studies are focused on putative serotonin (5-HT) mechanisms and serotonergic therapy (Veasey, 2003). As of yet, no large-scale clinical trials for serotonin drugs against sleep apnea have been done because small trials in patients with obstructive sleep apnea have shown no clear benefit (Hanzel et al., 1991; Kraiczi et al., 1999; Stradling et al., 2003; Carley et al., 2007). Rat and dog models have yielded more promising results (Radulovacki et al., 1998; Veasey et al., 1999; Carley and Radulovacki, 1999; Carley et al., 2001; Veasey et al., 2001; Carley and Radulovacki, 2005). Mouse gene targeting offers a new approach to the identification of susceptibility factors to sleep-related apnea (Nakamura et al., 2007; Real et al., 2007) and allows investigation of the efficacy of serotonin agents in genetically defined models of central sleep apnea, as shown with the serotonin synthesis inhibitor

∗ Corresponding author at: Sérotonine et Neuropharmacologie, EA 3544, Faculté de pharmacie, 5 rue Jean-Baptiste Clément, 92296 Châtenay-Malabry cedex, France. Tel.: +33 1 46 83 53 78; fax: +33 1 46 83 53 55. E-mail address: [email protected] (C. Real). 1569-9048/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.resp.2009.07.003

parachlorophenylalanine in a mouse strain lacking monoamine oxidase type A (MAOA) (Real et al., 2007). Here, we extend this work by investigating the effects of ondansetron and fluoxetine on sleep apnea in this same strain. Levels of extracellular 5-HT in the central nervous system appear to be lower during sleep than during wake (Portas et al., 1998; Jacobs and Fornal, 1999). Given that 5-HT can stimulate breathing (Hodges and Richerson, 2008), including through 5-HT2A and 5-HT2C receptors on respiratory motoneurons (Fenik and Veasey, 2003), it has long been hypothesized that insufficient 5-HT activity during sleep might predispose to apnea (Hanzel et al., 1991). Along this line, fluoxetine and paroxetine, two serotonin selective reuptake inhibitors (SSRIs), were shown to reduce the rate of sleep apnea during NREMS in patients with obstructive sleep apnea, but with a wide variability in individual responses (Hanzel et al., 1991; Berry et al., 1999; Kraiczi et al., 1999). In a study in rats, a low-dose regimen of fluoxetine had no consistent effect on central sleep apnea (Carley and Radulovacki, 2005). On the other hand, ondansetron, an inhibitor of peripheral 5-HT3 receptors, had no effect on obstructive sleep apnea in a 10patient study (0.15 mg kg−1 , per os) (Stradling et al., 2003) but was shown to reduce mixed (central and obstructive) sleep apnea in the English bulldog (2 mg kg−1 , per os) and central sleep apnea in the Sprague–Dawley rat (1 mg kg−1 , intraperitoneally) during rapid eye movement sleep (REMS) (Radulovacki et al., 1998; Veasey et al., 2001; Carley and Radulovacki, 2008). Interestingly, the effect of ondansetron on sleep apnea in this rat model appears to be poten-

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

231

Fig. 1. Following electrode implantation (Day 1, D1, implantation), mice were allowed 10 days of recovery. Then, after baseline recording (D11, baseline), each animal was studied after saline (D13, saline) and ondansetron (D15, ondan) injections. Next, mice received daily injections of fluoxetine and were tested on the 1st (D16, fluox1) and 13th (D28, fluox13) days of treatment. On the 14th day, fluoxetine and ondansetron were co-injected (D29, fluox14 + ondan).

tiated by chronic fluoxetine (Carley and Radulovacki, 2005). This suggests that activation of the 5-HT system by fluoxetine combines with inhibition of peripheral 5-HT3 receptors to reduce sleep apnea. However, other interpretations can be suggested because fluoxetine is known to have both serotonergic agonist and antagonist properties (Ni and Miledi, 1997; Eisensamer et al., 2003). Thus, the influence of fluoxetine on sleep breathing needs to be evaluated in genetic conditions of low, moderate, and high concentrations of 5-HT. In the present study, we tested the effects of fluoxetine and ondansetron on sleep apnea in mice. We have recently shown that mice lacking monoamine oxidase A (MAOA), an important enzyme in the degradation of 5-HT and norepinephrine (NE) (Cases et al., 1995; Chen et al., 2004), are a model of central sleep apnea (Real et al., 2007), whereas the prevalence of sleep apnea in humans lacking MAOA has not been examined (Tuinier et al., 1995; Vossler et al., 1996). Adult MAOA knock-out (KO) mice with a C3H wild-type background display increased levels of 5-HT and NE (Cases et al., 1995) and increased rates of apnea in NREMS and REMS (Real et al., 2007). Their sleep apnea indices can be normalized by reducing 5-HT synthesis with parachlorophenylalanine (Real et al., 2007), suggesting that excess endogenous 5-HT facilitates sleep apnea, without excluding putative cofactors such as increased norepinephrine levels and perinatal effects of increased monoamine levels (Burnet et al., 2001). Here, we found that acute ondansetron, as well as chronic fluoxetine, induced a significant decrease in NREMS apnea in MAOA KO mice. Thus, in the presence of high endogenous baseline levels of serotonin, central sleep apnea in humans may respond favorably to fluoxetine or ondansetron.

2. Methods 2.1. Animals Experiments were performed on mice of the C3H/HeOuJ strain (C3H, control mice) and its transgenic Tg(H2-IFN-␤)8 strain (Tg8, MAOA-deficient mice). Tg8 mice were obtained following the injection of an interferon-beta cassette into a one-cell C3H embryo, leading to the insertional deletion of two essential exons in the MAOA locus (Cases et al., 1995). C3H and Tg8 mice were bred and raised under standard housing conditions in the transgenic animal facility of Paris-Sud University at Châtenay-Malabry (France). We used 2- to 3-month-old C3H and Tg8 males (20–25 g body weight). Because of the frequent fighting initiated by Tg8 males (Cases et al., 1995), both mutant and wild-type males were housed in individual cages (20 cm × 20 cm × 30 cm) from the age of 6 weeks, maintained in a ventilated cabinet with a 12:12-h light-dark cycle (lights on at 7:00 AM) and a temperature of 23 ± 1 ◦ C, with food and water available ad libitum. All procedures involving animals and their care were conducted in conformity with institutional guidelines

