| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Departments of Neurology and 2Cellular and Molecular Physiology, Yale University School of Medicine, New Haven, Connecticut; and 3Neurology Service, Veteran's Affairs Medical Center, West Haven, Connecticut
Submitted 21 May 2006; accepted in final form 18 July 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
; Cavelier et al. 2005; Hamann et al. 2002; Nusser and Mody 2002; Rossi et al. 2003; Semyanov et al. 2004; Soltesz and Nusser 2001) and that this is caused by activation of high-affinity, extrasynaptic GABAA receptors (Brickley et al. 1999, 2001; Mody 2001; Rossi and Hamann 1998; Stell and Mody 2002; Stell et al. 2003) by the low levels of ambient GABA normally present in the brain (Lerma et al. 1986; Tossman et al. 1986). The source of GABA responsible for tonic inhibition, however, remains unclear. The most common explanation is that some of the GABA released into the synaptic cleft on vesicular fusion is not recovered by GABA transporters. It escapes from the synapse and diffuses to extrasynaptic sites. It is likely that this "spillover" (Isaacson et al. 1993; Rossi and Hamann 1998) does occur, given observations that tonic inhibition is often increased in response to pharmacological blockade of GAT1 (Keros and Hablitz 2005; Overstreet and Westbrook 2001; Rossi et al. 2003; Semyanov et al. 2003, 2004; Wall and Usowicz 1997), and is greater in slices prepared from GAT1 knockout mice (Chiu et al. 2005; Jensen et al. 2003). However, it is also likely that there are additional nonvesicular sources of GABA because tonic inhibition can still occur after blocking vesicular GABA release with tetanus toxin, concanamycin, or removal of extracellular calcium (Hamann et al. 2002; Rossi et al. 2003; Wu et al. 2001, 2003).
It has recently been proposed that GABA transporters help determine the level of tonic inhibition by attempting to clamp ambient [GABA] at the level at which transporters are at equilibrium (Richerson and Wu 2003). This equilibrium is determined by the stoichiometry of GABA transporters. GAT1 translocates two sodium ions, one chloride ion, and one uncharged GABA molecule with each thermodynamic reaction cycle. Because of this stoichiometry, it is predicted that GAT1 is near equilibrium under normal physiological conditions and would reverse with a relatively small increase in membrane potential or cytosolic [GABA] (Richerson and Wu 2003). In support of this hypothesis, there is a large increase in tonic GABA current in rat hippocampal cultures that have been depolarized with a small increase in extracellular [K+] or that have been treated with the GABA transaminase inhibitor vigabatrin, which increases cytosolic [GABA]. In both cases the increase in tonic current is not prevented by blocking vesicular GABA release, and is decreased by GABA transporter antagonists (Wu et al. 2001, 2003).
Here we used cultured hippocampal neurons and found that there was a significant amount of tonic GABA current in the absence of vigabatrin treatment. The amount of tonic current was unaffected by removing extracellular Ca2+ but was decreased by blocking GAT1. The amount of tonic current was sensitive to the transmembrane sodium gradient—increasing in response to manipulations that decrease the normally inward sodium gradient and decreasing in response to others that increase the inward sodium gradient. Similar results were obtained in slices, except that a GAT1-independent nonvesicular mechanism was found to play a larger role in GABA release. We conclude that when vesicular GABA release is reduced, GABA transporters will reverse in an attempt to reach their equilibrium. Under normal conditions, there is probably net uptake of synaptically released GABA by transporters, but they will only reduce extracellular [GABA] to the level at which they are at equilibrium, which may still be sufficiently high to induce tonic inhibition. Thus the source of GABA release may be conceptually less important in determining the level of tonic inhibition than the extracellular [GABA] at which GABA transporters are at equilibrium.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Primary cell cultures of hippocampal neurons and glia were prepared from neonatal (P0–P2) Sprague-Dawley rats as described previously (Gaspary et al. 1998; Wu et al. 2003). Hippocampi were dissociated and plated on coverslips coated with poly-L-ornithine and incubated in culture medium containing 63% modified Eagle's medium (MEM), 7% fetal bovine serum (FBS), and 30% Neurobasal medium/B27 supplement, with the following additives: 1 ng/ml basic fibroblast growth factor (bFGF), 10 ng/ml FGF-5, 100 U/ml penicillin, and 100 µg/ml streptomycin. At 24 h, one-half of the medium was changed to 100% Neurobasal/B27 with the same additives. Cultures were fed with half-medium changes every 1–4 wk. Cytosine 
-D-arabino-furanoside hydrochloride (Ara-C; 0.2–0.3 µM) was added to the original plating medium to control glial growth but was not included in subsequent medium changes. Cultures were grown for 4–10 wk before recording, unless indicated. The age of the cultures used are noted as the number of days grown in vitro (DIV).
Brain slices
Acute brain slices were prepared using standard methods. Juvenile Sprague-Dawley rats (12–18 days old) were killed by decapitation, and the brain was rapidly removed. The brain was mounted on a vibratome and submerged in ice cold Ringer solution that contained (in mM) 124 NaCl, 3 KCl, 2 CaCl2, 2 MgCl2, 1.3 NaH2PO4, 26 NaHCO3, 10 dextrose, and 1 kynurenic acid. Two hundred-micrometer-thick slices of the hippocampus were prepared and incubated for 2 h in the same solution at room temperature before recording. Bath solutions were bubbled with 5% CO2-95% O2 and had a pH of 7.4.
