α2A adrenoceptor-mediated presynaptic inhibition of GABAergic transmission in rat tuberomammillary nucleus neurons

Authors

  • Michiko Nakamura,

    1. Department of Pharmacology, School of Dentistry, Kyungpook National University, Daegu, Republic of Korea
    Search for more papers by this author
  • Kyungho Suk,

    1. Department of Pharmacology, School of Medicine, Kyungpook National University, Daegu, Republic of Korea
    2. Brain Science & Engineering Institute, Kyungpook National University, Daegu, Republic of Korea
    Search for more papers by this author
  • Maan-Gee Lee,

    1. Department of Pharmacology, School of Medicine, Kyungpook National University, Daegu, Republic of Korea
    2. Brain Science & Engineering Institute, Kyungpook National University, Daegu, Republic of Korea
    Search for more papers by this author
  • Il-Sung Jang

    Corresponding author
    1. Brain Science & Engineering Institute, Kyungpook National University, Daegu, Republic of Korea
    • Department of Pharmacology, School of Dentistry, Kyungpook National University, Daegu, Republic of Korea
    Search for more papers by this author

Address correspondence and reprint requests to Il-Sung Jang, PhD, Department of Pharmacology, School of Dentistry, Kyungpook National University, 188-1, Samduk 2 ga-dong, Jung-gu, Daegu 700-412, Republic of Korea. E-mail: jis7619@mail.knu.ac.kr

Abstract

Histaminergic neurons within the tuberomammillary nucleus (TMN) play an important role in the regulation of sleep-wakefulness. Here, we report the adrenergic modulation of GABAergic transmission in rat TMN histaminergic neurons using a conventional whole-cell patch clamp technique. Norepinephrine (NE) reversibly decreased the amplitude of action potential-dependent GABAergic inhibitory post-synaptic currents (IPSCs) and increased the paired pulse ratio. The NE-induced inhibition of GABAergic IPSCs was mimicked by clonidine, a selective α2 adrenoceptor agonist. However, cirazoline and isoproterenol, nonselective α1 and β adrenoceptor agonists, respectively, had no effect on GABAergic IPSCs. The NE-induced inhibition of GABAergic IPSCs was significantly blocked by BRL44408, a selective α2A adrenoceptor antagonist, but not imiloxan or JP1302, a selective α2B and α2C adrenoceptor antagonists. The extent of NE-induced inhibition of GABAergic IPSCs was inversely proportional to the extracellular Ca2+ concentration. Pharmacological agents affecting the activities of adenylyl cyclase or G-protein-coupled inwardly rectifying K+ channels did not affect the NE-induced inhibition of GABAergic IPSCs. However, NE had no effect on the frequency and amplitude of GABAergic miniature IPSCs. These results suggest that NE acts on presynaptic α2A adrenoceptor to inhibit action potential-dependent GABA release via the inhibition of Ca2+ influx from the extracellular space to GABAergic nerve terminals, and that this α2A adrenoceptor-mediated modulation of GABAergic transmission may be involved in regulating the excitability of TMN histaminergic neurons.

Abbreviations used
[Ca2+]o

extracellular Ca2+ concentration

AC

adenylyl cyclase

BRL44408

2-[(4,5-dihydro-1H-imidazol-2-yl)methyl]-2,3-dihydro-1-methyl-1H-isoindole maleate

GIRK channels

G-protein-coupled inwardly rectifying K+ channels

I h

hyperpolarization- and cyclic nucleotide-activated cation currents

IPSCs

inhibitory post-synaptic currents

JP1302

N-[4-(4-methyl-1-piperazinyl)phenyl]-9-acridinamine HCl

K–S test

Kolmogorov–Smirnov test

LC

locus coeruleus

mIPSCs

spontaneous miniature inhibitory post-synaptic currents

NE

norepinephrine

NREM

non-rapid eye movement

PPR

paired pulse ratio

REM

rapid eye movement

SQ22536

9-(tetrahydro-2-furanyl)-9H-purin-6-amine

SR95531

6-imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid HBr

TMN

tuberomammillary nucleus

TTX

tetrodotoxin

VDCC

voltage-dependent Ca2+ channel

VLPO

ventrolateral preoptic nucleus

Histaminergic neurons within the tuberomammillary nucleus (TMN) are regarded as one of the regulatory centers related to the modulation of sleep-wakefulness (Haas et al. 2008), as they exhibit a spontaneous firing pattern in a pacemaker fashion during wakefulness (Haas and Reiner 1988; Steininger et al. 1999). In addition, such a regular firing pattern decreases during non-rapid eye movement (NREM) sleep, and histaminergic neurons cease discharge during rapid eye movement (REM) sleep (Haas and Reiner 1988; Steininger et al. 1999). The firing pattern as well as neuronal excitability of histaminergic neurons would be regulated by acting in concert with other noradrenergic nuclei, serotonergic nuclei, cholinergic nuclei, and hypocretin/orexin neurons, and the ventrolateral preoptic nucleus (VLPO) (Brown et al. 2001; Haas et al. 2008). Of these, the VLPO contains GABAergic neurons and projects GABAergic inhibitory axon terminals to the TMN (Sherin et al. 1996, 1998; Steininger et al. 2001), and it is regarded as an important sleep center, as lesion in the VLPO causes insomnia and the impairment of NREM sleep (Lu et al. 2000). Therefore, GABAergic transmission from the VLPO to TMN neurons probably plays a crucial role in the regulation of sleep-wakefulness.