in compliance with national and international laws and policies (Council Directive # 87-848 of 19 October 1987, French Ministry of Agriculture and Forestry). 2.2. Surgical procedure Mice were anesthetized with ketamine (100 mg kg−1 ) plus xylazine (20 mg kg−1 ) and enameled-insulated nichrome electrodes (150 ␮m in diameter) were implanted for polygraphic sleep monitoring, as previously described (Léna et al., 2004). Briefly, two electroencephalogram (EEG) electrodes were placed over the right cortex (2 mm lateral and 2 mm posterior to bregma) and over the cerebellum (2 mm posterior to lambda, at midline), two electrooculogram (EOG) electrodes were positioned subcutaneously on each side of the left eye, and two electromyogram (EMG) electrodes were inserted into the neck musculature. All electrodes were fixed to the skull with Super-Bond C&B (GACD, France) and the acrylic cement Dentalon Plus (GACD, France), and soldered to a connector also embedded in cement. The animals were allowed 10 days to recover from surgery before baseline recording (Fig. 1). 2.3. Sleep recording and scoring Recordings were made during the light phase from 10:00 to 18:00. The night before experiments, mice were placed in the barometric chamber (see below) and connected to the recording cables. To allow freedom of movement to the animal, a slip ring was placed at the connection of the electrodes to the lines outside of the plethysmograph. The animal had food and water ad libitum and ambient temperature was maintained at 24 ◦ C. The EEG, EMG, and EOG signals were amplified and sampled by an EMBLA system (Medcare, Reykjavik, Iceland) and fed into a computer at a sampling frequency of 200 Hz for EEG and 100 Hz for neck EMG and EOG. Sleep recordings were scored in 5-s epochs by visual inspection of EEG, EOG, and neck EMG signals (Somnologica2 software, Medcare, Reykjavik, Iceland) using standard criteria to define wake, NREMS and REMS (Léna et al., 2004; Real et al., 2007). Vigilance state amounts were expressed as percent of the 8-h recording period. 2.4. Measurement of ventilation by whole body plethysmography We used double-chamber whole-body plethysmography (Jacky, 1978). One chamber (500 ml, 10-cm internal diameter) was used as a barometric chamber where the mouse was placed, and reference pressure was measured in the other, each chamber being continuously flushed with room air at a rate of 700 ml min−1 and outlet gas being monitored for O2 and CO2 to ensure that the outlet fractions of O2 and CO2 remain close to ambient air values (Elisa Duo, Engström, Denmark). The plethysmograph was placed in a circulating water bath set at 24 ◦ C, in a ventilated room with a temperature

232

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

Fig. 2. Examples of the 2 types of apnea are shown on breathing tracings recorded during NREMS and REMS in untreated C3H and Tg8 mice (inspiration is up on the pressure trace). During NREMS, both strains of mice had spontaneous and post-sigh apneas, whereas they only had spontaneous apneas during REMS. Post-sigh apnea was typified by an interruption of flow preceded by a sigh within the previous 10 s, whereas spontaneous apnea was characterized by a sudden interruption of flow during quiet breathing. A 10-s reference baseline for a spontaneous apnea in a Tg8 mouse is underlined.

of 21 ± 1 ◦ C and a relative humidity of 55%. Plethysmographic signals were recorded as pressure changes between the chambers by use of a differential pressure transducer. Amplified signals were fed into a computer together with EEG, EOG, and EMG signals as described above. 2.5. Definition of apnea Apneas and sighs in NREMS and REMS were identified visually, as previously described (Real et al., 2007). Apnea was defined as a flat plethysmographic signal for at least twice the average respiratory cycle duration calculated over a 10-s period of regular breathing within the 20 s preceding the apnea (Nakamura et al., 2003; Real et al., 2007). Sighs were defined as a respiratory cycle with amplitude at least 50% higher than the average amplitude calculated over a 10-s period of regular breathing preceding the sigh (Léna et al., 2004; Popa et al., 2005; Real et al., 2007). Apneas were classified as post-sigh (i.e., with a sigh in the preceding 10 s) or spontaneous (no sigh in the preceding 10 s) (Léna et al., 2004; Popa et al., 2005; Real et al., 2007) (Fig. 2). The apnea index, defined as the number of apneas per hour, and mean apnea duration were calculated separately for NREMS and REMS. The sigh index was defined as the number of sighs per hour of each sleep stage. 2.6. Pharmacological treatments The 5-HT3 receptor antagonist ondansetron (Zophran) and the SSRI fluoxetine (Sigma–Aldrich, Lyon, France) were prepared in physiological saline immediately before use and injected intraperitoneally (i.p.). Mice were divided into two groups. Group 1 mice (5 C3H and 6 Tg8) received 1 mg free base kg−1 ondansetron (0.125 mg ml−1 ) and 1 mg salt kg−1 fluoxetine (0.125 mg ml−1 ) (Carley and Radulovacki, 2005). Group 2 mice (6 C3H and 7 Tg8) received 2 mg free base kg−1 ondansetron (0.25 mg ml−1 ) and 16 mg salt kg−1 fluoxetine (2 mg ml−1 ) (El Yacoubi et al., 2003). All mice underwent electrode implantation on Day 1 and received saline on Day 13, ondansetron on Day 15, fluoxetine from Day 16 to Day 28 (13-day chronic treatment), and both fluoxetine and ondansetron on Day 29 (Fig. 1). Injections of saline and drugs were made at about 09:45 AM (i.e., 15 min before recording sleep and breathing on days 11, 13, 15, 28 and 29). Previous studies with these mice had indicated that repeated injections of saline do not significantly affect sleep apnea (Real et al., 2007). 2.7. Statistical analysis Data are expressed as mean ± standard deviation (SD). Significance was tested by a two-way analysis of variance (ANOVA) on 8-h recordings, with genotype and treatment as factors. In case of significance (p < 0.05), the ANOVA was followed by an unpaired t-test or a Mann–Whitney rank sum test when appropriate to compare genotypes, and by a paired t-test or a Wilcoxon signed-rank test

when appropriate to assess the effect of drugs (SigmaStat software, SPSS, Chicago). 3. Results Mice were divided into two groups, each receiving a different dose of ondansetron and fluoxetine (see Section 2), with a sequential schedule of administration (Fig. 1). The treatment effects on vigilance states and sleep apnea are reported separately below. 3.1. Vigilance states At baseline on Day 11, Tg8 mutant mice had the same amounts of vigilance states as wild-type C3H mice: on average, 29% wake, 67% NREMS, and 4% REMS (Table 1, Groups 1 and 2 pooled). C3H and Tg8 mice had the same duration of wake, NREMS and REMS episodes: 5.3 ± 1.5, 10.8 ± 2.4, and 2.2 ± 0.9 min, respectively (mean ± SD of both genotypes). In Group 1 mice (5 C3H and 6 Tg8), saline injection had no effect on sleep (data not shown). Administration of ondansetron (1 mg kg−1 ) or fluoxetine (1 mg kg−1 ), or their co-administration, did not induce any significant change in sleep-wake patterns as compared with saline injection (data not shown). In Group 2 mice (6 C3H and 7 Tg8), saline injection decreased the amount of REMS in C3H mice, with no change in REMS episode duration (35% decrease compared to baseline, t = 2.7, p = 0.043). This induced a significant difference in REMS percentage between C3H and Tg8 mice (t = −3.4, p = 0.006) (Fig. 3). Acute administration of ondansetron in Group 2 mice (Fig. 1; ondan; 2 mg kg−1 , i.p.) decreased the amount of REMS compared to saline (F(1,25) = 4.9, p = 0.038), but did not modify that of wake and NREMS nor episode durations. This decrease in REMS was observed in both C3H (t = 7.0, p < 0.001) and Tg8 (t = 3.2, p = 0.018) mice (Fig. 3). Acute treatment with fluoxetine (Fig. 1; fluox1; 16 mg kg−1 , i.p.) decreased the amount of REMS (F(1,25) = 47.9, p < 0.001 compared to saline) in Tg8 mice only (t = 13.2, p < 0.001 vs. t = 2.4, p = 0.059 for C3H mice) (Fig. 3). Wake and NREMS were not affected. Chronic fluoxetine treatment: at the 13th day of treatment (Fig. 1; fluox13; 16 mg kg−1 per day, i.p.), REMS was decreased in both C3H and Tg8 mice (t = 3.8, p = 0.013, vs. t = 2.6, p = 0.041, respectively) with no change in wake or NREMS (Fig. 3). Table 1 MAOA-deficient (Tg8) mice show wild-type (C3H) amounts of vigilance states at baseline. Genotype

