Antidepressant-like effect of bright light is potentiated by l -serine administration in a mouse model of seasonal affective disorder

Brain Research Bulletin 118 (2015) 25–33

Contents lists available at ScienceDirect

Brain Research Bulletin journal homepage: www.elsevier.com/locate/brainresbull

Research report

Antidepressant-like effect of bright light is potentiated by l-serine administration in a mouse model of seasonal affective disorder Misato Kawai a , Ryosei Goda a , Tsuyoshi Otsuka a , Ayaka Iwamoto a , Nobuo Uotsu b , Mitsuhiro Furuse a , Shinobu Yasuo a,∗ a b

Laboratory of Regulation in Metabolism and Behavior, Faculty of Agriculture, Kyushu University, Fukuoka 812-8581, Japan Research Institute, FANCL Co., Yokohama 244-0806, Japan

a r t i c l e

i n f o

Article history: Received 23 July 2015 Received in revised form 28 August 2015 Accepted 29 August 2015 Available online 1 September 2015 Keywords: Seasonal affective disorder Depression-like behavior Mouse Serotonin Raphe nuclei c-Fos

a b s t r a c t Bright light therapy is used as the primary treatment for seasonal affective disorder; however, the mechanisms underlying its antidepressant effect are not fully understood. Previously, we found that C57BL/6J mice exhibit increased depression-like behavior during a short-day condition (SD) and have lowered brain serotonin (5-HT) content. This study analyzed the effect of bright light on depression-like behaviors and the brain serotonergic system using the C57BL/6J mice. In the mice maintained under SD, bright light treatment (1000 lx, daily 1 h exposure) for 1 week reduced immobility time in the forced swimming test and increased intake of saccharin solution in a saccharin intake test. However, the light treatment did not modify 5-HT content and selective 5-HT uptake in the amygdala, or temporal patterns of core body temperature and wheel-running activity throughout a day. In the next experiment, we attempted to enhance the effect of bright light by using l-serine, a precursor of d-serine that acts as an N-methyl-daspartic acid receptor coagonist. Daily subcutaneous injection of l-serine for 2 weeks prior to the bright light strongly reduced the immobility time in the forced swimming test, suggesting a synergistic effect of light and l-serine. Furthermore, bright light increased the total number of 5-HT-immunoreactive cells and cells that had colocalized 5-HT and c-Fos immunosignals in several subregions of the raphe nuclei. These effects were potentiated by prior injection of l-serine. These data suggest that the bright light may elicit an antidepressant-like effect via enhanced 5-HT signals in the brain and l-serine can enhance these effects. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Seasonal changes in photoperiod regulate mammalian physiology and behavior. Seasonal affective disorder (SAD) is a subtype of major depressive or bipolar disorders that follow the seasonal pattern of major depressive episodes occurring at a specific time of the year (Rosenthal et al., 1984). Symptoms of SAD include depression, with the associated diminished pleasure or interest, feelings of worthlessness, and decreased ability to think or concentrate. Most SAD patients also report atypical symptoms including hypersomnia, hyperphagia, decreased energy levels, and carbohydrate craving (Rosenthal et al., 1984). Bright light therapy is used as the primary treatment for SAD (Lewy et al., 1982; Terman et al., 1989;

∗ Corresponding author at: Laboratory of Regulation in Metabolism and Behavior, Faculty of Agriculture, Kyushu-University, Hakozaki 6-10-1, Higashi-ku, Fukuoka 812-8581, Japan. Fax: +81 92 642 4426. E-mail address: [email protected] (S. Yasuo). http://dx.doi.org/10.1016/j.brainresbull.2015.08.010 0361-9230/© 2015 Elsevier Inc. All rights reserved.

Oldham and Ciraulo, 2014). Although mechanisms underlying SAD remain elusive, numerous studies suggest the involvement of the brain serotonergic system (Carlsson et al., 1980; Lambert et al., 2002; Gupta et al., 2013). Additionally, the circadian phase-shift hypothesis has been proposed based on the observation that the internal circadian rhythms of SAD patients are phase-delayed relative to the external clock or other rhythms, such as sleep-wake cycle (Lewy et al., 1987). In this theory, bright light therapy exerts an antidepressant effect through phase-advance of the clock. However, there is still some controversy regarding the involvement of circadian system (Oldham and Ciraulo, 2014). There are several proposed animal models of SAD, including diurnal rodents (fat sand rats: Einat et al., 2006 grass rats: Leach et al., 2013) and Siberian hamsters (Prendergast and Nelson, 2005). In grass rats, daytime light deficiency under 12 h light and 12 h darkness (12L12D) increases stress-induced immobility and decreases the number of serotonin (5-HT)-immunoreactive neurons (Leach et al., 2013), suggesting a link between light and the brain serotonergic system. However, mechanistic analysis has

