The Journal of Neurophysiology Vol. 79 No. 3 March 1998, pp. 1321-1328
Copyright ©1998 by the American Physiological Society

Changes in Intracellular Calcium Concentration Affect Desensitization of GABA_A Receptors in Acutely Dissociated P2-P6 Rat Hippocampal Neurons

Jerzy W. Mozrzymas and Enrico Cherubini

Biophysics Sector and Istituto Nazionale Fisica della Materia Unit, International School for Advanced Studies (SISSA), 34014 Trieste, Italy

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Mozrzymas, Jerzy W. and Enrico Cherubini. Changes in intracellular calcium concentration affect desensitization of GABA_A receptors in acutely dissociated P2-P6 rat hippocampal neurons. J. Neurophysiol. 79: 1321-1328, 1998. The whole cell configuration of the patch-clamp technique was used to study the effects of different cytosolic calcium concentrations [Ca²⁺]_i on desensitization kinetics of -aminobutyric acid (GABA)-activated receptors in acutely dissociated rat hippocampal neurons. Two different intrapipette concentrations of the calcium chelator 1,2-bis(2-aminophenoxy)ethane-N,N,N,N-tetraacetic acid (BAPTA; 11 and 0.9 mM, respectively) were used to yield a low (1.2 × 10⁸ M) or a high (2.2 × 10⁶ M) [Ca²⁺]_i. In low [Ca²⁺]_i, peak values of GABA-evoked currents (20 µM) evoked at 30 mV, were significantly larger than those recorded in high calcium [2,970 ± 280 (SE) pA vs. 1,870 ± 150 pA]. The extent of desensitization, assessed from steady-state to peak ratio was significantly higher in high calcium conditions (0.14 ± 0.007 vs. 0.11 ± 0.008). Similar effects of [Ca²⁺]_i on desensitization were observed with GABA (100 µM). Recovery from desensitization, measured at 30 s interval with double pulse protocol was significantly slower in high [Ca²⁺]_i than in low [Ca²⁺]_i (54 ± 3% vs. 68 ± 2%). The current-voltage relationship of GABA-evoked currents was linear in the potential range between 50 and 50 mV. The kinetics of desensitization process including the rate of onset, extent of desensitization, and recovery were voltage independent. The run down of GABA-evoked currents was faster with the higher intracellular calcium concentration. The run down process was accompanied by changes in desensitization kinetics: in both high and low [Ca²⁺]_i desensitization rate was progressively increasing with time as the slow component of the desensitization onset was converted into the fast one. In excised patches, the desensitization kinetics was much faster and more profound than in the whole cell configuration, indicating the involvement of intracellular factors in regulation of this process. In conclusion, [Ca²⁺]_i affects the desensitization of GABA_A receptors possibly by activating calcium-dependent enzymes that regulate their phosphorylation state. This may lead to modifications in cell excitability because of changes in GABA-mediated synaptic currents.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