C3H Tg8

Vigilance state (%) Wake

NREMS

REMS

28.8 ± 4.2 28.5 ± 7.9

66.8 ± 4.0 67.0 ± 9.4

4.4 ± 0.9 4.5 ± 2.4

Amounts of vigilance states (mean ± SD) are expressed as percent of total recording time (8-h recording from 10:00 to 18:00 on Day 11). There is no significant difference in state amounts between C3H (n = 11) and Tg8 (n = 13) mice.

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

233

Table 2 MAOA-deficient (Tg8) mice show higher indices of sleep apnea than wild-type (C3H) mice before and after saline injection. Sleep

Treatment

Genotype

Apnea index

NREMS

Baseline

C3H Tg8

0.57 ± 0.45 2.94 ± 1.42###

0.24 ± 0.23 1.35 ± 0.99##

0.33 ± 0.39 1.59 ± 1.20##

32.0 ± 4.9 32.8 ± 7.8

Saline

C3H Tg8

0.39 ± 0.43 2.53 ± 1.45###

0.17 ± 0.19 1.02 ± 0.69###

0.23 ± 0.30 1.51 ± 1.01###

34.6 ± 7.7 32.4 ± 11.2

Baseline

C3H Tg8

2.03 ± 3.10 12.84 ± 12.97#

2.03 ± 3.10 12.84 ± 12.97#

0.0 ± 0.0 0.0 ± 0.0

2.2 ± 2.4 0.5 ± 0.7#

Saline

C3H Tg8

2.59 ± 4.58 9.25 ± 9.40#

2.59 ± 4.58 9.25 ± 9.40#

0.0 ± 0.0 0.0 ± 0.0

0.7 ± 1.6 0.6 ± 1.5

Total

REMS

Sigh index Spontaneous

Post-sigh

Apnea and sigh indices (mean ± SD) are the number of (post-sigh, spontaneous, or total) apneas and sighs per hour of NREMS or REMS (8-h recording from 10:00 to 18:00), before and after saline injection (on Day 11 and Day 13, respectively). The mice are the same as for Table 1. There are significant differences between C3H and Tg8 mice: # p < 0.05. ## p < 0.01. ### p < 0.001.

Finally, we examined whether a combined treatment with fluoxetine and ondansetron (Fig. 1; fluox14 + ondan) would have a different effect from each drug separately. On the 14th day of chronic fluoxetine treatment, co-administration of fluoxetine (16 mg kg−1 ) and ondansetron (2 mg kg−1 ) induced a decrease in REMS similar to that obtained with fluoxetine or ondansetron alone (F(1,25) = 28.4, p < 0.001 compared to saline) in both C3H (t = 8.8, p < 0.001) and Tg8 mice (t = 7.9, p < 0.001), while the amounts of wake and NREMS remained unchanged (Fig. 3). Interstrain comparisons indicate that acute ondansetron tended to have a lesser effect in Tg8 than in C3H mice (a REMS decrease of 59.8 ± 47.2% and 85.3 ± 20.1%, respectively; p = 0.245). In contrast, acute fluoxetine had a larger effect in Tg8 than in C3H mice (a REMS decrease of 87.9 ± 12.1% and 51.1 ± 42.3%, respectively; p = 0.049). 3.2. Sleep apnea At baseline on Day 11, Tg8 mice significantly differed from C3H mice by a greater total (spontaneous plus post-sigh) apnea index, which was increased 5-fold during NREMS (t = −3.8, p = 0.001) and 6-fold during REMS (t = −2.6, p = 0.031) (Table 2, Groups 1 and 2 pooled). In NREMS, this increase in total apnea index involved both spontaneous and post-sigh apneas (t = −3.4, p = 0.002 and t = −3.2, p = 0.004, respectively). In REMS, this increase was due to spontaneous apneas only (Table 2). The two mouse strains did not differ with respect to the sigh occurrence index in NREMS (Table 2) or apnea duration in NREMS and REMS (Table 3). Saline injection had no effect on apnea and sigh indices of Tg8 and C3H mice during REMS and NREMS, regardless of whether Groups 1 and 2 were pooled (as in Table 2) or analyzed separately. In Group 1, saline-treated Tg8 mice had 3.4-fold higher apnea indices than saline-treated C3H mice in NREMS and REMS (Fig. 4), whereas in Group 2, they had 10.5- and 3.6-fold higher apnea indices in NREMS (Fig. 5) and REMS (data not shown), respectively, both spon-

taneous and post-sigh apneas being significantly increased in each group. In Group 1 mice, which received the lower dose of ondansetron and fluoxetine (1 mg kg−1 ), there was no significant effect of all 4 drug treatments on the total apnea index in REMS in either strain (Fig. 4). However during NREMS, Tg8 mice exhibited a significant decrease of this index after chronic treatment with fluoxetine (Fig. 1; fluox13; t = 3.1, p = 0.027 compared to saline) as well as after co-administration of ondansetron on the last day of chronic fluoxetine (fluox14 + ondan; t = 3.0, p = 0.030). The decrease in apnea index produced by this combined treatment was similar to that obtained with chronic fluoxetine alone. Concerning Group 2, the REMS apnea index was not calculated because of major decreases in REMS amounts with high drug doses (see Fig. 3). Therefore, quantification of apneas was performed only during NREMS. The drug treatments did not change the sigh index or apnea duration during NREMS in C3H and Tg8 mice (data not shown). Acute administration of ondansetron (Fig. 1; ondan; 2 mg kg−1 ) had a significant effect on the total (F(1,25) = 25.9, p < 0.001), spontaneous (F(1,25) = 8.3, p = 0.009) and post-sigh (F(1,25) = 23.9, p < 0.001) apnea indices, as compared with saline. These apnea indices decreased in Tg8 mice only (total, t = 5.1, p = 0.002; spontaneous; t = 3.5, p = 0.013; and post-sigh, t = 5.0, p = 0.002) (Fig. 5). Acute treatment with fluoxetine (Fig. 1; fluox1; 16 mg kg−1 ) decreased the total apnea index (F(1,25) = 21.5, p < 0.001) in Tg8 mice only (t = 4.0, p = 0.007), essentially through a decrease in post-sigh apneas (t = 3.9, p = 0.008), while the decrease in spontaneous apneas was statistically borderline (t = 2.3, p = 0.060) (Fig. 5). These effects were the same at the 13th day of fluoxetine treatment (16 mg kg−1 per day): i.e., a decrease of total and post-sigh apnea indices in Tg8 mice only (Fig. 5). Chronic fluoxetine administration with co-administration of ondansetron on the last day (Fig. 1; fluox14 + ondan) had a significant effect on the total apnea index (F(1,25) = 22.8, p < 0.001) with a

