| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Neurosurgery and Interdisciplinary Neuroscience Program, College of Medicine, Pennsylvania State University, Hershey, Pennsylvania; and 2Department of Psychiatry, Duke University Medical Center, Durham, North Carolina
Submitted 22 May 2007; accepted in final form 29 June 2007
 |
ABSTRACT |
|---|
|
-aminobutyric acid type A receptors (GABAARs), is alone capable of triggering robust action potentials to increase cell firing. This excitatory influence of GABAergic input results from the Cl– homeostasis that maintains intracellular Cl– at high levels. The GABAA excitation is often followed by a slow inhibition mediated by GABABRs that suppresses tonic firing. Interestingly, in a subpopulation of the cells, the GABAB inhibition exhibits a remarkably low threshold for synaptic activation in that low-strength GABAergic input often activates selectively the GABAB slow inhibition, whereas the GABAA excitation requires further increases in stimulus strength. Our study demonstrates that the dual activation of GABAergic excitation and inhibition through GABAARs and GABABRs generates distinct temporal patterns of cell firing that alter the cellular output in an activity-dependent manner. |
|
INTRODUCTION |
|---|
|
-aminobutyric acid type A and type B receptors (GABAARs and GABABRs) (Curtis et al. 1970
; Dutar and Nicoll 1988a; Hill and Bowery 1981). The GABAARs generate a fast hyperpolarization of postsynaptic cells by increasing Cl– conductance (Allen et al. 1977; Curtis et al. 1970; Gray and Johnston 1985). This GABAAR-mediated inhibition (GABAA inhibition) composes the fundamental—and oftentimes the sole—form of GABAergic inhibition in response to low-strength activation of GABAergic input. The GABABRs, on the other hand, are generally activated in response to either high-frequency or large-intensity stimulation of GABAergic input (Buhl et al. 1994; Isaacson et al. 1993; Miles and Wong 1984). The activation of GABABRs produces a slowly rising and longer-lasting membrane hyperpolarization through increases in K+ conductance (Dutar and Nicoll 1988b; Gahwiler and Brown 1985; Hill and Bowery 1981; Newberry and Nicoll 1984). Such differences in the activation threshold and timescale of the two types of GABAergic inhibition diversify the neurotransmitter action of GABA in regulating neural network activity. Moreover, the differential activation of GABAARs and GABABRs in an activity-dependent manner often determines the pattern of neural network oscillations as shown, for instance, in the thalamus (Kim and McCormick 1988a,b; Kim et al. 1995, 1997).
In contrast to the typical GABAergic inhibition just cited, a question has been raised recently as whether GABAergic input under any physiological condition is also capable of generating an excitation of postsynaptic cells. Relevant to this question, it is noteworthy that certain electrical properties of the postsynaptic membrane were shown to exploit the otherwise typical GABAergic inhibition to promote action potential generation. For example, as most vividly illustrated in the thalamus and the inferior olive (Crunelli and Leresche 1991; Deschenes et al. 1984; Jahnsen and Llinas 1984; Llinas and Yarom 1981), the membrane hyperpolarization by GABAergic inhibition can lead to a rebound activation of low-threshold Ca2+ current. This Ca2+ current, in turn, generates a transient membrane depolarization that induces a burst of high-frequency action potentials. More recently, a different kind of GABAergic excitation was demonstrated in layer V cells of the cerebral cortex (Gulledge and Stuart 2003). The resting membrane potential of these cells is markedly hyperpolarized and negative to the GABAA reversal potential. Therefore synaptic activation of GABAARs in these cells results in membrane depolarization toward the GABAA reversal potential. However, because the GABAA reversal potential in these cells is still 10–15 mV more negative than spike threshold, this membrane depolarization by GABAARs alone is incapable of triggering action potentials, even though it can boost the excitatory influence of glutamatergic input that is concurrently active.
Yet, the question still persists whether activation of GABAergic input alone—without relying on such distinct electrical properties of the postsynaptic membrane as mentioned earlier—can still exert an excitatory influence on the postsynaptic cell under physiological conditions in the mature mammalian brain. Here, we provide evidence in the habenula at postnatal days 18–25 supporting the GABAergic excitation that alone triggers robust action potentials.
The habenula is located on the posterior medial surface of the caudal thalamus. It serves a pivotal role in linking various limbic forebrain areas with the midbrain neuromodulatory centers such as the interpeduncular nucleus, substantia nigra pars compacta, ventral tegmental area, and dorsal raphe nucleus (Herkenham and Nauta 1977, 1979). This habenular link has been implicated in a variety of behaviors including reward and motivation, pain, nutrition, anxiety and learned helplessness, reproduction, and sleep–waking cycle (Ellison 1994; Klemm 2004; Lecourtier and Kelly 2007; Sutherland 1982). The habenula is generally divided into the medial and lateral nucleus (Andres et al. 1999; Geisler et al. 2003; Kim and Chang 2005). Each nucleus can be further divided into a number of subnuclei (Andres et al. 1999; Geisler et al. 2003), raising the possibility for the presence of multiple channels of information flow from the limbic forebrain to the midbrain via the habenula. As a step toward understanding how the information flow from the limbic forebrain to the midbrain is regulated, we first examined how GABAergic synapses control the habenular output from the medial nucleus. Our study showed that these GABAergic synapses produce a fast excitation of postsynaptic cells that is often followed by a slow inhibition. The excitation and inhibition are mediated by GABAARs and GABABRs, respectively, and produce a distinct regulation of tonic firings of habenular cells.
|
|
METHODS |
|---|
|
Sprague–Dawley rats (18- to 25 days old) were used (Charles Rivers Laboratories, Wilmington, MA). All experimental procedures were approved by the Animal Care and Use Committee at the Pennsylvania State University College of Medicine, in accordance with the National Institutes of Health Guide for the Use and Care of Laboratory Animals. Animals were deeply anesthetized with an intraperitoneal injection of a lethal dose of sodium barbiturate. Brain slices (400 µm thick) in the coronal plane were generated from the habenula of the epithalamus. The slices were preincubated before experiments at room temperature for 2 h in artificial cerebrospinal fluid (aCSF) containing (in mM): 125 NaCl, 2.5 KCl, 26 NaHCO3, 1.25 NaH2PO4, 1.2 MgSO4, 2 CaCl2, 15 dextrose, saturated with 95% O2-5% CO2 (pH 7.4). When needed for recording, individual slices were transferred to a submerged-type recording chamber that was fixed to the stage of a BX50WI microscope (Olympus, Tokyo, Japan) and was superfused at 3 ml/min with aCSF.