Norepinephrine (NE) is a neurotransmitter implicated in the regulation of several physiological responses in the central nervous system, including pain, depression, and sleep-wakefulness (Rajkowska 2000; Pertovaara 2006; Berridge 2008). NE exerts these regulatory effects by activating two major receptor subtypes, namely, α (α1 and α2) and β (β1, β2 and β3) adrenoceptors, which comprised seven transmembrane proteins and are coupled to G-proteins (Bylund et al. 1994). In general, α1 and β adrenoceptors are thought to be expressed on the post-synaptic membrane, but α2 adrenoceptors exist on both post-synaptic and presynaptic membranes of central neurons (see below). α2 Adrenoceptors are further divided into three subtypes, α2A, α2B, and α2C receptors, and are widely expressed not only in the peripheral and central nervous systems but also in peripheral tissues, including cardiac and smooth muscles (Bylund 1995; Gilsbach and Hein 2008).

NE and its receptors are closely involved in the regulation of sleep-wakefulness. For example, the local application of clonidine, a selective α2 adrenoceptor agonist, to the medial preoptic area produces arousal, but yohimbine, a selective α2 adrenoceptor antagonist, produces sleep (Ramesh et al. 1995). In addition, previous studies have shown that noradrenergic nuclei of the brainstem project strongly to the TMN region (Ericson et al. 1989; Lee et al. 2005), and that NE and clonidine inhibit GABAergic transmission in TMN neurons (Stevens et al. 2004). However, which α2 adrenoceptor subtypes are responsible for the presynaptic inhibition and how presynaptic α2 adrenoceptors regulate GABAergic synaptic transmission remain unknown. In this study, we have attempted to identify the presynaptic α2 adrenoceptor subtypes involved and the mechanisms underlying NE-induced presynaptic inhibition of GABA release in TMN histaminergic neurons.

Materials and methods

Preparation

All experiments complied with the guiding principles for the care and use of animals approved by the Council of the Physiological Society of Korea and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and every effort was made to minimize both the number of animals used and their suffering.

Sprague Dawley rats of either sex (11–16 days old; Samtako, Osan, Korea) were decapitated under pentobarbital anesthesia (50 mg/kg, i.p.). The brain was dissected and transversely or sagittally sliced at a thickness of 400 μm using a microslicer (VT1000S; Leica, Nussloch, Germany) in a cold artificial cerebrospinal fluid (ACSF; 120 NaCl, 2 KCl, 1 KH2PO4, 26 NaHCO3, 2 CaCl2, 1 MgCl2, and 10 glucose, saturated with 95% O2 and 5% CO2). The brain slices containing the TMN were kept in an ACSF saturated with 95% O2 and 5% CO2 at 22–25°C for at least 1 h before electrophysiological recording. Thereafter, a slice containing the TMN was transferred into a recording chamber, and histaminergic neurons in TMN were identified under an upright microscope (E600FN; Nikon, Tokyo, Japan) with a water-immersion objective (X40). Images of slices and neurons were obtained using a digital microscope camera (ProgRes® MF; Jenoptik, Jena, Germany). The ACSF routinely contained 3 mM kynurenic acid to block ionotropic glutamate receptors. In experiments with Ba2+, KH2PO4 in ACSF was replaced with equimolar KCl. The bath was perfused with ACSF at 2 mL/min by the use of a peristaltic pump (MP-1000; EYELA, Tokyo, Japan).

Electrical measurements

All electrophysiological measurements were performed using conventional whole-cell patch recording mode at a holding potential (VH) of 0 mV (Axopatch 200B; Molecular Devices, Union City, CA, USA). Patch pipettes were made from borosilicate capillary glass (1.5 mm outer diameter, 0.9 mm inner diameter; G-1.5; Narishige, Tokyo, Japan) by use of a pipette puller (P-97; Sutter Instrument Co., Novato, CA, USA). The resistance of the recording pipettes filled with internal solution (in mM; 140 CsMeHSO3, 5 TEA-Cl, 5 CsCl, 2 EGTA, 2 Mg-ATP and 10 Hepes, pH 7.2 with Tris-base) was 4–6 MΩ. The liquid junction potential (~ −11 mV, measured by exchanging bath solution from internal solution to standard external solution) and pipette capacitance were compensated for. Neurons were viewed under phase contrast on an inverted microscope (TE2000; Nikon). Membrane currents were filtered at 2 kHz, digitized at 5 kHz, and stored on a computer equipped with pCLAMP 10.0 (Molecular Devices). In whole-cell recordings, 10 mV hyperpolarizing step pulses (30 ms in duration) were periodically delivered to monitor the access resistance (15–20 MΩ), and recordings were discontinued if the access resistance changed by more than 15%. All experiments were performed at 22–25°C. To record action potential-dependent GABAergic inhibitory post-synaptic currents (IPSCs), a glass stimulation pipette (~ 10 μm diameter) filled with a bath solution was positioned around the TMN region, except where indicated. Brief paired pulses (500 μs, 100–200 μA, 10 Hz) were applied by the stimulation pipette at a frequency of 0.1 Hz using a stimulator (SEN-7203; Nihon Kohden, Tokyo, Japan) equipped with an isolator unit (SS-701J; Nihon Kohden).