26

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

not been advanced in detail because of a lack of inbred laboratory animal models. This is due to a conventional premise that laboratory mice and rats, being non-seasonal breeders, are inappropriate for the investigation of photoperiod-related functions. Recently, however, we clarified that compared to C57BL/6J mice under long-day condition (LD), those under short-day conditions (SD) exhibit an increased immobility in the forced swimming test (FST), a depression-like behavior, and reduced intake of saccharin, a depression-related anhedonic behavior, with lowered 5-HT content in the amygdala (Otsuka et al., 2014). These mice also showed increased intake of sucrose and high sensitivity of 5-HT synthesis to glucose, which may reflect carbohydrate craving in patients with SAD (Otsuka et al., 2014). From these findings, C57BL/6J mice may serve as a useful tool for elucidating the mechanisms of SAD. Predictive validity is a crucial criterion that animal models for psychiatric disorders should fulfill. This study first evaluated the antidepressant-like effect of bright light in C57BL/6J mice under SD. We further examined the effect of bright light on the brain serotonergic system and rhythm profiles of wheel-running activity and core body temperature. Because N-methyl-d-aspartic acid (NMDA) receptors are suggested to play a pivotal role in light signaling in the brain and 5-HT system (de Kock et al., 2006; Albrecht, 2012), we attempted to enhance the antidepressant-like effect of bright light by administration of l-serine, a precursor of d-serine that acts as an NMDA receptor coagonist. Finally, we examined bright lightand/or l-serine-induced modulation of serotonergic signals in the raphe nuclei by analyzing 5-HT and c-Fos immunoreactivity. 2. Materials and methods 2.1. Animals Male 4-week-old C57BL/6J mice were obtained from Japan SLC (Shizuoka, Japan). Mice were housed in a group of three or four animals. After acclimation for 1–2 weeks, they were exposed to SD [8 h of light (5 lx), 16 h of darkness] or LD [16 h of light (100 lx), 8 h of darkness] as indicated below. The light intensity under SD sufficiently caused increased depression-like behavior in our previous study (Otsuka et al., 2014). Light was supplied by a white LED light bulb (ELG-01B(W), light ranging from 400 to 700 nm, peak: 440–460 nm, Asahi Electric Corporation, Osaka, Japan). The animal boxes were placed in a room at a temperature of 25 ± 1 ◦ C. Water and a standard diet for laboratory rodents (MF, Oriental Yeast, Tokyo, Japan) were available ad libitum. All animal experiments were conducted in accordance with the Guidelines for Animal Experiments of the Faculty of Agriculture at Kyushu University, as well as the Law (No. 105) and Notification (No. 6) of the Japanese Government. All experiments were approved by Animal Care and Use Committee of Kyushu University under permission number A24-044-0. 2.2. Experiment 1: Effect of bright light on behaviors and brain 5-HT contents and uptake To clarify the effect of photoperiod and bright light on depression- and anxiety-like behaviors and on the brain serotonergic system, mice were randomly divided into 3 weight-matched groups (n = 12–13) after acclimation. The first and second groups were maintained under SD and LD for 3 weeks, respectively. The third group was maintained under SD for 2 weeks, followed by bright light treatment (LP) under SD for 1 week. Bright light treatment involved daily exposure to 1000 lx light (LDA4N-H-G570, light ranging from 400 to 700 nm, peak: 440–460 nm, Asahi Electric Corporation) under SD at Zeitgeber time (ZT, ZT0 represents light onset) 1–2 with timing controlled by an automatic scheduler mod-

ule attached to the light bulb. Next, behavioral tests were started. Mice were maintained under SD, LD, and LP conditions until the end of the experiment including behavioral test periods. For behavioral testing, the open field test (OFT, a test for spontaneous activity in a novel environment and anxiety-like behavior) and the FST (test for depression-like behavior) were performed 2 days apart. Tests were performed during light periods, i.e., ZT2.5–4, under white light (5 lx). Four days after the FST, mice were euthanized by decapitation under deep anesthesia with isoflurane gas, and the amygdala samples were dissected at ZT2, 10, and 18 (n = 4–5) for analysis of levels of 5-HT and its major metabolite, 5-hydroxyindoleacetic acid (5-HIAA), by high performance liquid chromatography (HPLC). Euthanasia during the dark phase was performed under dim red light. Analysis was performed using the amygdala samples, because (1) in humans, seasonal variations in 5-HT transporter binding are reported in a corticolimbic circuit comprising the amygdala and medial prefrontal cortex (Praschak-Rieder et al., 2008), (2) brightlight intervention negatively affected threat-related amygdala and prefrontal reactivity in humans (Fisher et al., 2014), and (3) our previous study using mice showed that the photoperiod regulates 5-HT and 5-HIAA levels in the amygdala, but not in other brain regions such as the hypothalamus (Otsuka et al., 2014). Another batch of animals was maintained under LD, SD, and LP (n = 9–11) in the same way as above, and used for the saccharin intake test (a test for anhedonia/depression-like behavior) during the dark period (ZT22-23). This timing was selected based on a previous study that showed photoperiodic changes in the intake of sweet solutions without changes in food and water intake (Otsuka et al., 2014). For the 5-HT uptake assay, mice maintained under LD, SD, and LP were decapitated under deep anesthesia with isoflurane gas at ZT2, 6, and 10 (n = 4), and tissue samples of the amygdala (two samples per animal, from left and right hemispheres) were punched out from 0.5-mm-thick brain slices using a needle of 2.2 mm diameter. These timings were selected to determine temporal 5-HT uptake after bright light treatment. 2.3. Experiment 2: Effect of bright light on the rhythms of wheel-running activity and core body temperature To determine the effect of photoperiod and bright light on the phase relationship between rhythms of wheel-running activity and core body temperature, mice were intraperitoneally implanted with thermo loggers (Thermochron SL, KN Laboratories, Osaka, Japan) under anesthesia with isoflurane, and individually housed in a cage equipped with a running wheel. Temporal patterns of wheel-running activity were measured using a computer system (Chronobiology Kit, Stanford Software Systems, Palo Alto, CA). They were maintained under LD, SD, or LP (n = 8) for 3 weeks, and their rhythm profiles of wheel-running activity and core body temperature were analyzed during the last 7 days. ClockLab software (Actimetrics, Evanston, IL) was used to determine the rhythm profiles. Dark onset in each lighting condition was aligned to compare the rhythm profiles between lighting conditions. 2.4. Experiment 3: Effect of bright light and l-serine on behaviors and 5-HT neurons This experiment was conducted to analyze the synergistic effect of bright light and l-serine administration. After acclimation under SD for 3 weeks, mice were randomly divided into 4 weight-matched groups (n = 8–10). The first and second groups were maintained under SD with a daily subcutaneous (s.c.) injection of saline (vehicle) or l-serine (5 mmol/kg), respectively, 15 min prior to ZT1. The third and fourth groups were exposed to LP (1000 lx light at ZT1-2 under SD) with a daily s.c. injection of saline or l-serine,