Electrophysiological recordings
All recordings were performed at 34–35°C. Whole cell and gramicidin-perforated patch recordings were obtained with visual control using a x40 water-immersion objective lens with an infrared differential interference contrast filter. Recording pipettes were pulled from borosilicate glass capillaries (OD 1.5 mm, ID 0.84 mm). For whole cell patch recording, the pipette was filled with internal solution composed of (in mM): 130 K-gluconate, 10 HEPES, 1 EGTA, 2 MgCl2, 0.1 CaCl2, 4 Na2-ATP, 0.3 GTP (pH 7.2, 280 mOsm/kg). Typical electrode resistance was 3–5 M
, with access resistance in the range of 5–10 M. The liquid junction potential was calculated (15 mV) and the membrane potential corrected accordingly.
For perforated patch recording, gramicidin (Sigma, St. Louis) was dissolved in dimethylsulfoxide (DMSO, 10 mg/ml). The gramicidin solution was then diluted to a final concentration of 5–10 µg/ml with internal solution composed of 140 mM potassium gluconate, 4 mM KCl, and 10 mM HEPES (pH 7.2, 280 mOsm). When bumetanide—the sulfamoylbenzoic acid "loop" diuretic that inhibits the Na-K-Cl cotransporter—was tested for its effect on the GABAA reversal potential, KCl in the internal solution was increased to 40 mM, whereas potassium gluconate was decreased to 104 mM. The liquid junction potential was calculated (15 and 12 mV, respectively, with the 4 and 40 mM KCl internal solutions) and Vm corrected accordingly. Perforated patches developed steadily for 20–30 min after gigaseal formation. Typical series resistance was 20–30 M. Whole cell and perforated patch recordings were made using an Axoclamp 2B amplifier (Axon Instruments, Forest City, CA).
For extracellular recording, the patch pipette was filled with aCSF (3–5 M). Single-unit extracellular signals were recorded with an Xcell-3 amplifier (FHC, Bowdoinham, ME), sampled at 20 kHz.
Synaptic stimulation
Stimulating electrode (tip diameter 10–15 µm) was fabricated from a theta glass capillary and filled with aCSF. To evoke synaptic potentials, a train of stimuli at 50–200 µA was delivered at 1–20 Hz every 60 s. Each stimulus was 100 µs in duration. To isolate the GABAergic response, all synaptic potentials were recorded in the presence of 20 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 100 µM D-2-amino-5-phosphonovaleric acid (D-APV) [selective blockers of 
-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/kainate and N-methyl-D-aspartate (NMDA)–subtype glutamate receptors, respectively], 1 µM LY 341495 (antagonist of group II metabotropic glutamate receptors), and 2 µM mecamylamine (antagonist of nicotinic cholinergic receptors).
Data analysis
Data were collected and analyzed using pClamp8 software by an A/D converter (Digidata 1320, Axon Instruments) and are presented as means ± SD. All differences stated in the text are statistically significant (paired t-test, P < 0.05).
Drugs
For bath application of LY 341495, CGP 55845 (Tocris, St. Louis), and bumetanide (Sigma), they were initially dissolved in DMSO and then added to aCSF. The final concentration of DMSO was 0.01%. The bath infusion of 0.01% DMSO alone in aCSF did not alter the electrical and synaptic property of habenular cells. CNQX, APV, bicuculline, and mecamylamine (Sigma) were dissolved directly in aCSF for bath application.
GABA immunohistochemistry
Animals were perfused through the heart with 2% glutaraldehyde for 20 min. Brains were cut into 50-µm-thick sections with a Vibratome 1000plus (Vibratome, St. Louis). The sections were preincubated with 10% normal goat serum in 10 mM phosphate-buffered saline (PBS, pH 7.4) for 1 h to block nonspecific staining. Sections were then incubated overnight at room temperature in the primary GABA antibody (raised in rabbit; Chemicon, Temecula, CA) diluted 1:2,000–4,000 in PBS containing 1% normal goat serum. After rinse, sections were incubated in biotinylated anti-rabbit IgG (raised in goat) diluted 1:200 (Vector, Burlingame, CA) for 1 h. After rinse, sections were incubated in the avidin and biotinylated HRP solution (Vectastatin elite ABC kit) and were then reacted with 0.05% 3,3-diaminobenzidine (DAB) in 0.1 M phosphate buffer (pH 7.2) in the presence of 0.001% hydrogen peroxide.
GABABR and KCC2 immunohistochemistry
Animals were perfused with 4% paraformaldehyde for 40 min. Brain sections (50 µm thick) were preincubated with 10% normal goal serum in PBS for 1 h. The sections were incubated overnight in either the primary GABABR antibody (raised in guinea pig; Chemicon) diluted 1:1,000–2,000 in PBS containing 1% normal goat serum or KCC2 antibody (raised in rabbit; Chemicon) diluted to a final concentration of 20 µg/mL in PBS containing 1% normal goat serum. After rinse, sections for GABABR immunohistochemistry were incubated for 1 h in biotinylated anti-guinea pig IgG (raised in goat) diluted 1:200. For KCC2 immunohistochemistry, the sections were incubated in biotinylated anti-rabbit IgG (raised in goat) diluted 1:200 (Vector) for 1 h. After rinse, sections were incubated in the avidin and biotinylated HRP solution (Vectastatin elite kit) and were then reacted with DAB.
|
|
RESULTS |
|---|
|
All recordings were made from neurons in the ventral half of the medial habenula. By using extracellular and whole cell patch recordings, we first examined the electrical property of the cells. Medial habenular cells at rest generate tonic firing at frequencies of 0.5 to 12 Hz (5.0 ± 2.1 Hz; n = 30 cells) (Fig. 1 A). The tonic firing continued in the presence of 20 µM CNQX and 100 µM D-APV, the antagonists of AMPA/kainate- and NMDA-type glutamate receptors, respectively. This indicates that the tonic firing results from the intrinsic property of the cell membrane. To measure cell input resistance, we injected steady negative current at minimum amplitudes (5–10 pA) that suppressed tonic firing. Under this condition, a negative current pulse (10–20 pA, 300 ms) was superimposed to elicit a transient membrane hyperpolarization. The cell input resistance calculated from such voltage response was 1.2 ± 0.4 G (n = 20 cells). The detailed description of the intrinsic and morphological properties of habenular cells was provided in our recent papers (Chang and Kim 2004; Kim and Chang 2005).
