Electrophysiological and pharmacological properties of GABAergic cells in the dorsal raphe nucleus

The dorsal raphe nucleus (DRN) is the origin of the central serotonin [5-hydroxytryptamine (5-HT)] system and plays an important role in the regulation of many physiological functions such as sleep/arousal, food intake and mood. In order to understand the regulatory mechanisms of 5-HT system, characterization of the types of neurons is necessary. We performed electrophysiological recordings in acute slices of glutamate decarboxylase 67–green fluorescent protein knock-in mice. We utilized this mouse to identify visually GABAergic cells. Especially, we examined postsynaptic responses mediated by 5-HT receptors between GABAergic and serotonergic cells in the DRN. Various current responses were elicited by 5-HT and 5-HT1A or 5-HT2A/2C receptor agonists in GABAergic cells. These results suggested that multiple 5-HT receptor subtypes overlap on GABAergic cells, and their combination might control 5-HT cells. Understanding the postsynaptic 5-HT feedback mechanisms may help to elucidate the 5-HT neurotransmitter system and develop novel therapeutic approaches.


Introduction
The dorsal raphe nucleus (DRN), which is the origin of the central serotonin (5-hydroxytryptamine; 5-HT) system, plays an important role in regulating many physiological functions, such as sleep/arousal, food intake, and mood. The DRN provides serotonergic innervation to several brain regions, including the cerebral cortex, basal forebrain, mesencephalon, hypothalamus, and thalamus, and it receives reciprocal inputs from these broad brain regions. The DRN is composed of heterogeneous neuronal groups that differ in cellular morphology, electrophysiological properties, and the expression of neurotransmitters, such as 5-HT, gamma-aminobutyric acid (GABA), and glutamate [1]. This heterogeneous organization may correspond to a variety of physiological functions that are involved in serotonergic neurotransmission and complicate the understanding of DRN functions. In addition to 5-HT cells, GABAergic cells consist of another major DRN cell group [1][2][3]. GABAergic cells function as interneurons in local circuits with serotonergic projection neurons and regulate their output [4]. Therefore, characterization of the properties of each cell type, especially of GABAergic cells, and their interactions may help to elucidate the various functions of the DRN.
To date, many investigations have described the properties of DRN 5-HT cells through in vivo or in vitro electrophysiological recordings with neurochemical identification [5][6][7][8]. However, these studies did not directly characterize the properties of DRN GABAergic cells but described GABAergic cells as putative non-5-HT cells that do not exhibit tryptophan hydroxylase (TPH) immunoreactivity. Since identifying GABAergic cells under electrophysiological recordings is difficult, the properties of DRN GABAergic cells have been poorly characterized.
In order to characterize GABAergic cells, we used a knockin mouse line in which the expression of green fluorescent protein (GFP) was controlled by the glutamate decarboxylase (GAD) 67 endogenous promoter in order to specifically visualize GABAergic cells [9]. This transgenic mouse line (GAD67 ?/GFP ) allows for the identification and characterization of GABAergic cells in the brain [10][11][12]. The electrophysiological and pharmacological properties of directly identified GABAergic cells in the DRN have not yet been characterized. Thus, this study aimed to characterize the differences in electrophysiological and pharmacological properties in GABAergic and non-GABAergic cells in the DRN of GAD67 ?/GFP mice with whole-cell patch-clamp recording techniques and immunohistochemistry. The properties of GABAergic cells differed from those of non-GABAergic cells in the DRN. Furthermore, the DRN GABAergic cells were heterogeneous in their postsynaptic responses to 5-HT and selective agonists of 5-HT 1A and 5-HT 2A/2C receptors.

Animals
The generation of GAD67-GFP knock-in mice have been described previously [9], and these mice, used in the present study, were termed GAD67 ?/GFP . The mice were maintained with a genetic background of C57BL/6 at our animal facility. In accordance with a protocol approved by the Ethics Review Committee of Nippon Medical School, we made efforts to minimize the number of animals used and their suffering.

