Electrophysiological characteristics of inhibitory neurons of the prepositus hypoglossi nucleus as analyzed in Venus-expressing transgenic rats

Neuroscience 197 (2011) 89 –98

ELECTROPHYSIOLOGICAL CHARACTERISTICS OF INHIBITORY NEURONS OF THE PREPOSITUS HYPOGLOSSI NUCLEUS AS ANALYZED IN VENUS-EXPRESSING TRANSGENIC RATS M. SHINO,a,b1 R. KANEKO,a,c1 Y. YANAGAWA,a,d Y. KAWAGUCHId,e AND Y. SAITOa*

These results suggest that most inhibitory PHN neurons use either GABA or both GABA and glycine as neurotransmitters. Although the overall distribution of firing patterns in GABAergic neurons was similar to that of GABA&GLY neurons, only GABA&GLY neurons exhibited a firing pattern with a long first interspike interval. These differential electrophysiological properties will be useful for the identification of specific types of PHN neurons. © 2011 IBRO. Published by Elsevier Ltd. All rights reserved.

a

Department of Genetic and Behavioral Neuroscience, Gunma University Graduate School of Medicine, Maebashi, Gunma 371-8511, Japan b Department of Otolaryngology-Head and Neck Surgery, Gunma University Graduate School of Medicine, Maebashi, Gunma 371-8511, Japan c Institute of Experimental Animal Research, Gunma University Graduate School of Medicine, Maebashi, Gunma 371-8511, Japan

Key words: GABAergic, glycinergic, vesicular GABA transporter, afterhyperpolarization, firing pattern, whole-cell recording.

d

Japan Science and Technology Agency, Core Research for Evolutional Science and Technology (CREST), Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan e Division of Cerebral Circuitry, National Institute for Physiological Sciences, Okazaki, Aichi 444-8787, Japan

The identification and characterization of distinct neuronal populations, such as excitatory and inhibitory neurons, are significant steps toward understanding the overall structure and functions of particular brain regions. However, because excitatory and inhibitory neurons in the brainstem and spinal cord generally have intermingled distributions and heterogeneous morphological profiles, it is difficult to determine whether a particular neuron is excitatory or inhibitory, especially in living brain preparations. Therefore, specific features are needed to help with the identification of excitatory or inhibitory neurons. The prepositus hypoglossi nucleus (PHN) is a brainstem structure that is involved in the control of horizontal gaze holding (Büttner and Büttner-Ennever, 2006; McCrea and Horn, 2006). In the neural networks through the PHN, transient burst signals that are proportional to eye velocity are transformed into sustained signals that are proportional to eye position for gaze holdings (Robinson, 1975, 1989; Fukushima et al., 1992; Fukushima and Kaneko, 1995; Moschovakis, 1997). Because the transformation from velocity signals into position signals corresponds to a mathematical time integration, the PHN is regarded as an oculomotor neural integrator. In a previous study (Shino et al., 2008), we explored physiological markers for the identification of excitatory and inhibitory neurons in the PHN by investigating the relationship between intrinsic electrophysiological properties and neuronal phenotypes defined by neurotransmitter expression. Specifically, we studied afterhyperpolarization (AHP) profiles and firing patterns in response to current pulse injection using whole-cell patch clamp recordings in rat brainstem slice preparations in combination with reverse transcription-polymerase chain reaction (RT-PCR) analyses of mRNA expression. Using these techniques, we identified GABAergic and glutamatergic neurons in the PHN and found several GABAergic neuron-specific properties, such as an AHP without a slow

Abstract—The identification and characterization of excitatory and inhibitory neurons are significant steps in understanding neural network functions. In this study, we investigated the intrinsic electrophysiological properties of neurons in the prepositus hypoglossi nucleus (PHN), a brainstem structure that is involved in gaze holding, using whole-cell recordings in brainstem slices from vesicular GABA transporter (VGAT)-Venus transgenic rats, in which inhibitory neurons express the fluorescent protein Venus. To characterize the intrinsic properties of these neurons, we recorded afterhyperpolarization (AHP) profiles and firing patterns from Venus-expressing [Venus(ⴙ)] and Venus-non-expressing [Venus(ⴚ)] PHN neurons. Although both types of neurons showed a wide variety of AHP profiles and firing patterns, oscillatory firing was specific to Venus(ⴙ) neurons, while a firing pattern showing only a few spikes was specific to Venus(ⴚ) neurons. In addition, AHPs without a slow component and delayed spike generation were preferentially displayed by Venus(ⴙ) neurons, whereas a firing pattern with constant interspike intervals was preferentially displayed by Venus(ⴚ) neurons. We evaluated the mRNAs expression of glutamate decarboxylase (GAD65, GAD67) and glycine transporter 2 (GlyT2) to determine whether the recorded Venus(ⴙ) neurons were GABAergic or glycinergic. Of the 67 Venus(ⴙ) neurons tested, GlyT2 expression alone was detected in only one neuron. Approximately 40% (28/67) expressed GAD65 and/or GAD67 (GABAergic neuron), and the remainder (38/67) expressed both GAD(s) and GlyT2 (GABA&GLY neuron). 1 These authors contributed equally to this work. *Corresponding author. Tel: ⫹81-27-220-8041; fax: ⫹81-27-2208046. E-mail address: [email protected] (Y. Saito). Abbreviations: ADP, afterdepolarization; AHP, afterhyperpolarization; FIL, first interspike interval long; GAD, glutamate decarboxylase; GLY, glycinergic; GlyT2, glycine transporter 2; LFR, low firing rate; LTS, low-threshold calcium spike; MVN, medial vestibular nucleus; PHN, prepositus hypoglossi nucleus; RT-PCR, reverse transcription-polymerase chain reaction; VGAT, vesicular GABA transporter.