|
For recording the GABAergic synaptic response, we used gramicidin-perforated patch recording. This recording technique prevents the dialysis of the cytoplasm by the patch pipette solution, thus preserving the intracellular Cl– concentration and its physiological change in response to synaptic activation of GABAARs. To isolate GABAergic transmission, we blocked ionotropic glutamate receptors by including in the bathing medium 20 µM CNQX and 100 µM D-APV. A stimulating electrode was placed in the medial habenula 100–150 µm dorsal to the cells recorded.
In 10 cells recorded, a train of stimuli at 1–10 Hz (90–130 µA) produced both fast excitatory postsynaptic potentials (EPSPs) and a slowly rising inhibitory postsynaptic potential (IPSP). These two nonglutamatergic synaptic potentials exert a distinct effect on cell firing. The representative example is shown in Fig. 1B. In this cell, synaptic stimulation at 4 Hz generated fast EPSPs, which were successful only sporadically to trigger action potentials. This sporadic action potential generation was due to the counterbalance of the EPSPs by a slow IPSP, as subsequently detailed in Fig. 4. When stimulation rate was increased to 8 Hz, the EPSPs became more potent in triggering action potentials: all of the initial four to five EPSPs consistently activated action potentials (n = 6 trials of stimulation), thus briefly increasing cell firing to 8 Hz. The increase in stimulation rate also resulted in an increase in the amplitude of the slow IPSP. This IPSP appeared to prevent some of the EPSPs late in the sequence from reaching spike threshold. On termination of stimulation, the IPSP in response to 8- to 10-Hz stimulation decayed for the next 1–3 s and suppressed tonic firing.
|
|
|
Indeed, in 14 cells tested, the block of GABAARs resulted in an abolition of the EPSPs. The representative case is illustrated in Fig. 2. In this cell, a train of stimulation generated a sequence of EPSPs and a slow IPSP. The effect of these antagonistic synaptic potentials on cell firing again varied with the frequency and strength of stimulation. For example, 10-Hz stimulation at 70 µA generated EPSPs of moderate size that only sporadically triggered action potentials, probably as a result of the counterbalance by the slow IPSP (Fig. 2A). However, when stimulus intensity was increased to 90 µA, most stimuli generated large enough EPSPs that triggered action potentials, thus increasing firing rate to around 10 Hz. In addition, as shown in Fig. 2B, the increase in stimulation rate to 20 Hz further potentiated every EPSP to trigger the action potential. These EPSPs were blocked 20–60 s after bath infusion of 20 µM bicuculline, the selective antagonist of GABAARs (Fig. 2C) (n = 14 cells). Blocking of the EPSPs uncovered the slow IPSP that suppressed tonic firing. The slow IPSP was blocked by bath infusion of 1 µM CGP 55845, the selective antagonist of GABABRs (Fig. 2D) (n = 14 cells). Taken altogether, these observations indicate that the EPSPs were mediated by GABAARs and the slow IPSP by GABABRs.
In five cells analyzed, individual GABAA EPSPs reached their peak amplitude within 3–10 ms (6.7 ± 2.1, n = 50) after onset and exhibited a duration of 54.9 ± 19.5 ms. In response to five stimuli at 10–20 Hz, the GABAB IPSP showed a rise time of 264 ± 30 ms after onset with duration of 988 ± 282 ms (n = 30).
Cellular mechanism of GABAA excitation
To characterize the mechanism underlying the fast excitation mediated by GABAARs, we measured the reversal potential of the GABAAR-mediated synaptic response (GABAA reversal potential) (n = 10 cells).
For this study, the cell membrane was voltage-clamped at various levels from –75 mV in 15-mV increments (Fig. 3). At each holding potential, a train of GABAergic input produced fast synaptic currents followed by a slow outward current (Fig. 3Aa). The fast synaptic currents were inward in the entire subthreshold voltage range and were blocked by 20 µM bicuculline, indicating its production by GABAARs (Fig. 3Ab). The slow outward synaptic current was blocked by 1 µM CGP 55845, suggesting its generation by GABABRs (Fig. 3Ac). In four cells analyzed, the rise time of individual GABAA EPSCs was 4.8 ± 0.9 ms with a duration of 32.2 ± 5.4 ms (n = 40).
To measure the GABAA reversal potential, we plotted the fast GABAA synaptic current generated in response to the first stimulus of the train as a function of the holding potential (Fig. 3Ad). The GABAA reversal potentials measured in this manner (n = 3 cells) were not statistically different from those measured after GABABRs were selectively blocked by bath infusion of CGP (n = 7 cells), so they were pulled together. The GABAA reversal potential was in the range of –48 to –25 mV (–40 ± 7 mV; n = 10 cells). In these cells, the spike threshold estimated as the onset of the regenerative phase of the action potential was –48 ± 2 mV, suggesting that the GABAA reversal potential is more positive than spike threshold.
One possible explanation for such depolarized GABAA reversal potential is Cl– homeostasis that maintains intracellular Cl– at high levels. To test for this possibility, we examined how the block of the Na+,K+,Cl–-cotransporter (NKCC) affects the GABAA reversal potential. The NKCC is known to import and accumulate Cl– in the cytoplasm of neurons (Jang et al. 2001; Kakazu et al. 1999; Russell 2000). In four cells tested, bath infusion of 20 µM bumetanide, the selective blocker of NKCC, resulted in a hyperpolarizing shift of the GABAA reversal potential by 5 to 9 mV (7 ± 2, P < 0.02) (Fig. 3B).
GABAB inhibition counterbalancing GABAA excitation
Our data so far showed that GABAergic input onto medial habenular cells generates a dual synaptic response of excitation and inhibition mediated by GABAARs and GABABRs, respectively. We show below that the relative strength of the GABAA excitation versus GABAB inhibition varies widely among different cells.