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

respectively, as described above. These treatments continued until the end of the experiment. Two weeks after the onset of the s.c. injection/light pulse, behavioral tests were started. OFT and FST were performed 2 days apart during light periods, i.e., ZT2.5–4 under white light (5 lx). Four to five days after the FST, mice were deeply anesthetized with 2–4% isoflurane gas by using inhalational anesthesia system (NARCOBIT-E(II), Natsume, Tokyo, Japan) and fixed by transcardial perfusion with 4% paraformaldehyde in 0.1 M phosphate-buffered saline starting at ZT3 and 10 (n = 4–5) for immunofluorescence. Transcardial perfusion during the dark phase was performed under dim red light. In this experiment, we analyzed immunofluorescence of 5-HT and c-Fos in the raphe nuclei that are the major source of 5-HT innervation in the forebrain.

2.5. Behavioral tests The OFT was performed using an apparatus consisting of a black square base (40 × 40 cm) with walls 40 cm high. At the beginning of the test, a mouse was placed in the center of the apparatus and then allowed to move freely for 5 min. Open field behavior was analyzed with ANY-maze Software (Stoelting Co., Wood Dale, IL) by dividing the field into 25 squares (5 × 5 grid). The number of grid lines crossed was used for the evaluation of spontaneous activity under novel conditions, and the time spent in the central nine grids was used for assessing anxiety-like behavior. In the FST, mice were individually placed into plastic cylinders (27 cm high, 17 cm diameter) containing water 14.5 cm deep, maintained at 25 ± 1 ◦ C, for 7 min. Immobility time during the last 5 min was blindly analyzed. Mice were considered to be immobile when they floated in an upright position and made only small movements to keep their heads above water.

27

In the saccharin intake test, the mice were deprived of food, but not water, from ZT6 of the test day. The mice were transferred into individual cages 10 min prior to ZT22 under dim red light, and the 0.1% saccharin solution was administered with an additional bottle of water during ZT22-23 under darkness. On the next day, only a bottle of water was given in order to determine basal water intake, which was not affected by photoperiod or bright light (data not shown).

2.6. HPLC Tissue samples from the amygdala were punched out from 1.5mm-thick brain slices using a needle of 2.2 mm diameter, and frozen with liquid nitrogen. Samples were stored at −80 ◦ C and analyzed within 2 weeks of collection. The tissue samples were homogenized in ice-cold 0.2 M perchloric acid solution containing 0.01 mM EDTA·2Na and left for deproteinization on ice for 30 min. Then, they were centrifuged at 20,000 g for 15 min at 0 ◦ C. Supernatants were adjusted to pH 3 with 1 M sodium acetate and were filtrated using a 0.22 ␮m centrifugal filter unit (Millipore, Bedford, MA). Tissue concentrations of 5-HT and its metabolite 5-HIAA were analyzed using a HPLC system (Eicom, Kyoto, Japan) with a 150 × 3.0 mm octadecyl silane column (SC-50DS, Eicom) and an electrochemical detector (ECD-300, Eicom) at an applied potential of 0.85 V with an Ag/AgCl reference analytical electrode. The mobile phase, at pH 3.5, consisted of 0.1 M sodium acetate, 0.1 M citric acid, 1-octane sulfonate, and EDTA·2Na (5 mg/ml). The retention time and height of the peaks in tissue homogenates were measured and compared to samples of the external calibrating standard solution. The protein content of each tissue was determined using the

Fig. 1. Effects of photoperiod and bright light treatment on anxiety- and depression-like behaviors and the brain serotonergic system in C57BL/6J mice. Number of grid lines crossed (A) and time spent in centre area (B) in the open field test, immobility time (C) in the forced swimming test, and intake of saccharin solution (D) in the saccharin intake test were analyzed under short-day condition (SD, 8L16D, 5 lx light phase), long-day condition (LD, 16L8D, 100 lx light phase), or bright light treatment [LP, daily 1000 lx light at Zeitgeber time (ZT) 1–2 under SD] (mean + SEM, n = 9–13). Contents of serotonin (5-HT, E) and its major metabolite 5-hydroxyindoleacetic acid (5-HIAA, F) and selective [3 H]-5-HT uptake (G) in the amygdala were analyzed at indicated timings (mean + SEM, n = 4–5). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.001, Bonferroni multiple comparison test.

