The Journal of Neurophysiology Vol. 77 No. 4 April 1997, pp. 2202-2208
Copyright ©1997 by the American Physiological Society

Endogenous GABA Activates Small-Conductance K⁺ Channels Underlying Slow IPSCs in Rat Hippocampal Neurons

Yves De Koninck¹ and Istvan Mody²

¹ Department of Pharmacology and Therapeutics, McGill University, Montreal, Quebec H3G 1Y6, Canada; and ² Departments of Neurology and Physiology, UCLA School of Medicine, Los Angeles, California 90095

	ABSTRACT

Abstract Introduction Methods Results Discussion References

De Koninck, Yves and Istvan Mody. Endogenous GABA activates small-conductance K⁺ channels underlying slow IPSCs in rat hippocampal neurons. J. Neurophysiol. 77: 2202-2208, 1997. The objective of this study was to determine the properties of K⁺ channels activated by endogenously released trasmitter under synaptic conditions. First, the levels of -aminobutyric acid (GABA) were depleted in hippocampal nerve endings to establish the relative contribution of endogenously released GABA to the activation of GABA_B receptors mediating slow inhibitory postsynaptic currents (IPSCs). Inhibition of glutamic acid decarboxylase and GABA reuptake effectively depleted >85% of the releasable GABA pool, producing parallel reductions of GABA_A and GABA_B receptor-mediated IPSCs, indicating that both classes of receptors are activated synaptically by endogenously released GABA. Whole cell patch-clamp recordings of stimulus-evoked slow IPSCs at potentials hyperpolarized from the potassium reversal potential were consistent with the activation of a nonrectifying (n = 3) or slightly outwardly rectifying (n = 4) K⁺ conductance by the endogenously released GABA. Spectral analysis of the decay phase of GABA_B IPSCs revealed several time constants indicating complex underlying channel kinetics. Nonstationary variance analysis yielded a small unitary conductance in the range of 5-13 pS, consistent with a large number of channels activated during evoked currents. These results indicate that in granule cells of the dentate gyrus, GABA released synaptically from interneuron terminals activates an unusually small K⁺ conductance, with no or slight outward rectification. This conductance is therefore unlike those typically reported for neuronal G protein-coupled K⁺ channels or those activated by exogenously applied baclofen with larger, inwardly rectifying conductances.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Several experiments suggest that -aminobutyric acid (GABA) is the endogenous mediator of slow, GABA_B receptor-mediated inhibitory postsynaptic currents (IPSCs). For example, exogenously applied GABA activates a K⁺ conductance with a pharmacological profile similar to that of the slow IPSC (for reviews, see Misgeld et al. 1995; Thompson 1994) and GABA uptake inhibitors enhance the amplitude and prolong the duration of GABA_B inhibitory postsynaptic potentials/IPSCs (Isaacson et al. 1993; Thompson and Gähwiler 1992). Yet, activation of postsynaptic GABA_B receptors generally requires stimulation of the presynaptic terminals with a stimulus intensity or frequency greater than that necessary to elicit GABA_A responses (Destexhe and Sejnowski 1995; Misgeld et al. 1995; Mody et al. 1994; Thompson 1994). The larger stimulus intensities or faster stimulus frequencies may be required to 1) activate extrasynaptic GABA_B receptors (spillover of GABA) (Isaacson et al. 1993), 2) produce a sufficiently long activation of these receptors to raise the intracellular concentration of G proteins to the level required for activation of the K⁺ channels (Destexhe and Sejnowski 1995), or 3) cause the release of another neurotransmitter or cofactor necessary for the activation of these receptors (Mody et al. 1994). To determine the relative contribution of endogenous GABA versus that of other possible endogenous agents, we studied the effects of depleting the releasable pool of GABA from interneuron terminals. This was accomplished by inhibiting the GABA synthesizing enzyme glutamic acid decarboxylase (GAD) with isonicotinic acid hydrazide (isoniazid, INH) in combination with a reuptake blocker, nipecotic acid (NIP) (Geddes and Wood 1984; Maynert and Kaji 1962; Wood and Abrahams 1971; Wood et al. 1988).