Data analysis

The amplitudes of individual GABAergic IPSCs were measured by subtracting the baseline from the peak amplitude. Spontaneous miniature IPSCs (mIPSCs) were counted and analyzed using the MiniAnalysis program (Synaptosoft, Inc., Decatur, GA, USA) as described previously (Jang et al. 2002). Briefly, mIPSCs were screened automatically using an amplitude threshold of 10 pA, and then visually accepted or rejected based on the rise and decay times. Basal noise levels during voltage-clamp recordings were less than 10 pA. The average values of both the frequency and amplitude of mIPSCs during the control period (5–10 min) were calculated for each recording, and the frequency and amplitude of all the events during the NE application (5 min) were normalized to these values. The effects of these different conditions were quantified as a percentage increase in mIPSC frequency compared to the control values. The continuous curves for the concentration-inhibition relationship of NE were fitted using a least-squares fit to the following equation:

display math

where I is the NE-induced inhibition of GABAergic IPSCs, C is the concentration of NE, EC50 are the concentrations for the half-effective response, and n is the Hill coefficient.

Drugs

The drugs used in this study were tetrodotoxin (TTX), 6-imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid HBr (SR95531), EGTA, cirazoline, BRL44408 (2-[(4,5-dihydro-1H-imidazol-2-yl)methyl]-2,3-dihydro-1-methyl-1H-isoindole maleate), imiloxan, JP1302 (N-[4-(4-methyl-1-piperazinyl)phenyl]-9-acridinamine HCl) (from Tocris, Bristol, UK), kynurenic acid, ATP-Mg, norepinephrine (NE), clonidine, isoproterenol, SQ22536 (from Sigma, St. Louis, MO, USA). All drugs were applied by bath application (2 mL/min).

Statistics

Numerical values are expressed as the mean ± standard error of the mean (SEM) using values normalized to the control. Significant differences in the mean amplitude of GABAergic IPSCs were tested using Student's paired two-tailed t-test, except where indicated, using absolute values rather than normalized ones. The inter-event intervals and amplitudes of a large number of GABAergic mIPSCs obtained from the same neuron were examined by constructing cumulative probability distributions and compared using the Kolmogorov–Smirnov (K–S) test with Stat View software (SAS Institute, Inc., Cary, NC, USA). Values of p < 0.05 were considered significant.

Results

NE acts presynaptically to inhibit GABAergic transmission

Our previous study showed that large neurons in the TMN region are positive for histidine decarboxylase, a marker of histaminergic neurons, but small ones are negative (Yum et al. 2008), indicating that large neurons are histaminergic neurons. To further verify whether large neurons are histaminergic neurons, we examined the electrophysiological property of these neurons. Histaminergic neurons are known to have the hyperpolarization- and cyclic nucleotide-activated cation currents (Ih), which participate in the rhythmic and burst firing of histaminergic neurons (Haas et al. 2008). In this study, hyperpolarizing step pulses elicited the Ih, in a voltage-dependent manner, in large neurons, but not in small neurons (Figure S1a). These results support the notion that morphologically identified large neurons are histaminergic neurons. In the following experiments, all electrophysiological recordings were made from large and Ih-positive neurons. In the presence of 3 mM kynurenic acid, which blocks ionotropic glutamate receptors, action potential-dependent synaptic currents were recorded from TMN histaminergic neurons at a VH of 0 mV by electrical stimulation through a glass pipette. These synaptic currents were completely and reversibly blocked by 10 μM SR95531, a selective GABAA receptor antagonist (Figure S1b), indicating that these are inhibitory post-synaptic currents (IPSCs) mediated by GABAA receptors.

