Chronic buspirone treatment normalizes regional serotonin synthesis in the olfactory bulbectomized rat brain: An autoradiographic study

Brain Research Bulletin 69 (2006) 101–108

Chronic buspirone treatment normalizes regional serotonin synthesis in the olfactory bulbectomized rat brain: An autoradiographic study Arata Watanabe 1 , Shu Hasegawa 2 , Kyoko Nishi 3 , Khnah Q. Nguyen, Mirko Diksic ∗ Cone Neurological Research Laboratory, Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Que., Canada H3A 2B4 Received 16 September 2005; received in revised form 12 November 2005; accepted 17 November 2005 Available online 9 December 2005

Abstract The effects of chronic buspirone treatments, administered by minipump at doses of 10 and 20 mg/(kg day) for 14 days, on brain 5-HT synthesis in olfactory bulbectomized (OBX) rats were evaluated. The ␣-[14 C]methyl-l-tryptophan autoradiographic method was used. We compared the synthesis in the buspirone treated OBX rats (administered either 10 mg/(kg day) (OBX-10) or 20 mg/(kg day) (OBX-20)) to that of the saline treated OBX rats (OBX-SAL), and the sham operated rats (SHX) treated with saline. In addition, OBX-10 rats were compared to SHX rats treated with 10 mg/(kg day) (SHX-10) of buspirone. All treatments were carried out for 14 days. Adult Sprague–Dawley rats were used. Two weeks following the OBX or SHX procedures, the rats were assigned to the OBX-10, OBX-20, OBX-SAL, SHX-10, or SHX-SAL groups, respectively. The 5-HT synthesis rates R (pmol/(g/min)) were calculated from the trapping constant of ␣-[14 C]MTrp (K*; ml/(g min)) and the plasma concentration of the plasma non-protein-bound tryptophan (Cp; pmol/ml) using the lumped constant (LC) measured previously in the rat brain. There was no significant difference in the plasma free or total tryptophan among these groups. The overall synthesis in the OBX-10 group was not statistically different from the OBX-SAL group, but it was different from the OBX-20 and SHX-SAL groups. The OBX-20 rats had an overall significant reduction in 5-HT synthesis, when compared to the OBX-SAL group, but did not differ from the SHX-SAL group, which did not differ from the SHX-10 group. These results suggest that 10 mg/(kg day) of buspirone for 14 days in the OBX rats did not produce a significant alteration in 5-HT synthesis, but 20 mg/(kg day) for 14 days resulted in an overall significant reduction in brain 5-HT synthesis. The latter treatment brought the synthesis to the level found in the sham operated rats, i.e., a normal level. These results suggest that normalization (reduction to the level found in the SHX-SAL rats) of 5-HT synthesis in the OBX requires a greater dose of buspirone (20 mg/(kg day)) than that needed to produce a desensitization of the 5-HT1A receptors in the sham operated rats (10 mg/(kg day)). This probably indicates that 5-HT1A receptors have different functionality in the OBX rats than that found in the intact or sham operated rats. Furthermore, our results support the hypothesis that 5-HT1A receptors mediate the antidepressant-like effect of 5-HT1A agonists, as the chronic 5-HT1A agonist treatment in the depression model known to be sensitive to antidepressants resulted in the normalization of 5-HT synthesis. © 2005 Elsevier Inc. All rights reserved. Keywords: Serotonin; Synthesis rate; Olfactory bulbectomy; Buspirone; ␣-[14 C]methyl-l-tryptophan; Autoradiography; Rat

1. Introduction

∗

Corresponding author. Tel.: +1 514 398 8526. E-mail address: [email protected] (M. Diksic). 1 Permanent address: Department of Neurosurgery, University of Yamanashi, 1110 Shimokato Tamaho-cho, Nakakoma-gun, Yamanashi 409-3898, Japan. 2 Permanent address: Department of Neurosurgery, Kumamoto University School of Medicine, 1-1-1 Honjo, Kumamoto, Kumamoto 860-8556 Japan. 3 Permanent address: Department of Neurological Surgery, School of Medicine, The University of Tokushima, 3-18-15, Kuramoto-cho, Tokushima 770-8505, Japan. 0361-9230/$ – see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2005.11.008

Buspirone is a 5-HT1A receptor partial agonist at postsynaptic receptors and a full agonist at pre-synaptic receptors [6,15,34]. It also displays antagonistic properties at the D2 receptors [24,33] and is metabolized to the ␣2-adrenergic receptor antagonist, 1-(2-pyrimidinyl-piperazine) [3,11]. It has been reported that buspirone has anxiolytic and possibly antidepressant effects [18,19]. The olfactory bulbectomized (OBX) rat model is a model of a form of agitated depression that, in many respects, resembles agitated human depression [17]. Many full and partial agonists of 5-HT1A sites are beneficial in the treatment of depressive symptoms in animal models of