To date, little is known about the properties of the K⁺ channels activated by endogenously released GABA, i.e., under synaptic conditions. The GABA_B receptors activated by their endogenous agonist may engage channels distinct from those activated by synthetic GABA_B agonists such as baclofen or may favor distinct conductance or kinetic states for the same channel (for review, see Misgeld et al. 1995). Our previous findings on synaptically elicited GABA_B IPSCs have shown the activation of an outwardly rectifying K⁺ channel (Otis et al. 1993), in contrast to inwardly rectifying conductances activated by exogenous agonists (Misgeld et al. 1995). Therefore, to determine the unitary conductance of the K⁺ channels linked to GABA_B receptors during their synaptic activation, we used nonstationary fluctuation analysis of synaptic currents (De Koninck and Mody 1994; Traynelis et al. 1993). A similar approach was used recently to derive the unit conductance of Ca²⁺-dependent K⁺ channels underlying the afterhyperpolarization of vagal motoneurons (Sah 1995). Preliminary results of this study have been published in abstract form (De Koninck and Mody 1995).

	METHODS

Abstract Introduction Methods Results Discussion References

Slice preparation

Recordings were made from granule cells of the dentate gyrus in coronal half-brain slices (400 µm thick) obtained from adult male Wistar rats (postnatal day 60; 200-400 g). Methods for the slice preparation and recordings have been described in detail previously (De Koninck and Mody 1994; Otis et al. 1993). Briefly, after pentobarbital sodium anesthesia (60 mg/kg ip), animals were decapitated, and the brain was quickly dissected and immersed for 1-2 min in cold (4°C) artificial cerebrospinal fluid (ACSF) solution. Slices were prepared with a Vibratome (Lancer Series 1000), were hemisected, and stored oxygenated at 32°C in a storage chamber until individually transferred to the recording chamber (34-35°C) at the start of an experiment.

The ACSF contained (in mM) 126 NaCl, 2.5 KCl, 2 CaCl₂, 2 MgCl₂, 26 NaHCO₃, 1.25 NaH₂PO₄, 10 glucose, 0.04 D-2-amino-5-phosphonovaleric acid (D-AP5), and 0.01 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). For noise analysis experiments, in which only monosynaptic GABA_B IPSCs were isolated, 10 µM picrotoxin was added (Davies et al. 1990). The remaining current could be completely and reversibly blocked by addition of CGP 35348 (200-800 µM; Ciba-Geigy), as previously described (Otis et al. 1993). The solutions were continuously bubbled with 95% O₂-5% CO₂ (pH 7.35 ± 0.05, mean ± SE). Drugs such as picrotoxin (10 µM), bicuculline (10-50 µM), INH (10 mM), and NIP (2 mM) were added to the ACSF. All chemicals were purchased from Sigma except CNQX and D-AP5 (Tocris Neuramin).

INH has been used to inhibit GAD activity in the brain and from synaptosomal preparations (Casey and Wood 1973; Geddes and Wood 1984; Maynert and Kaji 1962; Wood and Kurylo 1984; Wood and Peesker 1972; Wood et al. 1988), preferentially to other GAD synthesis inhibitors (Wood and Abrahams 1971). INH specifically reduced GABA levels from synaptosomes with accompanying increase in glutamate levels, as expected from block of GAD activity, and had no effect on the levels of other amino acids (aspartate, glutamine, taurine) (Geddes and Wood 1984; Wood and Kurylo 1984). Finally, the proconvulsant action of INH was shown to be specifically due to its effect on GABA metabolism and not on that of glutamate or any other amino acids (Geddes and Wood 1984).

NIP was used to prevent the reuptake of GABA that would have impaired the ability of INH to deplete GABA from the terminals (Wood et al. 1988).