To investigate the action of NE on the GABAergic transmission to TMN histaminergic neurons, we investigated whether NE affects the amplitude or paired pulse ratio (PPR) of GABAergic IPSCs evoked by pairing stimulation at an interval of 100 ms (10 Hz). In most (144 of 151 neurons, 95%) of the neurons tested, bath applied NE (1 μM) strongly inhibited GABAergic transmission. In nine neurons in which NE effect was fully analyzed, NE (1 μM) reversibly decreased the first IPSC (IPSC1) amplitude to 21 ± 5% of the control (p < 0.01, Fig. 1a and b), and increased the PPR (IPSC2/IPSC1) from 1.38 ± 0.18 to 4.41 ± 1.37 (p < 0.05, Fig. 1a and b), suggesting that NE acts presynaptically to decrease the probability of GABA release. In addition, NE reduced the amplitude of GABAergic IPSCs in a concentration-dependent manner with an EC50 value of 122 nM (Fig. 1c). NE showed the inhibitory effect even at a concentration as low as 10 nM (88 ± 3% of the control, n = 6, p < 0.05, Fig. 1c).

Figure 1.

Norepinephrine (NE) acts presynaptically to inhibit the amplitude of GABAergic inhibitory post-synaptic currents (IPSCs). (a) A typical time course of 1st IPSC (IPSC1) amplitude (upper) and paired pulse ratio (PPR; lower) before, during, and after the application of 1 μM NE. Insets represent typical traces of the numbered region. (b) NE-induced changes in IPSC1 amplitude (left) and PPR (right). Each column and error bar represents the mean ± SEM from nine neurons. *p < 0.05, **p < 0.01. (c) Concentration–response relationships of NE. The EC50 value calculated from curve fitting result was 122 nM. Each point and error bar represents the mean and SEM from six to nine neurons.

NE acts on presynaptic α2A adrenoceptors to inhibit GABAergic transmission

Adrenoceptors are divided into three subtypes, α1, α2, and β receptors, which are coupled to Gq/11-, Gi/o- and Gs-proteins, respectively (Kobilka et al., 1987; Bylund et al. 1994). Therefore, we next examined the effect of selective agonists for adrenoceptors on the GABAergic transmission to identify which receptor subtypes are responsible for the NE-induced inhibition of GABAergic IPSCs. A selective α1 adrenoceptor agonist, cirazoline (10 μM), and a selective β adrenoceptor agonist, isoproterenol (30 μM) had no effect on the amplitude of GABAergic IPSCs (cirazoline: 89 ± 6% of the control, n = 7, p = 0.11; isoproterenol: 106 ± 10% of the control, n = 6, p = 0.10, Fig. 2b). In contrast, the application of 1 μM clonidine, a selective α2 adrenoceptor agonist, greatly inhibited GABAergic IPSCs to 22 ± 6% of the control (n = 5, p < 0.01, Fig. 2a and b). These results indicate that NE inhibits GABA release via presynaptic α2 adrenoceptors on GABAergic terminals projecting onto histaminergic neurons.

Figure 2.

α2A adrenoceptors are responsible for the norepinephrine (NE)-induced inhibition of GABAergic inhibitory post-synaptic currents (IPSCs). (a) A typical time course of IPSC1 amplitude before, during, and after the application of 1 μM NE and 1 μM clonidine, a selective α2 adrenoceptor agonist. Insets represent typical traces of the numbered region. (b) Adrenoceptor agonists (cirazoline for α1, clonidine for α2, and isoproterenol for β adrenoceptors)-induced changes in IPSC1 amplitude. Each column and error bar represents the mean ± SEM from five to seven neurons. **p < 0.01, NS; not significant. (c) A typical time course of IPSC1 amplitude (upper) and PPR (lower) before, during, and after the application of 1 μM NE in the absence and presence of 3 μM BRL44408, a selective α2A adrenoceptor antagonist. Insets represent typical traces of GABAergic IPSCs of the numbered region. (d) Dose-dependent inhibition of NE-induced decrease of eIPSC amplitude by BRL44408. Each column and error bar represents the mean ± SEM from six neurons. *p < 0.05, **p < 0.01. (e) Non-effect of imiloxan (ILX) or JP1302, an α2B, α2C adrenoceptor blocker, respectively, on NE-induced decrease in action potential-dependent GABA release. Each column and error bar represents the mean ± SEM from seven neurons. **p < 0.01.

We further examined the effects of subtype selective α2 adrenoceptor antagonists on NE-induced inhibition of GABAergic IPSCs. In the presence of 1 μM BRL44408, a selective α2A receptor antagonist (Young et al. 1989; Callado and Stamford 1999), NE (1 μM) decreased IPSC1 amplitude (68 ± 5% of the BRL44408 condition, n = 6, p < 0.05), but the extent of NE-induced inhibition of GABAergic IPSCs was greatly reduced (Fig. 2c and d). However, NE (1 μM) failed to decrease IPSC1 amplitude in the presence of 3 μM BRL44408 (95 ± 6% of the BRL44408 condition, n = 6, p = 0.54, Fig. 2c and d). In contrast, NE (1 μM) still decreased IPSC1 amplitude in the presence of either 3 μM imiloxan, a selective α2B receptor antagonist (Michel et al. 1990), or 3 μM JP1302, a selective α2C receptor antagonist (Sallinen et al. 2007) (28 ± 6% of the imiloxan condition, n = 7, p = 0.13, and 9 ± 2% of the JP1302 condition, n = 7, p = 0.49, Fig. 2e). These results indicate that NE acts on presynaptic α2A adrenoceptors, and not α2B or α2C adrenoceptors, to inhibit GABAergic transmission. BRL44408 by itself had no effect on the basal amplitude of GABAergic IPSCs (Fig. 2c), suggesting that there was no tonic activation of presynaptic α2 adrenoceptors in our recording conditions.