As illustrated earlier (Figs. 1B and 2), in 59% of cells analyzed (16 of 27 cells), at least three of the initial five stimuli at 5–20 Hz produced GABAA EPSPs that are potent enough to trigger action potentials. In the other 41% of cells (11 of 27), however, GABAB IPSP production was so prominent that it appreciably hyperpolarized the membrane near to the K+ equilibrium potential (–100 mV). In this manner, the GABAB IPSP effectively prevented all GABAA EPSPs except the first from reaching spike threshold. The representative case is shown in Fig. 4A. In this cell, five stimuli were applied at 6 Hz (Fig. 4Aa). The first stimulus activated the EPSP that triggered the action potential, which was evident from the spike interval that was much shorter than those between spontaneous action potentials. The first stimulus also activated a prominent, slow IPSP that hyperpolarized the membrane to –70––75 mV. The slow IPSP by the next stimuli further hyperpolarized the membrane to the peak (–95 to –100 mV), suppressing not only tonic firing but also countervailing the EPSPs. Bath infusion of 20 µM bicuculline selectively blocked the EPSPs, uncovering the slow IPSP in full (Fig. 4Ab). The slow IPSP was blocked by further infusion of 1 µM CGP 55845 (Fig. 4Ac).
The GABAB slow IPSP, to counteract the action potential generation by GABAA EPSPs, is more vividly demonstrated by selectively blocking GABABRs (n = 5 cells). The representative example is presented in Fig. 4B. In this cell, a train of three stimuli at 7 Hz generated the slow IPSP that hyperpolarized the cell membrane to –100 mV, suppressing tonic firing (Fig. 4Ba). When the GABAB IPSP was blocked by 1 µM CGP 55845, the stimulation produced robust action potentials (Fig. 4Bb). This robust action potential generation was blocked by 20 µM bicuculline (Fig. 4Bc).
Interestingly, in five of those 11 cells that generated prominent membrane hyperpolarization by the GABAB IPSP, the train of stimulation at near-minimal strengths resulted in a selective activation of the GABAB IPSP, whereas activation of GABAA EPSPs required further increases in stimulus strength. The representative case is shown in Fig. 5. In this cell, when a train of 15 stimuli was applied at 6 Hz in the current amplitude of 53 µA, the slow IPSP in response to the first five stimuli hyperpolarized the membrane to around –80 mV at peak (Fig. 5Aa). This peak hyperpolarization was maintained by the rest of the stimulation. No EPSP was activated in response to this low-strength stimulation. When stimulus intensity was raised to 73 µA, it took only the first three stimuli to produce the slow IPSP that hyperpolarized the membrane to the peak at around –95 mV (Fig. 5Ab). This peak hyperpolarization was maintained by the remaining stimulation. In contrast to the stimulation at 53 µA, each stimulus at this increased intensity generated an EPSP. The bath infusion of 20 µM bicuculline and 1 µM CGP 55845 blocked both EPSPs and slow IPSP, indicative of their generation by GABAARs and GABABRs, respectively. This feature of selective activation of GABAB IPSP is further illustrated by the plot of how the amplitudes of GABAA EPSP and GABAB IPSP change depending on stimulus strength (Fig. 5B).
|
Morphological correlates of GABAergic transmission in the habenula
Our Nissl staining with cresyl violet showed that the medial habenula is densely packed with cells (Fig. 6 A). The strategic role of the habenula that connects the limbic forebrain to the midbrain neuromodulatory centers is manifested in the two conspicuous fiber bundles in the diencephalon: the stria medullaris and the fascicles retroflexus (Fig. 6B). The stria medullaris forms the afferent pathway to the habenula from the limbic forebrain, whereas the fascicles retroflexus forms the efferent pathway from the habenula to the midbrain (Herkenham and Nauta 1977, 1979).
|
Our staining with GABABR antibody revealed a strikingly high density of GABABRs in the medial habenula (Fig. 6Da). The expression density was much higher than that in any other brain areas including the lateral habenula, thalamus, hippocampus, and cerebral cortex. On the subcellular level, GABABR immunolabels were observed outlining the somatic membrane of medial habenular cells and in the periphery of the cytoplasm (Fig. 6D, b and c).
Our immunohistochemistry with antibody to the type II K+,Cl–-cotransporter (KCC2), the most abundant isoform of KCC in the mammalian brain, revealed the lack of its expression in the medial habenula (Fig. 6E). The KCC2 is known to extrude Cl– and function to maintain Cl– concentration low in the cytoplasm of neurons (Fiumelli et al. 2005; Rivera et al. 1999, 2004; Woodin et al. 2003).
|
|
DISCUSSION |
|---|
|
GABAA excitation
In contrast to the typical production of fast inhibition in numerous brain areas, our study showed that GABAARs in the medial habenula generate a fast excitation of postsynaptic cells. This GABAergic excitation is produced in the entire subthreshold voltage range and capable of triggering robust action potentials (Figs. 1 and 2). The GABAergic excitation results from the GABAA reversal potential that is more positive than spike threshold (Fig. 3). In consequence, synaptic activation of GABAARs produces an inward current that depolarizes the membrane past spike threshold toward the GABAA reversal potential, thus triggering action potentials. This depolarized GABAA reversal potential appears to be generated, at least in part, by intracellular Cl– homeostasis, based on two observations. First, the block of NKCC resulted in a hyperpolarizing shift of the GABAA reversal potential (Fig. 3B), suggesting that Cl– import by NKCC contributes to the formation of the depolarized GABAA reversal potential. Second, our immunohistochemistry disclosed the lack of KCC2 expression in the mature medial habenula. The KCC2 is known to extrude Cl–, thus decreasing the intracellular Cl– concentration (Fiumelli et al. 2005; Rivera et al. 1999, 2004). It is reasonable to expect that the lack of KCC2 expression would help to maintain intracellular Cl– at high levels in medial habenular cells, thus contributing to the formation of the GABAA reversal potential more positive than spike threshold. Consistent with our immunohistochemistry, a previous study showed the lack of KCC2 mRNA expression—but a significant density of NKCC mRNAs—in the medial habenula of the adult rat brain (Kanaka et al. 2001).