Recordings

Whole cell voltage-clamp recordings were obtained with the use of borosilicate glass capillaries with an inner filament (KG-33, 1.12 mm ID, 1.5 mm OD, Garner Glass) pulled to 1.5- to 3-µm outer tip diameters (0.5-1 µm diameter of the lumen) with the use of a two-stage vertical Narishige PP-83 puller. Intracellular solution consisted of (in mM) 140 potassium gluconate, 10 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid, 2 MgCl₂, 2 MgATP, and 0.2 tris(hydroxymethyl)aminomethane-guanosine 5-triphosphate, pH titrated to 7.2 with KOH, total osmolality 255-285 mosM.

Postsynaptic currents were evoked with the use of bipolar stimulating electrodes that were placed in the middle of the molecular layer of the dentate gyrus, 1-2 mm from the recording site. Constant current stimulus intensities ranging from 100 to 400 µA and 50-200 µs in duration reliably evoked GABA_B currents in most granule cells. As reported previously (Otis et al. 1993), no differences in the kinetics of the GABA_B current or in its sensitivity to antagonists were seen for different stimulus intensities.

Data analysis and curve fitting

All analog DC signals were digitized at 44 or 88 kHz (Neurocorder, NeuroData) and stored on a videotape for later analysis. Off-line, the recordings were low-pass filtered at 2 kHz (3 dB, 8-pole Bessel, Frequency Devices 9002), sampled at 10 kHz, and analyzed with the use of software developed by Y. De Koninck as described previously (De Koninck and Mody 1994; Otis et al. 1993).

For experiments comparing the effects of INH plus NIP on the GABA_A versus GABA_B responses, to be able to quantify the effect of GABA depletion on both components in the same cell, we used a subtraction protocol (Fig. 1). Because a delay occurs before the rising phase of the GABA_B IPSC (Otis et al. 1993), a single-exponential was fit to the initial decay phase of the response to estimate the decay kinetics of the GABA_A component. The extrapolated decay was subtracted from the remaining part of the response to isolate the late GABA_B component (Fig. 1C). Similarly, the GABA_B component was fitted with a fourth-power exponential activation and double-exponential decay kinetics (Otis et al. 1993) of the form where I(t) is current as a function of time, w_h1 and w_h2 (where A = Aw_h1 + Aw_h2) are weighting factors, _m is the time constant of the activation parameter, and _h1 and _h2 are the time constants of the inactivation parameters (Otis et al. 1993). Curves were fitted by a least-square method based on a simplex algorithm (De Koninck and Mody 1994).

View larger version (16K): [in this window] [in a new window]   FIG. 1. Isolation of -aminobutyric acid-A (GABAA) and GABAB components for quantitative measurements. A: in the presence of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and D-2-amino-5-phosphonovaleric acid (D-AP5), electrical stimulation in the molecular layer of the dentate gyrus produces a mixed GABAA and GABAB receptor-mediated inhibitory postsynaptic current (IPSC). B: the 2 IPSCs can be distinguished on the basis of their time course and reversal potential. We used a subtraction protocol (Otis et al. 1993) to properly quantify the area corresponding to each component. C: according to our previous studies (Otis et al. 1993), there is a delay before in the rising phase of the GABAB IPSC. Thus the initial phase of the IPSC consists only of a GABAA component. We can therefore fit a single exponential to the early decay phase of the 1st component and subtract the extrapolated curve from the remaining part of the response to isolate the late GABAB component. Similarly, the GABAB component can be fit with a 4th-power exponential activation and double-exponential decay kinetics. Subtraction of the fitted curve from the original trace effectively isolates the GABAA component.