Electrophysiology
Brain slices and coverslips with cultured cells were placed in a chamber on a fixed stage upright light microscope (Axioskop FS, Carl Zeiss). Neurons were visualized with a 40x water immersion objective using DIC imaging. Cells and slices were superfused at 3–4 ml/min with Ringer that contained (in mM) 124 NaCl, 3 KCl, 2 CaCl2, 2 MgCl2, 1.3 NaH2PO4, 26 NaHCO3, and 10 dextrose. For all recordings except those noted in Figs. 1 and 2 and those in Fig. 5, the bath solution was changed to zero calcium Ringer, which was the same solution except that CaCl2 was omitted and EGTA (1 mM) was added. In all cases, recordings were not made until 
5 min after switching to zero calcium Ringer. For experiments in Fig. 5, the control solution contained (in mM) 141 NaCl, 1.7 KCl, 2 MgCl2, 1.3 KH2PO4, 10 HEPES, 1 EGTA, and 10 dextrose (pH adjusted to 7.4 with NaOH), and the zero sodium solution contained (in mM) 141 N-methyl-D-glucamine (NMDG), 141 HCl, 1.7 KCl, 2 MgCl2, 1.3 KH2PO4, 10 HEPES, 1 EGTA, and 10 dextrose (pH adjusted to 7.4 with NMDG). For all recordings, GABAB receptors were blocked with CGP-55845 (1 µM), and glutamate receptors were blocked with either kynurenic acid (1 mM) or 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 µM) and (±)-2-amino-5-phosphonopentanoic acid (AP-5; 50 µM). All changes in the bath solution were made by switching a valve supplying the chamber from a gravity feed system. Exchange of the bath solution occurred within 20 s.
|
|
|
|
|
Patch-clamp recording electrodes (1.4–3.0 M
) were fabricated from thin-walled, borosilicate glass tubing (Diamond General, Ann Arbor, MI) with a micropipette puller (Model P-87, Sutter Instruments, Novato, CA). Whole cell recordings were made using an electrode solution that contained (in mM) 124 CsCl, 10 N-(2,6-dimethlphenylcarbamoylmethyl)triethylammonium chloride (QX-314), 10 HEPES, and 10 EGTA, except for those shown in Fig. 2A, in which case the electrode solution contained (in mM) 130 KOH, 135 methanesulfonic acid, 5 NaOH 2 KCl, 5 HEPES, 1 EGTA, and 5 GABA. Perforated patch-clamp recordings were made using methods described previously (Rae et al. 1991; Wu et al. 2001). One set of recordings was made using either amphotericin (n = 3) or gramicidin (n = 2) as the ionophore with a solution that contained (in mM) 135 KOH, 135 methanesulfonic acid, 5 KCl, 5 HEPES, and 1 EGTA. A second set of recordings was made using gramicidin (n = 5) with a solution that contained (in mM) 135 KCl, 10 HEPES, and 1 EGTA. Electrode solutions were adjusted to pH 7.2 with either KOH or CsOH and to an osmolarity of 270 ± 5 mOsm with dH2O. Initial seal resistance was >1 G. Access resistance was typically 5–15 M during whole cell recordings and 15–30 M during perforated patch recordings.
Recordings were performed using a MultiClamp 700A amplifier (Axon Instruments, Foster City, CA). Most recordings were made in voltage-clamp mode. During whole cell recordings holding potential was –60 mV and the calculated Nernst potential for Cl– was –1 mV, so all GABAA receptor–mediated currents were inward. During perforated-patch recordings, holding potential was –40 or –50 mV. The recording in Fig. 1C was made in current-clamp mode. Paired recordings between synaptically coupled neurons in Fig. 2A were made with the presynaptic neuron in current-clamp mode and the postsynaptic neuron in voltage-clamp mode at a holding potential of –50 mV. Recordings (except those in Figs. 1C and 2) were filtered (at 100 Hz), digitized (at 200 samples/s), and stored on computer using a commercially available data acquisition system (Digidata 1322A and PClamp software, Axon Instruments). Those in Fig. 1C and Fig. 2 were filtered at 1 KHz and sampled at 2,000 samples/s.
Sources for chemicals
Neurobasal medium, B27 supplement, and bFGF were purchased from Gibco BRL Products (Gaithersburg, MD). MEM was purchased from JRH Biosciences (Lenexa, KS). QX-314 and TTX were purchased from Alomone Labs (Jerusalem, Israel). SKF-89976a and CGP-55845 were purchased from Tocris Cookson (Ballwin, MO). Veratridine, kynurenic acid, AP-5, CNQX, bicuculline methiodide, gabazine, SNAP-5114, FBS, FGF-5, and all other chemicals not listed were purchased from Sigma Chemical (St. Louis, MO).
Data analysis
Tonic GABAA receptor–mediated current was quantified by measuring the change in holding current induced after exchanging the bath solution with one that contained bicuculline methiodide (50 µM) or gabazine (10 µM). Commercially available software (Clampfit, Axon Instruments) was used to subtract the baseline holding current from the steady-state current during bicuculline or gabazine application to obtain the mean GABAA receptor–mediated current. During recordings with calcium, tonic current was calculated using baseline current measured between inhibitory postsynaptic currents (IPSCs). When miniature IPSCs occurred, there was no attempt to measure tonic current between these mIPSCs, but the error in measurement of the tonic current in normal calcium Ringer ranged from 0 to 20 pA. This error was even smaller for zero Ca2+ Ringer (see RESULTS).