Activation of α2A adrenoceptors reduces the Ca2+ influx into GABAergic nerve terminals

In general, the activation of Gi/o-protein coupled receptors, such as α2A adrenoceptors, can induce (i) opening of G-protein coupled inwardly rectifying K+ (GIRK) channels, (ii) inhibition of adenylyl cyclase (AC) to reduce the intracellular cAMP concentration, and (iii) inhibition of voltage-dependent Ca2+ channels (VDCCs) to reduce the intracellular Ca2+ concentration (Wu and Saggau 1997; Brown and Sihra 2008). To verify the mechanism underlying the α2A adrenoceptor-mediated inhibition of GABAergic transmission, we first examined the involvement of G-protein-coupled inwardly rectifying K+ channels (GIRK channels) and the AC-cAMP system in the NE-induced inhibition of GABAergic IPSCs. The application of 1 mM Ba2+, a general GIRK channel blocker (Gerber et al. 1989), had no effect on the basal amplitude of GABAergic IPSCs. In the presence of 1 mM Ba2+, NE still decreased IPSC1 amplitude to 10 ± 4% of the Ba2+ condition (n = 6, p < 0.01, Fig. 3a and c). Next, the application of 100 μM SQ22536, an AC inhibitor (Turcato and Clapp 1999), had no effect on the basal amplitude of GABAergic IPSCs. In the presence of 100 μM SQ22536, NE still decreased IPSC1 amplitude to 11 ± 1% of the SQ22536 condition (n = 6, p < 0.01, Fig. 3b and c). These results suggest that neither GIRK channels nor AC-cAMP pathways is involved in the α2A adrenoceptor-mediated inhibition of GABAergic transmission.

Figure 3.

G-protein-coupled inwardly rectifying K+ channel (GIRK channel) and AC/cAMP signal transduction pathway are not involved in the norepinephrine (NE)-induced inhibition of GABAergic inhibitory post-synaptic currents (IPSCs). (a) A typical time course of IPSC1 amplitude before, during and after the application of 1 μM NE in the absence and presence of 1 mM Ba2+. (b) A typical time course of IPSC1 amplitude before, during and after the application of 1 μM NE in the absence and presence of 100 μM SQ22536. (c) NE-induced changes in IPSC1 amplitude in the absence and presence of Ba2+ (n = 6, left) and SQ22536 (n = 6, right). Each column and error bar represents the mean ± SEM. **p < 0.01, NS; not significant.

It has been established that Gβγ subunits can act on multiple types of VDCCs to inhibit the Ca2+ influx from the extracellular space and decrease neurotransmitter release (Wu and Saggau 1997; Brown and Sihra 2008). Since the neurotransmitter release is directly proportional to the extent of Ca2+ influx from the extracellular space and the resultant changes in the intra-terminal Ca2+ concentration (Wu and Saggau 1997), we examined whether the NE-induced inhibition of GABAergic IPSCs is dependent on the extracellular Ca2+ concentration ([Ca2+]o). To verify the influence of [Ca2+]o on the NE-induced inhibition of GABAergic IPSCs, we used NE at 100 nM, a concentration that approximates the EC50 value of NE, as shown in Fig. 1c, because 1 μM NE almost completely abolished GABAergic IPSCs. The change of [Ca2+]o from 2 mM to 0.5 mM decreased IPSC1 amplitude to 18 ± 2% of the control (n = 6, p < 0.05, Fig. 4a) and increased the PPR (415 ± 113% of the control, n = 6, p < 0.05, data not shown), consistent with a decrease in Ca2+ driving force at lower concentrations of extracellular Ca2+. In these conditions, the extent of NE-induced inhibition of GABAergic IPSCs was significantly increased (44 ± 6% of the 2 mM Ca2+ condition and 25 ± 4% of the 0.5 mM Ca2+ condition, n = 6, p < 0.01, anova test, Fig. 4a and c). In contrast, the change of [Ca2+]o from 2 mM to 5 mM increased IPSC1 amplitude to 158 ± 13% of the control (n = 6, p < 0.05, Fig. 4b), and decreased the PPR (83 ± 14% of the control (n = 6, p < 0.05, data not shown). In these conditions, the extent of NE-induced inhibition of GABAergic IPSCs was significantly reduced (44 ± 6% of the 2 mM Ca2+ condition and 65 ± 5% of the 5 mM Ca2+ condition, n = 6, p < 0.01, anova test, Fig. 4b and c).

Figure 4.