GABA is the major inhibitory neurotransmitter in mammalian CNS. It inhibits neuronal firing by activating three pharmacologically distinct classes of receptors designated as -aminobutyric acid-A (GABA_A), -aminobutyric acid-B (GABA_B), and -aminobutyric acid-C (GABA_C). Although both GABA_A and GABA_C receptors are integral ion channels, members of the same protein receptor superfamily (Bormann and Feigenspan 1995; Kaila 1994; Sivilotti and Nistri 1991), GABA_B receptors do not contain an integral ion channel but are coupled with cationic channels via guanosine 5-triphosphate (GTP)-binding proteins and intracellular messengers (Misgeld et al. 1995). Most of the fast synaptic inhibition is accomplished by GABA_A receptors (the role of GABA_C receptors in synaptic transmission is still unclear), whereas the "slow" synaptic inhibition results from the activation of GABA_B receptors. After an initial peak, GABA_A-mediated fast inhibitory postsynaptic currents (IPSC) decay with one or two time constants. In different synapses, the values of these time constants vary substantially ranging from a few to tens of milliseconds (Borst et al. 1994; Edwards et al. 1990; Puia et al. 1994; Weiss et al. 1988). Changes in IPSC time course may influence several physiological and pathological processes such as prevention of neuronal firing in response to excitatory inputs, short- and long-term plasticity processes, therapeutic effects of anesthetics and anticonvulsants, seizure activity, etc. Recently it was shown that desensitization, the process whereby receptors become inactive in the presence of the agonist, plays an important role in shaping synaptic currents (Jones and Westbrook 1995). Thus the processes underlying modulation of desensitization by various intracellular messengers are relevant for better understanding GABA_A receptor function. There is a large body of evidence indicating that GABA_A mediated responses can be regulated by kinases, phosphatases as well as intracellular calcium [Ca²⁺]_i (Stelzer 1992). In particular, it was demonstrated that activation of kinases such as the calcium calmodulin dependent protein kinase II (CaM-KII) or tyrosine kinase enhance the responses to GABA (Moss et al. 1995; Wang et al. 1995), whereas activation of phosphatase II B (calcineurin) induces a suppression of GABA-evoked currents (Chen and Wong 1995; Stelzer and Shi 1994). Moreover a rise in [Ca²⁺]_i through a ligand-gated channel such as the N-methyl-D-aspartate (NMDA) receptor induces a suppression of GABA-evoked currents, an effect that is abolished by calcineurin inhibitors. This suggests that the rise in the intracellular calcium activates calcineurin, that in turn suppresses GABA responses (Chen and Wong 1995; Stelzer and Shi 1994). Thus it appears that up and down regulation of GABA-evoked currents is regulated by calcium-dependent phosphorylation-dephosphorylation processes. Interestingly, the potentiating effects of CaM-KII on GABA responses is associated with a reduction of the extent of desensitization (Wang et al. 1995) and a similar effect was observed when blocking calcineurin with the cyclosporin A-cyclophilin A complex (Martina et al. 1996a). Because both CaM-KII and calcineurin are calcium dependent enzymes, the role of [Ca]_i on desensitization kinetics seems particularly interesting.

Therefore in this work we studied how different intracellular calcium concentrations, affect GABA_A receptor desensitization in acutely dissociated rat hippocampal neurons. It was found that changes in [Ca²⁺]_i significantly affect the extent of desensitization as well as the recovery process. Moreover, in long-lasting experiments, it was observed that run down of GABA-evoked currents was associated with modification in desensitization kinetics.

	METHODS

Abstract Introduction Methods Results Discussion References

Acutely dissociated cells were prepared according to the method described by Kay and Wong (1986). Hippocampal slices were obtained from postnatal (P) day P2-P6 Wistar rats. Slices were incubated for 40-60 min in oxygenated Krebs solution and then transferred for 10-12 min to a low chloride solution [82 Na₂SO₄, 5 MgCl₂, 30 K₂SO₄, 5 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), 1 NaHCO₃, and 10 glucose] containing 1 mg/ml pronase E (Sigma). After enzymatic treatment, the slices were washed for at least 30 min in the low chloride solution, mechanically dissociated by using Pasteur pipettes and plated into the recording chamber. GABA-evoked currents were recorded in the whole cell configuration of the patch-clamp technique by using the EPC-7 amplifier (List Medical, Darmstadt, Germany) from cells that on the basis of their shapes resembled pyramidal neurons. Series resistance (R_s) was in the range of 4-8 M and 50-70% of R_s compensation was accomplished. Two different intrapipette solutions containing low or high [Ca²⁺]_i (referred to later on as low and high calcium solutions) were used. The "low calcium" solution contained (in mM) 137 CsCl, 1 CaCl₂, 2 MgCl₂, 111 , 2 - bis(2 - aminophenoxy)ethane - N, N, N, N - tetraacetic   acid(BAPTA), 2 adenosine 5-triphosphate (ATP), and 10 HEPES (pH 7.2 with CsOH); the "high calcium" solution differed from the low calcium solution only in its content of BAPTA (0.9 mM instead of 11 mM). In this solution, the osmolarity was adjusted to 290 mOsm with sucrose. In some experiments ethylene glycol-bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid (EGTA) instead of BAPTA was used. Because GABA responses obtained with BAPTA or EGTA were similar in all respects, data from these experiments were pooled together. EQCAL (Biotools 1988) software was used to calculate the free calcium concentration present in our experimental conditions. These concentrations were 1.2 × 10⁸ and 2.2 × 10⁵ M for low and high calcium solutions, respectively. In our analysis we included only the experiments in which recordings were performed with both low and high Ca²⁺ intrapipette solutions on cells from the same preparation. The external solution contained (in mM) 137 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 20 glucose, and 10 HEPES (pH 7.4 with NaOH). GABA (20 or 100 µM) was applied by gravity with RSC-200 rapid perfusion system (Bio-Logic, Grenoble, France). The head of this system was modified to allow fast drug application on cells adhering to the bottom of Petri dishes. With this system, a complete exchange of the solution around the patched cell was achieved in <30 ms. The decaying phase of the normalized currents was fitted with a biexponential function in the form