102

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108

depression [19]. The antidepressant effects of buspirone have also been documented in double-blind, placebo-controlled clinical trials in humans [12]. In addition, it has been suggested that buspirone enhances the effect of SSRIs in rodents [25], and may enhance the effect of SSRIs, particularly in refractory patients [2,10]. Okazawa et al. [23] reported that a single dose of buspirone induces a significant decrease of 5-HT synthesis, while a 5-HT1A antagonist, WAY-100635, increases 5-HT synthesis throughout the brain of Sprague–Dawley rats [30]. This suggests the control of 5-HT synthesis via 5-HT1A receptors/autoreceptors. However, chronic treatment with buspirone did not induce significant differences in 5-HT synthesis in the rat brain of normal rats. It was suggested that the unaltered 5-HT synthesis rates in the chronic treatment study reflects a normalization of this parameter due to a desensitization of the 5-HT1A autoreceptors on the 5-HT cell bodies as well as the heteroceptors in the terminal area. It has also been shown that antidepressant action (e.g., behaviour changes) is different in normal rats than in the OBX model of depression [12], necessitating a study of antidepressant effects in animal models of depression. The olfactory bulbectomized rat model is widely accepted as a model of depression that bears many similarities to the agitated form of human depression [16,38,39]. The removal of the olfactory bulb bilaterally in rats results in changes in behavioural, endocrine, neurochemical and neuronal connections. Many of these changes are reversed by chronic, not acute, administration of antidepressants [12,38]. We reported widespread, but not uniform, increases of regional 5-HT synthesis throughout the rat brain following olfactory bulbectomy [14,36]. This elevation in 5-HT synthesis was brought to the levels found in the sham operated rats (normal levels) by a 14-day treatment with citalopram [14]. It was suggested that this large and widely spread increase in synthesis results in the creation of non-physiological circuitry which could [29], at least in part, be responsible for behavioural and neurochemical changes observed in the OBX rats [12]. Dysfunction of the brain serotonergic system has been shown to relate to behaviour and mood related disorders (e.g., [28]). To obtain a better understanding of the biochemical processes that underlie the neurochemical and behavioural changes in the OBX rats and the corresponding alleviation with drugs possessing antidepressant properties (e.g., citalopram normalized 5-HT synthesis [14]), we determined that it was important to understand how buspirone, a drug with anxiolytic and possibly antidepressant properties, affects 5-HT synthesis in this rat model of depression. The importance of this understanding is highlighted by the observations that 5-HT synthesis could be modulated by drugs acting through 5-HT1A receptors [23,30]. Further, it is important to know if a normalization (return to the level of an intact rat) of 5-HT synthesis can be achieved by buspirone treatment in the OBX rats. We report, in this study, the effects of chronic buspirone treatments on brain regional 5-HT synthesis in OBX rats. The synthesis in the treated rats was compared both to the sham and OBX groups treated with saline or buspirone. Two different doses of buspirone were used in the treatment of the OBX rats because the lower dose did not nor-

malize 5-HT synthesis. The sham rats were injected with one dose of buspirone or saline. The higher dose of buspirone was not used in the sham operated rats because it was expected that the receptors in the sham operated rats would be desensitized with 10 mg/(kg day) of buspirone, as was the case with the intact animals [23]. Because of this, the existence of a sham operated group treated with a higher dose of buspirone would not add any additional information to our study. The measurements were performed using the ␣-[14 C]methyl-l-tryptophan autoradiographic method, which permits measurements in many brain structures with a resolution of approximately 0.1 mm. This method has been widely validated through its ability to show the effects of drugs known to influence 5-HT synthesis (reviewed in [7,9]). 2. Materials and methods 2.1. Animals Male 180–200 g Sprague–Dawley rats were used for this experiment (Charles River Canada, St. Constant, Quebec, Canada). They were housed two per cage at the animal facility of the Montreal Neurological Institute for at least 3 days before the OBX surgery. Under 1% halothane anesthesia, a cranial window, the posterior edge of which was placed 5.2 mm from the Bergma, was created in the frontal bone [14,32,36]. The olfactory bulbs were cut and aspirated. Care was taken not to cause damage to the frontal cortex. The sham operations were performed in the same manner, but the bulbs were left intact. To prevent blood loss from the cranial window, a hemostatic sponge was filled to the dead space. Following the surgery, the rats were given buprenorphine (0.03 mg/kg; S.C.) to reduce any pain with a close observation until they recovered from anesthesia. The rats, housed two per cage, were then given 2 weeks to recover from the surgical procedure and to develop the OBX syndrome. When the brains were removed, the rats with any residual bulb and/or any frontal cortex damage were excluded from the final data analysis. In the experiments reported here, one rat was removed from each of the OBX groups because of surgery related damage. Excluded animals are not included in the number of rats per group reported in Table. All animal use procedures were in strict accordance with the Canadian Council on Animal Care guidelines, and were approved by the Animal Care Committee of McGill University.

2.2. Drug Two weeks following the olfactory bulbectomy or sham operation, either 10 or 20 mg/(kg day) of buspirone (obtained from Sigma–Aldrich Canada Ltd., Oakville, ont.) was administered subcutaneously for 14 days with a model 2ML2 minipump, which delivers solution at a constant rate of 0.12 mL/day. The buspirone was dissolved in saline. Under 1% halothane anesthesia, the minipumps were implanted. The rats were returned to their cages and housed in pairs for 2 weeks. The control animals, both in the sham and OBX groups, were administered 2 mL of saline by the minipump. There were two groups of the sham operated rats. One (N = 11) was administered saline (SHX-SAL) and the second (N = 8) was administered 10 mg/(kg day) of buspirone for 14 days (SHX-10). The OBX rats were divided into three groups and were administered either saline (OBX-SAL; N = 11) or buspirone (either 10 mg/(kg day) (OBX-10; N = 14) or 20 mg/(kg day) (OBX-20; N = 10)) in saline for 14 days. The rats were randomly assigned to different groups.