Nonstationary spectral and variance analysis

For noise analysis experiments, monosynaptic GABA_B IPSCs were studied in isolation (10 µM picrotoxin added). The technique used for the nonstationary noise analysis has been described in detail previously (De Koninck and Mody 1994). Briefly, with each cell, 50-300 postsynaptic currents were used for the noise analysis. For the variance analysis, a segment of a fixed length was selected from the peak of the current to past the end of the decay. The amplitude of the current was divided into bins (30-40) of equal size on the amplitude scale. Within each bin, the variance of the IPSC around the scaled average was computed and plotted against the mean current value within the scaled average current bin. The data were fitted with the relationship ² = iI_m  I²_m/N, where ² is the variance, I_m is the mean current, and i is the unitary current. For the spectral analysis, the average current was subtracted from each IPSC after scaling. Segments of 1,024-2,048 points (depending on the sampling frequency; chosen to be 5-10 times longer than the decay time constant of the ensemble average) were Parzen windowed and the one-sided power spectrum was computed by fast Fourier transform (De Koninck and Mody 1994). The power spectrum was fitted with the sum of n Lorentzian functions of the form where L(f) is the spectral density at frequency f, S_i is the power of the spectrum at f = 0, and f_ci is the cutoff frequency at which the spectral power is half. The corresponding time constants (_i) are derived with the relationship _i = 1/(2f_ci). The resulting power of the spectrum was not corrected for the power of the mean time course (Sigworth 1981), because the focus was only on evaluating the corner frequencies; therefore no attempt was made to derive the single-channel conductance from the power of the spectrum.

The goodness of fit was evaluated on the basis of fitting subsets of points drawn from the whole set of data points as well as from evaluation of the reduced ² (²) where the factor  = N  n is the number of degrees of freedom left after fitting N data points to the n parameters. The necessity of introducing additional Lorentzian components to the fits was judged first on the basis of visual inspection of the fitted curves superimposed onto the data. When the merit of additional components was not obvious, an F test was used to assess how the additional component improved the value of the reduced ² The F follows the F distribution with df₁ = 1 and df₂ = . The critical value for the merit of additional components was set at a low level (P < 0.001) to favor parsimony of the fitted function (De Koninck and Mody 1994).

	RESULTS

Abstract Introduction Methods Results Discussion References

Depletion of releasable GABA

INH (10 mM), an inhibitor of the pyridoxine cofactor of GAD (Wood et al. 1988), and NIP (2 mM), a GABA uptake blocker, were perfused onto the slices (Fig. 2). In all seven slices tested, multiple population spikes characteristic of a diminished GABAergic inhibition appeared within 45-60 min of perfusion (Fig. 2A). These epileptic field potentials were further enhanced by only 15% on block of postsynaptic GABA_A receptors with picrotoxin (100 µM) or bicuculline (100 µM) (Fig. 2, Ab and Ac). Similar values were obtained whether measurements of area or contour line (not shown) were used, consistent with reduced GAD activity and GABA uptake effectively depleting >85% of the releasable GABA pool.

View larger version (19K): [in this window] [in a new window]   FIG. 2. Depletion of the releasable GABA pools depresses both GABAA- and GABAB-mediated inhibitions. A: isoniazid and nipecotic acid depress GABAA receptor-mediated inhibition of field responses. As the drugs were perfused onto the slices, multiple population spikes characteristic of a diminished GABAergic inhibition appeared. The epileptic fields were enhanced by only 18 and 14% on block of postsynaptic GABAA receptors with picrotoxin (PTX) (10 µM) or bicuculline (BMI) (50 µM), respectively. The field responses were either quantified by measuring the area under the curve or with the use of a contour line (not shown) indicator. Similar results were obtained with both methods. B: isoniazid and nipecotic acid depress GABAA- and GABAB-mediated IPSCs, but not responses to exogenous baclofen. Each trace represents the average of 5 responses obtained by stimulating regularly at 30-s intervals. In the absence of drugs, this type of stimulation routinely yielded stable recordings lasting for several h without any rundown. The traces in Ba and Bb represent the subtracted GABAA and GABAB components, respectively (see Fig. 1C). Most of the evoked responses were typically abolished within <1 h, yet responses to exogenously applied baclofen and/or GABA always persisted, confirming that the neurons did not deteriorate and continued to respond to GABAB receptor activation. The holding potential was 60 mV. C: depressions of GABAA and GABAB IPSCs follow a parallel time course. Responses were quantified as the area under the curve for each isolated component (GABAA and GABAB, respectively) (points represent means ± SE, obtained from 4 slices).