Along with KCC2, three other isoforms of KCC (KCC1, -3, and -4) are known to be present in the mammalian brain (Clayton et al. 1998; Kanaka et al. 2001; Karadsheh et al. 2004; Payne et al. 1996; Pearson et al. 2001). In comparison to KCC2, these isoforms are expressed either at a much lower density or only in restricted areas of the rat brain. For example, KCC1 mRNA expression was shown to be generally low at all developmental stages, whereas the moderate density is restricted only to the olfactory bulb, hippocampus, cerebellum, and choroid plexus epithelial cells (Clayton et al. 1998; Kanaka et al. 2001). KCC3 is expressed at high levels only in the myelinated white matter tracts such as the corpus callosum, the dorsal column of the spinal cord, and Purkinje neurons and their axons in the cerebellum (Pearson et al. 2001). KCC4 is perhaps the least abundant because only the minimal expression level was detected in the forebrain, whereas high expression levels were restricted to the spinal cord and peripheral nerves (Karadsheh et al. 2004). Unlike the brain-specific expression of KCC2, the three other isoforms just cited are expressed in other body tissues as well. KCC1 and KCC4 were shown to be activated by cell swelling, whereas the brain-specific KCC2 is not responsive but is uniquely active under basal isotonic conditions (Gillen and Forbush 1999; Mercado et al. 2000; Payne 1997; Song et al. 2002). For these reasons, KCC2 is considered to play a major role in regulating Cl– in neurons, whereas the other isoforms may be more important in cell volume regulation. However, it remains to be seen whether and, if so, to what extent these other KCCs are involved in Cl– homeostasis and thus in GABAergic transmission in the brain.
Our finding that GABAA excitation in the medial habenula arises, at least in part, from intracellular Cl– homeostasis is similar to what was observed in the neonatal rat brain: until postnatal days 3–5, activation of GABAARs in response to exogenous agonist generates a transient membrane depolarization in numerous areas (Ben-Ari 2002; Chen et al. 1996; Cherubini et al. 1991; Payne et al. 2003). This GABAA depolarization in the infant brain seems to arise from the maintenance of intracellular Cl– at high levels as the result of a much higher density of NKCC expression than KCC2. After postnatal days 3–5, however, KCC2 expression increases dramatically to maintain intracellular Cl– at low levels (Lu et al. 1999; Yamada et al. 2004). As a result, GABAARs begin to generate membrane hyperpolarization that inhibits neural activity.
More recently, a GABAergic excitation similar to what we report here was demonstrated in a limited domain of mature neurons (Szabadics et al. 2006). In layer 2/3 pyramidal cells of the cerebral cortex at postnatal days 20–35, synaptic activation of GABAARs in the membrane of the axon initial segment produced a depolarizing postsynaptic potential that triggered the action potential; in contrast, activation of GABAARs in the somatodendritic membrane of these cells generated the typical IPSP. In these cells, the expression density of KCC2 in the membrane of the axon initial segment was shown to be much lower than that in the somatodendritic membrane. However, the GABAA reversal potential was not measured in this study and it remains to be seen whether the low-density expression of KCC2 indeed underlies the GABAA excitation in the axon initial segment by virtue of the formation of the GABAA reversal potential more positive than spike threshold. In addition, the GABAA excitation in the axon initial segment was not countervailed by GABAB slow inhibition.
The GABAA excitation in the medial habenula is different from that shown in mature hippocampal pyramidal cells (Michelson and Wong 1991; Staley et al. 1995). In these cells, tetanic stimulation of GABAergic synapses produces a biphasic sequence of fast hyperpolarization and depolarization mediated solely by GABAARs. The early hyperpolarizing response is mediated by the typical Cl– influx through the open GABAAR channels. The sustained Cl– influx, however, results in dissipation of the transmembrane Cl– electrochemical gradient, whereas HCO3– continues to move out of the cell through the open GABAAR channels and thus generates the late depolarizing response. In contrast, the GABAA excitation in the medial habenula is not preceded by the fast hyperpolarization and does not require tetanic stimulation of GABAergic input. In fact, this GABAA excitation can be generated in response to a single GABAergic input as indicated by the membrane depolarization and the action potential generation in response to the first GABAergic input in the train of stimulation (Figs. 1B, 2, and 4).
One question that still remains is whether and, if so, how this GABAergic excitation in the medial habenula changes as the brain becomes fully mature beyond the juvenile postnatal days 18–25. We were not able to address this question because our gramicidin-perforated recording from habenular cells at postnatal days 35–60 was highly unstable because the membrane seal spontaneously ruptured at the very early stage of perforation.
GABAB inhibition in the medial habenula
The GABAA excitation in the medial habenula is countervailed by the slow inhibition mediated by GABABRs. The relative strength of the GABAA excitation versus GABAB inhibition appears to vary widely among different cells. In those cells in which the GABAB inhibition was most prominent, the GABAB IPSP produced a large membrane hyperpolarization that prevented most of the GABAA EPSPs from triggering action potentials (Figs. 4 and 5).
Such prominent activation of GABAB inhibition may result from the high density of GABABR expression in the medial habenula. Previous in situ hybridization studies revealed that the medial habenula is among the brain areas that express the highest density of GABABR mRNAs (Jones et al. 1998; Kaupmann et al. 1998; Kuner et al. 1999). Consistent with these studies, our GABABR immunohistochemistry revealed a remarkably high density of GABABRs in the somatodendritic membrane of medial habenular cells. Yet, it is unclear how the low-strength stimulation of GABAergic input selectively activates GABAB inhibition as shown in a subpopulation of medial habenular cells (Fig. 5). Considering the prevailing view that GABABRs are localized mostly in the extrasynaptic membrane (Kulik et al. 2003; Mody 1994; Scanziani 2000), one possibility is that the GABA uptake system in the medial habenula is exceptionally poor in its efficiency. Accordingly, GABA would diffuse out of the synaptic cleft relatively easily and reach the GABABRs localized in the extrasynaptic membrane of neighboring cells, thus generating only the GABAB IPSP in these cells. An alternative possibility, considering the high-density expression of GABABRs in the somatodendritic membrane of medial habenular cells, is that some of these receptors are colocalized with GABAARs in the postsynaptic membrane apposing GABAergic terminals. The affinity of GABABRs for GABA is known to be much higher than that of GABAARs (Sodickson and Bean 1996). Therefore minimum release of GABA and its low concentration within the synaptic cleft below the affinity level of GABAARs may lead to a selective binding of GABA to GABABRs in the postsynaptic membrane, whereas increases in GABA release by increasing stimulus intensity result in GABA binding to both GABAARs and GABABRs. It also cannot be ruled out that some GABAergic synapses in the medial habenula contain only GABABRs in the active zone of the postsynaptic membrane and our low-strength stimulation activated only these synapses. It remains to be seen whether any of the above-cited possibilities indeed underlies the selective activation of GABAB inhibition in the medial habenula. It is noteworthy that the selective activation of GABAB slow inhibition was reported previously in a few brain areas such as the ventral tegmental area, lateral amygdala nucleus, CA1 hippocampal pyramidal neurons, and layer V pyramidal neurons of somatosensroy cortex (Benardo 1994; Nurse and Lacaille 1997; Sugita et al. 1992; Williams and Lacaille 1992). In these areas, activation of local GABAergic interneurons often generated either the GABAA fast inhibition or GABAB slow inhibition alone in the postsynaptic cell, raising the possibility that synaptic inputs to GABAARs and GABABRs may originate from separate populations of interneurons.