2.3. Tracer study Two weeks following the implantation of the minipumps, the rats were used for the tracer experiment. They were fasted overnight, with water given ad libitum the night before the experiment to stabilize the amino acid concentration in the blood. The rats were cannulated into femoral arteries and veins with 1% halothane anesthesia. After the cannulation, the rats were placed in a loose fitting plaster cast and allowed to awaken. Arterial PO2 , PCO2 , pH, and hematocrit were

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108 checked at least twice in each experiment. The temperature was maintained with heating lamps. To avoid the possible influence of the circadian rhythm on the results, the tracer was always injected between noon and 2 p.m. Note that the minipumps remained implanted during the entire experiment. To determine the time activity curve of the plasma tracer input, blood samples (14–15) were taken at progressively increased time periods from the start of the injection until decapitation. Following centrifugation for 5 min at 9300 × g, 20 ␮L of plasma was transferred into 9 mL of scintillating solvent and 1 ml of distilled water was added. The radioactivity was measured by liquid scintillation counting to obtain the plasma input function. An additional six samples were taken at different times. Three of these six samples were used for the measurements of plasma free tryptophan (with an ultrafiltrate using a 10,000 molecular weight cut off point) and another three were used following deproteinization with an equal volume of trichloroacetic acid for a determination of the plasma total tryptophan. In each group, approximately half of the rats were decapitated 60 min following the tracer injection and the other half at 150 min following the tracer injection. The brains were quickly extracted and frozen in a cold iso-pentane (at approximately −20 ◦ C), and stored in a freezer at −84 ◦ C until the brain slices were cut. Using cryotome (at about −22 ◦ C), 30 ␮m thick sections were cut, fixed onto glass-slides, and dried at approximately 60 ◦ C for at least 120 min. The brains of the OBX rats were closely inspected for the complete removal of the olfactory bulbs or frontal lobe damage. The brain slices were contacted onto X-ray film for 3 weeks along with 14 C-plastic standards calibrated to the tissue equivalent. The films were digitized using a microcomputer based image analysis system (MCID, Image Research Co., St. Catherine’s, Ont., Canada). The optical densities were converted into tissue tracer concentration based on an appropriate standard calibration curve made up from 14 C standards. The 5-HT synthesis rates R (pmol/(g/min)) were calculated from the trapping constant of ␣-[14 C]MTrp (K*; ml/(g min)), calculated as previously described [14,21,36], and the plasma concentration of the plasma non-protein-bound tryptophan Cp (pmol/ml) as; R = Cp × K*/LC [21]. The LC (stands for the lumped constant) was measured in vivo in the rat brain and it was estimated to be 0.42 ± 0.07 [35].

2.4. Statistical comparisons The primary global comparisons in which the effect of buspirone on 5-HT synthesis in the OBX rats was performed by comparing the OBX-10 and OBX20 groups to the OBX-SAL group. The effect of buspirone action on 5-HT synthesis in the sham operated rats was tested by comparing the synthesis in the SHX-10 and SHX-SAL groups. The objective of this comparison was to confirm that 10 mg/(kg day) of buspirone in the sham operated rats produced the desensitization of 5-HT1A receptors (no effect on synthesis) as was previously found in the intact animals [23]. At the same time, this confirmed the expectation that there is no need to use a higher dose of buspirone in the sham operated rats. To confirm the previous observations that OBX increases global and regional synthesis [14,36], we compared the OBX-SAL and SHX-SAL groups. We tested whether buspirone treatment produces synthesis normalization (a return to the level of the saline treated sham operated rats) in the OBX rats by assessing the differences between the OBX-10 and OBX-20 groups relative to the SHXSAL. The synthesis in the SHX-SAL group was assumed to be at the level of a “normal” rat. The global effects were evaluated by comparing the mean ratio of the 5-HT synthesis rates in the control (note that the control group depends on the comparisons made) and treatment groups using the one-group t-test with a null hypothesis of a ratio equal to 1 and a standard deviation (S.D.) of 0 (an ideal ratio). The synthesis in each of the brain structures for all five groups was compared by using a one-way ANOVA with Bonferroni post-hoc corrections for eight possible comparisons. All of the synthesis rates are presented as mean ± S.D. values, with the S.D. calculated from the variance reported by the least-squares method [21].

3. Results The PO2 , PCO2 , pH and Hematocrit (actual data not provided) of the rats were within normal levels, and there were no differences between the groups (ANOVA with Bonferroni correction). The values of these parameters did not vary from those