We then investigated GABA_A and GABA_B receptor-mediated monosynaptic IPSCs in whole cell recordings during the depletion of GABA. Consistent with the above findings, INH plus NIP produced a parallel reduction in the size of GABA_A IPSCs by 96% and GABA_B IPSCs by 81% (Fig. 2, B and C). In separate experiments, CGP 35348 completely blocked GABA_B IPSCs (not shown). Responses to exogenous application of baclofen persisted after ablation of the synaptic responses. Taken together, these results show that the rundown of the evoked GABA-mediated responses was due to effective depletion of GABA levels from presynaptic terminals and not to the toxicity of the drugs.

Conductance properties of K⁺ channels underlying the GABA_B IPSCs

Experiments related to conductance measurements, current-voltage relationships, and nonstationary noise analysis were performed in control slices on pharmacologically isolated (10 µM picrotoxin; 10 µM CNQX; 40 µM D-AP5) monosynaptic GABA_B IPSCs.

As reported previously (Otis et al. 1993), current-voltage relationships were linear (n = 9) in the range of 45 to 110 mV when the extracellular K⁺ concentration was 2.5 mM. With an extracellular K⁺ concentration of 6.8 mM, a small outward rectification was observed in four of seven cells tested (no rectification in the remaining 3), with a slope conductance for the outward portion of 2.69 ± 0.52 nS vs. 1.52 ± 0.16 nS for linear current-voltage plots (not shown).

To obtain information on the underlying single-channel conductance and kinetics, we performed spectral and variance analysis on the decay phase of evoked GABA_B IPSCs. Ensemble averages (100-300 events) were subtracted from individual traces (Fig. 3A) and noise analysis was performed on the difference trace as previously described (De Koninck and Mody 1994).

View larger version (32K): [in this window] [in a new window]   FIG. 3. Conductance and kinetics of K+ channels linked to GABAB receptors. A: subtraction of the ensemble average from individual records allowed noise analysis to be performed. Top record: superimposition of 5 individual recordings over the ensemble average (taken from 95 traces). Bottom record: example on a larger amplitude scale of subtracted records to emphasize the noise. B: spectral analysis of the decay phase of the current revealed several corner frequencies corresponding to the indicated time constants. C: nonstationary variance analysis yielded a small unitary conductance () of 9.1 pS, consistent with a large number of channels (N) activated.

Spectral analysis on the decay phase of evoked current (n = 3) revealed four major corner frequencies corresponding to time constants of 140 ± 52 ms, 46.2 ± 12.4 ms, 6.95 ± 2.15 ms, and 0.85 ± 0.24 ms at 35°C (Fig. 3B). This analysis resolved more time constants than the simple exponential curve fitting of the decay phases of the currents. The shortest corner frequency corresponds to the shorter of the two decay time constants (110.2 ± 6.8 ms). The very long decay time constant described in our previous study (516.2 ± 52.5 ms; 16% fraction) is likely to fall outside the resolution of the spectral analysis given the length of the samples used (Fig. 3A). Similarly, this will tend to yield an overestimation of the first corner frequency. Nevertheless, the additional corner frequencies revealed by the spectral analysis likely indicate complex intraburst kinetics of the underlying K⁺ channels such as those observed for the G protein-coupled single K⁺ channels in locus coeruleus neurons (Grigg et al. 1996).

Nonstationary variance analysis (n = 5) yielded an average estimate for the single-channel conductance () of 8.6 ± 2.8 pS (range 5.2-12.6 pS) (Fig. 3C). Accordingly, a large number of these channels (177 ± 51) was estimated to contribute to the peak of the evoked GABA_B IPSCs (Fig. 3C). Comparable estimates were obtained in two cells at room temperature (22°C).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Endogenous GABA as a mediator of both GABA_A and GABA_B IPSCs

The results obtained after depletion of GABA from inhibitory terminals indicate that both GABA_A and GABA_B receptors are activated synaptically by endogenously released GABA. It establishes that GABA is a necessary transmitter in the response, yet it does not completely rule out the possibility that another cotransmitter (albeit not sufficient by itself) participates in GABA_B responses.