Table 3 MAOA-deficient (Tg8) mice show the same duration of individual apneas as wild-type (C3H) mice at baseline. Sleep

Genotype

Apnea duration Total

Spontaneous

Post-sigh

NREMS

C3H Tg8

1.87 ± 0.67 2.13 ± 0.48

1.74 ± 0.78 2.14 ± 0.47

1.87 ± 0.76 2.07 ± 0.65

REMS

C3H Tg8

1.58 ± 0.33 1.81 ± 0.27

1.58 ± 0.33 1.81 ± 0.27

– –

Duration of (post-sigh, spontaneous, or total) apnea in NREMS and REMS is expressed in seconds (mean ± SD; 8-h recording from 10:00 to 18:00 on Day 11). The mice are the same as for Tables 1 and 2.

234

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

Fig. 4. Effects of the lower dose of ondansetron (1 mg kg−1 ) and fluoxetine (1 mg kg−1 ) on the total apnea index during NREMS and REMS in C3H mice (n = 5) and Tg8 mice (n = 6) (Group 1). Apnea indices are expressed as the number of apneas per hour of NREMS or REMS (mean ± SD; between 10:00 and 18:00). Chronic fluoxetine (Fluox13) and final co-treatment (Fluox14 + ondan), but not acute ondansetron (Ondan) and acute fluoxetine (Fluox1), induced a significant decrease in apnea index in Tg8 mice. All these treatments had no effect in C3H mice. Saline values in NREMS are 0.50 ± 0.60 for C3H mice and 1.70 ± 1.29 in Tg8 mice. Saline values in REMS are 0.78 ± 1.74 for C3H mice and 2.63 ± 2.71 for Tg8 mice. *p < 0.05, difference between saline and each drug treatment, paired t-test. # p < 0.05 and ## p < 0.01, difference between genotypes, unpaired t-test.

Fig. 3. Effects of the higher dose of ondansetron (2 mg kg−1 ) and fluoxetine (16 mg kg−1 ) on the amount of wake, NREMS or REMS in C3H (n = 6) and Tg8 mice (n = 7) (Group 2). Vigilance state amounts are expressed as percent of total recording time (mean ± SD; between 10:00 and 18:00). Acute ondansetron (Ondan), acute fluoxetine (Fluox1), chronic fluoxetine (Fluox13) and final co-treatment (Fluox14 + ondan) induced a significant decrease in REMS amounts in Tg8 mice, and in C3H mice alike except for acute fluoxetine. *p < 0.05 and ***p < 0.001, difference between saline and drug treatment, paired t-test. ## p < 0.01, difference between genotypes, unpaired t-test.

significant decrease in Tg8 mice only (t = 5.4, p = 0.002), concerning both spontaneous and post-sigh apneas (t = 2.4, p = 0.050 and t = 5.7, p = 0.001, respectively) (Fig. 5). The spontaneous apnea index after the treatment fluox14 + ondan was less affected than after the treatment ondan (0.38 ± 0.27 vs. 0.14 ± 0.10; paired t-test, p = 0.044), suggesting that chronic fluoxetine can limit the effect of acute ondansetron on spontaneous apneas in Tg8 mice. 4. Discussion This study shows that acute and chronic administration of fluoxetine, as well as acute administration of ondansetron alone or combined with chronic fluoxetine treatment, can reduce sleep apnea by several-fold in MAOA-deficient (Tg8) mice, a model of

central sleep apnea associated with high endogenous levels of serotonin. Specifically, at the lower dose evaluated, chronic fluoxetine and the combined treatment decreased the apnea index during NREMS, without changing NREMS and REMS amounts. At the higher dose, all four treatments decreased the apnea index during NREMS, and also reduced REMS. 4.1. Vigilance states Monoamines participate in sleep-wake regulation, and particularly, 5-HT has a tonic inhibitory influence on the expression of REMS (Pace-Schott and Hobson, 2002; Ursin, 2002), an effect mainly mediated through (post-synaptic) 5-HT1A receptors (Tissier et al., 1993; Monaca et al., 2003). However, sleep-wake patterns were similar in wild-type (C3H) and MAOA KO (Tg8) mice, even though 5-HT and norepinephrine concentrations in the brain of Tg8 mice are increased 1.5- and 2-fold, respectively (Cases et al., 1995; Real et al., 2007). This similarity in patterns is in agreement with previous data (Boutrel et al., 2000; Real et al., 2007) and may result from adaptive mechanisms in response to the increased levels of 5-HT and norepinephrine in Tg8 mice. For example, it has been shown that these mice exhibit down-regulation of the 5-HT transporter and partial desensitization of 5-HT1A autoreceptors, as well as reduced firing of 5-HT neurons (Evrard et al., 2002; Owesson et al., 2002).

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

235

might be related to the enhanced levels of peripheral (and central) 5-HT in the mutants, which would require more ondansetron to block peripheral (and central) 5-HT3 receptors. Alternatively, this might be due to a possible desensitization of 5-HT3 receptors in Tg8 mice. Fluoxetine at 16 but not 1 mg kg−1 decreased REMS amounts in our mice, in agreement with most studies in animals and humans that described a reduction of this sleep state for equivalent doses (Hendrickse et al., 1994; Maudhuit et al., 1994; Monaca et al., 2003), sometimes associated with an enhancement of NREMS and wake (Ursin, 2002). The mechanisms underlying the larger effect of acute fluoxetine (16 mg kg−1 ) in Tg8 compared to C3H mice (87.9% vs. 51% decrease, p = 0.049) could be related to the 4- to 10-fold higher SSRIinduced increase of 5-HT in the extracellular space compared to C3H mice (Evrard et al., 2002). This greater increase in extracellular 5-HT in Tg8 mice might lose amplitude or efficacy with chronic treatment, given that chronic fluoxetine tended to have a lesser effect than acute fluoxetine on REMS amounts in Tg8 mice (51.7% vs. 87.9%, p = 0.085). Finally, given that Tg8 mice exhibit increased levels of norepinephrine (Cases et al., 1995), the more pronounced effect of fluoxetine in mutants might also be due to its inhibition of norepinephrine uptake (Sánchez and Hyttel, 1999); however, acute fluoxetine did not appear to block 5-HT uptake by the norepinephrine transporter in vivo in Tg8 pups (Cases et al., 1998). 4.2. Sleep apnea

Fig. 5. Effects of the higher dose of ondansetron (2 mg kg−1 ) and fluoxetine (16 mg kg−1 ) on the (total, spontaneous, or post-sigh) apnea index during NREMS in C3H mice (n = 6) and Tg8 mice (n = 7) (Group 2). Apnea indices are expressed as the number of apneas per hour of NREMS (mean ± SD; between 10:00 and 18:00). Acute ondansetron (Ondan), acute fluoxetine (Fluox1), chronic fluoxetine (Fluox13), and final co-treatment (Fluox14 + ondan) induced a significant decrease in apnea index in Tg8 mice. All these treatments had no significant effect in C3H mice. Saline values for the total apnea index are 0.31 ± 0.26 for C3H and 3.25 ± 1.24 for Tg8 mice. *p < 0.05, **p < 0.01 and ***p < 0.001, difference between saline and drug treatment, paired t-test.