103

of previous similar experiments in our laboratory. No rats were removed from any of the groups because of abnormal blood gasses. In the OBX rats, the plasma concentration of free Trp was 12.0 ± 3.4 nmol/mL in the saline treated group (OBX-SAL), 12.5 ± 3.6 nmol/mL in the group treated with 10 mg/(kg day) of buspirone (OBX-10), and 9.5 ± 3.5 nmol/mL in the group treated with 20 mg/(kg day) of buspirone (OBX-20). In the SHX rats, the plasma concentration of free Trp was 9.7 ± 3.4 nmol/mL in the SHX-SAL group and 9.6 ± 3.5 nmol/mL in the group treated with 10 mg/(kg day) of buspirone (SHX-10). The total plasma Trp was 75 ± 14 nmol/mL in the OBX-SAL group, 75 ± 14 nmol/mL in the OBX-10 group, and 74 ± 17 nmol/mL in the OBX-20 group. In the SHX study, the total plasma Trp was 89 ± 13 nmol/mL in the SHX-SAL group and 76 ± 9 nmol/mL in the SHX-10 group. There were no significant differences (p > 0.05) between the plasma-free or total Trp concentrations between any of these groups (ANOVA). The Trp concentrations in all of the groups were within the range found in other experiments performed in the laboratory. A set of representative autoradiographic images representing 5-HT synthesis at several cross-sections in all five groups are shown in Fig. 1. It can be seen that the synthesis was not uniform throughout all of the anatomical brain structures (e.g., cortical layer VI, medial and lateral caudate). There was also a clear delineation of the dorsal and median raphe, two brain areas with the greatest synthesis of 5-HT. The greatest synthesis of 5-HT was found in the pineal body, the structure found outside of the brain, but inside the cranium. The 5-HT synthesis rates of the OBX and SHX rats are provided in Table 1 for selected brain structures. The values are provided for the SHX-SAL, SHX-10, OBX-SAL, OBX-10, and OBX-20 groups for the 14 days experiments. In the SHX-10 rats, the 5-HT synthesis rates of all of the brain areas investigated were not significantly different from those of the SHX-SAL group. The test of the mean ratio indicates that there was no global effect of buspirone treatment; globally, the synthesis is not significantly different between the SHX-SAL and SHX-10 groups (the mean ratio of SHX-10 and SHX-SAL is 1.04 ± 0.19; N = 31, t = 1.20; p > 0.2). The one sample t-test of the synthesis ratios between the OBX-SAL, the OBX-10 and OBX-20 groups indicated that 10 mg/(kg day) of buspirone did not have an overall significant influence on brain 5-HT synthesis in the OBX-10 group. The mean ratio for the OBX-10 to OBX-SAL groups was 0.98 ± 0.120 (t = −0.56; p > 0.5; N = 31), which means that the mean ratio was not significantly different from one, suggesting no overall effect of treatment. Despite not having any overall influence on 5-HT synthesis, there are a few structures in which 5-HT synthesis was significantly reduced with this treatment (OBX-10 relative to OBX-SAL): the frontal (37%), cingulate (40%), sensory-motor (50%), and parietal (73%) cortices, and the magnus raphe nucleus (50%). There are also two brain structures (CA3 layer of the ventral tegmental area) in which 5-HT synthesis was significantly greater in the OBX-10 group as compared to the OBX-SAL group. When the mean ratio of the synthesis in the OBX-20 group was compared to the OBX-SAL,

104

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108

Fig. 1. Representative autoradiograms obtained in the chronic buspirone treatments of SHX and OBX rats with saline or buspirone. The autoradiograms shown here are taken from the rat brain 150 min followiong the tracer injection. The brain slices were exposed for 3 weeks. The meaning of the structure abbreviations are: DR, raphe dorsal; MR, median; AN, accumbens; VCx, cortex visual; PCx, cortex parietal; SMCx, cortex sensory-motor; HT, hypothalamus; CL, caudate lateral; CM, caudate medial; TV, thalamus ventral; DL, thalamus dorsal; DHi, hippocampus dorsal; VHi, hippocampus ventral; PB, pineal body; SN, substantia nigra; AC, nucleus accumbens; VI, layer six of the cortex.

a significant influence of this treatment was found. The mean ratio is 0.58 ± 0.15 (t = −15.6; p < 10−16 ; N = 31), indicating a significant global influence on 5-HT synthesis (see Table 1). The statistical evaluation of the differences in the discrete brain areas found a significant reduction of the synthesis in all areas of the OBX-20 group except the dorsal and median raphe, ventral tegmental area and superior colliculus. To assess the effect of OBX on the 5-HT synthesis, a comparison of the OBX-SAL and SHX-SAL groups was performed. This comparison showed that the synthesis in the OBX-SAL group was significantly greater than in the SHX-SAL group (ratio = 1.70 ± 0.33; N = 31; t = 11.8; p < 10−13 ), which is in accordance with previously reported data [14,36]. To assess whether 20 mg/(kg day) of buspirone normalized 5-HT synthesis to the level found in the sham operated rats, we compared the OBX-20 rats to the SHX-SAL rats (the synthesis in the SHX-SAL rats was assumed to be normal and the same as in rats without any pharmacological treatment). The overall synthesis in the OBX-20 rats was not significantly different from the SHX-SAL rats, suggesting normalization (reduction to the levels in normal rats) and possibly desensitization of the 5-HT1A receptors, as previously reported for intact animals [23]. The mean value of the ratios of the regional synthesis in the OBX20 group relative to the SHX-SAL group, used as a control, was 0.97 ± 0.23 (p > 0.4; one sample t-test; N = 31; t = −0.82), which is not significantly different than 1 ± 0. A similar evaluation in which the OBX-10 group was compared to both the SHX-10 and SHX-SAL groups shows that 10 mg/(kg day) of buspirone did not bring the 5-HT synthesis rate to the level of the sham operated rats: the mean ratios were 1.58 ± 0.22 (p < 10−15 ; N = 31;

t = 14.6) and 1.62 ± 0.25 (p < 10−14 ; N = 31; t = 13.9), respectively.

4. Discussion The primary finding of this study is that buspirone treatment delivered continuously for 14 days significantly reduces brain 5-HT synthesis in the OBX rat model of depression, when 20 mg/(kg day) is delivered, as opposed to 10 mg/(kg day). The 20 mg/(kg day) dose of buspirone in the OBX-20 rats, when compared to the OBX-SAL group, reduced 5-HT synthesis in all of the brain areas except the ventral tegmental area, the dorsal and median raphe, and the superior colliculus. This dose also lowers the global 5-HT synthesis rates throughout the brain. In contrast, a dose of 10 mg/(kg day) delivered for 14 days in the OBX-10 group, relative to the OBX-SAL group, does not have a significant overall effect on brain 5-HT synthesis, as the ratio is not significantly different from 1 ± 0. However, there are a few structures in which the reduction in 5-HT synthesis (e.g., cingulate, frontal, sensory-motor, and parietal cortices, and magnus raphe nucleus; Table 1) has been observed. There are also two brain areas (the ventral tegmental area and CA3) in which 5-HT is greater in the OBX-10 group than in the OBX-SAL group. We do not have a good explanation for this, except that it may simply be by chance. There are no significant differences between the SHX-10 and SHX-SAL groups, indicating that chronic treatment with buspirone (10 mg/(kg day)) similarly desensitizes the receptors in the SHX rats, as has been previously observed in the intact rat [23]. This desensitization with a dose of 10 mg/(kg day)