Immunohistochemistry
Male heterozygous GAD67 ?/GFP mice aged 24 days (n = 2) were deeply anaesthetized with pentobarbital and transcardially perfused with freshly prepared 4 % paraformaldehyde in phosphate-buffered saline (PBS). The midbrain was removed, post-fixed at 4°C overnight, and cryoprotected in 20 % sucrose in PBS at 4°C overnight. The midbrain was cut into serial transverse cryosections with a cryostat (Leica, Tokyo, Japan). Immunohistochemistry was performed on free-floating sections, as previously described [13]. Briefly, transverse cryosections (20 lm) of the midbrain were incubated with a rabbit anti-GFP polyclonal antibody (1:2,000; EMD Millipore, MA, USA) and a sheep anti-TPH polyclonal antibody (1:2,000; Life Technologies, NY, USA) for 3 days at 4°C. The sections were then incubated with secondary antibodies labeled with Alexa Fluor 594 (1:2,000; Life Technologies) for GFP and Alexa Fluor 488 (1:2,000; Life Technologies) for TPH. After electrophysiological recording, slices were immersion-fixed overnight in 4 % paraformaldehyde prepared in 0.1 M PBS and then stored in PBS. Fixed sections were incubated with a sheep anti-TPH polyclonal antibody (1:1,000; Life Technologies) for 3 days at 4°C. Subsequently, these sections were incubated with an Alexa Fluor 488 (1:1,000)-conjugated donkey anti-sheep secondary antibody and streptavidin-conjugated Alexa Fluor 594 (1:1,000) to visualize the immunohistochemical labeling and biocytin. Images were captured with a high-resolution digital camera (Olympus, Tokyo, Japan).
Individual slices were transferred to a recording chamber attached to a microscope stage, continuously perfused with oxygenated ACSF at a flow rate of 1.5 mL/min, and maintained at room temperature (*27°C). Borosilicate glass-patch electrodes (World Precision Instruments, FL, USA) with a resistance of 6-12 MX when filled with an internal solution of 150 mM potassium methanesulfonate, 1.0 mM KCl, 0.2 mM K-EGTA, 20 mM HEPES, 3.0 mM MgATP 2 , and 0.4 mM Na-GTP (pH 7.38) were used for whole-cell recordings of DRN cells. Cells that were visualized under a blue light were considered GFP-positive (GFP(?)), and these were designated GABAergic cells. Cells without GFP were considered GFP-negative (GFP (-)), and these were designated non-GABAergic cells. Whole-cell patch-clamp recordings were acquired and controlled with Axon 700B Multiclamp amplifier and pClamp acquisition software (Molecular Devices, CA, USA). In the pharmacological experiments, the amplitudes of membrane current induced by serotonergic agonists were measured from the baseline before application of agonist to peak current amplitude of their responses.

Analysis and statistics
The data are presented as the mean ± standard error of the mean, and n represents the number of independent experiments. Statistical differences were evaluated with the Tukey-Kramer test. P values \0.05 were considered statistically significant.

Discussion
Utility of GAD67 ?/GFP mice Present study is the first that characterized differences in electrophysiological and pharmacological properties of GABAergic and non-GABAergic cells in the DRN by using GAD67 ?/GFP mice. To the best of our knowledge, there has been only 1 previous report about the electrophysiological properties of DRN GABAergic neurons [15].
Other studies that have used TPH immunoreactivity for cell typing have reported that the resting membrane potential, AP threshold, and input resistance did not statistically differ [6,7]. This was probably because the tested non-5-HT neurons potentially contained heterogeneous cells and were not a uniform population. In this study, we demonstrated that GABAergic cells and 5-HT cells within the DRN were localized in distinct subregions (Fig. 1), which was similar to the findings of a previous report [3]. Furthermore, we showed that GFP-negative and TPH-immunonegative small cells were located in the lateral DRN, which was similar to the localization of GFP(?) cells. These cells were putatively glutamatergic because the neuronal cell types in the DRN, other than 5-HT cells, are mainly glutamatergic or GABAergic [2,3,16]. However, the properties of DRN glutamatergic cells remain to be elucidated. Altogether, the identification of cell types with GAD67 ?/GFP mice enabled a clearer discrimination of the properties of DRN cells.