Here, we found that ondansetron at the dose of 2 but not 1 mg kg−1 significantly decreased REMS amounts without affecting NREMS or wakefulness (Fig. 3). A similar trend has been observed previously in Sprague–Dawley rats, where the dose of 1 mg kg−1 produced a statistically non-significant reduction of REMS amounts (Radulovacki et al., 1998). Ondansetron poorly crosses the blood–brain-barrier (BBB) (Simpson et al., 1992) and the 2 mg kg−1 dose may not suffice to inhibit central 5-HT3 receptors in mice (Scott et al., 2006). However, a significant effect of ondansetron in BBB-leaky areas such as the area postrema, cannot be excluded. Interestingly, 2 mg kg−1 of ondansetron tended to be less efficient at decreasing REMS in Tg8 than in C3H mice. This

Sleep apneas in C3H and Tg8 mice have a central origin (Real et al., 2007). Post-sigh apneas only occurred during NREMS, whereas spontaneous apneas occurred during both NREMS and REMS, which is in agreement with our previous report (Real et al., 2007) and with studies using other mouse strains (Nakamura et al., 2003; Popa et al., 2005). It is thought that post-sigh apneas result from hypocapnia following the augmented breath (Dempsey et al., 1996), and also from increased pulmonary stretch receptor activity (Nakamura et al., 2003). The origin of spontaneous apneas in wild-type mice is unknown (Nakamura et al., 2003). In the mouse strain 129/SvEvTac, a single intraperitoneal injection of the 5-HT2A receptor antagonist MDL100907 can specifically reduce the index of delayed post-sigh apneas (Popa et al., 2005). Of note, most post-sigh apneas in mouse studies, including ours, are delayed post-sigh apneas (Nakamura et al., 2003; Real et al., 2007) and have commonly been categorized as spontaneous apneas in rat studies (e.g., Saponjic et al., 2007). A current hypothesis is that peripheral endogenous 5-HT may predispose to apnea, whereas 5-HT activity in the central nervous system (CNS) may protect against apnea (Carley et al., 2007). In accordance with this view, the SSRIs fluoxetine and paroxetine, which increase 5-HT extracellular concentrations in the CNS (and at the periphery), have benefited some patients with sleep apnea in small trials (Hanzel et al., 1991; Kraiczi et al., 1999), whereas intraperitoneal injection of 5-HT in conscious Sprague–Dawley rats can produce a 3-fold increase in the apnea index during REMS (Carley and Radulovacki, 1999). The latter effect may be attributed to the activation of peripheral 5-HT receptors because peripherally administered 5-HT does not cross the BBB. Furthermore, peripheral administration of the 5-HT3 antagonist ondansetron, which poorly crosses the BBB (Simpson et al., 1992; Scott et al., 2006), fully blocked the stimulatory effect of exogenous 5-HT on sleep apnea expression in the Sprague–Dawley rat (Carley and Radulovacki, 1999). In the present study, Tg8 mice were found to exhibit more sleep apnea episodes than wild-type mice, which is analogous to our previous finding (Real et al., 2007). These mutant mice have elevated concentrations of 5-HT in the CNS (Cases et al., 1995; Lajard et al., 1999; Real et al., 2007), but also in the venous blood (2fold increase) and lungs, compared to wild-type mice (unpublished

236

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

data). Therefore, according to the hypothesis above, the elevated apnea index in Tg8 mice may be attributed to peripheral rather than central 5-HT, acting through 5-HT3 receptors. In C3H wild-type mice, we found no statistically significant effect of ondansetron on sleep apneas but only a trend towards a decrease of post-sigh apnea frequency during NREMS at the higher dose (2 mg kg−1 ). In a study in the Sprague–Dawley rat, ondansetron (1 mg kg−1 , i.p.) had a significant suppressant effect on spontaneous central apneas during REMS and less so during NREMS (Radulovacki et al., 1998; Carley and Radulovacki, 2008). The lack of significant effects in C3H mice may be related to possible differences in 5-HT activity between the two rodent strains, and may also be due to the low apnea indices in drug-free C3H mice compared to Sprague–Dawley rats (e.g., 6 and 7 of the 11 C3H mice, at baseline and after saline treatment, respectively, had no apneas during REMS). In Tg8 mice, ondansetron at the dose of 2 but not 1 mg kg−1 reduced the frequency of spontaneous and post-sigh apneas during NREMS. Since ondansetron poorly crosses the BBB (but may have access to BBB-leaky areas such as the area postrema), these data suggest that peripheral 5-HT3 receptors modulate both spontaneous and post-sigh apneas in Tg8 mice. At least two peripheral sensory pathways are involved in the control of breathing during sleep and appear to contain 5-HT3 receptors: (i) the carotid body receptors/glossopharyngeal petrosal ganglion pathway (Zhong et al., 1999), and (ii) the pulmonary receptors/vagal nodose ganglion pathway. The first pathway is excitatory for the medullary integrator of sensory inputs, but hypocapnia is believed to reduce its activation (Takakura et al., 2006). The second sensory pathway is inhibitory for the medullary integrator (Moreira et al., 2007a,b). Combination of pulmonary receptor activation and hypocapnia can trigger central sleep apneas in NREMS (Smith et al., 2003, 2007; Dempsey, 2005). 5-HT3 receptors in nodose and petrosal sensory neurons may amplify these signals, but not in the presence of ondansetron. With regard to fluoxetine, acute or chronic treatment induced no statistically significant effect on apnea expression in C3H mice. This resembles the results of one study in rats with 1 mg kg−1 i.p. fluoxetine (Carley and Radulovacki, 2005). In Tg8 mice, chronic but not acute administration of 1 mg kg−1 fluoxetine reduced the number of sleep apneas (Fig. 4). Such building-up of an effect would be accounted for by progressive adaptation, at possibly 5-HT2 receptors as demonstrated after repeated SSRI administration (Cadogan et al., 1993; Li et al., 1997), or by the fact that a single injection of 1 mg kg−1 fluoxetine may be insufficient to obtain therapeutic plasma levels of fluoxetine and norfluoxetine. Indeed, after a single injection of 16 mg kg−1 , a commonly used dose for acute effects, the decrease in sleep apnea was clear-cut in the mutants, and this effect persisted during chronic treatment (Fig. 5). The effect of fluoxetine on sleep apnea may be mainly mediated by the central 5-HT system. This system displays a large increase in extracellular 5-HT after SSRI treatment in adult Tg8 mice (Evrard et al., 2002) and in mouse pups lacking both MAOA and the serotonin transporter (Mössner et al., 2006). However, a peripheral role for fluoxetine cannot be ruled out since the diffusion and clearance of extracellular 5-HT in the peripheral nervous system may vary between the different endothelia (Linder et al., 2008). If it is postulated that the elevated sleep apnea index in Tg8 mice at baseline is linked to high extracellular concentrations of 5-HT (as suggested by the effect of parachlorophenylalanine; Real et al., 2007), then it is puzzling that a further augmentation of these levels by SSRI treatment induces a decrease in apnea index. This paradoxical result could be explained by regional 5-HT depletion (e.g., depletion from blood platelets), by desensitization of 5-HT receptors (e.g., 5-HT3 receptors), or by properties of fluoxetine other than inhibition of the serotonin transporter (SERT), includ-