(1)

where A_s is the steady-state current, A_fast is the fraction of the faster component and _fast and _slow are the fast and the slow time constants. The fraction of the slow component (A_slow) was calculated in Eq. 1 from the normalization condition: A_slow = 1  A_s  A_fast. To study the recovery process, a double-pulse protocol was used. The magnitude of recovery was calculated with the formula

(2)

where I₁ and I₂ are the peaks of currents evoked by the first and the second GABA pulse, respectively, and I_s1 and I_s2 are the currents at the end of drug application.

GABA was purchased from Sigma. Unless otherwise stated, data are expressed as means ± SE. Student's unpaired t-test was used for comparison of unpaired data.

	RESULTS

Abstract Introduction Methods Results Discussion References

Changes in [Ca²⁺]_i affect desensitization kinetics of GABA-evoked currents

As it will be discussed in detail later, the peak of GABA-evoked whole cell currents and the kinetics of desensitization undergo progressive changes during long-lasting experiments. Thus to eliminate the variability resulting from these long-term effects, we analyzed first only the responses to GABA obtained 4 min after breaking into the whole cell configuration. To investigate the effect of intracellular calcium concentration on desensitization kinetics, two different intrapipette solutions containing low and high [Ca²⁺]_i were used. Currents were evoked either by GABA (20 µM), the concentration close to EC₅₀ value for GABA_A receptor in the hippocampus (Celentano and Wong 1994; Schönrock and Bormann 1993), or by a subsaturating dose of GABA (100 µM). The time to peak (10-90%) of the responses evoked by GABA (20 µM) with the low or the high intracellular calcium solution ranged between 40 and 70 ms and for 100 µM GABA from 20 to 50 ms. Figure 1 shows a typical example of currents evoked by GABA (20 µM) with the low or the high intracellular calcium solution, recorded at 30 mV holding potential. GABA-evoked currents were characterized by a rapid rising phase followed by a biexponential decay tending to a steady-state value. By using the standard protocol of hyperpolarizing voltage pulses we have seen that the time course of the conductance during GABA applications paralleled that of GABA-induced currents, indicating that the current decay is a result of the ongoing desensitization of GABA receptors (not shown). It is clear from Fig. 1 that in comparison to high calcium solution, the current recorded in low calcium is characterized by a higher peak amplitude and lower extent of desensitization. As summarized in Table 1, peak values of GABA-evoked currents obtained with low-calcium solution were significantly (P < 0.05) larger (2,970 ± 250 pA, n = 11) than those recorded with high intracellular calcium solution (1,870 ± 150 pA, n = 10). It should be stressed, however, that with GABA (100 µM), the peak of the currents was often >6 nA. Because at most 70% of the series resistance could be compensated, the error in the clamping potential could affect the amplitude of the peak current. Therefore only those responses whose amplitude was <6 nA were included in the present study. Because of the elimination of larger amplitude responses, peak currents evoked by GABA (100 µM) obtained in low or high calcium solutions were not considered. The extent of desensitization induced by GABA (20 µM), assessed from the plateau to peak ratio was significantly higher with the high than with the low calcium solutions (see Table 1). In these conditions, no significant (P > 0.19) differences in the fast and slow time constants and relative areas were observed (Table 1). Similarly, when applying 100 µM GABA, the extent of desensitization was significantly higher with the high calcium solution (see Table 1). In contrast to the results obtained with GABA (20 µM), when a subsaturating concentration of GABA (100 µM) was applied, the percentage of the slow component (A_slow) was significantly (P < 0.05) smaller in the presence of a high calcium solution (Fig. 1B and Table 1).

View larger version (10K): [in this window] [in a new window]   FIG. 1. Changes in intracellular calcium concentration affects peak amplitude and desensitization kinetics of whole cell -aminobutyric acid (GABA)-evoked currents. Typical examples of currents evoked by GABA (20 µM, bars) from a holding potential 30 mV, with an intrapipette solution containing either a low (A) or a high (B) intracellular calcium solution. Decay of currents was fitted with a biexponential function (Eq. 1, continuous lines superimposed on current traces). Curve fitting parameters in A are peak current 2,950 pA, fast 0.91 s, slow 4.8 s, As 0.16, Afast 0.23, and Aslow 0.61. Parameters in B are peak current 2,040 pA, fast 0.72 s, slow 3.7 s, As 0.08, Afast 0.39, and Aslow 0.53. C: traces presented in A and B were normalized and superimposed to better illustrate differences in desensitization onset.