Probability values were determined using a Mann-Whitney rank sum test (SigmaStat). NS refers to a difference that is not statistically significant. Values expressed as "n" are the total number of neurons recorded. All values are means ± SE, and all error bars are SE.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
We have previously shown in hippocampal cultures that vigabatrin induces a large amount of nonvesicular GABA release that is mediated in part by reversal of GABA transporters (Wu et al. 2001, 2003). Here we performed similar experiments and found that there was also GABA-mediated tonic current in hippocampal cultures naïve to drug treatment. During gramicidin (n = 2) and amphotericin (n = 3) perforated-patch recordings in which the electrode solution contained 5 mM KCl, gabazine (n = 2) and bicuculline (n = 3) blocked an outward, or inhibitory, current at a holding potential of –40 mV (Fig. 1A, top). As in vigabatrin-treated cultures (Wu et al. 2001, 2003), the GAT1 antagonist SKF-89976a (40 µM) partially decreased the tonic current (Fig. 1A, bottom) by an average of 76% (n = 5), indicating that it was caused in part by nonvesicular GABA release through reversal of GAT1. To be certain that the tonic current was inhibitory under our experimental conditions, these recordings were repeated (n = 5) using gramicidin perforated-patch recordings containing 135 mM KCl. Gramicidin is permeable to monovalent cations, but not to chloride, so that neurons maintain their normal cytosolic [Cl–]. Therefore if an outward current is blocked by bicuculline that would indicate that the reversal potential for GABAA receptors is more negative than the holding potential and that GABA would induce an inhibitory current. The use of a high [Cl–] in the pipette guards against inadvertent breakthrough of the perforated patch, because if that occurred then GABAA receptor activation would be depolarizing. Under these conditions, bicuculline decreased a tonic outward current during voltage-clamp recordings (holding potential = –50 mV; Fig. 1B; n = 5) and led to an increase in action potential firing during current-clamp recordings (Fig. 1C; n = 5).
When whole cell voltage-clamp recordings were made in Ringer containing 2 mM calcium, bicuculline induced a decrease in tonic current in most neurons (Fig. 1D). In this case, the current was inward, because the electrode solution contained a high [Cl–]. The amplitude of this current did not change in response to removal of extracellular Ca2+ (Fig. 1E). On average, the amplitude of tonic current in 19 neurons (age = 41 ± 1 DIV) recorded in 2 mM Ca2+ was 167 ± 25 pA, and it was 163 ± 11 pA in the same neurons in zero Ca2+ solution (NS; Fig. 1F). These results are similar to those seen previously in slices from the cerebellum after vesicular release was decreased by removal of extracellular Ca2+ or depletion of vesicular GABA with concanamycin (Hamann et al. 2002; Rossi et al. 2003).
Zero calcium Ringer solution blocked evoked IPSCs and significantly reduced spontaneous synaptic events
To confirm that zero Ca2+ solution was effective in reducing vesicular GABA release, recordings were made from pairs of synaptically-coupled neurons in culture after blocking glutamate receptors with CNQX and AP-5. In a subset of recordings, stimulation of presynaptic neurons induced an outward current in postsynaptic neurons that was blocked by bicuculline (50 µM; n = 8). In 39 neurons, a postsynaptic response was induced by stimulating action potentials in the presynaptic neuron at 20–50 Hz, and in all 39 cases, these IPSCs were completely blocked by changing the bath solution to zero Ca2+ solution (Fig. 2A).
Llano et al. (2000) have reported that mIPSCs can occur in cerebellar Purkinje cells in the absence of extracellular calcium. We also observed mIPSCs in zero Ca2+ solution with TTX in cultured hippocampal neurons. However, the frequency of synaptic events was much lower than it was in normal Ringer. Changing bath solution from normal Ringer (with 2 mM Ca2+) to zero Ca2+ Ringer significantly reduced the number of postsynaptic events to 25 ± 3% of control and the total charge transfer to 18 ± 3% of control (n = 10 neurons; age = 26 ± 2 DIV). Comparing recordings (Fig. 2, B and C) and amplitude histograms (Fig. 2D) under the two conditions revealed that removing extracellular Ca2+ had a greater effect on larger events—probably IPSCs mediated by calcium influx during action potentials. Consistent with this assumption, TTX (0.4 µM) had little additional effect on the frequency or size distribution of synaptic events (Fig. 2, B–D). The frequency of synaptic events in zero calcium solution with TTX was 104 ± 15% of that in zero calcium solution (NS), and the total charge transfer was 120 ± 23% (NS).
Amplitude of tonic current increased with age in culture
The size of the response to bicuculline was dependent on the number of days that neurons were grown in culture (Fig. 3). This increase in tonic current was caused in part by an increase in membrane surface area (assessed by cell capacitance) and in part by an increase in current density (Table 1). The mechanism for the latter was not studied further, but could potentially be caused by an increase in extracellular [GABA], an increase in GABAA receptor density, or a developmental switch to GABAA receptor subunits with higher affinity. The increase in tonic current with age was not caused by a change in the reversal potential of GABAA receptors with maturation (Cherubini et al. 1991; Rivera et al. 1999), because the transmembrane chloride gradient of the recorded cell was clamped by the whole cell patch-clamp recording method.
|
TTX induced a decrease in tonic current
The amount of tonic current was less when recordings were made in solutions containing TTX (Fig. 4). To quantify this difference, we first measured the response to bicuculline during recordings from neurons (51 ± 5 DIV) in zero calcium Ringer and then from the same neurons in zero calcium Ringer containing TTX (0.2–0.4 µM). There was 356 ± 78 pA of tonic current in zero calcium Ringer (n = 10), and this decreased to 161 ± 39 pA in TTX (P < 0.05). On average, TTX led to a decrease in tonic current to 47 ± 6% of control (P < 0.001).