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108

105

Table 1 5-HT synthesis rates (pmol/(g/min)) measured in the OBX rats treated with 10 (OBX-10) and 20 (OBX-20) mg/(kg day) of buspirone or saline (OBX-SAL) for 14 daysa Structures

SHX-SAL (N = 11)

Accumbens nucleus Cingulate cortex Caudate (medial) Caudate (lateral) Caudoputamen at the level of GP Thalamus (dorsal) Thalamus (ventral) Amygdala Hippocampus (dorsal) Hippocampus (ventral) Field CA3 Hypothalamus Frontal cortex Parietal cortex Sensory-motor cortex Auditory cortex Visual cortex Globus pallidus Dorsal raphe nucleus Mediam raphe nucleus Magnus raphe nucleus Raphe pontine nucleus Claustrium Lateral geniculate body Medial forebrain bundle Substantia nigra compacta Substantia nigra reticulate Ventral tegmental area Superior colliculus Locus coeruleus Pineal body

43 20 33 31 37 28 29 41 38 40 35 34 20 19 16 24 22 33 134 97 40 32 38 31 29 27 19 32 25 31 680

a b o + † ‡ *

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

9+ 6+ 6+ 7+ 6+ 6+ 7+ 9+ 8+ 8+ 7+ 9+ 6+ 6 6 6+ 7+ 5+ 14 15 6 5+ 8+ 6+ 9+ 6+ 8+ 6+ 8+ 10+ 45

SHX-10 (N = 8) 47 25 38 32 32 25 27 37 33 38 38 30 23 22 22 30 26 31 119 86 26 22 45 25 32 28 24 31 29 38 712

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

12 10 6 9 8 11 11 11 9 11 12 12 7 9 11 9 9 7 11 10 9 9 9 10 11 11 7 12 10 16 55

OBX-SAL (N = 11) 71 49 57 50 66 43 40 60 59 61 50 57 41 45 33 45 44 50 145 115 69 59 72 43 48 51 40 40 35 50 720

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

12o 9o 10o 10o 10o 11o 12 17o 11o 17o 16 15o 9o 10o 6o 11o 10o 10o 10 o 8 14o 13o 12o 12o 11o 9o 10o 13 15 16o 55

OBX-10 (N = 14) 82 35 62 50 61 41 47 72 60 75 65 50 30 26 22 37 43 47 149 113 46 47 71 42 54 43 39 51 42 62 714

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

8 8* 7 9 11 8 9 10 8 9 9* 10 5* 7* 7* 6 7 11 18 12 11* 10 7 7 9 8 9 9* 8 10 38

OBX-20b (N = 10) 50 26 37 28 28 18 18 39 36 42 34 28 20 18 17 33 28 28 138 116 29 41 37 22 26 24 20 30 23 23 713

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

9†,‡ 9† 8†,‡ 7†,‡ 8†,‡ 9†,‡ 7†,‡ 11†,‡ 9†,‡ 10†,‡ 16†,‡ 9†,‡ 8†,‡ 7† 8† 9† 7†,‡ 7†,‡ 28 26 7†,‡ 7†,‡ 9†,‡ 8†,‡ 10†,‡ 7†,‡ 9†,‡ 8‡ 10‡ 9†,‡ 28

The values are presented as mean ± S.D., with N being the number of rats in each of the groups. There is no significant differences between OBX-20 and SHX-SAL or SHX-10 groups. Significant difference (p < 0.05) between OBX-SAL and SHX-SAL (ANOVA with Bonferroni correction). Significant difference (p < 0.05) between OBX-10 and SHX-SAL (ANOVA with Bonferroni correction). Significant difference (p < 0.05) between OBX-20 and OBX-SAL (ANOVA with Bonferroni correction). Significant difference (p < 0.05) between OBX-20 and OBX-10 groups (ANOVA with Bonferroni correction). Significant difference (p < 0.05) between OBX-10 and OBX-SAL groups (ANOVA with Bonferroni correction).

also confirmed that a 20 mg/(kg day) group of sham rats is not needed, because the lower dose produced a desensitization. There are significant differences between the OBX-20 group, and both the SHX-SAL and SHX-10 groups (note, however, that there is no difference between the SHX-SAL and SHX-10 groups), suggesting that 20 mg/(kg day) of buspirone normalized the 5-HT synthesis rates and reduced synthesis in the OBX rats to the level found in the sham operated rats. These observations suggest that the normalization in the synthesis in the OBX rats occurs only after a higher dose of buspirone was administered. However, comparisons of individual brain structures (between the OBX-10 and SHX-10 groups) suggest that normalization in some structures (e.g., cingulate, parietal, sensory-motor and auditory cortices) occurs even following a 10 mg/(kg day) treatment (Table 1). This also suggests that in addition to evaluating behavioural changes, one should also measure some of the neurochemical parameters before obtaining a

good understanding of the drug action. The reduction in synthesis after a lower dose (OBX-10 group) may be of significance, because some of these structures (e.g., cingulated and frontal cortices) are important for behavioural changes [13]. It has been reported that even 3 mg/kg of buspirone administered for 21 days produces significant behavioural effects in the OBX rats [20]. Because the OBX rats have an elevated 5-HT synthesis [14,36], successful treatment is expected to reduce synthesis in the OBX rats and bring it to the level of the sham operated rats treated with saline. Indeed, the data presented here show that the administration of 20 mg/(kg day) of buspirone for 14 days normalizes the global and regional 5-HT synthesis rates approximately to the levels found in the sham operated rats (when comparing the SHX-SAL and OBX-20 groups). It can be hypothesized that buspirone, after reaching the brain, lowers 5-HT synthesis through its action on 5-HT1A sites [23], subsequently reducing the amount of extraneuronal 5-HT and