28

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

Fig. 2. Effects of photoperiod and bright light treatment on the temporal profiles of wheel-running activity and core body temperature in C57BL/6J mice. Mice were maintained under short-day condition (SD, 8L16D, 5 lx light phase), long-day condition (LD, 16L8D, 100 lx light phase), or bright light treatment [LP, daily 1000 lx light at Zeitgeber time (ZT) 1–2 under SD]. Profiles of wheel-running activity (A) and core body temperature (B) were shown (mean + SEM, n = 8). Means of both profiles in LD (C), SD (D), and LP (E) were separately superimposed. Black and white bars below the graphs indicate light and dark periods, respectively. Shadows in C–E indicate dark periods. White rectangles on the bars of LP indicate the timing of bright light treatment.

bicinchoninic acid method to normalize the data. The data were expressed as ng/mg protein. 2.7. 5-HT uptake assay The amygdala samples were preincubated with artificial cerebrospinal fluid (124 mM NaCl, 3 mM KCl, 26 mM NaHCO3 , 2 mM CaCl2 , 1 mM MgSO4 , 1.25 mM KH2 PO4 , and 10 mM glucose, gassed for more than 30 min) at 37 ◦ C for 5 min. Then, [3 H]-5-HT (final concentration 50 nM, NET498, PerkinElmer, Waltham, MA) was added to the artificial cerebrospinal fluid containing amygdala from the left hemisphere, and incubated at 37 ◦ C for 10 min. Radioactivity in the tissue was counted using a liquid scintillation counter. To determine the nonspecific [3 H]-5-HT uptake, a selective 5-HT reuptake inhibitor, paroxetine (final concentration 1 ␮M) was added to the artificial cerebrospinal fluid containing the amygdala from right hemisphere and incubated for 15 min, and the [3 H]-5-HT uptake assay was subsequently performed as described earlier. Specific [3 H]-5-HT uptake was calculated by subtracting the nonspecific uptake from the total uptake. 2.8. Immunofluorescence After transcardial perfusion, dissected brains were immersed for 4 h in the same fixative, cryoprotected by 20% sucrose, and cut into coronal frozen sections of raphe nuclei (20 ␮m thick) using a cryostat. They were incubated in 10% normal donkey serum for 30 min and then with rabbit polyclonal anti-c-Fos (1:10,000, F7799, Sigma–Aldrich, St. Louis, MO) and

guinea pig polyclonal anti-5-HT (1:1000, Protos Biotech Corporation, NY) overnight at room temperature. Immunosignals for c-Fos were detected by Alexa-Fluor-488 donkey anti-rabbit IgG (1:800, Jackson ImmunoResearch, West Grove, PA) and for 5HT were detected by Cy3-conjugated donkey anti-guinea pig IgG (1:400, Jackson ImmunoResearch). Slides were observed using a fluorescence microscopy system (DMI 6000B, Leica, Wetzler, Germany) equipped with bandpass filters (excitation, 480/40 nm and 560/40 nm; emission, 527/30 nm and 645/75 nm, respectively). As a negative control for immunostaining, primary antibodies for 5-HT and c-Fos were omitted from the first incubation and slices were just incubated with secondary antibodies as described above. No signals were observed in these sections. Positive 5-HT staining was used as regional markers. Since 5-HTpositive cells were concentrated on the rostral to middle region of the raphe nuclei, these regions were analyzed. The raphe nuclei were divided into distinguishable subregions, dorsomedial dorsal raphe nuclei (DM), ventromedial dorsal raphe nuclei (VM), lateral dorsal raphe nuclei (L), and median raphe nuclei (MR). Four images containing the rostral to middle region of the raphe nuclei were analyzed in each animal. The representative staining of 5-HT was used to delineate the borders of subregions of the raphe nuclei. Boundaries of the subregions were drawn on transparent films and superimposed on the captured images to count immunoreactive cells. In each subregion, the number of total c-Fos-immunoreactive cells, total 5-HT-immunoreactive cells, and c-Fos and 5-HT double immunolabeled cells were manually counted using an image analyzing system (ImageJ, National Institute of Health, Bethesda, MD) in a blind manner.

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

29

2.9. Statistics One-way ANOVA was used to analyze the effects of lighting conditions on behavioral parameters. Before performing ANOVA, homogeneity of variance was confirmed by the Brown–Forsythe test. Post hoc comparisons were performed with Bonferroni multiple comparison tests. Brain 5-HT and 5-HIAA levels, 5-HT reuptake, and temporal rhythms of wheel-running activity and core body temperature were analyzed using two-way ANOVA followed by Bonferroni multiple comparison tests. The effects of bright light and l-serine administration on behavioral parameters or on the number of immunoreactive cells were analyzed by one-way or two-way ANOVA followed by Dunnett’s test (comparison with vehicle-injected, no light-exposed group). Statistically significant differences were considered if p-values <0.05.

3. Results 3.1. Bright light treatment exerted antidepressant-like effects In Experiment 1, the number of grid lines crossed or time spent in the central area in the OFT were unaffected by photoperiod or bright light treatment (p > 0.05) (Fig. 1A and B). Immobility times in the FST were significantly modified by lighting conditions (p < 0.0001). Post hoc analysis revealed that immobility times in the mice under SD were significantly higher than those in mice under LD (p < 0.0001). Bright light treatment significantly lowered the immobility times compared to mice under SD (p < 0.01) (Fig. 1C), although they did not reach those of mice under LD (Fig. 1C). Intake of saccharin solution was significantly modified by lighting conditions (p < 0.01); the intake was significantly lower in the mice under SD compared to those under LD (p < 0.01) (Fig. 1D). Mice exposed to bright light treatment exhibited significantly higher intake of saccharin solution compared to mice under SD (p < 0.05) (Fig. 1D). 3.2. Bright light treatment did not modify 5-HT content and selective 5-HT uptake in amygdala Lighting condition significantly affected 5-HT levels in the amygdala (p = 0.0009) with a significant interaction with time (p = 0.018). 5-HT levels at ZT2 and 10 were significantly higher in mice under LD compared to mice under SD and/or LP (ZT2: LD vs. SD, p < 0.01; LD vs. LP, p < 0.001, ZT10: LD vs. LP, p < 0.05). That is, mice under LD exhibited specific patterns of 5-HT levels among three groups (Fig. 1E), and light treatment did not influence them. Levels of 5-HIAA in the amygdala were significantly affected by lighting condition (p = 0.021) without a significant interaction with time (Fig. 1F). Photoperiod or bright light treatment did not affect selective [3 H]-5-HT uptake (Fig. 1G). 3.3. Bright light treatment had little effect on rhythms of wheel-running activity and core body temperature In Experiment 2, temporal profiles of wheel-running activity (Fig. 2A) and core body temperature (Fig. 2B) were significantly influenced by lighting condition (wheel-running activity: p = 0.029, core body temperature: p = 0.0001) with a significant interaction with time in wheel-running activity (p = 0.0006). Temporal patterns of these parameters around dark onset, i.e., timing of wheelrunning activity onset, were similar between conditions, while small peaks were aligned to dark offset in each photoperiod (Fig. 2A and B). Bright light treatment under SD did not induce a major impact on the temporal profiles of these parameters (Fig. 2A and B). Phase-relationship between the wheel-running activity and core