There are two possible explanations for the reduction in tonic current by TTX. The most obvious one is that blocking action potentials led to a reduction in vesicular release. However, this is unlikely to explain the effect of TTX, because TTX does not reduce vesicular GABA release below the already low level that exists in zero calcium Ringer (Fig. 2, A–D).
An alternative explanation is that TTX reduces influx of Na+ into neurons, leading to a decrease in cytosolic [Na+], a possibility for which there is experimental evidence (see DISCUSSION). Together with the decrease in mean membrane potential that would occur because of blockade of action potentials, this would be expected to decrease the outward driving force for GABA transporters, reduce GABA efflux, and cause a decrease in the extracellular [GABA] at which GABA transporters are at equilibrium (Richerson and Wu 2003).
Tonic current was enhanced by removing extracellular Na+
To further test the hypothesis that ambient [GABA] is sensitive to the transmembrane sodium gradient, the amplitude of tonic current in zero calcium HEPES solution was compared with that in zero calcium HEPES solution in which sodium was replaced by NMDG. On switching the bath solution to zero sodium solution, there was an initial small decrease in inward current, followed by a larger increase in inward current (Fig. 5A). The delayed inward current was caused by activation of GABAA receptors, because it was blocked by bicuculline (Fig. 5A). After blocking GABAA receptors with bicuculline, the initial decrease in inward current remained and was probably caused by a decrease in sodium leak current. In five neurons (31 ± 0 DIV), zero sodium solution induced a peak inward current of –767 ± 126 pA, and this was changed to an outward current of +55 ± 24 pA by bicuculline (P < 0.01). The effect of bicuculline was mimicked by the selective GABAA receptor antagonist gabazine (control response = –599 ± 301 pA vs. response in gabazine = +72 ± 26 pA; n = 3; Fig. 5B). The GABA efflux was primarily caused by transporter reversal, because it was completely and reversibly blocked by SKF-89976a (Fig. 5C). In nine neurons (34 ± 0 DIV), zero sodium solution induced a peak inward current of –1330 ± 195 pA, and this decreased to –20 ± 36 pA (P < 0.001) in the presence of SKF-89976a (40 µM).
After extracellular sodium was replaced with NMDG, the GABA-mediated inward current decayed slowly to a new steady-state value, with a level of tonic current that was now larger than that in zero calcium Ringer (Fig. 5D). Thus the level of tonic current was 86 ± 11 pA (n = 27; 37 ± 1 DIV) in the presence of extracellular sodium, whereas in the same set of neurons, the level of tonic current was 251 ± 29 pA in the absence of extracellular sodium for 5 min (P < 0.001).
These results are consistent with an increase in GABA release through transporter reversal. A reduction in extracellular [Na+] would lead to a large outward driving force for GABA transporters. This would cause an initial surge of GABA release as the transporters respond to the large outwardly directed sodium gradient. The GABA efflux would be expected to decay as cytosolic sodium and GABA are depleted. Such a decrease in cytosolic [Na+] does occur in cultured hippocampal neurons in response to removal of extracellular Na+ (Rose and Ransom 1997). In contrast, these results are not consistent with an increase in vesicular GABA release that had not been blocked by zero calcium solution. In that case, zero sodium solution should hyperpolarize neurons and block action potentials, which would inhibit vesicular GABA release.
Tonic current was enhanced by preventing inactivation of sodium channels
We tested the hypothesis that the tonic current would increase in response to an increase in cytosolic [Na+] and membrane potential by exposing cultures to a low concentration of veratridine (which prevents inactivation of sodium channels) in zero calcium solution. Continuous exposure to veratridine (0.8 µM) induced a large increase in tonic current (Fig. 6A; n = 13). This effect of veratridine was concentration dependent (Fig. 6B). Thus the amount of tonic current was 134 ± 33 pA in control solution (n = 10; 27.3 ± 0.2 DIV), 146 ± 27 pA in 0.1 µM veratridine (n = 10; 27.3 ± 0.2 DIV), 421 ± 41 pA in 0.4 µM veratridine (n = 15; 32.2 ± 1.2 DIV), and 672 ± 109 pA in 0.8 µM veratridine (n = 13; 32.7 ± 1.0 DIV). These differences were statistically significant for 0.1 versus 0.4 µM (P < 0.001) and 0.4 versus 0.8 µM (P < 0.05). The effect of veratridine was prevented by coapplication of TTX (Fig. 6C; n = 7). Veratridine did not cause IPSCs or an increase in mIPSCs in zero calcium solution, indicating that the increase in tonic current was not caused by stimulation of vesicular GABA release.
|
) and membrane potential.
|
The purpose of the experiments presented here was to understand how changes in the thermodynamic equilibrium of GAT1 affect ambient [GABA]. For this purpose, tissue culture has the advantage that it allows better control of the extracellular solution. However, there may be important differences in tonic inhibition in culture compared with brain slices. For that reason, it was important to determine whether the conclusions we made from recordings in culture also hold for recordings from brain slices.