In the depletion experiments, the attenuation of the GABA_B-mediated IPSC paralleled, with a slower time course, the decline of the GABA_A-mediated response. This may reflect a greater affinity of the GABA_B receptor for GABA (Isaacson et al. 1993). Alternatively, it may be an indication that the GABA_B response involves another endogenous agent. As stressed above, this would indicate that although GABA is a necessary transmitter, it may not be sufficient for the activation of synaptic GABA_B receptors. The data are, however, insufficient to establish that this difference in time course is statistically significant.

Properties of K⁺ channels linked to synaptic GABA_B receptors

The properties and the relative number of postsynaptic K⁺ channels linked to GABA_B receptors activated during slow IPSCs were studied with the use of nonstationary fluctuation analysis. We have determined that these receptors are linked to small-conductance (5-12 pS) K⁺ channels with no or slight outward rectification. A large number of these channels is activated during an average-sized stimulus-evoked GABA_B IPSC.

The GABA_B receptors are coupled through pertussis toxin-sensitive G proteins to K⁺ channels (Thalmann 1988). The small-conductance channels reported here are unlike the more commonly reported G protein-linked K⁺ channels with larger, inwardly rectifying conductances (Brown and Birnbaumer 1990; Doupnik et al. 1995). Nevertheless, small-conductance (13 pS) G protein-activated K⁺ channels have been described in hippocampal neurons (VanDongen et al. 1988), and recently a small unit conductance has been resolved for apamin-insensitive Ca²⁺-dependent K⁺ channels with the use of nonstationary noise analysis (Sah 1995).

Two previous studies have addressed the unitary conductance of K⁺ channels linked to GABA_B receptors. Both used cell-attached recordings to measure single-channel activity in response to exogenous applications of baclofen (Wagner and Dekin 1993) and/or GABA (Premkumar et al. 1990) and reported the opening of large-conductance (70-100 pS) channels. Although the channels observed by Premkumar et al. (1990) had a maximum conductance of ~70 pS, they also displayed many lower conductance states that were integral multiples of 5-6 pS, and in several cells the single-channel currents became progressively larger by multiples of an elementary current of 5-6 pS. This small elementary current may correspond to the main conductance state activated during the synaptic events recorded in our study.

In previous studies researchers used nonstationary noise analysis to estimate the conductance linked to non-N-methyl-D-aspartate receptors and GABA_A receptors activated at synapses. Interestingly, in those cases it was found that the larger of the multiple conductance states revealed in single-channel recordings appeared to carry most of the synaptic current (De Koninck and Mody 1994; Traynelis et al. 1993). On the basis of our results and those of Premkumar et al. (1990), it would appear that a small-subconductance state predominates during synaptic GABA_B currents. The possibility has to be considered, however, that the unit conductance was underestimated by the nonstationary noise analysis. Synaptic currents and estimates from noise analysis are attenuated by filtering due to the electrotonic properties of the cell (De Koninck and Mody 1994; Traynelis et al. 1993). Calculations based on electrotonic properties of cerebellar granule cells revealed that the values of single-channel conductance derived from noise analysis of filtered excitatory postsynaptic potentials may have been attenuated by up to 10%. In our recent study of GABA_A receptor-mediated miniature IPSCs, our worst case scenario of filtering for most distant IPSCs (Soltesz et al. 1995) yielded a possible attenuation of up to 25% in the unit conductance measured with noise analysis. Applying this factor to the present case would mean that our average value of 8.6 pS (range 5.2-12.6) would correspond to 11.4 pS (range 6.9-16.8). This value remains in the range of that reported for small G protein-activated K⁺ conductances and is almost 1 order of magnitude smaller than for the more commonly reported G protein-linked K⁺ channels. The small unit conductance linked to synaptically activated GABA_B receptors measured in our study would either indicate that channels activated in these conditions are different from those stimulated by exogenous agonists or that it is a subconductance state that preferentially participates in synaptic currents. Distinct K⁺ conductances may also be associated with GABA_B receptors in the different cell types studied.