ing antagonism at GIRK channels (Kobayashi et al., 2003) and at 5-HT2C receptors (Ni and Miledi, 1997; Sánchez and Hyttel, 1999; Wang et al., 1999). Other studies have shown that fluoxetine can act as a functional antagonist at human and rodent 5-HT3 receptors (Eisensamer et al., 2003, 2005), making it plausible that both fluoxetine and ondansetron reduce sleep apnea by antagonizing peripheral 5-HT3 receptors. We next investigated the effect of combined treatments with fluoxetine and ondansetron. In a previous study in rats, a non-effective dose of chronic fluoxetine (1 mg kg−1 ) augmented the (reducing) effect of acute ondansetron (1 mg kg−1 ) on sleep apnea (Carley and Radulovacki, 2005). Similarly, we found here that a non-effective dose of ondansetron (1 mg kg−1 ) tends to augment the (reducing) effect of chronic fluoxetine (1 mg kg−1 ) on sleep apnea (Fig. 4). However, at higher doses of both drugs, fluoxetine (16 mg kg−1 ) appeared to limit the effect of acute ondansetron (2 mg kg−1 ) on spontaneous apneas (Fig. 5). There are limitations that deserve to be mentioned when interpreting these findings. We have previously determined that repeated exposure to i.p. injections (Real et al., 2007) or to the barometric chamber (unpublished observations), or to combination of both, does not reduce sleep apnea in Tg8 mice in the absence of drug. Yet, in the present study, the animal’s response to fluoxetine may have been affected by prior treatment with ondansetron. Another limitation of our study is the low apnea index of C3H mice at baseline and after saline treatment. This prevents accurate determination of whether drug treatment can reduce apnea indices in wild-type context. However, an increase in apnea indices could have been detected, which provided a rationale for examining these paradigms in C3H wild-type mice. It would be interesting to know how different engineered strains respond to ondansetron and fluoxetine, for example in orexin knockout mice because they express enhanced levels of spontaneous sleep apnea (Nakamura et al., 2007), and 5-HT3 KO and SERT KOs in order to elucidate whether ondansetron and fluoxetine act through 5-HT3 and SERT, respectively. The question arises whether a reduction in post-sigh apnea index with fluoxetine and ondansetron presents a health risk. Indeed, the frequency of sighs followed by an apnea was found to be reduced in a group of 11 infants who later died of the sudden infant death syndrome (SIDS) (Kahn et al., 1988). A subsequent study, performed on 18 infants who died of SIDS, failed to confirm this association: the frequency of obstructive and mixed apneas per hour sleep is greater in future SIDS victims than in control infants, but the frequency of isolated and pre-apneic sighs is not significantly different (Franco et al., 2003). In any case, in infants, fluoxetine should not be used, while ondansetron might be used for only short periods (Freedman et al., 2006), but it may help in the diagnosis and emergency treatment of sleep apnea. In adults, these two drugs might be useful, for example, in patients with central sleep apnea in which inhaled CO2 was effective (Badr et al., 1994; Xie et al., 1997). Ondansetron may also be tested in patients with obstructive apnea responding to fluoxetine (Hanzel et al., 1991). Acknowledgments This study was supported by the CNRS, INSERM, French Ministry of Research, and University of Paris-Sud. We thank Pauline Robert and Valérie Domergue for dedicated animal care. References Badr, M.S., Grossman, J.E., Weber, S.A., 1994. Treatment of refractory sleep apnea with supplemental carbon dioxide. Am. J. Respir. Crit. Care Med. 150, 561–564. Berry, R.B., Yamaura, E.M., Gill, K., Reist, C., 1999. Acute effects of paroxetine on genioglossus activity in obstructive sleep apnea. Sleep 22, 1087–1092.