0306-4522/11 $ - see front matter © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2011.09.017

89

90

M. Shino et al. / Neuroscience 197 (2011) 89 –98

component (monophasic AHP), a firing pattern with a delay in the generation of the first spike, a firing pattern with a transient burst, and a firing pattern with a prolonged initial interspike interval. We also identified glutamatergic neuron-specific properties, such as a firing pattern with a low firing rate. However, only a few glycinergic neurons were identified in the study, although previous anatomical and physiological studies had demonstrated the presence of glycinergic neurons in the PHN (Spencer et al., 1989; Yingcharoen et al., 1989; Rampon et al., 1996; Tanaka and Ezure, 2004). Therefore, the preferential electrophysiological properties of inhibitory neurons in the PHN have still not been characterized sufficiently. Whole-cell recordings followed by single-cell RT-PCR is one way to determine the neurotransmitter phenotypes of recorded neurons (Takazawa et al., 2004; Bagnall et al., 2007; Shino et al., 2008). An alternative and powerful strategy is to characterize neuronal phenotypes using transgenic animals in which a specific neuronal population is fluorescently labeled. Indeed, by using transgenic mouse lines in which glutamatergic, GABAergic, and glycinergic neurons selectively expressed fluorescent proteins, Kolkman et al. (2011) characterized the intrinsic physiological properties of neurons in the PHN and the medial vestibular nucleus (MVN). Recently, Uematsu et al. (2008) used a bacterial artificial chromosome construct to generate a transgenic rat line in which the expression of Venus, a fluorescent protein that is brighter than enhanced green fluorescent protein (GFP) (Nagai et al., 2002), is driven by the vesicular GABA transporter (VGAT) gene promoter. VGAT is a common vesicle transporter for GABA and glycine that is expressed in both GABAergic and glycinergic neurons (Sagné et al., 1997; Chaudhry et al., 1998; Wojcik et al., 2006; for review, see Gasnier, 2000). Therefore, the principal inhibitory neurons in the brain would be expected to express Venus in VGAT-Venus transgenic rats. In the present study, to find inhibitory neuron-specific properties, we investigated the preferential electrophysiological properties of inhibitory PHN neurons in VGAT-Venus rats. To further clarify the preferential electrophysiological properties of GABAergic and glycinergic neurons, we investigated whether the recorded Venus-expressing neurons were GABAergic or glycinergic by single-cell RT-PCR analysis.

EXPERIMENTAL PROCEDURES All experimental procedures were approved by the Animal Care and Experimentation Committee of Gunma University (approval number 07-028, 10-003). Every effort was made to minimize the number of animals used and to prevent their suffering.

Transgenic rats The rats used in the present study were obtained by mating VGAT-Venus transgenic male rats to wild-type Wistar female rats. The generation of VGAT-Venus transgenic rats was detailed previously (Uematsu et al., 2008). Of the two lines of the transgenic rats (VGAT-Venus-A and VGAT-Venus-B) generated previously, we used VGAT-Venus-B rats for our experiments because the relatively lower fluorescence of neuropils in the PHN enabled us to easily identify Venus(⫹) neurons in living brainstem slices under the microscope. Rats were genotyped by PCR using the following

Table 1. The primer sets for cloning the target fragments Probe

Sequences (5’⬎3’)

Size (bp)

Venus

CTGTTCACCGGGGTGGTGC AGAGTGATCCCGGCGGCG CGCCATTCAGGGCATGTTCG GCCATGAGCAGCGTGAAGAC ACTGGTCCTCTTCACCTCAG CAGATCTTGACCCAACCTCTC TGCGATACTGAGAGCTGATG CATGCCTTGTCAACGTACC

674

VGAT GAD67 GLYT2

960 940 952

primers: Venus-F: 5’-GTTCAGCGTGTCCGGCGA-3’ and Venus-R: 5’-GCGGTCACGAACTCCAGC-3’.

In situ hybridization histochemistry To confirm the expression of Venus in VGAT-containing PHN neurons, we performed in situ hybridization analysis. VGAT mRNA signals were expected to be detectable mainly in the cell soma, unlike VGAT protein, which localizes mainly to axonal terminals (Chaudhry et al., 1998; Dumoulin et al., 1999). In our preliminary fluorescent microscopic observations, we detected no Venus fluorescence in cryosections obtained from VGAT-Venus rats that were not perfused with fixative, although Venus fluorescence was observed in living slices and perfusion-fixed sections. The absence of Venus fluorescence in cryosections without perfusion-fixation is presumably due to the escape of Venus protein from the cytoplasm or its degradation. Because we used cryosections without perfusion-fixation for in situ hybridization as described below, we performed in situ hybridization for both VGAT and Venus mRNAs. To investigate the co-localization of glutamate decarboxylase (GAD) and glycine transporter 2 (GlyT2) in PHN neurons, we performed double in situ hybridization. There are two GAD isozymes, GAD65 and GAD67 (Bu et al., 1992; Esclapez et al., 1994). In this study, we investigated the expression of GAD67 mRNA because in situ hybridization for GAD67 mRNA has been consistently successful in our laboratory. The probes for Venus, GlyT2, VGAT, and GAD67 were cloned by PCR using the primers listed in Table 1. To detect Venus (Nagai et al., 2002, GenBank ID: AY928551, nucleotides 22– 695), VGAT (NM031782, nucleotides 467–1426), GAD67 (M76177, nucleotides 1024 –1963), and GlyT2 transcripts (L21672, nucleotides 5374 – 6325), we synthesized Venus and GlyT2 cRNA probes using a digoxigenin (DIG)-UTP RNA labeling Kit (Roche Diagnostics, Basel, Switzerland), and VGAT and GAD67 cRNA probes using a fluorescein (FITC)-UTP RNA labeling kit (Roche Diagnostics) according to the manufacturer’s instructions. The double in situ hybridization analyses for Venus and VGAT mRNAs and GAD67 and GlyT2 mRNAs were performed on tissues from two VGAT-Venus rats and two wild-type rats, respectively. VGAT-Venus or wild-type rats [20 –21 postnatal days old (PND)] were decapitated under deep isoflurane anesthesia, and the brain was removed quickly. A trimmed brain block that included the PHN was embedded in OCT compound (Sakura Finetek, Tokyo, Japan) and rapidly frozen in isopentane cooled with dry ice. The frozen block was cut into frontal sections (10-␮m thick) using a cryostat (Leica CM3050 S, Leica Microsystems Japan, Tokyo, Japan). The sections were thaw-mounted on MAScoated slides (Matsunami Glass, Osaka, Japan) and air-dried. The sections were fixed in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffered saline (PBS, pH 7.4) for 10 min. Thereafter, the sections were processed according to the methods for double in situ hybridization histochemistry (Watakabe et al., 2007, 2010; Katori et al., 2009) with a slight modification to the dilution of anti-FITC antibody conjugated with horseradish peroxidase (1: 4000) and the treatment time of the sections with TSA-Plus (dini-

M. Shino et al. / Neuroscience 197 (2011) 89 –98 trophenol) reagents (5 min). Finally, the sections were counterstained for 5 min with 4’,6-diamidino-2-phenylindole, dihydrochloride (DAPI, Invitrogen) diluted to 100 ng/ml in PBS and mounted in CC/Mount mounting medium (Diagnostic Biosystems, Pleasanton, CA, USA). Images of the sections were captured under a fluorescent microscope (Axioplan 2, Zeiss, Tokyo, Japan), and the brightness and contrast of the captured images were adjusted using the AxioVision 4.7.1 (Zeiss) and Photoshop CS4 (Adobe Systems). To quantify the colocalization of Venus and VGAT or GAD67 and GlyT2 in individual neurons, we counted the labeled neurons on either side of the PHN. The number of labeled neurons was obtained from a total of four sections (two sections from each animal, separated by 200 – 300 ␮m) containing the rostral-intermediate region of the PHN. The threshold for detection of the labeled neurons was determined arbitrarily.