To study the process of recovery from desensitization, a standard double pulse protocol was used. To avoid variability resulting from modifications occurring in long-lasting experiments, statistical analysis was performed only for the first pair of GABA (20 µM) stimuli. The recovery process, assessed at 30 s interval was significantly (P < 0.05) slower with the high calcium solution (0.68 ± 0.028, n = 11 and 0.54 ± 0.029, n = 10, in low and high calcium, respectively, Fig. 2).

View larger version (12K): [in this window] [in a new window]   FIG. 2. Changes in intracellular calcium concentration affects recovery from desensitization of GABA-evoked currents. A: pairs of GABA pulses (20 µM, bars) applied at 30 s intervals at a holding potential of 30 mV, with an intrapipette solution containing either a low (top) or a high (bottom) intracellular calcium solution. Note that recovery from desensitization is faster when recording with low calcium solution. B: each column represents averaged extent of recovery (calculated with Eq. 2) with low (LC) or high (HC) calcium solution (0.68 ± 0.028, n = 11 in low vs. 0.54 ± 0.029, n = 10 in high calcium). Bars are SE. *, P < 0.05.

Currents evoked by GABA (20 µM) at 30 mV in nearly symmetrical chloride solutions were inward. Their amplitude decreased with membrane depolarization and became outward at 2.9 ± 1.2 mV (n = 4) and 4.1 ± 2.2 mV (n = 3) in low and high calcium solutions, respectively (Fig. 3C). Moreover, the current-voltage relationship was linear in the potential range between 50 and 50 mV (Fig. 3C). The kinetics of the desensitization process including the rate of onset and extent of desensitization were voltage independent in the same potential range (Fig. 3). At 30 mV, fast and slow time constants were _fast = 1.05 ± 0.2 s, _slow = 4.6 ± 0.6 s in low calcium and _fast = 1.1 ± 0.3 s, _slow = 4.5 ± 0.8 s in high calcium solutions. The respective areas were A_s = 0.12 ± 0.04, A_fast = 0.36 ± 0.12, A_slow = 0.52 ± 0.09 and A_s = 0.09 ± 0.03, A_fast = 0.38 ± 0.09, A_slow = 0.53 ± 0.07 in low and high calcium solutions, respectively. These values are not significantly different from those observed at 30 mV (see Table 1).

View larger version (14K): [in this window] [in a new window]   FIG. 3. Membrane voltage does not affect desensitization kinetics. Currents evoked by GABA (20 µM, bars) at holding potentials of 30 and 30 mV, with an intrapipette solution containing either a low (A) or a high (B) intracellular calcium solution. Decay of currents was fitted with a biexponential function having the following kinetic parameters: in A at 30 mV, fast 1.15 s, slow 4.2 s, As 0.1, Afast 0.35, Aslow 0.55; in A at 30 mV, fast 1.21 s, slow 5.4 s, As 0.08, Afast 0.43; Aslow 0.49; In B at 30 mV, fast 1.1 s, slow 5.2 s, As 0.07, Afast 0.39, Aslow 0.54; in B at 30 mV, fast 1.02 s, slow 4.9 s, As 0.07, Afast 0.41, Aslow 0.52. C: current-voltage relationship for samples shown in A () and in B ().

To check whether or not the external calcium concentration [Ca²⁺]_o could affect the desensitization kinetics, in three experiments [Ca²⁺]_o was varied within the range of 1-5 mM. Changes in [Ca²⁺]_o did not modify the desensitization kinetics of GABA-evoked currents recorded with the low or the high [Ca²⁺]_i at different holding voltages (from 50 to 50 mV). Moreover, a train of depolarizing stimuli (2 Hz, from 60 to 0 mV, 50 ms pulse duration), applied for 5-10 s before GABA application (at [Ca²⁺]_o = 2 mM), did not affect the GABA-evoked currents (n = 2, data not shown).