In recent years, it has been indicated that the synaptic response mediated by GABAARs and GABABRs can be regulated to a differential degree by intracellular signaling through various kinase systems (Dutar and Nicoll 1988a; Hahner et al. 1991; Kano et al. 1992; Kuramoto et al. 2007; Nusser et al. 1999; Poisbeau et al. 1999; Song and Messing 2005). Such differential regulation, if occurring in the medial habenula, would alter the relative size of the GABAA excitation versus GABAB slow inhibition, thus varying the scale of the GABAergic excitation under different neuromodulatory conditions. It is an interesting possibility that the difference in the relative strength of the GABAA excitation versus GABAB inhibition among different medial habenular cells in our study may be due, at least in part, to the difference in the basal activity of those kinase systems among these cells.
In conclusion, our study demonstrated that the GABAergic regulation of the cellular output can be more versatile than hitherto considered in the mature mammalian brain and produces a distinct temporal pattern of cell firing.
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: U. Kim, Department of Neurosurgery and Interdisciplinary Neuroscience Program, College of Medicine, Pennsylvania State University, Hershey, PA 17033-0850 (E-mail: ukim{at}hmc.psu.edu)
|
|
REFERENCES |
|---|
|
Andres KH, von During M, Veh RW. Subnuclear organization of the rat habenular complexes. J Comp Neurol 407: 130–150, 1999.[CrossRef][Web of Science][Medline]
Benardo LS. Separate activation of fast and slow inhibitory postsynaptic potentials in rat neocortex in vitro. J Physiol 476: 203–215, 1994.
Ben-Ari Y. Excitatory actions of GABA during development: the nature or the nurture. Nat Rev Neurosci 3: 728–739, 2002.[CrossRef][Web of Science][Medline]
Buhl EH, Halasy K, Somogy P. Diverse sources of hippocampal unitary inhibitory postsynaptic potentials and the number of synaptic release sites. Nature 368: 823–828, 1994.[CrossRef][Medline]
Chang SY, Kim U. Ionic mechanism of long-lasting discharges of action potentials triggered by membrane hyperpolarization in the medial lateral habenula. J Neurosci 24: 2172–2181, 2004.
Chen G, Trombley PQ, van den Pol AN. Excitatory actions of GABA in developing rat hypothalamic neurons. J Physiol 494: 451–464, 1996.
Cherubini E, Gaiarsa JL, Ben-Ari Y. GABA: an excitatory transmitter in early postnatal life. Trends Neurosci 14: 515–519, 1991.[CrossRef][Web of Science][Medline]
Clayton GH, Owens GC, Wolff JS, Smith RL. Ontogeny of cation-Cl– cotransporter expression in rat neocortex. Dev Brain Res 109: 281–292, 1998.[Medline]
Crunelli V, Leresche N. A role for GABAB receptors in excitation and inhibition of thalamocortical cells. Trends Neurosci 14: 16–21, 1991.[CrossRef][Web of Science][Medline]
Curtis CR, Duggan AW, Felix D, Johnson GAR. GABA, bicuculline and central inhibition. Nature 226: 1222–1224, 1970.[CrossRef][Medline]
Deschenes M, Paradis M, Roy JP, Steriade M. Electrophysiology of neurons of lateral thalamic nuclei in cat: resting properties and burst discharges. J Neurophysiol 51: 1196–1219, 1984.
Dutar P, Nicoll RA. A physiological role for GABAB receptors in the central nervous system. Nature 332: 156–158, 1988a.[CrossRef][Medline]
Dutar P, Nicoll RA. Pre- and postsynaptic GABAB receptors in the hippocampus have different pharmacological properties. Neuron 1: 585–591, 1988b.[CrossRef][Web of Science][Medline]
Ellison G. Stimulant-induced psychosis, the dopamine theory of schizophrenia, and the habenula. Brain Res Brain Rev 19: 223–239, 1994.[CrossRef][Medline]
Fiumelli H, Cancedda L, Poo MM. Modulation of GABAergic transmission by activity via postsynaptic Ca2+-dependent regulation of KCC2 function. Neuron 48: 773–786, 2005.[CrossRef][Web of Science][Medline]
Gahwiler BH, Brown DA. GABAB-receptor activated K+ current in voltage-clamped CA3 pyramidal cells in hippocampal cultures. Proc Natl Acad Sci USA 82: 1558–1562, 1985.
Geisler S, Andres KH, Veh RW. Morphologic and cytochemical criteria for the identification and delineation of individual subnuclei within the lateral habenular complex of the rat. J Comp Neurol 458: 78–97, 2003.[CrossRef][Web of Science][Medline]
Gillen CM, Forbush B. Functional interaction of the K-Cl cotransporter (KCC1) with the Na-K-Cl cotransporter in HEK-293 cells. Am J Physiol Cell Physiol 276: C328–C336, 1999.
Gray R, Johnston D. Rectification of single GABA-gated chloride channels in adult hippocampal neurons. J Neurophysiol 54: 134–142, 1985.