Slice preparation and whole-cell recording The procedures for slice preparation and whole-cell patch clamp recording were similar to those described previously (Shino et al., 2008). Briefly, a VGAT-Venus rat (16 –21 PND) was decapitated under deep anesthesia with isoflurane, and the brain was removed quickly. Frontal slices (250-␮m thick) that included the PHN were cut with a Microslicer (Pro 7, Dosaka EM, Kyoto, Japan). Slices were incubated in an extracellular solution containing (in mM): 125 NaCl, 2.5 KCl, 2 CaCl2, 1 MgCl2, 1.25 NaH2PO4, 26 NaHCO3, and 25 glucose, aerated with 95% O2 and 5% CO2 (pH 7.4), for more than 1 h at room temperature. The slices were then transferred to a submerged recording chamber on an upright microscope (Leica DM LFS, Leica Microsystems Japan) and continuously perfused with the extracellular solution at a rate of 5 ml/min. The bath temperature was kept at 30 –32 °C with an in-line heater (SH-27A, Warner Instruments, Hamden, CT, USA). Venus(⫹) and Venus(⫺) neurons were identified under fluorescence optics, and whole-cell current-clamp recordings were performed under observation with Nomarski optics using an EPC-8 patch clamp amplifier (HEKA, Darmstadt, Germany). Pipettes were filled with an internal solution containing (in mM): 120 K-methylsulfate, 20 KCl, 0.2 EGTA, 2 MgATP, 0.3 NaGTP, 10 HEPES, and 0.1 spermine, pH adjusted to 7.3 with KOH. For single-cell RT-PCR analysis, patch pipettes were filled with an autoclaved internal solution containing (in mM): 140 K-methylsulfate, 0.2 EGTA, 2 MgCl2, 10 HEPES, and 0.1 spermine, pH adjusted to 7.3 with KOH. The osmolarity of the internal solution was 280 –290 mOsm/L, and the resistance of the electrodes in the bath solution was 3–7 M⍀. Voltage signals were low-pass filtered at 3 kHz and digitized at 10 kHz. The measured liquid junction potential (⫺5 mV) was corrected. Neurons with membrane potentials more negative than ⫺50 mV immediately after patch membrane rupture and action potential peaks higher than 0 mV were used for further analyses. Most PHN neurons recorded fired spontaneously. At the membrane potentials at which spontaneous firings occurred, differences in firing patterns were not obvious due to inactivation of voltage-gated ion channels that were responsible for distinct firing patterns. Therefore, depolarizing current pulses (400 ms in duration) were applied to the neurons at the membrane potential sufficiently below action potential threshold (⫺85 to ⫺75 mV) that was maintained by the injection of constant currents. Firing patterns were then examined by applying depolarizing current pulses to 200 pA in 40 pA steps. Firing patterns were analyzed from spike trains containing 5–10 action potentials, or 1–3 action potentials for a firing pattern with low firing late (see below). For analysis of AHP profiles, the current pulses were adjusted to induce one action potential during 400 ms, and two to three action potentials were recorded from each neuron. Data were acquired using a pClamp9 system (Molecular Devices, Foster City, CA, USA). Offline analysis was performed with Axograph software (Molecular Devices).

91

Table 2. Primers used for single cell RT-PCR

1st-round PCR GAD65-1F GAD65-1R GAD67-1F GAD67-1R GlyT2-1F GlyT2-2R VGAT-3F VGAT-1R 2nd-round PCR GAD65-2F GAD65-3R GAD67-2F GAD67-2R GlyT2-4F GlyT2-3R VGAT-1F VGAT-4R

Sequences

Product length (bp)

AGCGCCAGACTAGCAGAACC TAGAGTTGTTTGGCAGTGCGTC TTCCACGCCTTCGCCTGCAAC CCATGCCTTCCAGCAACTGGTG GAGGACGAGAACGTGAGTGTG CCAGTGCCAGCATCATCAAGTAAG GCACTGCGACGATCTCGAC GCTATGGCCACATACGAGTCC

522

GGTCCTTCGGATCTGAAGATGG TGCGGTTGGTCTGCCAATTCC CGTAGCCCATGGATGCACCAG GTCTTGCGGACATAGTTGAGGAG GGCACGCTGGAGCACAAC CGTTCTGGAAGGCCAGGTAG CTGAAATCGGAAGGCGAGCC TTGCCGGTGTAGCAGCACAC

443 331 396

436 291 259 305

Single-cell RT-PCR analysis Following whole-cell recordings, we performed single-cell RTPCR analyses using previously described procedures (Takazawa et al., 2004; Shino et al., 2008) with a slight modification. The primer sets used in this study were also modified (Table 2). The contents of the recorded cells were aspirated and expelled into a 0.5 ml tube containing 9 ␮l of a solution containing dNTPs (10 mM), random hexamer (25 ␮M), oligo-dT15 (2.5 ␮M), DTT (0.1 M), and RNase-free water. Next, 1 ␮l of RNase inhibitor and 1 ␮l of Sensiscript reverse transcriptase (Qiagen, Hilden, Germany) were added, and the mixture was incubated at 37 °C overnight. The cDNA fragments were amplified using a two-step protocol, as described previously (Takazawa et al., 2004). The products of the 2nd round PCR were separated on 1.5% agarose gels and stained with Ethidium Bromide. Each experiment contained a negative control in which RT enzyme was not added; no PCR product was detected for any negative control.