Whole cell patch-clamp recordings were made from neurons in the CA1 pyramidal cell layer in hippocampal slices from 12- to 18-day-old rats. There was not a significant amount of tonic current at baseline in either normal Ringer (n = 10) or zero calcium Ringer (n = 7). This was not unexpected, because others have reported that CA1 pyramidal neurons do not have a significant amount of tonic GABA current unless GAT1 antagonists are used (Frahm et al. 2001; Stell et al. 2003). There are many reasons that there could be more tonic GABA current at baseline in our cultures than in slices, such as the specific GABA receptor subunits expressed in culture, the older age of neurons in culture, or the greater heterogeneity of neurons in culture. It is also possible that our cultures contained a higher proportion of GABAergic neurons than in slices, although this is less likely because the majority of neurons in our cultures were glutamatergic as assessed by measuring evoked EPSCs during paired recordings.
Acute application of veratridine at concentrations effective in cultures (
1 µM) did not induce any tonic current during recordings in zero calcium solution (n = 3). However, application of 30 µM veratridine in zero calcium solution did induce a large inward current (peak current = –662 ± 61 pA; n = 33; Fig. 8A). This current was caused by an increase in ambient [GABA], because it was blocked by gabazine (10 µM; peak current = –13.3 ± 3.6 pA; n = 9; Fig. 8A) or bicuculline (50 µM; peak current = –18.5 ± 3.5 pA; n = 7). The increase in tonic current was also blocked by the GAT1 antagonists SKF-89976a (20–40 µM; peak current = –22.2 ± 1.3 pA; n = 5) or NO-711 (10 µM; peak current = –19 ± 6 pA; n = 6; Fig. 8B), indicating that it was caused by reversal of the GABA transporter.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
The data presented here show that there is tonic GABA mediated current in cultures of the rat hippocampus under control conditions. This result is in contrast to our previous experiments using a similar approach, in which it was necessary to pretreat cells with vigabatrin to induce a significant amount of tonic current in rat hippocampal cultures (Wu et al. 2001). The previous experiments were done in cultures that were younger than those used here, and recordings were made in Ringer that contained TTX. The experiments presented here that showed a decrease in tonic current with TTX and younger age (Figs. 3 and 4) suggest that these two differences may account for the smaller amount of tonic current in those previous experiments.
Role of spillover as a source of GABA for tonic inhibition
The tonic current that we recorded in culture was not dependent on extracellular Ca2+. In previous work, tonic inhibition in cerebellar slices was also not reduced by removing extracellular calcium, by blocking release of calcium from intracellular stores, or by depleting vesicular GABA content with concanamycin (Hamann et al. 2002; Rossi et al. 2003). Together, these results suggest that, in both preparations, continuous vesicular GABA release (i.e., spillover) is not essential as a source of ambient GABA for tonic inhibition.
Under normal conditions, spillover likely plays a major role as a source of ambient GABA. However, the data presented here suggest that when vesicular release is significantly reduced, GAT1 reversal can replace spillover in providing ambient GABA for tonic inhibition. This result is consistent with the hypothesis that GAT1 functions effectively in clamping extracellular [GABA] at the level at which GAT1 is at equilibrium (Richerson and Wu 2003). When spillover occurs, GAT1 likely operates in the forward direction, but in its absence, GAT1 likely operates in reverse, in both cases regulating extracellular [GABA] at the same level.
These data do not address the relative contribution of GAT1 reversal in comparison to other forms of nonvesicular GABA release. In fact, our data suggest that other nonvesicular forms of GABA release do exist in our preparation (Richerson and Wu 2003; Wu et al. 2001, 2005), although the mechanisms these other forms of GABA release have not yet been defined.
Level of tonic current was sensitive to the transmembrane sodium gradient
The amount of tonic GABA current was decreased by TTX and increased by veratridine. As above, this was not caused by residual vesicular GABA release that was enhanced by veratridine or reduced by TTX, because evoked IPSCs had already been completely blocked by zero Ca2+ solution and the infrequent synaptic events that remained were not affected by TTX (Fig. 2). In addition, if veratridine did increase tonic current by stimulating vesicular release, then any IPSCs that were induced should have been observed, which was not the case.
The most likely explanation for the effect of TTX and veratridine was that they altered cytosolic [Na+] and membrane potential, and this changed the driving force on GABA transporters. There is experimental evidence that these manipulations can change [Na+]i. For example, in neurons from Lymnaea, TTX reduces [Na]i under baseline conditions (Onizuka et al. 2004). In cultures of the hippocampus, there are spontaneous, transient increases in [Na]i of 5 mM in 27% of coverslips, and these are blocked by TTX (Rose and Ransom 1997). These Na+ transients are also blocked by 10 mM Mg2+, suggesting they are caused by synchronized bursting of neurons. Some of these hippocampal neurons respond to an increase in [K+]o with an increase in [Na+]i, and this is also blocked by TTX. Thus under a variety of conditions, TTX can decrease [Na+]i. Under the conditions of our experiments, TTX might have a particularly large effect on [Na+]i, because zero calcium solution can induce bursting of hippocampal neurons (Shuai et al. 2003). Thus the most likely explanation for the effect of TTX on tonic current is that [Na+]i was reduced and neurons were hyperpolarized, and this decreased the extracellular [GABA] at which GABA transporters are at equilibrium (Richerson and Wu 2003). This is consistent with an earlier report that TTX can reduce tonic GABA current (Wall and Usowicz 1997). It is frequently assumed that TTX induces a decrease in transmitter release solely because of blockade of synaptic activity, but the results obtained here indicate that caution should be used in making this conclusion, because TTX may also reduce transporter-mediated GABA release.