Modifications of the desensitization kinetics in long-term experiments

With both low and high calcium solutions, a progressive decrease in amplitude of GABA-evoked currents was observed. This phenomenon, known as run down, was faster when the intracellular pipette contained the high calcium solution (see Fig. 4). Thus after 30 min, the reduction in the peak amplitude of GABA responses was 49 ± 8% and 80 ± 10% in low and high calcium solution, respectively. These findings confirm previous observations indicating that run down is a calcium-dependent process (see Stelzer 1992). In our experiments, run down was associated with changes in desensitization kinetics. As shown in Fig. 5, A and B, with low [Ca²⁺]_i the plateau to peak ratio of responses evoked by GABA (20 µM) slowly declined to a steady state value within 30 min, whereas the plateau to peak ratio of GABA currents evoked at high [Ca²⁺]_i did not show any significant change with time. Both in low and in high [Ca²⁺]_i, no changes in the fast and in the slow time constants were observed throughout the experiment. Fast time constant values were 0.91 ± 0.082 and 0.87 ± 0.074 s and slow time constants were 3.84 ± 0.38 and 4.19 ± 0.56 s, in low and high [Ca²⁺]_i, respectively. These values were similar to those obtained 4 min after breaking into the whole cell configuration (see Table 1). However the percentages of the areas underlying the fast and slow kinetic components showed marked modifications with time: the percentage of the slow component (A_slow) decreased and the fast one (A_fast) increased (Fig. 5C). Although not significant (P > 0.2), there was a trend for the A_slow value to decrease with time to a larger extent in high [Ca²⁺]_i, than in the low calcium solution (44 ± 4% vs. 34 ± 6%, after 30 min, Fig. 5D). Altogether, during the recording period, the desensitization became faster because of the conversion of the slow component into the fast one. In contrast to what we observed for currents elicited by 20 µM GABA, the fast and slow time constants characterizing the desensitization onset of the currents evoked by 100 µM GABA, showed a progressive decrease. With low [Ca²⁺]_i, after 30 min, the value of _fast decreased by 41 ± 15%, whereas the value of _slow decreased by 37 ± 19% (n = 4). With high calcium these values were reduced by 40 ± 18% and 32 ± 14% (n = 3). As already mentioned, with 100 µM GABA concentration, the initial value of A_slow was larger in low than in high [Ca²⁺]_i (see Table 1). When using the low [Ca²⁺]_i, the A_slow component showed a progressive decrease with time, reaching after 20-30 min ~40% of the initial value (data not shown). With the high [Ca²⁺]_i, the A_slow parameter was as low as 0.16 ± 0.025 and did not show large variations during the recording period. Thus in the case of high GABA concentration, the desensitization rate was increased because of the shortening of the time constants and to the conversion of the slow desensitizing component into the fast one.

View larger version (11K): [in this window] [in a new window]   FIG. 4. Changes in intracellular calcium concentration affects run down of GABA-evoked currents. Each point in graph represents mean (from 5 to 9 cells) peak current amplitude evoked every 5 min by GABA (20 µM, normalized to amplitude of 1st GABA response) from a holding potential of 30 mV, with an intrapipette solution containing either a low () or a high () intracellular calcium solution. Bars are SE. Note faster run down with a high intracellular calcium solution.

View larger version (14K): [in this window] [in a new window]   FIG. 5. Changes in desensitization kinetics of GABAA receptors in long-lasting experiments. A: currents evoked by GABA (20 µM, bars) from a single neuron at a holding potential of 30 mV, 4 min (top) and 30 min (bottom) after breaking into whole cell configuration. Pipette was filled with low calcium solution. Decay of currents was fitted with biexponential functions (Eq. 1). Top: kinetic parameters were fast 0.8 s, slow 3.8 s, As 0.12, Afast 0.24, and Aslow 0.64. Bottom: fast 0.91 s, slow 3.5 s, As 0.07, Afast 0.53, and Aslow 0.4. B: same current traces as in A now normalized and superimposed to better visualize differences in desensitization kinetics. C: mean steady-state to peak ratios of GABA-evoked currents (n = 7), recorded using low () and high () intrapipette calcium solution, as a function of time. Bars are SE. D: time course of slow (Aslow, ) and fast (Afast, ) components of desensitization onset of currents evoked by GABA (20 µM) in a single neuron. Note conversion of slow into fast component with time. This recording was performed with low calcium intrapipette solution. E: time course of mean value of slow component (Aslow) of desensitization of GABA-evoked currents (20 µM) measured with low () and high () intrapipette calcium solution. Each point represents average of 7 experiments. Bars are SE. All recordings were made at a holding potential of 30 mV.