Fig. 3. Combined effect of bright light and l-serine injection on anxiety- and depression-like behaviors in C57BL/6J mice. Mice were maintained under the shortday condition (8L16D, 5 lx light phase) with or without bright light treatment [daily 1000 lx light at Zeitgeber time (ZT) 1–2]. In each condition, mice were subcutaneously injected with vehicle (Veh, saline) or l-serine (Ser, 5 mmol/kg) 15 min prior to ZT1. Number of grid lines crossed (A) and time spent in centre area (B) in the open field test, and immobility time (C) in the forced swimming test were analyzed (mean + SEM, n = 8–10). *p < 0.05, ***p < 0.001, Dunnett’s test vs. vehicle-treated, no light-group.

body temperature was similar between the mice under LD, SD, and LP (Fig. 2C–E). 3.4. l-serine administration prior to bright light elicited a strong antidepressant-like effect In Experiment 3, bright light treatment or l-serine administration did not influence the number of grid lines crossed or the time spent in the central area in the OFT (Fig. 3A and B). In the FST, the mice that had bright light treatment significantly reduced their immobility time compared to the vehicle-injected, no lightexposed mice (p < 0.05) (Fig. 3C). This effect was potentiated by a combination of bright light and l-serine administration (p < 0.001) (Fig. 3C). l-serine administration without bright light treatment had no significant effect on their immobility time (p > 0.05).

30

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

Fig. 4. Localization of serotonin (5-HT) and c-Fos immunoreactivities in the raphe nuclei of C57BL/6J mice. The raphe nuclei were divided into subregions, dorsomedial dorsal raphe nuclei (DM), ventromedial dorsal raphe nuclei (VM), lateral dorsal raphe nuclei (L), and median raphe nuclei (MR) (A), based on the localization of 5-HT immunolabeling (B and C). Representative images for double immunolabeling in the DM show clear signals for 5-HT (D, red), c-Fos (E, green), and their colocalization (F, arrowheads). There are also the cells with c-Fos immunosignals without 5-HT staining (F, arrows). Scale bars, B, C: 250 ␮m, D–F, 100 ␮m.

3.5. Bright light induced serotonergic signals and l-serine potentiated its effect Immunosignals for 5-HT were concentrated in the rostral to middle region of the raphe nuclei, which could be distinguished to four subregions, DM, VM, L, and MR (Fig. 4A–C). Double immunofluorescence of 5-HT and c-Fos showed several 5-HT-positive cells that had nuclear staining of c-Fos (Fig. 4D–F, representative images of DM, colocalization of 5-HT and c-Fos is shown by arrowheads in Fig. 4F). There were also several cells that had 5-HT-negative, c-Fos-positive signals (Fig. 4F, arrows). The total number of 5-HT- or c-Fos-immunoreactive cells and their colocalized cells after bright light treatment and/or l-serine administration were counted in each subregion at ZT3 and 10 (2 and 9 h after the start of the 1 h bright light pulse, respectively). The results are shown in Fig. 5. A bright light and vehicle injection induced 5-HT immunoreactive cells at ZT3 in the VM and L (p < 0.05). l-serine injection prior to the bright light significantly enhanced these effects; 5-HT-positive cells were strongly increased at ZT3 in the DM (p < 0.05), VM (p < 0.001), L (p < 0.001), and MR (p < 0.05), and at ZT10 in the DM (p < 0.05) and L (p < 0.05). lserine injection without bright light decreased the 5-HT-positive cell numbers in the DM (p < 0.05) and VM (p < 0.01) at ZT10. The total number of c-Fos-immunoreactive cells significantly increased after bright light treatment, regardless of l-serine injection, at ZT10 in the VM (p < 0.01) and MR (vehicle: p < 0.01; l-serine: p < 0.05). As a result of bright light with vehicle injection, the number of doublelabeled cells was significantly increased at ZT3 in the DM (p < 0.01) and L (p < 0.05), and at ZT10 in all regions examined (DM and L, p < 0.05; VM and MR, p < 0.001). Bright light with l-serine injection also increased the number of double-labeled cells with slightly different dynamics compared to vehicle-injected animals; the number of double-labeled cells was increased at ZT3 in the VM (p < 0.01) and at ZT10 in all regions examined (DM and L, p < 0.001; VM and MR, p < 0.01).