RESULTS Distribution of Venus-expressing neurons in the PHN Fluorescent microscopic observation of perfusion-fixed brainstem sections revealed that Venus(⫹) neurons were scattered within the PHN (Fig. 1). This distribution of Venus(⫹) neurons was seen throughout the rostrocaudal PHN. The fluorescently labeled somata and proximal dendrites of PHN neurons were easily detected (Fig. 1B). Prior to the electrophysiological analyses of PHN neurons in VGAT-Venus rats, we verified that Venus was specifically expressed in VGAT-positive PHN neurons using double in situ hybridization analysis. The green (VGAT) and red (Venus) fluorescent signals in PHN neurons whose somata were identified by DAPI staining were compared under fluorescent microscopy (Fig. 2). Of the 449 PHN neurons analyzed, VGAT was detected in all Venus(⫹) neurons (n⫽409). In contrast, Venus was not detected in 10% of VGAT-positive neurons (n⫽40). Although not all VGAT-positive neurons express Venus, these ob-

92

M. Shino et al. / Neuroscience 197 (2011) 89 –98

Fig. 1. Scattered distribution of Venus-expressing neurons in the PHN. (A) Fluorescent micrograph of the PHN in a perfusion-fixed tissue section (50-␮m thick) obtained from a VGAT-Venus transgenic rat. The dashed line shows the rough boundary of the PHN. Scale bar⫽100 ␮m. 4V, fourth ventricle; D, dorsal; L, lateral. (B) Highmagnification photomicrograph of the PHN in the area outlined by the rectangle in (A). Scale bar⫽50 ␮m.

few spikes during current injection despite a sufficient membrane depolarization (low firing rate; LFR). In addition to the five firing patterns classified previously, we added an oscillatory firing pattern to the classification. PHN neurons exhibiting oscillatory firings were first reported by Idoux et al. (2006), who referred to them as type D neurons. In our previous study (Shino et al., 2008), we also found PHN neurons that exhibited oscillatory firings but classified them as late-spiking neurons because they also exhibited a latespiking property. In the present study, to clarify the difference between Venus(⫹) and Venus(⫺) neurons, we classified neurons exhibiting oscillatory firings and late-spiking neurons separately. Fig. 3C2 shows the distribution of firing patterns in Venus(⫹) (open bars) and Venus(⫺) neurons (filled bars). The distribution of firing patterns of Venus(⫹) neurons was significantly different from that of Venus(⫺) neurons (P⬍0.0001, Fisher’s exact test). The

servations strongly suggest that all Venus(⫹) neurons in the PHN are VGAT positive. Electrophysiological properties of PHN neurons in VGAT-Venus rats In a previous study (Shino et al., 2008), we characterized the AHP profiles, firing patterns, and hyperpolarized response patterns of PHN neurons. Because the hyperpolarized response patterns in excitatory PHN neurons were similar to those in inhibitory neurons, we used AHP profiles and firing patterns to discriminate excitatory and inhibitory PHN neurons. According to this classification, we investigated the electrophysiological properties of Venus(⫹) (n⫽108) and Venus(⫺) PHN neurons (n⫽68) identified under fluorescent microscopy (Fig. 3A). Fig. 3B1 shows the voltage traces of three AHP profiles: (1) AHP with an afterdepolarization (ADP, arrow) [AHP(s⫹) with ADP], (2) AHP with a slow component [AHP(s⫹)], and (3) AHP without a slow component [AHP(s⫺)]. The AHP profiles of Venus(⫹) neurons (Fig. 3B2, open bars) was significantly different from that of Venus(⫺) neurons (Fig. 3B2, filled bars, P⬍0.0001, Fisher’s exact test). AHP(s⫹) with ADP and AHP(s⫹) were preferentially found in Venus(⫹) and Venus(⫺) PHN neurons, respectively, although both profiles were observed in both neuronal types. The most distinct difference between Venus(⫹) and Venus(⫺) neurons was the presence of AHP(s⫺), which were found in only a few Venus(⫺) neurons and a substantial number of Venus(⫹) neurons. Fig. 3C1 shows the different firing patterns observed in response to depolarizing current pulses: (1) a repetitive firing pattern with relatively constant interspike intervals (continuous spiking); (2) a firing pattern with a delay in the generation of the first spike due to transient hyperpolarization following the onset of the depolarizing pulse (late spiking); (3) a firing pattern exhibiting a cluster of at least two spikes due to a low-threshold calcium spike (LTS) (although we regarded this as “burst spiking” in our previous report, we refer to it here as “a firing pattern with LTS” for consistency with Idoux et al., 2006); (4) a firing pattern exhibiting a first interspike interval (ISI) that was longer than the second ISI (first interspike interval long; FIL); and (5) a firing pattern exhibiting only a

Fig. 2. Double in situ hybridization analysis of Venus and VGAT mRNAs. (A) Fluorescent micrograph of a frontal section of the brainstem stained with DAPI. The dashed line shows the rough boundary of the PHN. 4V, fourth ventricle. Right: High-magnification photomicrograph of the PHN in the outlined area. (B, C) Expression of VGAT (B) and Venus mRNAs (C) in PHN neurons. (D) Merged image of (A–C). Scale bars in (D): left⫽100 ␮m; right⫽50 ␮m.

M. Shino et al. / Neuroscience 197 (2011) 89 –98

93

Fig. 3. Intrinsic electrophysiological properties of PHN neurons in VGAT-Venus transgenic rats. (A1) Photomicrograph of the PHN in a living slice. (A2, 3) Photomicrographs of Venus(⫹) (2, arrow) and Venus(⫺) (3, arrow) neuron using epifluoscence (left) and Nomarski optics (right). Scale bar⫽10 ␮m. (B1, C1) Afterhyperpolarization (AHP) profiles (B1) and firing patterns (C1) of PHN neurons. (B2, C2) Distributions of AHP profiles (B2) and firing patterns (C2) in Venus(⫹) (open bars, n⫽108) and Venus(⫺) (filled bars, n⫽68) PHN neurons. The ordinate shows the percentage of Venus(⫹) or Venus(⫺) neurons exhibiting each property.