Also the recovery from desensitization of currents evoked by GABA (20 µM) showed some changes during long-term recordings. After 30 min, the recovery (measured at 30 s interval) decreased from 0.68 ± 0.028 (n = 11) and 0.54 ± 0.029 (n = 10) to 0.62 ± 0.019 (n = 7) and 0.525 ± 0.011 (n = 7) in low and high calcium solutions, respectively.

To test whether or not the observed long-term changes were the result of washout of some intracellular factors, in a few (n = 4) experiments GABA-evoked currents (100 µM) were recorded from the excised outside-out patches. Surprisingly, in comparison with the whole cell recordings, desensitization was extremely fast showing, in two of four cells, a monoexponential kinetics (time constants 0.72 and 0.69 s) and, in two cells, a biexponential time course was observed (A_fast = 0.36, _fast = 0.12 s; A_slow = 0.61, _slow = 0.84 s and A_fast = 0.48, _fast = 0.25 s; A_slow = 0.51, _slow = 1.23 s for the 2 cells). The steady state to peak ratio was close to zero (0.017 ± 0.01, n = 4), indicating almost complete desensitization.

Long-term modifications of GABA-evoked responses, such as those observed in our experiments, could be related to use-dependent mechanisms. To explore this possibility in three experiments (both with low and high [Ca²⁺]_i) GABA responses were evoked 4 and 30 min after breaking into the whole cell configuration. In these conditions, run down as well as the increase in the rate and the extent of desensitization were indistinguishable from those seen when GABA pulses were applied every 5 min (data not shown). This finding indicates that long-term effects are time rather than use dependent.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The present data clearly show that changes in [Ca²⁺]_i affect desensitization kinetics of GABA_A receptor-mediated responses. Thus with high [Ca²⁺]_i the extent of desensitization is higher and the recovery from desensitization slower than with low [Ca²⁺]_i. Moreover modifications in [Ca²⁺]_i alter the run down and the desensitization kinetics of GABA-induced currents in long-lasting experiments.

Changes in [Ca²⁺]_i affect desensitization kinetics of GABA-evoked currents

Several reports have demonstrated a crucial role of [Ca²⁺]_i in the up or down regulation of GABA_A receptor function. Hence increase in [Ca²⁺]_i suppresses GABA-evoked currents in dorsal root ganglion cells (Inoue et al. 1986), in acutely isolated hippocampal neurons (Chen and Wong 1995; Stelzer and Shi 1994), and in cultured porcine melanotrophs (Mouginot et al. 1991), whereas it potentiates GABA responses in cerebellar Purkinje cells (Llano et al. 1991). Increase in cytosolic calcium may be achieved through activation of voltage-dependent calcium channels (Inoue et al. 1986; Llano et al. 1991), NMDA receptor channels (Chen and Wong 1995; Stelzer and Shi 1994), or intracellular calcium stores (Mouginot et al. 1991). In the majority of these reports the issue of desensitization kinetics was not analyzed in detail. An examination of desensitization kinetics as function of membrane potential and [Ca²⁺]_i was performed by Frosch et al. (1992) in cultured cortical neurons. In contrast to our results, however, these authors did not observe any effect of [Ca²⁺]_i on desensitization kinetics of GABA receptors. This discrepancy may reflect differences in GABA receptor properties in cultured cortical neurons and in acutely dissociated hippocampal cells. Further evidence for different GABA receptor properties is given by the observation that in cortical neurons GABA receptor desensitization showed monoexponential decay; whereas, in our studies two kinetic components were clearly present. Moreover, as it will be discussed later, in contrast to what Frosch and co-workers (1992) observed, in our experiments desensitization was voltage independent.

In the present study, the effect of calcium on the desensitization process was achieved by dialyzing the cell with solutions containing different free calcium concentrations. In these conditions the peak amplitude of GABA-evoked currents measured with low calcium intracellular solution was significantly larger than that obtained with high calcium solution. This finding is consistent with previous reports that GABA-evoked responses are suppressed by increase in [Ca²⁺]_i, because of a decrease in receptor affinity (Inoue 1986; Martina et al. 1994). This mechanism cannot explain however the increased extent of desensitization with high [Ca²⁺]_i found in the present study. Because at increasing agonist concentrations, desensitization becomes faster and more profound (Akaike et al. 1986; Celentano and Wong 1994; Frosch et al. 1992; Oh and Dichter 1992) one would expect that a decrease in receptor affinity would give rise to a decrease rather than an increase in the extent of desensitization. Therefore it appears that intracellular calcium exerts its modulatory action on desensitization process by mechanisms that are not directly related to the affinity of the agonist binding sites.