Gulledge AT, Stuart GJ. Excitatory actions of GABA in the cortex. Neuron 37: 299–309, 2003.[CrossRef][Web of Science][Medline]
Hahner L, McQuilkin S, Harris A. Cerebellar GABAB receptors modulate function of GABAA receptors. FASEB J 5: 2466–2472, 1991.[Abstract]
Herkenham M, Nauta WJH. Afferent connections of the habenular nuclei in the rat. A horseradish peroxidase study, with a note on the fiber-of-passage problem. J Comp Neurol 173: 123–145, 1977.[CrossRef][Web of Science][Medline]
Herkenham M, Nauta WJH. Efferent connections of the habenular nuclei in the rat. J Comp Neurol 187: 19–48, 1979.[CrossRef][Web of Science][Medline]
Hill DR, Bowery NG. 3H-baclofen and 3H-GABA bind to bicuculline-insensitive GABAB sites in rat brain. Nature 290: 149–152, 1981.[CrossRef][Medline]
Isaacson JS, Solis JM, Nicoll RA. Local and diffuse synaptic actions of GABA in the hippocampus. Neuron 10: 165–175, 1993.[CrossRef][Web of Science][Medline]
Jahnsen H, Llinas R. Electrophysiological properties of guinea-pig thalamic neurons: an in vitro study. J Physiol 349: 205–226, 1984.
Jang IS, Jeong HJ, Akaike N. Contribution of the Na-K-Cl cotransporter on GABAA receptor-mediated presynaptic depolarization in excitatory nerve terminals. J Neurosci 21: 5962–5972, 2001.
Jones KA, Borowsky B, Tamm JA, Craig DA, Durkin MM, Dai M, Yao WJ, Johnson M, Gunwaldsen C, Huang LY, Tang C, Shen Q, Salon JA, Morse K, Laz T, Smith KE, Nagarathnam D, Noble SA, Branchek TA, Gerald C. GABAB receptors function as a heteromeric assembly of the subunits GABABR1 and GABABR2. Nature 396: 674–679, 1998.[CrossRef][Medline]
Kakazu Y, Akaike N, Komiyama S, Nabekura J. Regulation of intracellular chloride by cotransporters in developing lateral superior olive neurons. J Neurosci 19: 2843–2851, 1999.
Kanaka C, Ohno K, Okabe A, Kuriyama K, Itoh T, Fukuda A, Sato K. The differential expression patterns of messenger RNAs encoding K-Cl cotransporters (KCC1, 2) and Na-K-2Cl cotransporter (NKCC1) in the rat nervous system. Neuroscience 104: 933–946, 2001.[CrossRef][Web of Science][Medline]
Kano M, Kano M, Fukunaga K, Konnerth A. Synaptic excitation produces a long-lasting rebound potentiation of inhibitory synaptic signals in cerebellar Purkinje cells. Nature 356: 601–604, 1992.[CrossRef][Medline]
Karadsheh MF, Byun N, Mount DB, Delpire E. Localization of the KCC4 potassium-chloride cotransporter in the nervous system. Neuroscience 123: 381–391, 2004.[CrossRef][Web of Science][Medline]
Kaupmann K, Malitschek B, Schuler V, Heid J, Froestl W, Beck P, Mosbacher J, Bischoff S, Kulik A, Shigemoto R, Karschin A, Bettler B. GABAB-receptor subtypes assemble into functional heteromeric complexes. Nature 396: 683–687, 1998.[CrossRef][Medline]
Kim U, Bal T, McCormick DA. Spindle waves are propagating synchronized oscillations in the ferret LGNd in vitro. J Neurophysiol 74: 1301–1323, 1995.
Kim U, Chang SY. Dendritic morphology, local circuitry, and intrinsic electrophysiology of neurons in the rat medial and lateral habenular nuclei of the epithalamus. J Comp Neurol 483: 236–250, 2005.[CrossRef][Web of Science][Medline]
Kim U, McCormick DA. The functional influence of burst and tonic firing mode on synaptic interactions in the thalamus. J Neurosci 18: 9500–9516, 1998a.
Kim U, McCormick DA. Functional and ionic properties of a slow afterhyperpolarization in ferret perigeniculate neurons in vitro. J Neurophysiol 80: 1222–1235, 1998b.
Kim U, Sanchez-Vives MV, McCormick DA. Functional dynamics of GABAergic inhibition in the thalamus. Science 278: 130–134, 1977.
Klemm WR. Habenular and interpeduncularis nuclei: shared components in multiple-function networks. Med Sci Monit 10: RA261–RA273, 2004.[Web of Science][Medline]
Kulik A, Vida I, Lujan R, Haas CA, Lopez-Bendito G, Shigemoto R, Frotscher M. Subcellular localization of metabotropic GABAB receptor subunits GABAB1a/b and GABAB2 in the rat hippocampus. J Neurosci 23: 11026–11035, 2003.
Kuner R, Kohr G, Gruneward S, Eisenhardt G, Bach A, Kornau HC. Role of heteromer formation in GABAB receptor function. Science 283: 74–77, 1999.
Kuramoto N, Wilkins ME, Fairfax BP, Revilla-Sanchez R, Terunuma M, Tamaki K, Iemata M, Warren N, Couve A, Calver A, Horvath Z, Freeman K, Carling D, Huang L, Gonzales C, Cooper E, Smart TG, Pangalos MN, Moss SJ. Phospho-dependent functional modulation of GABA(B) receptors by the metabolic sensor AMP-dependent protein kinase. Neuron 53: 233–247, 2007.[CrossRef][Web of Science][Medline]
Lecourtier L, Kelly PH. A conductor hidden in the orchestra? Role of the habenular complex in monoamine transmission and cognition. Neurosci Biobehav Rev 31: 658–672, 2007.[CrossRef][Medline]
Llinas R, Yarom Y. Properties and distribution of ionic conductances generating electroresponsiveness of mammalian inferior olivary neurons in vitro. J Physiol 315: 569–584, 1981.
Lu J, Karadsheh M, Delphire E. Developmental regulation of the neuronal-specific isoform of K-Cl cotransporter KCC2 in postnatal rat brains. J Neurobiol 39: 558–568, 1999.[CrossRef][Web of Science][Medline]
Mercado A, Song L, Vazquez N, Mount DB, Gamba G. Functional comparison of the K+-Cl– cotransporters KCC1 and KCC4. J Biol Chem 275: 30326–30334, 2000.