Intrinsic membrane properties of DRN GABAergic cells
The membrane properties of DRN GABAergic cells had distinctive features: more depolarized resting membrane potential, steeper AP threshold, and higher input resistance. These results suggested that GABAergic cells were easily excitable compared to 5-HT neurons. Moreover, the lower overshoot amplitude and fast AP kinetics may contribute to the high-frequency firing of up to *25 Hz in the GABAergic cells. In fact, the input-output relationships induced by positive currents were linear until 100 pA of injection current, and they exhibited frequency adaptation over 100 pA (Fig. 2d).

Physiological and pharmacological implications of DRN GABAergic cells
In the course of pharmacological experiments shown in Fig. 3, we could not exclude synaptic activity-driven indirect actions of serotonergic agonists. However, in most cases, spontaneous postsynaptic currents obtained from recorded cells were not altered by each agonist. Therefore, it is thought that the serotonergic agonist-induced membrane currents reflect direct action of the postsynaptic membrane.
It has been established that feedback inhibition through somatodendritic 5-HT 1A autoreceptors located on DRN 5-HT cells (Fig. 3a2, b2) is an important mechanism for controlling 5-HT neuronal activity [17,18]. Like 5-HT cells, GABAergic cells play an important role in controlling 5-HT neuronal activity through 5-HT receptor-mediated modulatory actions [18,19]. We found that 5-HT elicited excitatory responses in most GABAergic cells, while inhibitory actions were observed in 14.8 % of the GABAergic cells. However, these results cannot be explained simply by the activation of 5-HT 1A or 5-HT 2A/2C receptors in the subsequent experiments with 8-OH-DPAT or DOI. In the 8-OH-DPAT experiment, outward currents were observed in 50 % of the tested cells, which was a much higher cell population than that in the 5-HT challenge (14.8 %). However, it was comparable to immunohistochemical and in situ hybridization results that have reported that 40-50 and 10-15 %, respectively, of non5-HT neurons express the 5-HT 1A receptor [15,20]. Therefore, the discrepancy between the 5-HT and 8-OH-DPAT results can be explained by the effects of other 5-HT receptor subtypes, such as the 5-HT 2A/2C receptor, which may overwhelm the response of 5-HT 1A receptor activation. Previous studies have shown that the balance between the 5-HT 1A and 5-HT 2A/2C receptors is important in the neuronal regulation by 5-HT [8,21]. Moreover, the inward currents elicited by 8-OH-DPAT are probably due to the activation of the 5-HT 7 receptor as 8-OH-DPAT has a moderate affinity for 5-HT 7 receptors [22,23].
For the selective 5-HT 2A/2C receptor-agonist experiment, DOI elicited inward currents in only 50 % of the tested GABAergic neurons (Fig. 3a3). Hence, only half of the 5-HT-induced inward currents were mediated by 5-HT 2A/2C receptor activation in GABAergic cells. The remaining half might be accounted for by the activation of the 5-HT 7 receptor [18]. That is, in addition to the 5-HT 1A and 5-HT 2A/2C receptors, 5-HT 7 receptors might be implicated in the postsynaptic effects of DRN GABAergic neurons. Actually, the 5-HT 7 receptor, which is expressed in some GABAergic neurons, contributes to GABA release to regulated 5-HT neurons in the DRN [24,25].
In heterogeneous DRN neurons, competitive/synergistic actions of multiple 5-HT receptor subtypes induce distinct signaling properties and contribute to a diverse 5-HT neurotransmitter system. These complex regulations have been implicated in the etiology and treatment of many common psychiatric disorders. More extensive explorations of 5-HT receptor subtypes, including their properties, distributions, and interactions, are necessary for the development of novel therapeutic strategies that target 5-HT receptor subtypes.