In summary, GABA_B receptors activated by endogenous transmitter are linked to small-conductance K⁺ channels with no or small outward rectification. These channels appear to have different properties than those stimulated by exogenous agonists, thus raising the possibility that the receptors or the K⁺ channels activated by synaptically released GABA are distinct from those activated experimentally by exogenous applications of agonists or that a small-subconductance state is the predominant charge carrier during synaptic GABA_B currents.

	ACKNOWLEDGEMENTS

  We thank Dr. T. S. Otis for participating in some of the experiments.

  This work was supported by National Institutes of Health Grants NS-30549 to I. Mody, NS-34022 To Y. De Koninck, and MT-12942 to Y. De Koninck from the Canadian Medical Research Council. Y. De Koninck is a Scholar of the Canadian MRC.

	FOOTNOTES

  Address for reprint requests: Y. De Koninck, Dept. of Pharmacology and Therapeutics, McGill University, 3655 Drummond St., Rm. 1242, Montreal, Quebec H3G 1Y6, Canada.

  Received 8 August 1996; accepted in final form 9 December 1996.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• BROWN, A. M., BIRNBAUMER, L. Ionic channels and their regulation by G protein subunits. Annu. Rev. Physiol. 52: 197-213, 1990.[Medline]
• CASEY, R. E., WOOD, J. D. Isonicotinic acid hydrazide-induced changes in the metabolism of gamma-aminobutyric acid in the brain of four species. Comp. Biochem. Physiol. B Comp. Biochem. 45: 741-748, 1973.[Medline]
• DAVIES, C. H., DAVIES, S. N., COLLINGRIDGE, G. L. Paired-pulse depression of monosynaptic GABA-mediated inhibitory postsynaptic responses in rat hippocampus. J. Physiol. Lond. 424: 513-531, 1990.[Abstract/Free Full Text]
• DE KONINCK, Y., MODY, I. Noise analysis of miniature IPSCs in adult rat brain slices; properties and modulation of synaptic GABA_A receptor channels. J. Neurophysiol. 71: 1318-1335, 1994.[Abstract/Free Full Text]
• DE KONINCK, Y., MODY, I. The postsynaptic conductance linked to GABA_B receptors activated by endogenously released transmitter. Soc. Neurosci. Abstr. 25: 1347, 1995.
• DESTEXHE, A., SEJNOWSKI, T. J. G protein activation kinetics and spillover of gamma-aminobutyric acid may account for differences between inhibitory responses in the hippocampus and thalamus. Proc. Natl. Acad. Sci. USA 92: 9515-9519, 1995.[Abstract/Free Full Text]
• DOUPNIK, C. A., DAVIDSON, N., LESTER, H. A. The inward rectifier potassium channel family. Curr. Opin. Neurobiol. 5: 268-277, 1995.[Medline]
• GEDDES, J. W., WOOD, J. D. Changes in the amino acid content of nerve endings (synaptosomes) induced by drugs that alter the metabolism of glutamate and gamma-aminobutyric acid. J. Neurochem. 42: 16-24, 1984.[Medline]
• GRIGG, J. J., KOZASA, T., NAKAJIMA, Y., NAKAJIMA, S. Single-channel properties of a G-protein-coupled inward rectifier potassium channel in brain neurons. J. Neurophysiol. 75: 318-328, 1996.[Abstract/Free Full Text]
• ISAACSON, J. S., SOLIS, J. M., NICOLL, R. A. Local and diffuse synaptic actions of GABA in the hippocampus. Neuron 10: 165-175, 1993.[Medline]
• MAYNERT, E. W., KAJI, H. K. On the relationship of brain gamma-aminobutyric acid to convulsion. J. Pharmacol. Exp. Ther. 137: 114-121, 1962.[Abstract/Free Full Text]