4. Discussion C57BL/6J maintained under SD showed high immobility in the FST and low intake of saccharin solution, i.e., high depression-like behaviors, compared to those under LD; results consistent with our previous study (Otsuka et al., 2014). Bright light treatment successfully reduced the levels of immobility and increased the saccharin intake under SD in the present study. These changes might not be a consequence of the changes in spontaneous activity or anxietylike behaviors, since parameters in the OFT were not modified by the lighting conditions. As the depressive symptoms of many SAD patients are ameliorated by bright light therapy (Lewy et al., 1982; Rosenthal et al., 1984; Terman et al., 1989; Oldham and Ciraulo, 2014), our data may suggest the predictive validity of C57BL/6J mice as an animal model of SAD. Although C57BL/6J mice cannot produce detectable levels of melatonin, an established photoperiodic messenger, our present and previous studies (Otsuka et al., 2012, 2014) suggest that stressand mood-related behavior and physiology are regulated by photoperiod via a melatonin-independent pathway. Furthermore, the antidepressant effects of light therapy in SAD patients appear to be independent of light-induced melatonin suppression (Rosenthal et al., 1986). There could also be an argument regarding the nocturnal habits of mice that differs from the diurnal habits of humans; nocturnal mice have a longer active phase under SD, whereas humans have a shorter active phase during winter. These differences also affect the sleep duration and quality (Wehr et al., 1993), which can modify mood-related behaviors. In fact, deprivation of rapid eye movement sleep decreases immobility in the FST in rats (Porsolt et al., 1978). Although we cannot exclude the possibility that our data resulted from secondary effects of sleep changes, the effects of photoperiod or bright light on mood-related behavior appear to be, at least in part, common between nocturnal and diurnal species, as our data are compatible with an antidepressant-like

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

31

Fig. 5. Activation of serotonergic signaling in the raphe nuclei by a combination of bright light and l-serine injection in C57BL/6J mice. Mice were maintained under short-day condition (8L16D, 5 lx light phase) with or without bright light treatment [daily 1000 lx light at Zeitgeber time (ZT) 1–2]. In each condition, mice were subcutaneously injected with vehicle (Veh, saline) or l-serine (Ser, 5 mmol/kg) 15 min prior to ZT1. Total numbers of 5-HT or c-Fos-immunopositive cells and colocalized cells of them were counted in dorsomedial dorsal raphe nuclei (DM), ventromedial dorsal raphe nuclei (VM), lateral dorsal raphe nuclei (L), and median raphe nuclei (MR). *p < 0.05, **p < 0.01, ***p < 0.001, Dunnett’s test vs. vehicle-treated, no light-group. Mean + SEM, n = 4–5.

effect of bright light in the diurnal fat sand rat (Ashkenazy et al., 2009). The brain serotonergic system is deeply involved in the pathogenesis of SAD (Gupta et al., 2013), given the abundant evidence, such as seasonal variations of 5-HT content in human post-mortem brain specimens (Carlsson et al., 1980), 5-HIAA levels in the cerebrospinal fluid in normal volunteers (Brewerton et al., 1988), 5-HT turnover in the brain (Lambert et al., 2002), and 5-HT transporter binding in the living human brain (Praschak-Rieder et al., 2008). The present study confirmed our previous finding that 5-HT content

in the amygdala was higher under LD than under SD in C57BL/6J mice (Otsuka et al., 2014), although present and previous studies did not identify 5-HT release in the synaptic cleft that is an essential parameter for neurotransmission. However, bright light treatment under SD did not mimic the effect of LD in the present study. These data suggest that bright light treatment may elicit an antidepressant-like effect through a specific pathway that is different from LD-induced processes. Clinical studies have suggested two possibilities for the mechanisms of light therapy: (1) regulation of 5-HT transporter function and (2) involvement of a circadian phase

32

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33

shift. The first possibility is supported by the correlation between light therapy-induced remissions from depression and normalizing 5-HT transporter functions that are in a hyperfunctional state during depression (Willeit et al., 2008). However, in the present study, photoperiod and bright light did not significantly influence the selective [3 H]-uptake in the amygdala of C57BL/6J mice. The second possibility is based on the theory that bright light resets the dysynchronization of the internal clock in SAD patients (Lewy et al., 1987). In the present study, bright light had little impact on the rhythm profiles and phase relationship between wheel-running activity and core body temperature rhythms. These observations are consistent with those of a study using grass rats that showed that bright light intensity during light phase under 12L12D has no major effect on temporal patterns of locomotor activity (Leach et al., 2013). Light has multiple impacts on the raphe nuclei, including the 5-HT systems (Gonzalez and Aston-Jones, 2008). In the present study, l-serine administration prior to the bright light exposure enhanced the effect of the bright light in the FST, while l-serine alone had little influence. These effects appear to be independent of spontaneous activity or anxiety-like behavior, as parameters in the OFT were unaffected by light and l-serine administration. These data suggest that l-serine and/or its metabolites potentiate bright light-induced signal transduction in the brain. This hypothesis is supported by the observation that the number of 5-HT immunoreactive cells was increased in the VM and L by bright light, and this effect was potentiated by l-serine injection. Further, bright light treatment activated 5-HT neurons in all subregions examined, as evaluated by 5-HT and c-Fos double immunolabeling, and the activation in the VM was enhanced by l-serine administration. VM, L, and DM consist of dorsal raphe nuclei that heavily innervate the brain areas involved in mood-related behaviors (Hensler, 2006). These data suggest that bright light might exert antidepressant-like effect through altered 5-HT signaling in the dorsal raphe nuclei. Our data are consistent with the results found in grass rats that bright light intensity during light phase is associated with a higher number of 5-HT-immunoreactive cells in the dorsal raphe nuclei (Leach et al., 2013). We also detected light-induced increases in the total number of c-Fos-immunopositive cells in several subregions, which is also in line with reports using grass rats (Adidharma et al., 2012) and Mongolian gerbils (Fite et al., 2005). However, the activated neurons in grass rat appear to be non-5-HT neurons (Adidharma et al., 2012). This discrepancy may be a result of a species-specific neuronal network for signaling from the retina to raphe nuclei; a direct retinal projection is present in several species including Mongolian gerbils (Fite et al., 1999), whereas it is not detected in mice (Hattar et al., 2006). The glutamatergic pathway plays a pivotal role in the neuromodulation of 5-HT in the dorsal raphe nuclei (Soiza-Reilly and Commons, 2011). Because administration of l-serine, which can be converted into d-serine, a coagonist of NMDA receptors, potentiated light-induced 5-HT signaling, the NMDA pathway may be involved in the light-induced signal transduction cascade within the dorsal raphe nuclei. In line with this notion, l-serine alone, without bright light, had little effect on 5-HT immunosignals or neuronal activity. Alternatively, the effects of light and l-serine administration may be integrated in the retinas and transmitted into the raphe nuclei, because d-serine can enhance excitatory currents elicited by NMDA or NMDA receptor-mediated component of light-evoked synaptic responses (Stevens et al., 2003). In addition to bright light, which has been established as a regulator of brain functions (Pail et al., 2011), l-serine has been implicated in the regulation of stress-related functions in experimental animals (Shigemi et al., 2008; Nagasawa et al., 2012). Our study clarified that l-serine administration in combination with bright light might elicit a strong antidepressant-like effect with