106

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108

re-establishing the physiological circuitry which, in turn, reestablishes the synthesis control. During the same process, the receptors become desensitized, as is the case in intact rats [1,23]. Following chronic treatment with buspirone, the agonist does not lower the synthesis below normal levels because the receptors controlling the synthesis become desensitized. The data presented suggest that the 5-HT1A receptors in the OBX rats have a different response to a chronic treatment with the agonist than intact rats [23]. This confirms the importance of evaluating drugs in animal models of depression rather than in intact animals. There is no significant difference (p > 0.05) in the plasma free Trp between the OBX-20 and OBX-SAL groups, however the plasma free Trp is approximately 21% lower in the OBX-20 group. As the plasma free Trp is probably related to brain 5HT synthesis [8,22,26], this difference could in part be related to the observed differences in 5-HT synthesis. However, it is unlikely that the differences in the plasma free Trp are solely responsible for the observed differences, because the differences in the synthesis in the majority of the brain areas are substantially greater than this difference in the plasma Trp (Table 1) (e.g., OBX-20 compared to OBX-SAL). We previously reported a widespread increase in regional 5-HT synthesis in the majority of the projection areas of the olfactory bulbectomized rats [14,36]. These increases in 5-HT synthesis accord with reports of others who reported an increased density of 5-HT transporters and tryptophan hydroxylase concentration in the frontal cortex, compared to the control rats [18]. This suggests a local and specific alteration of 5-HT neurotransmission in the frontal cortex that may be related to the long-lasting changes in behavioural responsiveness caused by OBX. Zhou et al. [39] reported that the OBX-associated axotomy of 5-HT fibers projecting into the olfactory bulb caused collateral sprouting and the regeneration and synaptogenesis of new 5-HT synapses, especially in the frontal cortex. Others have reported a reduction in 5-HT synthesis measured as an accumulation of 5-hydroxytryptophan following the inhibition of aromatic amino acid decarboxylase [31]. However, 5-HT turnover calculated from the data reported in that manuscript (see Table 1 in [31]) is significantly elevated in the OBX rats. These two sets of results from the same work contradict each other. The second set of data (5-HT turnover) would accord with the results presented here and those reported previously [14,36,39]. A comparison of the SHX-SAL and OBX-SAL groups confirms the observation of increased synthesis in the OBX rats (Table 1). Certainly, further experiments are required to obtain a full understanding of the reasons for this widespread elevation, but there is a possibility that as a result of the OBX, the 5-HT receptors, through which 5-HT synthesis is controlled (e.g., 5-HT1A , 5-HT1B ), are not functioning properly because of the new connections made after the removal of the olfactory bulbs. In human post-mortem studies and pharmacological experiments of the 5-HT1A receptors, a link is suggested between the decreased receptor function and depression [5]. The data presented here also support the hypothesis of an alteration in the functionality of the 5-HT1A receptors. It has been demonstrated that many of the behavioural, endocrine and neurochemical changes are reversed by chronic

but not acute treatments of antidepressants [12]. For example, Mar et al. [20] reported that chronic buspirone treatment of OBX rats showed an increasing rate of habituation, and treatments with various antidepressants attenuate the passive avoidance deficit associated with OBX [12]. The behaviour improvements could be related to a normalization of brain 5-HT synthesis resulting in less 5-HT in the brain tissue for the creation of non-physiological circuitry, and the subsequent normalization of neuronal circuitry [29]. It is not easy to envision why and how behaviour changes precede changes in the synthesis of this very important brain neurotransmitter. Because buspirone and its metabolite have an effect on serotonergic, dopaminergic and noradrenergic systems, one would expect that behaviour changes will follow neurochemical changes. There is considerable evidence to support the hypothesis that 5-HT1A agonists, including buspirone [4,37], possess antidepressant activity [18,19]. There are several hypotheses regarding the site of action of 5-HT1A agonists with antidepressantlike effects. It has also been suggested that both pre-synaptic and post-synaptic 5-HT1A receptors mediate antidepressant-like activities of 5-HT1A agonists [19]. A differential desensitization of pre-and post-synaptic 5-HT1A receptors has also been proposed [27], as well as the full agonist and partial agonist properties of buspirone and its metabolite at those receptor sites, respectively [3,6,15]. It was proposed that the desensitization of the dorsal raphe 5-HT1A autoreceptors with buspirone results in the same level of 5-HT synthesis in the buspirone and saline treated rats [23,27]. It is interesting to note that following a 14-day buspirone treatment of the OBX rats (OBX-20 and OBX-10), the synthesis in the DR (dorsal raphe) and MR (median raphe) are not significantly different than the OBX-SAL group (Table 1), while in a large majority of other structures, the synthesis is greater in the OBX-SAL group than the OBX-20 group. In our previous study with citalopram, we observed a similar trend [14]. This would suggest that the receptors in the DR and MR are possibly not responsive to the agonist and/or extracellular 5-HT, as the synthesis in the OBX rats is greater than in the SHX rats. However, in the cortices, such as the frontal, sensory-motor, and parietal, as well as in some deep brain structures (e.g., median caudate, thalamus, amygdala), the 5-HT synthesis rates were decreased in the OBX-20 group in comparison to the OBX-SAL group, but the synthesis is at the same level as the SHX-SAL group, which can be assumed to be a group with normal 5-HT synthesis. This may suggest that the receptors controlling the synthesis in the OBX rats have different sensitivities and/or functionality than those found in the sham or intact rats. It is likely that nonphysiological connections between the serotonergic neurons and other neurons, as well as new connections established after OBX, play a role in the observed differences in the synthesis and action of buspirone. 5. Conclusion We have demonstrated that 5-HT synthesis rates in the brain of OBX rats treated with 20 mg/(kg day) buspirone for 14 days were significantly lower than rates in the OBX-SAL rats, and