majority of Venus(⫹) and Venus(⫺) neurons exhibited late spiking and continuous spiking, respectively. LFR was never observed in Venus(⫹) neurons, whereas the oscillatory firing was never observed in Venus(⫺) neurons. Neurotransmitter types in Venus-expressing neurons To determine whether the Venus(⫹) neurons recorded were GABAergic or glycinergic, we evaluated the expression of specific mRNAs, such as GADs (GAD65, GAD67) and GlyT2, by single-cell RT-PCR analysis following whole-cell recordings. To confirm that a sufficient volume of Venus(⫹) neuron contents had been harvested, we also measured the expression of VGAT mRNA and investigated the expression of GADs and GlyT2 in VGAT-positive neurons (Fig. 4A). The expression of GADs and/or GlyT2 together with VGAT, was detected in 67 Venus(⫹) PHN neurons. Consistent with our previous study, in which only a few PHN neurons expressing GlyT2 alone were detected (Shino et al., 2008), the expression of GlyT2 alone was detected in only one Venus(⫹) PHN neuron. Approximately 40% (28/67) of the neurons expressed GADs

alone, and the remainder (38/67) expressed both GADs and GlyT2 (Fig. 4B). The expression patterns of GAD65, GAD67, and/or GlyT2 mRNAs in Venus(⫹) PHN neurons are summarized in Table 3. The existence of PHN neurons expressing both GADs and GlyT2 mRNAs was confirmed by double in situ hybridization analysis (Fig. 5). We analyzed the expressions of GAD67 and/or GlyT2 mRNAs in 623 PHN neurons that were positive for GAD67 and/or GlyT2 mRNAs (Table 3). Approximately 40% of the neurons expressed both GAD67 and GlyT2 mRNAs. Meanwhile, 49% of PHN neurons expressed GAD67 alone, and 11% expressed GlyT2 alone. Comparison of electrophysiological properties between GABAergic and GABA&GLY neurons In this study, we regarded the neurons expressing GADs alone and both GADs and GlyT2 as GABAergic and GABA&GLY neurons, respectively. Fig. 6A, B show the distributions of AHP profile (1) and firing pattern (2) in GABAergic and GABA&GLY neurons. Both GABAergic and GABA&GLY neurons included a wide variety of pop-

94

M. Shino et al. / Neuroscience 197 (2011) 89 –98

pattern identified were made up of GABAergic and GABA&GLY neurons, respectively (Fig. 6C). Although the overall cellular distributions of GABAergic neurons based on AHP profiles and firing patterns were not significantly different from those of GABA&GLY neurons (AHP profiles: P⫽0.36, firing patterns: P⫽0.17, Fisher’s exact test), neurons exhibiting continuous spiking and FIL were predominantly GABA&GLY neurons rather than GABAergic neurons (Fig. 6C2, Table 3). To compare the approximate cell sizes of GABAergic and GABA&GLY neurons, we estimated their membrane capacitance via transient capacitance current in response to a ⫺10 mV hyperpolarizing voltage step from a holding potential of ⫺70 mV. The membrane capacitance of GABAergic neurons (39.8⫾22.9 pA, n⫽28) was not significantly different from that of GABA&GLY neurons (47.4⫾22.7 pA, n⫽38, P⫽0.19, t-test). Fig. 4. Expression of GADs and GlyT2 mRNAs in Venus-expressing PHN neurons. (A) The mRNA expression patterns in three Venus(⫹) PHN neurons. Neurons expressing both GADs and GlyT2 mRNAs were regarded as GABA&GLY neurons. (B) Percentage of neurons expressing GADs and/or GlyT2 mRNAs.

ulations exhibiting various AHP profiles and firing patterns, and the overall distribution of GABAergic neurons was similar to that of GABA&GLY neurons. The proportion of neurons with late spiking profiles was markedly higher for GABAergic neurons than for GABA&GLY neurons, although this was the major common firing pattern in both types of neurons (Fig. 6A2). To clarify the electrophysiological characteristics of the GABAergic and GABA&GLY neurons, we also conducted the inverse analysis, determining what percentage of each AHP profile and firing

DISCUSSION In the present study, we investigated the intrinsic electrophysiological properties of PHN neurons in VGAT-Venus rats. VGAT-Venus rats were first generated by Uematsu et al. (2008) and have subsequently been used in the investigations of cortical and hippocampal GABAergic neurons (Fujiwara-Tsukamoto et al., 2010; Koyanagi et al., 2010). However, because VGAT is a vesicle transporter for both GABA and glycine (Sagné et al., 1997; Dumoulin et al., 1999), VGAT-Venus transgenic animals were expected to be useful for the identification of not only GABAergic but also glycinergic neurons. Both GABAergic and glycinergic neurons express Venus in VGAT-Venus transgenic mice

Table 3. Summary of electrophysiological and molecular properties of inhibitory PHN neurons (the number of neurons is shown) GABAergic Number of neurons exhibiting different AHP profile (n⫽67) AHP(s⫹) with ADP AHP(s⫹) AHP(s) Number of neurons exhibiting different firing pattern (n⫽67) Continuous spiking Late spiking LTS FIL LFR Oscillatory firing Number of neurons in which mRNAs were detected by RT-PCR (n⫽67) GAD65 GAD67 GAD65⫹GAD67 GlyT2 GAD65⫹GlyT2 GAD67⫹GlyT2 GAD65⫹GAD67⫹GlyT2 Number of neurons in which mRNAs were detected by in situ hybridization (n⫽623) GAD67 GlyT2 GAD67⫹GlyT2

Glycinergic

GABA&GLY

12 7 9

1 0 0

22 5 11

1 20 3 1 0 3

1 0 0 0 0 0

5 17 5 7 0 4

8 6 14 0 0 0 0

0 0 0 1 0 0 0

0 0 0 0 7 1 30

306 0 0

0 71 0

0 0 246

M. Shino et al. / Neuroscience 197 (2011) 89 –98

95

Fig. 5. Double in situ hybridization analysis of GAD67 and GlyT2 mRNA. (A) Fluorescent micrograph of a frontal section of the brainstem stained with DAPI. The dashed line shows the rough boundary of the PHN. 4V, fourth ventricle. Bottom: High-magnification photomicrograph of the PHN in the outlined area. (B, C) The expression of GAD67 (B) and GlyT2 mRNAs (C) in PHN neurons. (D) Merged image of (A–C). Arrows indicate PHN neurons expressing both GAD67 and GlyT2 mRNAs (white), GAD67 mRNA alone (green), and GlyT2 mRNA alone (red), respectively. Scale bars in (D): upper⫽200 ␮m; lower⫽50 ␮m.