Norepinephrine (NE)-induced inhibition of GABAergic inhibitory post-synaptic currents (IPSCs) is dependent on the concentration of extracellular Ca2+. (a) A typical time course of IPSC1 amplitude before, during, and after application of 100 nM NE in the presence of 2 mM and 0.5 mM extracellular Ca2+. Note that GABAergic IPSCs were reduced in the presence of 0.5 mM extracellular Ca2+. (b) A typical time course of IPSC1 amplitude before, during, and after application of 100 nM NE in the presence of 2 mM and 5 mM extracellular Ca2+. (c) NE-induced changes in IPSC1 amplitude in the presence of various concentrations of extracellular Ca2+. Open circles represent individual results, but closed circles represent the mean ± SEM from six to seven neurons. **p < 0.01. Note that the NE-induced inhibition of GABAergic IPSCs is inversely related to the [Ca2+]o.

We also examined the effect of NE on GABAergic mIPSCs in the presence of 300 nM TTX. The amplitude and frequency of GABAergic mIPSCs the application of Ca2+-free (plus 2 mM EGTA) bath solution did not affect the amplitude and frequency of GABAergic mIPSCs (n = 3, data not shown), suggesting that GABAergic mIPSCs recorded from TMN neurons are classical miniature currents that are independent of the Ca2+ influx from the extracellular space (Scanziani et al. 1992; Capogna et al. 1993). NE (1 μM) did not shift the distribution of inter-event interval (p = 0.09; K-S test, Fig. 5a and b) and current amplitude of GABAergic mIPSCs (p = 0.69; K-S test, Fig. 5a and b). In addition, NE (1 μM) did not change the mean mIPSC frequency (90 ± 5% of the control, n = 10, p = 0.21) and amplitude (105 ± 4% of the control, n = 10, p = 0.11) (Fig. 5b insets).

Figure 5.

Norepinephrine (NE) has no effect on the frequency and amplitude of GABAergic spontaneous miniature inhibitory post-synaptic currents (mIPSCs). (a) A typical time course of mIPSC amplitude before, during and after the application of 1 μM NE. Insets represent typical traces of mIPSCs with an expand time scale. Six hundred and fifty-six events were plotted. (b) Cumulative probability plots for inter-event interval (left, p = 0.09, K–S test) and current amplitude (right, p = 0.69, K-S test) of GABAergic mIPSCs. Three hundred and fifty-five events for control and 311 events for NE were plotted. Insets represent the mean ± SEM from 10 neurons. NS; not significant.

α2A adrenoceptors also inhibit GABAergic transmission in sagittal brain slices

Although TMN histaminergic neurons receive GABAergic inputs from various regions (Yang and Hatton, 1997), the GABAergic innervation from the VLPO to TMN may play a particularly important role in the regulation of sleep-wakefulness, as the VLPO is regarded as a sleep center (Sherin et al. 1996, 1998; Steininger et al. 2001). Therefore, we finally examined whether NE also affects GABAergic transmission originating from the VLPO region. To test this, we used sagittal brain slices containing both the VLPO and TMN regions, and GABAergic IPSCs were recorded from histaminergic neurons by paired pulse stimulation using a stimulating pipette placed in the VLPO region (Fig. 6a). All recordings were obtained from Ih-positive large neurons (Fig. 6a). The mean latency between the stimulus and the onset of monosynaptic GABAergic IPSCs was 12.3 ± 0.4 ms (n = 6), as our previous study had shown (see also Nakamura and Jang 2012). In these conditions, the application of NE (1 μM) reversibly decreased IPSC1 amplitude to 13 ± 5% of the control (n = 6, p < 0.05, Fig. 6b and c). The NE-induced inhibition of GABAergic IPSCs was completely blocked by the selective α2A adrenoceptor antagonist BRL44408 (3 μM) (94 ± 6% of the BRL44408 condition, n = 6, p = 0.46, Fig. 6b and c). BRL44408 (3 μM) by itself had no effect on IPSC1 amplitude (85 ± 16% of the control, n = 6, p = 0.33, Fig. 6c), consistent with a lack of tonic activation of α2A adrenoceptors.

Figure 6.

Norepinephrine (NE) acts on α2A adrenoceptors to inhibit the amplitude of GABAergic inhibitory post-synaptic currents (IPSCs) in sagittal brain slices. (a) A photograph of sagittal slice from rat brain (left). A stimulating pipette (Stim.) was placed to the ventrolateral preoptic nucleus (VLPO) region and IPSCs were recorded from histaminergic neurons within the tuberomammillary nucleus (TMN) region. All pharmacological experiments were performed with Ih-positive large neurons (right). (b) A typical time course of IPSC1 amplitude before, during, and after the application of 1 μM NE. Insets represent typical traces of the numbered region. (c) NE-induced changes in IPSC amplitude in the absence and presence of 3 μM BRL44408. Each column and error bar represents the mean ± SEM from six neurons. **p < 0.01, NS; not significant.