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238 Boutrel, B., Evrard, A., Seif, I., De Maeyer, E., Hamon, M., Adrien, J., 2000. Circadian rhythms of sleep-wakefulness, motor activity and body temperature in mutant mice lacking monoamine oxidase A. J. Sleep Res. 9, 24. Burnet, H., Bévengut, M., Chakri, F., Bou-Flores, C., Coulon, P., Gaytán, S., Pásaro, R., Hilaire, G., 2001. Altered respiratory activity and respiratory regulations in adult monoamine oxidase A-deficient mice. J. Neurosci. 21, 5212–5221. Cadogan, A.K., Marsden, C.A., Tulloch, I., Kendall, D.A., 1993. Evidence that chronic administration of paroxetine or fluoxetine enhances 5-HT2 receptor function in the brain of the guinea pig. Neuropharmacology 32, 249–256. Carley, D.W., Radulovacki, M., 1999. Role of peripheral serotonin in the regulation of central sleep apneas in rats. Chest 115, 1397–1401. Carley, D.W., Depoortere, H., Radulovacki, M., 2001. R-zacopride a 5-HT3 antagonist/5-HT4 agonist, reduces sleep apneas in rats. Pharmacol. Biochem. Behav. 69, 283–289. Carley, D.W., Radulovacki, M., 2005. Apnea suppression by ondansetron and fluoxetine alone and combined. Sleep 28 (Abstract Supplement), 12. Carley, D.W., Olopade, C., Ruigt, G.S., Radulovacki, M., 2007. Efficacy of mirtazapine in obstructive sleep apnea syndrome. Sleep 30, 35–41. Carley, D.W., Radulovacki, M., 2008. Pharmacology of vagal afferent influences on disordered breathing during sleep. Respir. Physiol. Neurobiol. 164, 197–203. Cases, O., Seif, I., Grimsby, J., Gaspar, P., Chen, K., Pournin, S., Müller, U., Aguet, M., Babinet, C., Shih, J.C., De Maeyer, E., 1995. Aggressive behavior and altered amounts of brain serotonin and norepinephrine in mice lacking MAOA. Science 268, 1763–1766. Cases, O., Lebrand, C., Giros, B., Vitalis, T., De Maeyer, E., Caron, M.G., Price, D.J., Gaspar, P., Seif, I., 1998. Plasma membrane transporters of serotonin, dopamine, and norepinephrine mediate serotonin accumulation in atypical locations in the developing brain of monoamine oxidase A knock-outs. J. Neurosci. 18, 6914–6927. Chen, K., Holschneider, D.P., Wu, W., Rebrin, I., Shih, J.C., 2004. A spontaneous point mutation produces monoamine oxidase A/B knock-out mice with greatly elevated monoamines and anxiety-like behavior. J. Biol. Chem. 279, 39645–39652. Dempsey, J.A., Smith, C.A., Harms, C.A., Chow, C., Saupe, K.W., 1996. Sleep-induced breathing instability. Sleep 19, 236–247. Dempsey, J.A., 2005. Crossing the apnoeic threshold: causes and consequences. Exp. Physiol. 90, 13–24. Eisensamer, B., Rammes, G., Gimpl, G., Shapa, M., Ferrari, U., Hapfelmeier, G., Bondy, B., Parsons, C., Gilling, K., Zieglgänsberger, W., Holsboer, F., Rupprecht, R., 2003. Antidepressants are functional antagonists at the serotonin type 3 (5-HT3) receptor. Mol. Psychiatry 8, 994–1007. Eisensamer, B., Uhr, M., Meyr, S., Gimpl, G., Deiml, T., Rammes, G., Lambert, J.J., Zieglgänsberger, W., Holsboer, F., Rupprecht, R., 2005. Antidepressants and antipsychotic drugs colocalize with 5-HT3 receptors in raft-like domains. J. Neurosci. 25, 10198–10206. El Yacoubi, M., Bouali, S., Popa, D., Naudon, L., Leroux-Nicollet, I., Hamon, M., Costentin, J., Adrien, J., Vaugeois, J.M., 2003. Behavioral, neurochemical, and electrophysiological characterization of a genetic mouse model of depression. Proc. Natl. Acad. Sci. U.S.A. 100, 6227–6232. Evrard, A., Malagié, I., Laporte, A.M., Boni, C., Hanoun, N., Trillat, A.C., Seif, I., De Maeyer, E., Gardier, A., Hamon, M., Adrien, J., 2002. Altered regulation of the 5-HT system in the brain of MAO-A knock-out mice. Eur. J. Neurosci. 15, 841–851. Fenik, P., Veasey, S.C., 2003. Pharmacological characterization of serotonergic receptor activity in the hypoglossal nucleus. Am. J. Respir. Crit. Care Med. 167, 563–569. Franco, P., Verheulpen, D., Valente, F., Kelmanson, I., de Broca, A., Scaillet, S., Groswasser, J., Kahn, A., 2003. Autonomic responses to sighs in healthy infants and in victims of sudden infant death. Sleep Med. 4, 569–577. Freedman, S.B., Adler, M., Seshadri, R., Powell, E.C., 2006. Oral ondansetron for gastroenteritis in a pediatric emergency department. N. Engl. J. Med. 354, 1698–1705. Hanzel, D.A., Proia, N.G., Hudgel, D.W., 1991. Response of obstructive sleep apnea to fluoxetine and protriptyline. Chest 100, 416–421. Hendrickse, W.A., Roffwarg, H.P., Grannemann, B.D., Orsulak, P.J., Armitage, R., Cain, J.W., Battaglia, J., Debus, J.R., Rush, A.J., 1994. The effects of fluoxetine on the polysomnogram of depressed outpatients: a pilot study. Neuropsychopharmacology 10, 85–91. Hodges, M.R., Richerson, G.B., 2008. Contributions of 5-HT neurons to respiratory control: neuromodulatory and trophic effects. Respir. Physiol. Neurobiol. 164, 222–232. Jacobs, B.L., Fornal, C.A., 1999. Activity of serotonergic neurons in behaving animals. Neuropsychopharmacology 21, 9S–15S. Jacky, J.P., 1978. A plethysmograph for long-term measurements of ventilation in unrestrained animals. J. Appl. Physiol. 45, 644–647. Kahn, A., Blum, D., Rebuffat, E., Sottiaux, M., Levitt, J., Bochner, A., Alexander, M., Grosswasser, J., Muller, M.F., 1988. Polysomnographic studies of infants who subsequently died of sudden infant death syndrome. Pediatrics 82, 721–727. Kobayashi, T., Washiyama, K., Ikeda, K., 2003. Inhibition of G protein-activated inwardly rectifying K+ channels by fluoxetine (Prozac). Br. J. Pharmacol. 138, 1119–1128. Kraiczi, H., Hedner, J., Dalhöf, P., Ejnell, H., Carlson, J., 1999. Effect of serotonin uptake inhibition on breathing during sleep and daytime symptoms in obstructive sleep apnea. Sleep 22, 61–67. Lajard, A.M., Bou, C., Monteau, R., Hilaire, G., 1999. Serotonin levels are abnormally elevated in the fetus of the monoamine oxidase-A-deficient transgenic mouse. Neurosci. Lett. 261, 41–44.