The response to veratridine may have an opposite explanation. Veratridine can increase cytosolic [Na+] (Rose and Ransom 1997) and causes depolarization, both of which would increase the extracellular [GABA] at which GABA transporters are at equilibrium (Richerson and Wu 2003). Consistent with this conclusion, Szerb (1979) reported that 50 µM veratridine induces nonvesicular GABA release from cortical slices.
Replacing sodium with NMDG in the absence of extracellular calcium also increased tonic GABA current. This could not be explained by stimulation of calcium-independent vesicular GABA release, because replacing Na+ with NMDG would block action potentials and hyperpolarize neurons. On the other hand, this manipulation would reverse the sodium gradient and cause an outward driving force on GABA transporters (Richerson and Wu 2003). Removing extracellular Na+ would hyperpolarize neurons, which would reduce the outward driving force on GAT1, but would simultaneously reverse the [Na+] gradient, which would increase the outward driving force. Because this manipulation led to an increase in tonic current, it is apparent that changes in the [Na+] gradient can influence GABA transport independent of changes in membrane potential. Consistent with the observed decay of tonic current with prolonged exposure to zero sodium solution (Fig. 5), GABA efflux should wane with time as intracellular Na+ (Onizuka et al. 2004; Rose and Ransom 1997) and GABA are depleted. Reversal of GABA transporters has previously been proposed as the mechanism for increased GABA release during ischemia in rats after sodium substitution with NMDG (Phillis et al. 1999).
Effect of changes in the [Na+] gradient on tonic GABA current in culture is through GAT1
Confirming that the increases in tonic current induced in culture by veratridine and zero sodium solution were mediated entirely by reversal of GAT1, they were both completely blocked by SKF-89976a. The dominant role of GAT1 reversal under these conditions might be caused by the fact that neurons are favored as the primary source of nonvesicular GABA release. GAT1 is present on glia (Conti et al. 2004; Kinney and Spain 2002; Minelli et al. 1995; Ribak et al. 1996; Vitellaro-Zuccarello et al. 2003), but it is more heavily expressed in neurons. There is probably relatively little cytosolic GABA in glia at baseline, because of their high expression of GABA transaminase. Because vigabatrin blocks GABA transaminase, GABA efflux from glia may play a bigger role after treatment with vigabatrin (Wu et al. 2001). In fact, there is evidence that vigabatrin increases GABA release by both GAT1-dependent and GAT1-independent mechanisms (Wu et al. 2003, 2005). However, the low [GABA] that would be expected in glia in the current paradigm would not favor GAT1-independent GABA release from glia. In addition, veratridine acts on fast transient sodium channels, which can be found on glia (Sontheimer et al. 1991), but are more abundant on neurons. Veratridine increases cytosolic [Na+] in neurons (Rose and Ransom 1997) but has not been shown to alter cytosolic [Na+] in glia. For these reasons, there may be a larger effect of GAT1 blockade on tonic current in our culture protocol than would occur under other circumstances.
Comparison of tonic current in culture and brain slices
The major conclusions from the experiments in culture were supported by data from hippocampal slices. Ambient [GABA] was increased by veratridine and zero sodium solution, and the GABA release responsible for this was not calcium dependent. Thus tonic GABA current was modulated by changes in the transmembrane sodium gradient. This tonic GABA current depended in part on GAT1-mediated GABA release. We believe that the GABA release induced by veratridine in slices may have also come primarily from neurons, because veratridine raises intracellular [Na+] in neurons but not glia (Rose and Ransom 1997). In addition, Na+ channels and GAT1 are both expressed more on neurons than glia, and this GABA release was completely blocked by SKF-89976a.
There were also differences between brain slices and culture. For example, there was not a significant amount of tonic GABA current in slices at baseline. It was also much more difficult to exchange solution in brain slices, which was likely to be responsible for the need to use a higher concentration of veratridine in slices than in culture, the difficulty in getting full recovery from block by gabazine and SKF-89976a (Fig. 8), and the differences in the kinetics of the response to zero sodium solution (Fig. 5 vs. Fig. 8).
Although some of these differences between slices and culture may have been simply methodological, there were also some differences suggesting that alternative mechanisms of tonic inhibition exist in slices. For example, SKF-89976a did not block nonvesicular GABA release induced by zero sodium solution in brain slices. Thus there is a GAT1-independent mechanism of nonvesicular GABA release in slices. There are many alternative mechanisms of nonvesicular release of neurotransmitters (Hell et al. 1991; Vautrin et al. 2000; Wang et al. 2002; Ye et al. 2003), and several of these may be particularly relevant to brain slices, including GAT3 reversal (Kinney 2005; Kirmse and Kirischuk 2006), swelling-activated anion channels (Kimelberg et al. 1990), and leakage from damaged cells (Wall and Usowicz 1997). For example, there is cell damage caused by trauma and ischemia during brain slice preparation, which could induce the latter two forms of release. Alternatively, because normal neuron–glia interactions are better preserved in slices than in culture, glial release of GABA through GAT3 could be preserved more faithfully in slices. Our experiments showed that the GAT3 antagonist SNAP-5114 did not block the response to zero sodium solution (Fig. 8D). We interpret these results as showing that GAT1 and GAT3 are not required for nonvesicular GABA release. It is possible that there is a contribution from GAT1 or GAT3 reversal (Kinney 2005; Kirmse and Kirischuk 2006), but neither of these are the sole mechanism of nonvesicular GABA release under these conditions. The mechanism of the additional source(s) of nonvesicular GABA release that is GAT1 and GAT3 independent remains unclear.