enhanced 5-HT signals in mice. This finding provides a novel strategy of therapeutics for psychiatric diseases including SAD, i.e., optimization of treatment efficiency by utilizing the synergistic effect of a functional nutrient and light therapy. In conclusion, we found that bright light treatment in C57BL/6J mice under SD might induce an antidepressant-like effect that is potentiated by prior injection of l-serine. Bright light and l-serine appear to increase 5-HT signals and neuronal activation in the raphe nuclei. These data suggest the predictive validity of C57BL/6J mice as an animal model of SAD, providing a nutritional enhancer of bright light treatment. Acknowledgements We thank Dr. Ryuichi Tatsumi and Dr. Wataru Mizunoya for the use of fluorescent microscopy system. This work was supported by Grants-in-Aid for Young Scientists (B) (No. 24780286) to S.Y., Challenging Exploratory Research (No. 24650490) and Scientific Research (A) (No. 23248046) to M.F. from the Japanese Society for the Promotion of Science. References Adidharma, W., Leach, G., Yan, L., 2012. Orexinergic signaling mediates light-induced neuronal activation in the dorsal raphe nucleus. Neuroscience 220, 201–207. Albrecht, U., 2012. Timing to perfection: the biology of central and peripheral circadian clocks. Neuron 74, 246–260. Ashkenazy, T., Einat, H., Kronfeld-Schor, N., 2009. Effects of bright light treatment on depression- and anxiety-like behaviors of diurnal rodents maintained on a short daylight schedule. Behav. Brain Res. 201, 343–346. Brewerton, T.D., Berrettini, W.H., Nurnberger Jr., J.I., Linnoila, M., 1988. Analysis of seasonal fluctuations of CSF monoamine metabolites and neuropeptides in normal controls: findings with 5HIAA and HVA. Psychiatry Res. 23, 257–265. Carlsson, A., Svennerholm, L., Winblad, B., 1980. Seasonal and circadian monoamine variations in human brains examined post mortem. Acta Psychiatr. Scand. Suppl. 280, 75–85. de Kock, C.P., Cornelisse, L.N., Burnashev, N., Lodder, J.C., Timmerman, A.J., Couey, J.J., Mansvelder, H.D., Brussaard, A.B., 2006. NMDA receptors trigger neurosecretion of 5-HT within dorsal raphe nucleus of the rat in the absence of action potential firing. J. Physiol. 577, 891–905. Einat, H., Kronfeld-Schor, N., Eilam, D., 2006. Sand rats see the light: short photoperiod induces a depression-like response in a diurnal rodent. Behav. Brain Res. 173, 153–157. Fisher, P.M., Madsen, M.K., Mc Mahon, B., Holst, K.K., Andersen, S.B., Laursen, H.R., Hasholt, L.F., Siebner, H.R., Knudsen, G.M., 2014. Three-week bright-light intervention has dose-related effects on threat-related corticolimbic reactivity and functional coupling. Biol. Psychiatry 76, 332–339. Fite, K.V., Janusonis, S., Foote, W., Bengston, L., 1999. Retinal afferents to the dorsal raphe nucleus in rats and Mongolian gerbils. J. Comp. Neurol. 414, 469–484. Fite, K.V., Wu, P.S., Bellemer, A., 2005. Photostimulation alters c-Fos expression in the dorsal raphe nucleus. Brain Res. 1031, 245–252. Gonzalez, M.M., Aston-Jones, G., 2008. Light deprivation damages monoamine neurons and produces a depressive behavioral phenotype in rats. Proc. Natl. Acad. Sci. U. S. A. 105, 4898–4903. Gupta, A., Sharma, P.K., Garg, V.K., Singh, A.K., Mondal, S.C., 2013. Role of serotonin in seasonal affective disorder. Eur. Rev. Med. Pharmacol. Sci. 17, 49–55. Hattar, S., Kumar, M., Park, A., Tong, P., Tung, J., Yau, K.W., Berson, D.M., 2006. Central projections of melanopsin-expressing retinal ganglion cells in the mouse. J. Comp. Neurol. 497, 326–349. Hensler, J.G., 2006. Serotonergic modulation of the limbic system. Neurosci. Biobehav. Rev. 30, 203–214. Lambert, G.W., Reid, C., Kaye, D.M., Jennings, G.L., Esler, M.D., 2002. Effect of sunlight and season on serotonin turnover in the brain. Lancet 360, 1840–1842. Leach, G., Adidharma, W., Yan, L., 2013. Depression-like responses induced by daytime light deficiency in the diurnal grass rat (Arvicanthis niloticus). PLoS One 8, e57115. Lewy, A.J., Kern, H.A., Rosenthal, N.E., Wehr, T.A., 1982. Bright artificial light treatment of a manic-depressive patient with a seasonal mood cycle. Am. J. Psychiatry 139, 1496–1498. Lewy, A.J., Sack, R.L., Miller, L.S., Hoban, T.M., 1987. Antidepressant and circadian phase-shifting effects of light. Science 235, 352–354. Nagasawa, M., Ogino, Y., Kurata, K., Otsuka, T., Yoshida, J., Tomonaga, S., Furuse, M., 2012. Hypothesis with abnormal amino acid metabolism in depression and stress vulnerability in Wistar Kyoto rats. Amino Acids 43, 2101–2111. Oldham, M.A., Ciraulo, D.A., 2014. Bright light therapy for depression: a review of its effects on chronobiology and the autonomic nervous system. Chronobiol. Int. 31, 305–319.