Michelson HB, Wong RKS. Excitatory synaptic responses mediated by GABAA receptors in the hippocampus. Science 253: 1420–1423, 1991.
Miles R, Wong RK. Unitary inhibitory synaptic potentials in the guinea-pig hippocampus in vitro. J Physiol 356: 97–113, 1984.
Mody I, De Koninck Y, Otis TS, Soltesz I. Bridging the cleft at GABA synapses in the brain. Trends Neurosci 17: 517–525, 1994.[CrossRef][Web of Science][Medline]
Newberry NR, Nicoll RA. Direct hyperpolarizing action of baclofen on hippocampal pyramidal cells. Nature 308: 450–452, 1984.[CrossRef][Medline]
Nurse S, Lacaille JC. Do GABAA and GABAB inhibitory postsynaptic responses originate from distinct interneurons in the hippocampus? Can J Physiol Pharmacol 75: 520–525, 1997.[CrossRef][Web of Science][Medline]
Nusser Z, Sieghar W, Mody I. Differential regulation of synaptic GABAA receptors by cAMP-dependent protein kinase in mouse cerebellar and olfactory bulb neurons. J Physiol 521: 421–435, 1999.
Payne JA. Functional characterization of the neuronal-specific K-Cl cotransporter: implication for [K+]o regulation. Am J Physiol Cell Physiol 273: C1516–C1525, 1997.
Payne JA, Rivera C, Voipio J, Kaila K. Cation-chloride co-transporters in neuronal communication, development and trauma. Trends Neurosci 26: 199–206, 2003.[CrossRef][Web of Science][Medline]
Payne JA, Stevenson TJ, Donaldson LF. Molecular characterization of a putative K-Cl cotransporter in rat brain. J Biol Chem 271: 16245–16252, 1996.
Pearson MM, Lu J, Mount DB, Delpire E. Localization of the K-Cl cotransporter, KCC3, in the central and peripheral nervous system: expression in the choroid plexus, large neurons and white matter tracts. Neuroscience 103: 481–491, 2001.[CrossRef][Web of Science][Medline]
Poisbeau P, Cheney MC, Browning MD, Mody I. Modulation of synaptic GABAA receptor function by PKA and PKC in adult hippocampal neurons. J Neurosci 15: 674–683, 1999.
Rivera C, Voipio J, Payne JA, Ruusuvuori E, Lahtinen H, Lamsa K, Pirvola U, Saarma M, Kaila K. The K+/Cl– co-transporter KCC2 renders GABA hyperpolarizing during neuronal maturation. Nature 397: 251–255, 1999.[CrossRef][Medline]
Rivera C, Voipio J, Thomas-Crusells J, Li H, Emri Z, Sipilä S, Payne JA, Minichiello L, Saarma M, Kaila K. Mechanism of activity-dependent downregulation of the neuron-specific K-Cl cotransporter KCC2. J Neurosci 24: 4683–4691, 2004.
Russell JM. Sodium-potassium-chloride cotransport. Physiol Rev 80: 211–276, 2000.
Scanziani M. GABA spillover activates postsynaptic GABAB receptors to control rhythmic hippocampal activity. Neuron 25: 673–681, 2000.[CrossRef][Web of Science][Medline]
Sodickson DL, Bean BP. GABAB receptor-activated inwardly rectifying potassium current in dissociated hippocampal CA3 neurons. J Neurosci 16: 6374–6385, 1996.
Song L, Mercado A, Vazquez N, Xie Q, Desai R, George AL, Gamba G, Mount DB. Molecular, functional, and genomic characterization of human KCC2, the neuronal K-Cl cotransporter. Brain Res Mol Brain Res 103: 91–105, 2002.[Medline]
Song M, Messing RO. Protein kinase C regulation of GABAA receptors. Cell Mol Life Sci 62: 119–127, 2005.[CrossRef][Web of Science][Medline]
Staley KJ, Soldo BL, Proctor WR. Ionic mechanisms of neuronal excitation by inhibitory GABAA receptors. Science 269: 977–981, 1995.
Sugita S, Johnson SW, North RA. Synaptic inputs to GABAA and GABAB receptors originate from discrete afferent neurons. Neurosci Lett 134: 207–211, 1992.[CrossRef][Web of Science][Medline]
Sutherland RJ. The dorsal diencephalic conduction system: a review of the anatomy and functions of the habenular complex. Neurosci Biobehav Rev 6: 1–13, 1982.[CrossRef][Web of Science][Medline]
Szabadics J, Varga C, Molnar G, Olah S, Barzo P, Tamas G. Excitatory effect of GABAergic axo-axonic cells in cortical microcircuits. Science 311: 233–235, 2006.
Williams S, Lacaille JC. GABAB receptor-mediated inhibitory postsynaptic potentials evoked by electrical stimulation and by glutamate stimulation of interneurons in stratum lacunosum-moleculare in hippocampal CA1 pyramidal cells in vitro. Synapse 11: 249–258, 1992.[CrossRef][Web of Science][Medline]
Woodin MA, Ganguly K, Poo MM. Coincident pre- and postsynaptic activity modifies GABAergic synapses by postsynaptic changes in Cl– transporter activity. Neuron 39: 807–820, 2003.[CrossRef][Web of Science][Medline]
Yamada J, Okabe A, Toyoda H, Kilb W, Luhmann HJ, Fukuda A. Cl– uptake promoting depolarizing GABA actions in immature rat neocortical neurons is mediated by NKCC1. J Physiol 557: 829–841, 2004.
This article has been cited by other articles:
|
|
 |
|
  Y. Kim and L. O. Trussell Negative Shift in the Glycine Reversal Potential Mediated by a Ca2+- and pH-Dependent Mechanism in Interneurons J. Neurosci., September 16, 2009; 29(37): 11495 - 11510. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
) was measured from spike threshold. To measure the amplitude of GABAA EPSP (
), individual EPSPs by the last 5 stimuli of the train in all the trials were pulled together and averaged. It is noteworthy that the lowest stimulus intensity activated only the GABAB IPSP. Ca: cell is voltage-clamped at –70 mV. Stimulation at 60 µA generates only slow outward currents. Cb: increase in stimulus intensity to 80 µA generates both fast inward and slow outward currents. Cc: these fast and slow currents are blocked by bicuculline and CGP.