237

Léna, C., Popa, D., Grailhe, R., Escourrou, P., Changeux, J.P., Adrien, J., 2004. Beta2containing nicotinic receptors contribute to the organization of sleep and regulate putative micro-arousals in mice. J. Neurosci. 24, 5711–5718. Li, Q., Muma, N.A., Battaglia, G., Van de Kar, L.D., 1997. Fluoxetine gradually increases [125I]DOI-labelled 5-HT2A/2C receptors in the hypothalamus without changing the levels of Gq- and G11-proteins. Brain Res. 775, 225–228. Linder, A.E., Ni, W., Szasz, T., Burnett, R., Diaz, J., Geddes, T.J., Kuhn, D.M., Watts, S.W., 2008. A serotonergic system in veins: serotonin transporter-independent uptake. J. Pharmacol. Exp. Ther. 325, 714–722. Maudhuit, C., Jolas, T., Lainey, E., Hamon, M., Adrien, J., 1994. Effects of acute and chronic treatment with amoxapine and cericlamine on the sleep-wakefulness cycle in the rat. Neuropharmacology 33, 1017–1025. Monaca, C., Boutrel, B., Hen, R., Hamon, M., Adrien, J., 2003. 5-HT1A/1B receptormediated effects of the selective serotonin reuptake inhibitor, citalopram, on sleep: studies in 5-HT1A and 5-HT1B knockout mice. Neuropsychopharmacology 28, 850–856. Moreira, T.S., Takakura, A.C., Colombari, E., West, G.H., Guyenet, P.G., 2007a. Inhibitory input from slowly adapting lung stretch receptors to retrotrapezoid nucleus chemoreceptors. J. Physiol. 580, 285–300. Moreira, T.S., Takakura, A.C., Colombari, E., Guyenet, P.G., 2007b. Activation of 5hydroxytryptamine type 3 receptor-expressing C-fiber vagal afferents inhibits retrotrapezoid nucleus chemoreceptors in rats. J. Neurophysiol. 98, 3627–3637. Mössner, R., Simantov, R., Marx, A., Lesch, K.P., Seif, I., 2006. Aberrant accumulation of serotonin in dopaminergic neurons. Neurosci. Lett. 401, 49–54. Nakamura, A., Fukuda, Y., Kuwaki, T., 2003. Sleep apnea and effect of chemostimulation of breathing instability in mice. J. Appl. Physiol. 94, 525–532. Nakamura, A., Zhang, W., Yanagisawa, M., Fukuda, Y., Kuwaki, T., 2007. Vigilance state-dependent attenuation of hypercapnic chemoreflex and exaggerated sleep apnea in orexin knockout mice. J. Appl. Physiol. 102, 241–248. Ni, Y.G., Miledi, R., 1997. Blockage of 5HT2C serotonin receptors by fluoxetine (Prozac). Proc. Natl. Acad. Sci. U.S.A. 94, 2036–2040. Owesson, C.A., Hopwood, S.E., Callado, L.F., Seif, I., McLaughlin, D.P., Stamford, J.A., 2002. Altered presynaptic function in monoaminergic neurons of monoamine oxidase-A knockout mice. Eur. J. Neurosci. 15, 1516–1522. Pace-Schott, E.F., Hobson, J.A., 2002. The neurobiology of sleep: genetics, cellular physiology and subcortical networks. Nat. Rev. Neurosci. 3, 591–605. Popa, D., Léna, C., Fabre, V., Prenat, C., Gingrich, J., Escourrou, P., Hamon, M., Adrien, J., 2005. Contribution of 5-HT2 receptor subtypes to sleep-wakefulness and respiratory control and functional adaptations in knock-out mice lacking 5-HT2A receptors. J. Neurosci. 25, 11231–11238. Portas, C.M., Bjorvatn, B., Fagerland, S., Gronli, J., Mundal, V., Sorensen, E., Ursin, R., 1998. On-line detection of extracellular levels of serotonin in dorsal raphe nucleus and frontal cortex over the sleep/wake cycle in the freely moving rat. Neuroscience 83, 807–814. Radulovacki, M., Trbovic, S.M., Carley, D.W., 1998. Serotonin 5-HT3-receptor antagonist GR 38032F suppresses sleep apneas in rats. Sleep 21, 131–136, Erratum for Figure 3. Real, C., Popa, D., Seif, I., Callebert, J., Launay, J.M., Adrien, J., Escourrou, P., 2007. Sleep apneas are increased in mice lacking monoamine oxidase A. Sleep 30, 1295–1302. Sánchez, C., Hyttel, J., 1999. Comparison of the effects of antidepressants and their metabolites on reuptake of biogenic amines and on receptor binding. Cell. Mol. Neurobiol. 19, 467–489. Saponjic, J., Radulovacki, M., Carley, D.W., 2007. Monoaminergic system lesions increase post-sigh respiratory pattern disturbance during sleep in rats. Physiol. Behav. 90, 1–10. Scott, J.A., Wood, M., Flood, P., 2006. The pronociceptive effect of ondansetron in the setting of P-glycoprotein inhibition. Anesth. Analg. 103, 742–746. Simpson, K.H., Murphy, P., Colthup, P.V., Whelan, P., 1992. Concentration of ondansetron in cerebrospinal fluid following oral dosing in volunteers. Psychopharmacology (Berl.) 109, 497–498. Smith, C.A., Nakayama, H., Dempsey, J.A., 2003. The essential role of carotid body chemoreceptors in sleep apnea. Can. J. Physiol. Pharmacol. 81, 774–779. Smith, I., Lasserson, T.J., Wright, J., 2006. Drug therapy for obstructive sleep apnoea in adults. Cochrane Database Syst. Rev. 19, CD003002. Smith, C.A., Chenuel, B.J., Henderson, K.S., Dempsey, J.A., 2007. The apneic threshold during non-REM sleep in dogs: sensitivity of carotid body vs. central chemoreceptors. J. Appl. Physiol. 103, 578–586. Stradling, J., Smith, D., Radulovacki, M., Carley, D.W., 2003. Effect of ondansetron on moderate obstructive sleep apnoea, a single night, placebo-controlled trial. J. Sleep Res. 12, 169–170. Takakura, A.C., Moreira, T.S., Colombari, E., West, G.H., Stornetta, R.L., Guyenet, P.G., 2006. Peripheral chemoreceptor inputs to retrotrapezoid nucleus (RTN) CO2sensitive neurons in rats. J. Physiol. 572, 503–523. Tissier, M.H., Lainey, E., Fattaccini, C.M., Hamon, M., Adrien, J., 1993. Effects of ipsapirone, a 5-HT1A agonist, on sleep/wakefulness cycles: probable postsynaptic action. J. Sleep Res. 2, 103–109. Tuinier, S., Verhoeven, V.M.A., Scherders, M.J.W.T., Fekkes, D., Pepplinkhuizen, L., 1995. Neuropsychiatric and biochemical characteristics of X-linked MAO-A deficiency syndrome. A single case intervention study. New Trends Exp. Clin. Psychiatry 11, 99–107. Ursin, R., 2002. Serotonin and sleep. Sleep Med. Rev. 6, 55–67. Veasey, S.C., Fenik, P., Panckeri, K., Pack, A.I., Hendricks, J.C., 1999. The effects of trazodone with l-tryptophan on sleep-disordered breathing in the English bulldog. Am. J. Respir. Crit. Care Med. 160, 1659–1667.

238

C. Real et al. / Respiratory Physiology & Neurobiology 168 (2009) 230–238

Veasey, S.C., Chachkes, J., Fenik, P., Hendricks, J.C., 2001. The effects of ondansetron on sleep-disordered breathing in the English bulldog. Sleep 24, 155–160. Veasey, S.C., 2003. Serotonin agonists and antagonists in obstructive sleep apnea: therapeutic potential. Am. J. Respir. Med. 2, 21–29. Vossler, D.G., Wyler, A.R., Wilkus, R.J., Gardner-Walker, G., Vlcek, B.W., 1996. Cataplexy and monoamine oxidase deficiency in Norrie Disease. Neurology 46, 1258–1261. Wang, D., Post, M., Cutz, E., 1999. Expression of serotonin receptor 2c in rat type II pneumocytes. Am. J. Respir. Cell Mol. Biol. 20, 1175–1180.

White, D.P., 2005. Pathogenesis of obstructive and central sleep apnea. Am. J. Respir. Crit. Care Med. 172, 1363–1370. Xie, A., Rankin, F., Rutherford, R., Bradley, T.D., 1997. Effects of inhaled CO2 and added dead space on idiopathic central sleep apnea. J. Appl. Physiol. 82, 918–926. Zhong, H., Zhang, M., Nurse, C.A., 1999. Electrophysiological characterization of 5HT receptors on rat petrosal neurons in dissociated cell culture. Brain Res. 816, 544–553.