It should be stressed that when, as in the present experiments, large and long-lasting GABA responses are evoked, the current decay could be affected by a shift in the chloride reversal potential (E_Cl) resulting from a redistribution of chloride ions. However, the decay of the membrane conductance paralleled the decrease of the current. In a similar preparation Huguenard and Alger (1986) have demonstrated that an intracellular chloride concentration of 68 mM (much smaller than in our experiments) was sufficient to "buffer" the effects of the chloride currents on the E_Cl. Thus the decay of GABA-evoked currents, recorded in our experimental conditions represents decrease in conductance resulting from GABA receptor desensitization. Interestingly, when applying 20 or 100 µM GABA, differences in the effect of [Ca²⁺]_i on desensitization kinetics were observed. Hence while at 20 µM GABA only the plateau to peak ratio was affected, at 100 µM GABA also the desensitization onset was accelerated as indicated by the increase in the area of the fast component. This suggests that intracellular calcium affects the dependence of the desensitization process on GABA concentration.

In the present study, desensitization process was found to be voltage independent and this property was not affected neither by changes in [Ca²⁺]_i nor by the dose of GABA. This finding is in contrast with other studies on cultured neurons in which the voltage dependence of desensitization was clearly observed (e.g., Frosch et al. 1992; Oh and Dichter 1992; Yoon 1994). It is possible that this discrepancy reflects differences in GABA receptors in different preparations and/or different experimental conditions. Interestingly, Frosch et al. (1992) have observed that the voltage dependence of desensitization was lost on patch excision, indicating that this process is critically regulated by soluble intracellular messengers. It is worth noting, however, that in rat retinal ganglion cells, the membrane potential does not influence the rate of desensitization (Tauck et al. 1988). In a recent study on recombinant GABA_A receptors, Dominguez-Perrot et al. (1996) have demonstrated that, in receptors lacking the ₂ subunit, desensitization is voltage independent. We do not know the subunits composition of native receptors in acutely dissociated hippocampal neurons. However in the same preparation we have found that GABA_A receptors were sensitive to zinc (Martina et al. 1996b) indicating the lack of the ₂ subunit (Draguhun et al. 1990; Smart et al. 1994 for review).

Changes in [Ca²⁺]_o (ranging from 1 to 5 mM) in the presence of low or high [Ca²⁺]_i, different holding voltages (from 50 to 50 mV) as well as the application of conditioning depolarizing pulses before GABA application, did not change the desensitization kinetics. These observations indicate that an increase in the chemical gradient for calcium and calcium entry through voltage-dependent calcium channels are unable to sufficiently modify [Ca²⁺]_i to affect GABA receptors (see also Stelzer and Shi 1994).

Modifications of desensitization kinetics in long-term experiments

In agreement with other studies (Stelzer 1992; Stelzer et al. 1988; Stelzer and Shi 1994) our data clearly demonstrate that cytosolic calcium plays an important role in long-term suppression of GABA_A-mediated conductance. In addition, we have observed that, desensitization kinetics undergoes modifications in time becoming faster and more profound and that this process is affected by calcium. Our finding that the plateau to peak ratio, determined with low [Ca]_i, reaches that seen in high [Ca]_i only after ~20 min (see Fig. 5B) suggests that intracellular calcium is involved in mechanisms controlling the extent of desensitization. However a gradual decrease in steady-state to peak ratio at low [Ca]_i indicates that also other factors (progressively diluted because of cell dialysis) may be involved in this process. An increase in desensitization rate of GABA receptors was already reported for other preparations (Frosch et al. 1992; Gyenes et al. 1994).

Interestingly, the acceleration of the desensitization onset observed in our long-term experiments took place via the conversion of the slow into the fast component. This finding suggests that these two kinetic components are not related to two different types of receptors but rather they refer to a single receptor population with multiple rates of desensitization. A similar conclusion was proposed by Celentano and Wong (1994) who observed an interconversion between fast and slow phases of GABA receptor desensitisation in guinea pig hippocampal neurons. Our data indicate that run down and the modification of desensitization kinetics are not use- but time-dependent processes suggesting a progressive washout of some intracellular factors. This possibility is further supported by the observation that a large increase in the rate and the extent of desensitization was seen on patch excision. A similar acceleration of the desensitization onset after detaching the patch from the cell was reported by Frosch et al. (1992). A time-dependent run down was also observed in acutely dissociated guinea pig hippocampal neurons (Stelzer et al. 1988).