Physiological significance of the response of GABA transporters to the transmembrane sodium gradient
Some of the changes in sodium used here would not realistically occur under physiological conditions. However, they provide evidence that GABA transporters are influenced by the transmembrane sodium gradient and that changes in sodium concentration might alter the level of tonic inhibition under normal physiological conditions. In fact, because there are two sodium ions cotransported with each GABA molecule, the driving force on GAT1 is more steeply dependent on the sodium gradient than it is on either the GABA or chloride gradients or membrane potential. There is a small amount of sodium influx with each action potential, and during a burst of action potentials, there is an even greater rise in intracellular [Na+]. Thus even modest changes in neuronal activity would be expected to alter the driving force on GAT1 and may favor GAT1 reversal.
The influence of sodium levels on tonic GABA inhibition may be even greater under pathological conditions. When energy demand exceeds supply, such as during seizures or brain ischemia, there will be a reduction in activity of the Na/K ATPase. The resulting rise in intracellular [Na+] would be larger than during normal physiological activity, and this would be predicted to cause a large amount of nonvesicular GABA efflux. Thus it is likely that the effect of sodium on tonic inhibition defined here are relevant to brain function under physiological and/or pathological conditions.
GABA transporters regulate the level of tonic inhibition even when they do not reverse
It is not clear how commonly GABA transporters reverse in vivo. Under the conditions of these experiments, when vesicular release is blocked by zero calcium solution and extracellular GABA is continuously washed away by the flowing bath solution, it is more likely that GABA transporters would continuously operate in reverse in an attempt to bring extracellular [GABA] back up to the equilibrium level. However, even when GABA transporters are not reversing, they may still play an important role in regulating the amount of tonic inhibition. We believe that it is conceptually less important to determine whether GABA transporters are operating in the forward or reverse direction than to define the extracellular [GABA] at which GABA transporters are at their equilibrium. Based on both experimental evidence and theoretical modeling (Richerson and Wu 2003), GABA transporters can be viewed as attempting to clamp extracellular [GABA] at a level determined by the membrane potential and the transmembrane gradients of Na+, Cl–, and GABA. If there is a large amount of vesicular GABA release, GABA transporters will operate in the forward direction, but only as much as is needed to drive extracellular [GABA] down to the level at which they are at equilibrium. If there is no vesicular GABA release, or there is significant clearance of GABA by some mechanism other than GABA transport (e.g., diffusion away from the tissue in vitro), GABA transporters will operate in reverse in an attempt to raise extracellular [GABA] to the equilibrium level. In this view, GABA transporters would play an important role in determining the level of tonic inhibition by regulating extracellular [GABA], regardless of the direction of transport. We propose that GABA transporters attempt to clamp extracellular [GABA] at the level at which they are at or near their thermodynamic equilibrium, and under some conditions, this level is high enough to induce tonic inhibition.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: G. B. Richerson, Neurology, LCI-712B, Yale University School of Medicine, 15 York St., PO 208018, New Haven, CT 06520-8018 (E-mail: George.Richerson{at}Yale.Edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Brickley SG, Cull-Candy SG, and Farrant M. Single-channel properties of synaptic and extrasynaptic GABAA receptors suggest differential targeting of receptor subtypes. J Neurosci 19: 2960–2973, 1999.
Brickley SG, Revilla V, Cull-Candy SG, Wisden W, and Farrant M. Adaptive regulation of neuronal excitability by a voltage- independent potassium conductance. Nature 409: 88–92, 2001.[CrossRef][Medline]
Cavelier P, Hamann M, Rossi D, Mobbs P, and Attwell D. Tonic excitation and inhibition of neurons: ambient transmitter sources and computational consequences. Prog Biophys Mol Biol 87: 3–16, 2005.[CrossRef][ISI][Medline]
Cherubini E, Gaiarsa JL, and Ben-Ari Y. GABA: an excitatory transmitter in early postnatal life. Trends Neurosci 14: 515–519, 1991.[CrossRef][ISI][Medline]
Chiu CS, Brickley S, Jensen K, Southwell A, Mckinney S, Cull-Candy S, Mody I, and Lester HA. GABA transporter deficiency causes tremor, ataxia, nervousness, and increased GABA-induced tonic conductance in cerebellum. J Neurosci 25: 3234–3245, 2005.
Conti F, Minelli A, and Melone M. GABA transporters in the mammalian cerebral cortex: localization, development and pathological implications. Brain Res Brain Res Rev 45: 196–212, 2004.[CrossRef][Medline]
Frahm C, Engel D, and Draguhn A. Efficacy of background GABA uptake in rat hippocampal slices. Neuroreport 12: 1593–1596, 2001.[CrossRef][ISI][Medline]
Gaspary HL, Wang W, and Richerson GB. Carrier-mediated GABA release activates GABA receptors on hippocampal neurons. J Neurophysiol 80: 270–281, 1998.
Hamann M, Rossi DJ, and Attwell D. Tonic and spillover inhibition of granule cells control information flow through cerebellar cortex. Neuron 33: 625–633, 2002.[CrossRef][ISI][Medline]
Hell JW, Edelmann L, Hartinger J, and Jahn R. Functional reconstitution of the 
-aminobutyric acid transporter from synaptic vesicles using artificial ion gradients. Biochemistry 30: 11795–11800, 1991.[CrossRef][Medline]
Isaacson JS, Solis JM, and Nicoll RA. Local and diffuse synaptic actions of GABA in the hippocampus. Neuron 10: 165–175, 1993.[CrossRef][ISI][Medline]
Jensen K, Chiu CS, Sokolova I, Lester HA, and Mody I. GABA transporter-1 (GAT1)-deficient mice: differential tonic activation of GABAA versus GABAB receptors in the hippocampus. J Neurophysiol 90: 2690–2701, 2003.