that were generated using the same VGAT-Venus BAC vector (Wang et al., 2009). Therefore, VGAT-Venus transgenic animals are useful for the identification of inhibitory neurons and the clarification of their overall distribution in the brainstem and spinal cord, in which inhibitory synaptic transmissions are mediated by the release of GABA and/or glycine from synaptic vesicles (see Legendre, 2001). PHN neurons were classified based on several aspects of their intrinsic electrophysiological properties (see Eugène et al., 2011). Based on action potential waveforms and firing patterns, Idoux et al. (2006) classified PHN neurons into four types: neurons exhibiting monophasic AHP (type A), biphasic AHP (type B), biphasic AHP with

low threshold spike (type B with LTS), and oscillatory firing (type D). Later, we classified these neurons based on their AHP profiles and firing patterns (Shino et al., 2008). Neurons exhibiting AHP(s⫺) and AHP(s⫹) seem to correspond to type A neurons (see Rössert and Straka, 2011), and neurons exhibiting AHP(s⫹) with ADP correspond to type B neurons. Neurons exhibiting LTS and oscillatory firing correspond to type B with LTS and type D, respectively. The other firing patterns observed in our study may represent subcategories of type A and B neurons. In the present study, only Venus(⫹) neurons exhibited the oscillatory firing corresponding to type D neurons, indicating that type D neurons are primarily inhibitory. In

Fig. 6. Distribution of PHN neurons as classified by AHP profile and firing pattern. (A, B) Distribution of GABAergic Venus(⫹) PHN neurons (A, n⫽28) and GABA&GLY neurons (B, n⫽38) classified by AHP profiles (1) and firing patterns (2). (C) Percentage of GABAergic (closed bars) and GABA&GLY neurons (open bars) among neurons exhibiting each AHP profile (1) [AHP(s⫹) with ADP (n⫽20), AHP(s⫹) (n⫽12), and AHP (s⫺) (n⫽20)] and firing pattern (2) [continuous spiking (n⫽6), late spiking (n⫽37), LTS (n⫽8), FIL (n⫽8), Osc (n⫽7)].

96

M. Shino et al. / Neuroscience 197 (2011) 89 –98

contrast, LFR was observed only in Venus(⫺) neurons. Because our previous study showed that LFR was a preferential property of glutamatergic neurons in the PHN (Shino et al., 2008), this finding indicates that PHN neurons exhibiting LFR are most likely excitatory. Except for LFR and oscillatory firing, all types of AHP profiles and firing patterns were observed in both Venus(⫹) and Venus(⫺) neurons; however, there were differences in the percentages of neurons exhibiting each AHP profile and firing pattern. A marked difference in the proportion of neurons exhibiting AHP(s⫺) was observed between Venus(⫹) and Venus(⫺) neurons. The present finding that most neurons exhibiting AHP(s⫺) were Venus(⫹) neurons supports our previous finding that GABAergic neurons exhibit AHP(s⫺) (Shino et al., 2008). Together, these findings strongly indicate that PHN neurons exhibiting AHP(s⫺) are inhibitory. The comparison of firing patterns between Venus(⫹) and Venus(⫺) PHN neurons suggests that late and continuous spiking are preferential properties of inhibitory and excitatory PHN neurons, respectively. Preferential firing patterns, such as continuous spiking in excitatory neurons and late spiking in inhibitory neurons, might be a common feature in the brain regions involved in oculomotor integration; this relationship is observed not only in the PHN but also in the MVN (Takazawa et al., 2004), which also acts as a horizontal oculomotor integrator (Cheron and Godaux, 1987; Escudero et al., 1992; McFarland and Fuchs, 1992; Mettens et al., 1994). In the oculomotor integrator, mutual inhibition between bilateral integrator regions is suggested to be important for the integrator function (Cannon et al., 1983; Galiana and Outerbridge, 1984; Cannon and Robinson, 1985; Arnold and Robinson, 1997). Therefore, PHN and MVN neurons exhibiting late spiking might contribute to commissural projections in the integrator circuits. One of the main purposes of the present study was to characterize glycinergic PHN neurons; only a few glycinergic PHN neurons had been detected in our previous study (Shino et al., 2008). In the present study, we sampled Venus(⫹) neurons and modified the PCR primers used to amplify GlyT2 mRNA. As a result, PHN neurons expressing GlyT2 mRNA were detected more frequently. However, only one neuron expressing GlyT2 alone was identified; most PHN neurons that expressed GlyT2 also expressed GADs. The co-localization of GlyT2 and GAD67 mRNAs in single PHN neurons was previously reported by Tanaka and Ezure (2004). They conducted a double in situ hybridization analysis of neurons in the midbrain, pons, and cerebellum and graded the co-localization: no doublelabeled neurons (category I), intermingled double-labeled and single-labeled neurons (category II), and all doublelabeled neurons (category III). The PHN belonged to category II, and PHN neurons expressing GlyT2 and those expressing GAD67 were distributed within the same area (Tanaka and Ezure, 2004). In our in situ hybridization analysis, the co-localization of GlyT2 and GAD67 mRNAs was observed in approximately 40% of labeled PHN neurons. Together, these findings indicate the existence of a

neuronal population within the PHN that expresses both GlyT2 and GAD67 mRNAs. In situ hybridization analysis revealed that about 10% of the labeled PHN neurons expressed GlyT2 mRNA alone, whereas only one PHN neuron exhibiting GlyT2 mRNA alone was detected by single-cell RT-PCR analysis. The differences in detection may be due to the detection of not only GAD67 but also GAD65 mRNAs by singlecell RT-PCR analysis. Indeed, PHN neurons expressing GlyT2 and GAD65 were identified by single-cell RT-PCR analysis (Table 3). If the in situ hybridization probe also recognized both GAD65 and GAD67 mRNAs, the proportion of neurons exhibiting GlyT2 mRNA but not GAD(s) would be expected to decrease. In combination with our previous findings, these data suggest that most glycinergic neurons (as defined by the expression of GlyT2) co-express GADs and that a substantial number of GABAergic neurons co-express GlyT2. Because VGAT is a common vesicular transporter of GABA and glycine, Venus(⫹) PHN neurons that express both GADs and GlyT2 may be capable of releasing both GABA and glycine, as is the case with inhibitory neurons in other regions of the central nervous system (Jonas et al., 1998; O’Brien and Berger, 1999; Russier et al., 2002; Dugué et al., 2005). A previous study in cats provided evidence in favor of glycine as a neurotransmitter in inhibitory PHN neurons that project to the abducens nucleus (Spencer et al., 1989). Therefore, whether GABA or glycine is used as a neurotransmitter may depend on the expression or density of postsynaptic GABA and/or glycine receptors of the subsynaptic membrane. The overall distributions of GABA&GLY PHN neurons, as classified by AHP profiles and firing patterns, were similar to those of GABAergic neurons. Thus, GABAergic and GABA&GLY neuron populations may be composed of neurons with similar intrinsic properties. Continuous spiking and FIL firing patterns were exhibited more frequently by GABA&GLY neurons than by GABAergic neurons. In particular, FIL neurons were much more likely to be GABA&GLY neurons than GABAergic neurons. Thus, among inhibitory neurons in the PHN, continuous spiking and FIL may be useful markers for distinguishing GABA&GLY neurons from GABAergic neurons. The present study revealed that the AHP profiles and firing patterns of Venus(⫹) PHN neurons overlapped with those of Venus(⫺) PHN neurons, although some AHP profiles and firing patterns were specific to either group. In our previous study of the PHN and MVN, neurons could not be accurately classified based solely on their intrinsic electrophysiological properties (Takazawa et al., 2004; Saito and Ozawa, 2007; Shino et al., 2008; Saito et al., 2008). There is no doubt that transgenic animals in which specific neuronal populations are labeled greatly improve the identification of neuronal phenotypes, especially in electrophysiological studies using in vitro preparations in which fluorescently labeled neurons can be observed directly. However, transgenic animals are not necessarily useful for studies that do not permit the direct visualization of neurons, such as in vivo recordings and blind patch