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108

they were, overall, not significantly different from the SHX-SAL or SHX-10 groups. There were no global reductions in the 5-HT synthesis in the OBX-10 rats in comparison to the OBX-SAL rats. This suggests that there is, throughout the brain, a normalization of 5-HT synthesis only after the 20 mg/(kg day) treatment. These results suggest that in the olfactory bulbectomized rats, following chronic buspirone treatment with 20 mg/(kg day) for 14 days, 5-HT1A autoreceptors/receptors are probably at a similar functional state as in the sham operated rats treated with saline. Because even smaller doses of buspirone showed improvement in the behaviour of the OBX rats [20], we can suggest that buspirone may have had an antidepressant effect in the OBX rats by producing changes in 5-HT synthesis.

[15] [16]

[17]

[18]

[19]

[20]

Acknowledgments The research reported here was supported in part by the Canadian Institute for Health Research (MOP-42438). M.D. is a Killam Scholar at the Montreal Neurological Institute of McGill University. We would also like to thank Ms. Valerie-Ann Cherneski for the editorial help.

[21]

References

[23]

[1] P. Blier, C. de Montigny, Modification of 5-HT neuron properties by sustained administration of the 5-HT1A agonist gepirone: electrophysiological studies in the rat brain, Synapse 1 (1987) 470–480. [2] C. Bouwer, D.J. Stein, Buspirone is an effective augmenting agent of serotonin selective re-uptake inhibitors in severe treatment-refractory depression, S. Afr. Med. J. 87 (1997) 534–537. [3] S. Caccia, I. Conti, G. Vigano, S. Garattini, 1-(2-Pyrimidinyl)-piperazine as active metabolite of buspirone in man and rat, Pharmacology 33 (1986) 46–51. [4] L. Cervo, G. Grignaschi, R. Samanin, Different effects of intracerebral and systemic administration of buspirone in the forced swimming test: involvement of a metabolite, Life Sci. 43 (1988) 2095–2102. [5] J. Coplan, S. Wolk, D. Klein, Anxiety and the Serotonin 1A Receptor, Raven Press, New York, 1995. [6] M. De Vivo, S. Maayani, Characterization of the 5-hydroxytryptamine1a receptor-mediated inhibition of forskolin-stimulated adenylate cyclase activity in guinea pig and rat hippocampal membranes, J. Pharmacol. Exp. Therap. 238 (1986) 248–253. [7] M. Diksic, Labelled ␣-methyl-l-tryptophan as tracer to study brain serotonergic system, J. Psychiatry Neurosci. 26 (2001) 293–303. [8] M. Diksic, S. Nagahiro, T. Chaly, T.L. Sourkes, Y.L. Yamamoto, W. Feindel, Serotonin synthesis rate measured in living dog brain by positron emission tomography, J. Neurochem. 56 (1991) 153–162. [9] M. Diksic, S.N. Young, Study of the brain serotonergic system with labelled ␣-methyl-l-tryptophan, J. Neurochem. 78 (2001) 1–17. [10] E.C. Dimitriou, C.E. Dimitriou, Buspirone augmentation of antidepressant therapy, J. Clin. Psychopharmacol. 18 (1998) 465–469. [11] G. Engberg, A metabolite of buspirone increases locus coeruleus activity via alpha 2-receptor blockade, J. Neural. Transm. 76 (1989) 91– 98. [12] L.F. Fabre, Buspirone in the management of major depression: a placebocontrolled comparison, J. Clin. Psychiatry 51 (Suppl.) (1990) 55–61. [13] G. Grecksch, D. Zhou, C. Franke, U. Schroder, B. Sabel, A. Becker, G. Huether, Influence of olfactory bulbectomy and subsequent imipramine treatment on 5-hydroxytryptaminergic presynapses in the rat frontal cortex: behavioural correlates, Br. J. Pharmacol. 122 (1997) 1725– 1731. [14] S. Hasegawa, A. Watanabe, K.Q. Nguyen, G. Debonnel, M. Diksic, Chronic administration of citalopram in olfactory bulbectomy rats

[22]

[24]

[25]

[26]

[27]

[28] [29]

[30]

[31]

[32]

[33] [34]

[35]