M. Kawai et al. / Brain Research Bulletin 118 (2015) 25–33 Otsuka, T., Goto, M., Kawai, M., Togo, Y., Sato, K., Katoh, K., Furuse, M., Yasuo, S., 2012. Photoperiod regulates corticosterone rhythms by altered adrenal sensitivity via melatonin-independent mechanisms in Fischer 344 rats and C57BL/6J mice. PLoS One 7, e39090. Otsuka, T., Kawai, M., Togo, Y., Goda, R., Kawase, T., Matsuo, H., Iwamoto, A., Nagasawa, M., Furuse, M., Yasuo, S., 2014. Photoperiodic responses of depression-like behavior, the brain serotonergic system, and peripheral metabolism in laboratory mice. Psychoneuroendocrinology 40, 37–47. Pail, G., Huf, W., Pjrek, E., Winkler, D., Willeit, M., Praschak-Rieder, N., Kasper, S., 2011. Bright-light therapy in the treatment of mood disorders. Neuropsychobiology 64, 152–162. Porsolt, R.D., Anton, G., Blavet, N., Jalfre, M., 1978. Behavioural despair in rats: a new model sensitive to antidepressant treatments. Eur. J. Pharmacol. 47, 379–391. Praschak-Rieder, N., Willeit, M., Wilson, A.A., Houle, S., Meyer, J.H., 2008. Seasonal variation in human brain serotonin transporter binding. Arch. Gen. Psychiatry 65, 1072–1078. Prendergast, B.J., Nelson, R.J., 2005. Affective responses to changes in day length in Siberian hamsters (Phodopus sungorus). Psychoneuroendocrinology 30, 438–452. Rosenthal, N.E., Sack, D.A., Gillin, J.C., Lewy, A.J., Goodwin, F.K., Davenport, Y., Mueller, P.S., Newsome, D.A., Wehr, T.A., 1984. Seasonal affective disorder: a description of the syndrome and preliminary findings with light therapy. Arch. Gen. Psychiatry 41, 72–80.

33

Rosenthal, N.E., Sack, D.A., Jacobsen, F.M., James, S.P., Parry, B.L., Arendt, J., Tamarkin, L., Wehr, T.A., 1986. Melatonin in seasonal affective disorder and phototherapy. J. Neural. Transm. Suppl. 21, 257–267. Shigemi, K., Tsuneyoshi, Y., Hamasu, K., Han, L., Hayamizu, K., Denbow, D.M., Furuse, M., 2008. l-Serine induces sedative and hypnotic effects acting at GABA(A) receptors in neonatal chicks. Eur. J. Pharmacol. 599, 86–90. Soiza-Reilly, M., Commons, K.G., 2011. Glutamatergic drive of the dorsal raphe nucleus. J. Chem. Neuroanat. 41, 247–255. Stevens, E.R., Esguerra, M., Kim, P.M., Newman, E.A., Snyder, S.H., Zahs, K.R., Miller, R.F., 2003. d-serine and serine racemase are present in the vertebrate retina and contribute to the physiological activation of NMDA receptors. Proc. Natl. Acad. Sci. U. S. A. 100, 6789–6794. Terman, M., Terman, J.S., Quitkin, F.M., McGrath, P.J., Stewart, J.W., Rafferty, B., 1989. Light therapy for seasonal affective disorder. A review of efficacy. Neuropsychopharmacology 2, 1–22. Wehr, T.A., Moul, D.E., Barbato, G., Giesen, H.A., Seidel, J.A., Barker, C., Bender, C., 1993. Conservation of photoperiod-responsive mechanisms in humans. Am. J. Physiol. 265, R846–857. Willeit, M., Sitte, H.H., Thierry, N., Michalek, K., Praschak-Rieder, N., Zill, P., Winkler, D., Brannath, W., Fischer, M.B., Bondy, B., Kasper, S., Singer, E.A., 2008. Enhanced serotonin transporter function during depression in seasonal affective disorder. Neuropsychopharmacology 33, 1503–1513.