M. Shino et al. / Neuroscience 197 (2011) 89 –98

clamp recordings in adult slice preparations. Therefore, at present, intrinsic electrophysiological properties that are “preferential” to particular neuronal phenotypes are still used as clues for cell identification during electrophysiological recordings, although other subtype-specific cell properties should be identified in future studies.

CONCLUSIONS The present study using VGAT-Venus transgenic rats suggests that continuous spiking and LFR are preferential properties of presumptive excitatory PHN neurons, whereas AHP(s⫺), late spiking, and oscillatory firing are preferential properties of inhibitory PHN neurons. Most Venus(⫹) PHN neurons expressed either GADs or both GADs and GlyT2. Given that only a few neurons expressed GlyT2 alone, this suggests that most inhibitory neurons in the PHN either release GABA alone or both GABA and glycine. The intrinsic electrophysiological properties of GABAergic PHN neurons were mostly similar to those of GABA&GLY PHN neurons; however, continuous spiking and FIL were more common in GABA&GLY neurons. Although the functional significance of these preferential electrophysiological properties remains to be clarified, these properties can be used to identify excitatory and inhibitory neurons and GABA&GLY neurons in the PHN. Acknowledgments—We thank Seiji Ozawa for helpful comments on this manuscript and the members of our laboratory for technical assistance and comments. We also thank Atsushi Miyawaki for providing pCS2-Venus, Minato Nakazawa for discussion of our analysis, and the staff at the Institute of Experimental Animal Research, Gunma University Graduate School of Medicine for technical help. This work was supported by Grant-in-Aid for Scientific Research on Priority Areas from MEXT, Grant-in-Aid for Scientific Research from JSPS, and a grant from the Cooperative Study Program of National Institute for Physiological Sciences, Japan.

REFERENCES Arnold DB, Robinson DA (1997) The oculomotor integrator: testing of a neural network model. Exp Brain Res 113:57–74. Bagnall MW, Stevens RJ, du Lac S (2007) Transgenic mouse lines subdivide medial vestibular nucleus neurons into discrete, neurochemically distinct populations. J Neurosci 27:2318 –2330. Bu DF, Erlander MG, Hitz BC, Tillakaratne NJ, Kaufman DL, WagnerMcPherson CB, Evans GA, Tobin AJ (1992) Two human glutamate decarboxylases, 65-kDa GAD and 67-kDa GAD, are each encoded by a single gene. Proc Natl Acad Sci U S A 89:2115–2119. Büttner U, Büttner-Ennever JA (2006) Present concepts of oculomotor organization. Prog Brain Res 151:1– 42. Cannon SC, Robinson DA (1985) An improved neural-network model for the neural integrator of the oculomotor system: more realistic neuron behavior. Biol Cybern 53:93–108. Cannon SC, Robinson DA, Shamma S (1983) A proposed neural network for the integrator of the oculomotor system. Biol Cybern 49:127–136. Chaudhry FA, Reimer RJ, Bellocchio EE, Danbolt NC, Osen KK, Edwards RH, Storm-Mathisen J (1998) The vesicular GABA transporter, VGAT, localizes to synaptic vesicles in sets of glycinergic as well as GABAergic neurons. J Neurosci 18:9733–9750.

97

Cheron G, Godaux E (1987) Disabling of the oculomotor neural integrator by kainic acid injections in the prepositus-vestibular complex of the cat. J Physiol 394:267–290. Dugué GP, Dumoulin A, Triller A, Dieudonné S (2005) Target-dependent use of co-released inhibitory transmitters at central synapses. J Neurosci 25:6490 – 6498. Dumoulin A, Rostaing P, Bedet C, Lévi S, Isambert MF, Henry JP, Triller A, Gasnier B (1999) Presence of the vesicular inhibitory amino acid transporter in GABAergic and glycinergic synaptic terminal boutons. J Cell Sci 112:811– 823. Esclapez M, Tillakaratne NJ, Kaufman DL, Tobin AJ, Houser CR (1994) Comparative localization of two forms of glutamic acid decarboxylase and their mRNAs in rat brain supports the concept of functional differences between the forms. J Neurosci 14: 1834 –1855. Escudero M, de la Cruz RR, Delgado-García JM (1992) A physiological study of vestibular and prepositus hypoglossi neurones projecting to the abducens nucleus in the alert cat. J Physiol 458:539 –560. Eugène D, Idoux E, Beraneck M, Moore LE, Vidal PP (2011) Intrinsic membrane properties of central vestibular neurons in rodents. Exp Brain Res 210:423– 436. Fujiwara-Tsukamoto Y, Isomura Y, Imanishi M, Ninomiya T, Tsukada M, Yanagawa Y, Fukai T, Takada M (2010) Prototypic seizure activity driven by mature hippocampal fast-spiking interneurons. J Neurosci 30:13679 –13689. Fukushima K, Kaneko CR (1995) Vestibular integrators in the oculomotor system. Neurosci Res 22:249 –259. Fukushima K, Kaneko CR, Fuchs AF (1992) The neuronal substrate of integration in the oculomotor system. Prog Neurobiol 39:609 – 639. Galiana HL, Outerbridge JS (1984) A bilateral model for central neural pathways in vestibuloocular reflex. J Neurophysiol 51:210 –241. Gasnier B (2000) The loading of neurotransmitters into synaptic vesicles. Biochimie 82:327–337. Idoux E, Serafin M, Fort P, Vidal PP, Beraneck M, Vibert N, Mühlethaler M, Moore LE (2006) Oscillatory and intrinsic membrane properties of guinea pig nucleus prepositus hypoglossi neurons in vitro. J Neurophysiol 96:175–196. Jonas P, Bischofberger J, Sandkühler J (1998) Corelease of two fast neurotransmitters at a central synapse. Science 281:419 – 424. Katori S, Hamada S, Noguchi Y, Fukuda E, Yamamoto T, Yamamoto H, Hasegawa S, Yagi T (2009) Protocadherin-alpha family is required for serotonergic projections to appropriately innervate target brain areas. J Neurosci 29:9137–9147. Kolkman KE, Moghadam SH, du Lac S (2011) Intrinsic physiology of identified neurons in the prepositus hypoglossi and medial vestibular nuclei. J Vestib Res 21:33– 47. Koyanagi Y, Yamamoto K, Oi Y, Koshikawa N, Kobayashi M (2010) Presynaptic interneuron subtype- and age-dependent modulation of GABAergic synaptic transmission by beta-adrenoceptors in rat insular cortex. J Neurophysiol 103:2876 –2888. Legendre P (2001) The glycinergic inhibitory synapse. Cell Mol Life Sci 58:760 –793. McCrea RA, Horn AK (2006) Nucleus prepositus. Prog Brain Res 151:205–230. McFarland JL, Fuchs AF (1992) Discharge patterns in nucleus prepositus hypoglossi and adjacent medial vestibular nucleus during horizontal eye movement in behaving macaques. J Neurophysiol 68:319 –332. Mettens P, Godaux E, Cheron G, Galiana HL (1994) Effect of muscimol microinjections into the prepositus hypoglossi and the medial vestibular nuclei on cat eye movements. J Neurophysiol 72:785– 802. Moschovakis AK (1997) The neural integrators of the mammalian saccadic system. Front Biosci 2: D552–D577. Nagai T, Ibata K, Park ES, Kubota M, Mikoshiba K, Miyawaki A (2002) A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat Biotechnol 20:87–90.