• MISGELD, U., BIJAK, M., JAROLIMEK, W. A physiological role for GABAB receptors and the effects of baclofen in the mammalian central nervous system. Prog. Neurobiol. 46: 423-462, 1995.[Medline]
• MODY, I., DE KONINCK, Y., OTIS, T. S., SOLTESZ, I. Bridging the cleft at GABA synapses in the brain. Trends Neurosci. 17: 517-525, 1994.[Medline]
• OTIS, T. S., DE KONINCK, Y., MODY, I. Characterization of synaptically elicited GABA_B responses using patch-clamp recordings in rat hippocampal slices. J. Physiol. Lond. 463: 391-407, 1993.[Abstract/Free Full Text]
• PREMKUMAR, L. S., CHUNG, S. H., GAGE, P. W. GABA-induced potassium channels in cultured neurons. Proc. R. Soc. Lond. B Biol. Sci. 241: 153-158, 1990.[Medline]
• SAH, P. Properties of channels mediating the apamin-insensitive afterhyperpolarization in vagal motoneurons. J. Neurophysiol. 74: 1772-1776, 1995.[Abstract/Free Full Text]
• SIGWORTH, F. J. Interpreting power spectra from nonstationary membrane current fluctuations. Biophys. J. 35: 289-300, 1981.[Abstract/Free Full Text]
• SOLTESZ, I., SMETTERS, D. K., MODY, I. Tonic inhibition originates from synapses close to the soma. Neuron 14: 1273-1283, 1995.[Medline]
• THALMANN, R. H. Evidence that guanosine triphosphate (GTP)-binding proteins control a synaptic response in brain: effect of pertussis toxin and GTPgammaS on the late inhibitory postsynaptic potential of hippocampal CA3 neurons. J. Neurosci. 8: 4589-4602, 1988.[Abstract]
• THOMPSON, S. M. Modulation of inhibitory synaptic transmission in the hippocampus. Prog. Neurobiol. 42: 575-609, 1994.[Medline]
• THOMPSON, S. M., GÄHWILER, B. H. Effects of the GABA uptake inhibitor tiagabine on inhibitory synaptic potentials in rat hippocampal slice cultures. J. Neurophysiol. 67: 1698-1701, 1992.[Abstract/Free Full Text]
• TRAYNELIS, S. F., SILVER, R. A., CULL-CANDY, S. G. Estimated conductance of glutamate receptor channels activated during EPSCs at the cerebellar mossy fiber-granule cell synapse. Neuron 11: 279-289, 1993.[Medline]
• VANDONGEN, A. M., CODINA, J., OLATE, J., MATTERA, R., JOHO, R., BIRNBAUMER, L., BROWN, A. M. Newly identified brain potassium channels gated by the guanine nucleotide binding protein Go. Science Wash. DC 242: 1433-1437, 1988.[Abstract/Free Full Text]
• WAGNER, P. G., DEKIN, M. S. GABAb receptors are coupled to a barium-insensitive outward rectifying potassium conductance in premotor respiratory neurons. J. Neurophysiol. 69: 286-289, 1993.[Abstract/Free Full Text]
• WOOD, J. D., ABRAHAMS, D. E. The comparative effects of various hydrazides on -aminobutyric acid and its metabolism. J. Neurochem. 18: 1017-1025, 1971.[Medline]
• WOOD, J. D., KURYLO, E. Amino acid content of nerve endings (synaptosomes) in different regions of brain: effects of gabaculine and isonicotinic acid hydrazide. J. Neurochem. 42: 420-425, 1984.[Medline]
• WOOD, J. D., KURYLO, E., LANE, R. Gamma-aminobutyric acid release from synaptosomes prepared from rats treated with isonicotinic acid hydrazide and gabaculine. J. Neurochem. 50: 1839-1843, 1988.[Medline]
• WOOD, J. D., PEESKER, S. J. The effect on GABA metabolism in brain of isonicotinic acid hydrazide and pyridoxine as a fuction of time after administration. J. Neurochem. 19: 1527-1537, 1972.[Medline]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society