107

restores brain 5-HT synthesis rates: an autoradiographic study, Psychopharmacology (Berl.) 179 (2005) 781–790. D. Hoyer, H.W. Boddeke, Partial agonists, full agonists, antagonists: dilemmas of definition, Trend Pharmacol. Sci. 14 (1993) 70–75. J.A. Jesberger, J.S. Richardson, Brain output dysregulation induced by olfactory bulbectomy: an approximation in the rat of major depressive disorder in humans? Int. J. Neurosci. 38 (1988) 241–265. J.P. Kelly, A.S. Wrynn, B.E. Leonard, The olfactory bulbectomized rat as a model of depression: an update, Pharmacol. Ther. 74 (1997) 299– 316. G.A. Kennett, C.T. Dourish, G. Curzon, Antidepressant-like action of 5-HT1A agonists and conventional antidepressants in an animal model of depression, Eur. J. Pharmacol. 134 (1987) 265–274. I. Lucki, A. Singh, D.S. Kreiss, Antidepressant-like behavioral effects of serotonin receptor agonists, Neurosci. Biobehav. Rev. 18 (1994) 85– 95. A. Mar, E. Spreekmeester, J. Rochford, Antidepressants preferentially enhance habituation to novelty in the olfactory bulbectomized rat, Psychopharmacology (Berl.) 150 (2000) 52–60. S. Nagahiro, A. Takada, M. Diksic, T.L. Sourkes, K. Missala, Y.L. Yamamoto, A new method to measure brain serotonin synthesis in vivo. II. A practical autoradiographic method tested in normal and lithiumtreated rats, J. Cereb. Blood Flow Metab. 10 (1990) 13–21. S. Nishizawa, C. Benkelfat, S.N. Young, M. Leyton, S. Mzengeza, C. De Montigny, P. Blier, M. Diksic, Differences between males and females in rates of serotonin synthesis in human brain, Proc. Nat. Acad. Sci. U.S.A. 94 (1997) 5308–5313. H. Okazawa, F. Yamane, P. Blier, M. Diksic, Effects of acute and chronic administration of the serotonin1A agonist buspirone on serotonin synthesis in the rat brain, J. Neurochem. 72 (1999) 2022–2031. M.F. Piercey, M.W. Smith, J.T. Lum-Ragan, Excitation of noradrenergic cell firing by 5-hydroxytryptamine1A agonists correlates with dopamine antagonist properties, J. Pharmacol. Exp. Ther. 268 (1994) 1297– 1303. J.P. Redrobe, M. Bourin, Dose-dependent influence of buspirone on the activities of selective serotonin reuptake inhibitors in the mouse forced swimming test, Psychopharmacology 138 (1998) 198–206. M. Salter, R.G. Knowles, C.L. Aogson, How dose displacement of albumin-bound tryptophan cause sustained increases in the free tryptophan concentration in plasma and 5-hydorxytryptamine synthesis in the brain? Biochem. J. 262 (1989) 365–368. L.J. Sim-Selley, L.J. Vogt, R. Xiao, S.R. Childers, D.E. Selley, Regionspecific changes in 5-HT(1A) receptor-activated G-proteins in rat brain following chronic buspirone, Eur. J. Pharmacol. 389 (2000) 147–153. J.C. Soares, J.J. Mann, The functional neuroanatomy of mood disorders, J. Psychiatr. Res. 31 (1997) 343–393. M.R. Spoont, Modulation role of serotonin in neuronal information processing: implications for human psychopathology, Psychol. Bull. 112 (1992) 330–350. Y. Tohyama, F. Yamane, M.F. Merid, M. Diksic, Effects of selective 5-HT1A antagonists on regional serotonin synthesis in the rat brain: an autoradiographic study with ␣-[14 C]methyl-l-tryptophan, Eur. Neuropsychopharmacol. 11 (2001) 193–202. H.M. van der Stelt, M.E. Breuer, B. Olivier, H.G.M. Westenberg, Permanent deficits in serotonergic functioning of olfactory bulbectomized rats: an in vivo microdialysis study, Biol. Psychiatry 57 (2005) 1061–1067. H. van Riezen, B.E. Leonard, Effects of psychotropic drugs on the behavior and neurochemistry of olfactory bulbectomized rats, Pharmacol. Ther. 47 (1990) 21–34. I. van Wijngaarden, M.T. Tulp, W. Soudijn, The concept of selectivity in 5-HT receptor research, Eur. J. Pharmacol. 188 (1990) 301–312. C.P. VanderMaelen, G.K. Matheson, R.C. Wilderman, L.A. Patterson, Inhibition of serotonergic dorsal raphe neurons by systemic and iontophoretic administration of buspirone, a non-benzodiazepine anxiolytic drug, Eur. J. Pharmacol. 129 (1986) 123–130. M. Vanier, K. Tsuiki, M. Grdisa, K. Worsley, M. Diksic, Determination of the lumped constant for the ␣-methyltryptophan method of estimating the rate of serotonin synthesis, J. Neurochem. 64 (1995) 624–635.

108

A. Watanabe et al. / Brain Research Bulletin 69 (2006) 101–108

[36] A. Watanabe, Y. Tohyama, K.Q. Nguyen, S. Hasegawa, G. Debonnel, M. Diksic, Regional brain serotonin synthesis is increased in the olfactory bulbectomy rat model of depression: an autoradiographic study, J. Neurochem. 85 (2003) 469–475. [37] S. Wieland, I. Lucki, Antidepressant-like activity of 5-HT1A agonists measured with the forced swim test, Psychopharmacology 101 (1990) 497–504.

[38] P. Willner, Animal models of depression: an overview, Pharmacol. Ther. 45 (1990) 425–455. [39] D. Zhou, G. Grecksch, A. Becker, C. Frank, J. Pilz, G. Huether, Serotonergic hyperinnervation of the frontal cortex in an animal model of depression, the bulbectomized rat, J. Neurosci. Res. 54 (1998) 109– 116.