Keros S and Hablitz JJ. Subtype-specific GABA transporter antagonists synergistically modulate phasic and tonic GABAA conductances in rat neocortex. J Neurophysiol 94: 2073–2085, 2005.
Kimelberg HK, Goderie SK, Higman S, Pang S, and Waniewski RA. Swelling-induced release of glutamate, aspartate, and taurine from astrocyte cultures. J Neurosci 10: 1583–1591, 1990.[Abstract]
Kinney GA. GAT-3 transporters regulate inhibition in the neocortex. J Neurophysiol 94: 4533–4537, 2005.
Kinney GA and Spain WJ. Synaptically evoked GABA transporter currents in neocortical glia. J Neurophysiol 88: 2899–2908, 2002.
Kirmse K and Kirischuk S. Ambient GABA constrains the strength of GABAergic synapses at Cajal-Retzius cells in the developing visual cortex. J Neurosci 26: 4216–4227, 2006.
Lerma J, Herranz AS, Herreras O, Abraira V, and Martin dR. In vivo determination of extracellular concentration of amino acids in the rat hippocampus. A method based on brain dialysis and computerized analysis. Brain Res 384: 145–155, 1986.[CrossRef][ISI][Medline]
Llano I, Gonzalez J, Caputo C, Lai FA, Blayney LM, Tan YP, and Marty A. Presynaptic calcium stores underlie large-amplitude miniature IPSCs and spontaneous calcium transients. Nat Neurosci 3: 1256–1265, 2000.[CrossRef][ISI][Medline]
Minelli A, Brecha NC, Karschin C, DeBiasi S, and Conti F. GAT-1, a high-affinity GABA plasma membrane transporter, is localized to neurons and astroglia in the cerebral cortex. J Neurosci 15: 7734–7746, 1995.[Abstract]
Mody I. Distinguishing between GABAA receptors responsible for tonic and phasic conductances. Neurochem Res 26: 907–913, 2001.[CrossRef][ISI][Medline]
Nusser Z and Mody I. Selective modulation of tonic and phasic inhibitions in dentate gyrus granule cells. J Neurophysiol 87: 2624–2628, 2002.
Onizuka S, Kasaba T, Hamakawa T, Ibusuki S, and Takasaki M. Lidocaine increases intracellular sodium concentration through voltage-dependent sodium channels in an identified lymnaea neuron. Anesthesiology 101: 110–120, 2004.[CrossRef][ISI][Medline]
Overstreet LS and Westbrook GL. Paradoxical reduction of synaptic inhibition by vigabatrin. J Neurophysiol 86: 596–603, 2001.
Phillis JW, Song D, and O'Regan MH. Effects of hyperosmolarity and ion substitutions on amino acid efflux from the ischemic rat cerebral cortex. Brain Res 828: 1–11, 1999.[CrossRef][ISI][Medline]
Rae J, Cooper K, Gates P, and Watsky M. Low access resistance perforated patch recordings using amphotericin B. J Neurosci Methods 37: 15–26, 1991.[CrossRef][ISI][Medline]
Ribak CE, Tong WM, and Brecha NC. GABA plasma membrane transporters, GAT-1 and GAT-3, display different distributions in the rat hippocampus. J Comp Neurol 367: 595–606, 1996.[CrossRef][ISI][Medline]
Richerson GB and Wu Y. The dynamic equilibrium of neurotransmitter transporters: not just for reuptake anymore. J Neurophysiol 90: 1363–1374, 2003.
Rivera C, Voipio J, Payne JA, Ruusuvuori E, Lahtinen H, Lamsa K, Pirvola U, Saarma M, and Kaila K. The K+/Cl- co-transporter KCC2 renders GABA hyperpolarizing during neuronal maturation. Nature 397: 251–255, 1999.[CrossRef][Medline]
Rose CR and Ransom BR. Regulation of intracellular sodium in cultured rat hippocampal neurones. J Physiol 499: 573–587, 1997.[CrossRef][ISI]
Rossi DJ and Hamann M. Spillover-mediated transmission at inhibitory synapses promoted by high affinity 
6 subunit GABAA receptors and glomerular geometry. Neuron 20: 783–795, 1998.[CrossRef][ISI][Medline]
Rossi DJ, Hamann M, and Attwell D. Multiple modes of GABAergic inhibition of rat cerebellar granule cells. J Physiol 548: 97–110, 2003.
Semyanov A, Walker MC, and Kullmann DM. GABA uptake regulates cortical excitability via cell type-specific tonic inhibition. Nat Neurosci 6: 484–490, 2003.[ISI][Medline]
Semyanov A, Walker MC, Kullmann DM, and Silver RA. Tonically active GABAA receptors: modulating gain and maintaining the tone. Trends Neurosci 27: 262–269, 2004.[CrossRef][ISI][Medline]
Shuai J, Bikson M, Hahn PJ, Lian J, and Durand DM. Ionic mechanisms underlying spontaneous CA1 neuronal firing in Ca2+-free solution. Biophys J 84: 2099–2111, 2003.
Soltesz I and Nusser Z. Neurobiology—background inhibition to the fore. Nature 409: 24–27, 2001.[Medline]
Sontheimer H, Ransom BR, Cornell-Bell AH, Black JA, and Waxman SG.</