Discussion

NE acts on α2A adrenoceptors to inhibit GABAergic transmission onto TMN neurons

Adrenoceptors, like other G-protein coupled receptors, can act presynaptically to modulate neurotransmitter release (for a review, see Gilsbach and Hein 2008). Although α1 and β adrenoceptors act post-synaptically to modulate neuronal excitability, some previous studies have shown that α1A and β2 adrenoceptors increase the probability of release of various neurotransmitters including NE, GABA, and glutamate at peripheral and central synapses (Trendelenburg et al. 2000; Braga et al. 2004; Kolaj and Renaud 2007; Lee et al. 2007). In this study, however, we found that cirazoline and isoproterenol, selective α1 and β adrenoceptor agonists, respectively, had no facilitatory effect on GABAergic IPSCs, indicating that functional α1 and β adrenoceptors might not be expressed on GABAergic nerve terminals projecting to TMN neurons. In contrast, the activation of presynaptic α2 adrenoceptors inhibits the release of various neurotransmitters, such as GABA, glutamate, and NE itself, as heteroreceptors or autoreceptors at central and peripheral synapses. Among α2 adrenoceptors, functional presynaptic α2B and α2C adrenoceptors in central neurons are still unknown (Gilsbach and Hein 2008), although α2C adrenoceptors are found in axon terminals (Olave and Maxwell 2004). However, α2C adrenoceptors are known to serve as feedback regulators to inhibit catecholamine release from adrenal chromaffin cells (Brede et al. 2003; Moura et al. 2006). In central neurons, α2A adrenoceptor subtypes seem to be mainly involved in the presynaptic modulation of neurotransmitter release at central synapses (Boehm 1999; Kawasaki et al. 2003; Chong et al. 2004).

This study provides additional evidence that α2A adrenoceptors are responsible for the presynaptic inhibition of GABAergic transmission onto TMN histaminergic neurons. We found that exogenously applied NE, an endogenous agonist of adrenoceptors, decreased GABAergic IPSC1 amplitude in a concentration-dependent manner, and increased the PPR of the two successive IPSCs. In addition, since NE did not affect the amplitude of GABAergic mIPSCs, NE is likely to decrease the presynaptic probability of GABA release without affecting the sensitivity of post-synaptic GABAA receptors to GABA. We also found that NE-induced inhibition of GABAergic IPSCs was concentration-dependently blocked by BRL44408, a selective α2A adrenoceptor antagonist (pKi = 8.0, Young et al. 1989; Callado and Stamford 1999; Alexander et al. 2009), but not by imiloxan and JP1302, which are selective α2B (pKi = 7.3, Michel et al. 1990; Alexander et al. 2009) and α2C adrenoceptor antagonists (pKi = 7.8, Sallinen et al. 2007; Alexander et al. 2009), respectively. Given that imiloxan is about 55-fold more sensitive to α2B than α2A adrenoceptor subtypes (Michel et al. 1990) and that JP1302 is about 100-fold more sensitive to α2C than α2A adrenoceptor subtypes (Sallinen et al. 2007), the present pharmacological results suggest that α2A adrenoceptors are functionally expressed on GABAergic nerve terminals projecting to TMN histaminergic neurons. It should also be noted that NE had no effect on GABAergic IPSC1 amplitude after the blockade of α2A adrenoceptors with BRL44408, which is consistent with no involvement of α1 and β adrenoceptors in the noradrenergic modulation of GABAergic transmission onto TMN neurons.

Mechanisms underlying the α2A adrenoceptor-mediated inhibition of GABA release

All three α2 adrenoceptor subtypes are coupled to the intracellular second messenger system via Gi/o proteins, therefore, the activation of presynaptic α2 adrenoceptors could reduce GABAergic transmission onto histaminergic neurons by causing the reduction of cAMP levels because of the inhibition of AC, which is mediated by Gα subunits, and the opening of GIRK channels and the inhibition of VDCCs, which is mediated by Gβγ subunits (Wu and Saggau 1997; Brown and Sihra 2008). In addition, the inhibition of neurotransmitter release could be mediated by the direct modulation of synaptic release machinery downstream of the Ca2+ influx (Wu and Saggau 1997; Brown and Sihra 2008). Although the physiological roles of α2 adrenoceptors in the peripheral or central nervous system have been studied extensively, only a few previous studies have described the mechanisms underlying the α2 adrenoceptor-mediated inhibition of neurotransmitter release. For example, the activation of α2 adrenoceptors reduces NE release from chick sympathetic neurons via the inhibition of ω-conotoxin GVIA-sensitive N-type VDCCs in chick sympathetic neurons (Boehm and Huck 1996). In the case of VLPO neurons, the activation of α2 adrenoceptors reduces the probability of spontaneous GABA release in an AC/cAMP-dependent manner (Matsuo et al. 2003).