98

M. Shino et al. / Neuroscience 197 (2011) 89 –98

O’Brien JA, Berger AJ (1999) Cotransmission of GABA and glycine to brain stem motoneurons. J Neurophysiol 82:1638 –1641. Rampon C, Luppi PH, Fort P, Peyron C, Jouvet M (1996) Distribution of glycine-immunoreactive cell bodies and fibers in the rat brain. Neuroscience 75:737–755. Robinson DA (1975) Oculomotor control signals. In: Basic mechanisms of ocular motility and their clinical implications (Lennerstrand G, Bach-y-Rita P, eds), pp 337–374. Oxford: Pergamon. Robinson DA (1989) Integrating with neurons. Annu Rev Neurosci 12:33– 45. Rössert C Straka H (2011) Interactions between intrinsic membrane and emerging network properties detemine signal processing in central vestibular neurons. Exp Brain Res 210:437– 449. Russier M, Kopysova IL, Ankri N, Ferrand N, Debanne D (2002) GABA and glycine co-release optimizes functional inhibition in rat brainstem motoneurons in vitro. J Physiol 541:123–137. Sagné C, El Mestikawy S, Isambert MF, Hamon M, Henry JP, Giros B, Gasnier B (1997) Cloning of a functional vesicular GABA and glycine transporter by screening of genome databases. FEBS Lett 417:177–183. Saito Y, Ozawa S (2007) Membrane properties of rat medial vestibular nucleus neurons in vivo. Neurosci Res 59:215–223. Saito Y, Takazawa T, Ozawa S (2008) Relationship between afterhyperpolarization profiles and the regularity of spontaneous firings in rat medial vestibular nucleus neurons. Eur J Neurosci 28:288 –298. Shino M, Ozawa S, Furuya N, Saito Y (2008) Membrane properties of excitatory and inhibitory neurons in the rat prepositus hypoglossi nucleus. Eur J Neurosci 27:2413–2424. Spencer RF, Wenthold RJ, Baker R (1989) Evidence for glycine as an inhibitory neurotransmitter of vestibular, reticular, and prepositus hypoglossi neurons that project to the cat abducens nucleus. J Neurosci 9:2718 –2736.

Takazawa T, Saito Y, Tsuzuki K, Ozawa S (2004) Membrane and firing properties of glutamatergic and GABAergic neurons in the rat medial vestibular nucleus. J Neurophysiol 92:3106 –3120. Tanaka I, Ezure K (2004) Overall distribution of GLYT2 mRNAcontaining versus GAD67 mRNA-containing neurons and colocalization of both mRNAs in midbrain, pons, and cerebellum in rats. Neurosci Res 49:165–178. Uematsu M, Hirai Y, Karube F, Ebihara S, Kato M, Abe K, Obata K, Yoshida S, Hirabayashi M, Yanagawa Y, Kawaguchi Y (2008) Quantitative chemical composition of cortical GABAergic neurons revealed in transgenic Venus-expressing rats. Cereb Cortex 18:315–330. Wang Y, Kakizaki T, Sakagami H, Saito K, Ebihara S, Kato M, Hirabayashi M, Saito Y, Furuya N, Yanagawa Y (2009) Fluorescent labeling of both GABAergic and glycinergic neurons in vesicular GABA transporter (VGAT)-Venus transgenic mouse. Neuroscience 164:1031–1043. Watakabe A, Ichinohe N, Ohsawa S, Hashikawa T, Komatsu Y, Rockland KS, Yamamori T (2007) Comparative analysis of layer-specific genes in Mammalian neocortex. Cereb Cortex 17:1918 –1933. Watakabe A, Komatsu Y, Ohsawa S, Yamamori T (2010) Fluorescent in situ hybridization technique for cell type identification and characterization in the central nervous system. Methods 52:367–374. Wojcik SM, Katsurabayashi S, Guillemin I, Friauf E, Rosenmund C, Brose N, Rhee JS (2006) A shared vesicular carrier allows synaptic corelease of GABA and glycine. Neuron 50:575–587. Yingcharoen K, Rinvik E, Storm-Mathisen J, Ottersen OP (1989) GABA, glycine, glutamate, aspartate and taurine in the perihypoglossal nuclei: an immunocytochemical investigation in the cat with particular reference to the issue of amino acid colocalization. Exp Brain Res 78:345–357.

(Accepted 9 September 2011) (Available online 16 September 2011)