In this study, we tested for all the possible mechanisms of α2A adrenoceptor-mediated inhibition of GABA release in histaminergic neurons. Our results that NE still decreased GABAergic IPSCs even in the presence of Ba2+ or SQ22536, suggested no involvement of GIRK channels and AC/cAMP-dependent pathways in the NE-induced inhibition of GABAergic IPSCs. We also found that the NE-induced inhibition of GABAergic IPSCs was highly dependent on the [Ca2+]o, and the inhibitory action of NE was inversely related to the [Ca2+]o. These results suggest that the NE-induced inhibition of GABAergic IPSCs might be related to the reduction of presynaptic Ca2+ influx, presumably via the inhibition of presynaptic VDCCs, because the release probability is proportional to the change in the intra-terminal Ca2+ concentration (Wu and Saggau 1997). Further studies are needed to reveal whether the inhibition of presynaptic Ca2+ entry is mediated by the α2 adrenoceptor-mediated inhibition of presynaptic VDCCs and which VDCC subtypes are involved in the NE-induced inhibition of GABAergic IPSCs.

We found that NE had no effect on the frequency and amplitude of GABAergic mIPSCs. Similarly, a previous study had shown that the activation of α2 adrenoceptors inhibits action potential-dependent glutamate release from primary afferent terminals but has no effect on the amplitude and frequency of miniature excitatory post-synaptic currents in rat substantia gelatinosa neurons (Kawasaki et al. 2003). One explanation of the differential effects of NE on action potential-dependent and action potential-independent GABA release onto TMN neurons would be that presynaptic α2A adrenoceptors might be located on pre-terminal regions that are not close to the release sites of GABAergic nerve terminals, although there is no evidence for the detailed localization of α2A adrenoceptors involved in the presynaptic regulation of GABAergic transmission. Alternatively, since GABAergic mIPSCs recorded from TMN neurons were independent of the Ca2+ influx from the extracellular space, the activation of presynaptic α2A adrenoceptors might not affect the synaptic release machinery downstream of the Ca2+ influx in GABAergic nerve terminals (see Brown and Sihra 2008).

Physiological implications

Noradrenergic inputs from the locus coeruleus (LC), as well as A1 and A2 noradrenergic neurons, to the basal forebrain, including the medial septal and medial preoptic areas, are related to the noradrenergic promotion of arousal (Koyama and Kayama, 1993; Berridge 2008). Since the discharge rates of LC neurons are highest during wakefulness and low during sleep (Foote et al. 1980; Aston-Jones and Bloom 1981), NE would play pivotal roles in the regulation of arousal/behavioral state (Foote and Morrison 1987). In the basal forebrain, α1 and β adrenoceptors are likely to be involved in the noradrenergic modulation of arousal (Berridge and Foote 1996; Berridge et al. 2003, 2005). In the medial preoptic area, however, the activation of α2 adrenoceptors seems to produce arousal (Ramesh et al. 1995; Ramesh and Kumar 1998). On the other hand, the VLPO and TMN receive dense noradrenergic innervations from the LC and noradrenergic A1 and A2 cell groups (Ericson et al. 1989; Chou et al. 2002; Lee et al. 2005), suggesting that the excitability of neurons within these two areas are regulated by the noradrenergic system. In fact, NE hyperpolarizes sleep-active VLPO neurons to decrease their excitability (Gallopin et al. 2000).

In this study, we have shown that NE acts on presynaptic α2A adrenoceptors to reduce GABAergic transmission onto TMN histaminergic neurons. Furthermore, we could identify that the activation of presynaptic α2A adrenoceptors inhibits monosynaptic GABAergic input from the VLPO region to TMN neurons using sagittal brain slices, although we could not verify whether the present monosynaptic GABAergic IPSCs originate from sleep-active VLPO neurons (see Nakamura and Jang 2012). Given that GABAergic transmission from sleep-active VLPO neurons to TMN histaminergic neurons play a crucial role in the regulation of sleep-wakefulness (Sherin et al. 1996, 1998; Steininger et al. 2001), our present results suggest that NE might increase the excitability of TMN neurons by inhibiting GABAergic transmission; therefore, the noradrenergic system could contribute to the promotion of arousal by acting on the TMN histaminergic system. This speculation is generally consistent with the noradrenergic promotion of arousal (Berridge 2008). Given that the discharge rates of LC neurons are highest during waking (Foote et al. 1980; Aston-Jones and Bloom 1981), NE is expected to modulate GABAergic transmission onto TMN neurons in a behavioral state-dependent manner.

In conclusion, we have shown that NE acts on α2A adrenoceptors on GABAergic nerve terminals innervating histaminergic neurons to inhibit action potential-dependent GABA release by reducing the presynaptic Ca2+ influx. This α2A adrenoceptor-mediated modulation of GABAergic transmission onto TMN neurons may affect the excitability of these neurons and thus might contribute to the behavioral state-dependent consolidation of wakefulness.

Acknowledgement

This study was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012R1A2A2A02046812 and 2012-0009327).

Conflicts of interest

None.

Ancillary