The observation that in our experiments run down and modification of desensitization kinetics were concomitant and that both of them were calcium dependent, might raise the possibility that the decrease in amplitude of GABA-induced current is caused by the increased desensitization onset. This hypothesis, however, is unlikely because run down, as pointed out above, is a time-dependent process that can proceed in the absence of GABA, indicating that the underlying mechanisms are not directly related to desensitization. Secondly, the values of the time constants (not affected by high [Ca²⁺]_i) describing desensitization onset are much slower than the risetime of GABA-evoked currents, ruling out any considerable impact of desensitization on the peak current. Moreover the possibility that an ultrafast desensitization component (beyond the resolution of our system) would affect the peak current is unlikely because the fastest desensitization time constants reported so far, with concentrations of GABA similar to those used in our experiments, are tens of milliseconds (see, e.g., Celentano and Wong 1994). Such component should be detectable by our system (see METHODS). Altogether, run down and desensitization appear to be two independent phenomena. A similar conclusion was proposed by Stelzer et al. (1988). More recently, Harata et al. (1997) have demonstrated that desensitization can be enhanced without increasing run down of GABA-evoked currents, further indicating the lack of interdependence between these processes.

Altogether, our data obtained in long-term experiments show that the responsiveness to exogenously applied GABA is diminishing in time not only because of the reduction of current amplitude (run down) but also because of the increase in the rate and the extent of desensitization and that these processes are affected by calcium.

In conclusion, it appears that [Ca²⁺]_i plays a crucial role not only in regulating the peak amplitude of GABA-evoked currents but also in the modulation of receptor desensitization at different times during the recording period. Calcium may activate calcium-dependent enzymes that would affect the phosphorylation state of GABA_A receptors (Stelzer et al. 1988; Stelzer 1992). In particular activation of calcium calmodulin-dependent phosphatase 2B would affect these processes and a recent study from this laboratory has shown that inhibition of calcineurin by the cyclosporin-cyclophilin complex indeed results in reduction of GABA receptor desensitization (Martina et al. 1996a). Thus a short or long-lasting impairment of GABA receptor function would lead to an increased cell excitability and this may be relevant for long-term changes in synaptic efficacy such those occurring during long-term potentiation or epilepsy.

	ACKNOWLEDGEMENTS

  This work was partially supported by a grant from Consiglio Nazionale delle Ricerche (CNR 96.03039) to E. Cherubini. J. W. Mozrzymas is on leave from Department of Biophysics, Academy of Medicine, Wroclaw, Poland.

	FOOTNOTES

  Address for reprint requests: J. W. Mozrzymas, Biophysics Sector, International School for Advanced Studies, Via Beirut 2-4, 34014 Trieste, Italy.

  Received 2 June 1997; accepted in final form 29 October 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• AKAIKE, N., INOUE, M., KRISHTAL, O. A. "Concentration-clamp" study of -aminobutyric-acid-induced chloride current kinetics in frog sensory neurons. J. Physiol. (Lond.) 379: 171-185, 1986.[Abstract/Free Full Text]
• BORMANN, J., FEIGENSPAN, A. GABA_C receptors. Trends Neurosci. 18: 515-519, 1995.[Medline]
• BORST, G. G., LODDER, J. C., KITS, K. S. Large amplitude variability of GABAergic IPSC in melanotrophs from Xenopus laevis: evidence that quantal size differs between synapses. J. Neurophysiol. 71: 639-655, 1994.[Abstract/Free Full Text]
• CELENTANO, J. J., WONG, R.K.S. Multiphasic desensitization of the GABA_A receptor in outside-out patches. Biophys. J. 66: 1039-1050, 1994.[Abstract/Free Full Text]
• CHEN, Q. X., WONG, R.K.S. Suppression of GABA_A receptor responses by NMDA application in hippocampal neurones acutely isolated from the adult guinea-pig. J. Physiol. (Lond.) 482: 353-362, 1995.[Medline]
• DOMINGUEZ-PEROT, C., FELTZ, P., POULTER, M. O. Recombinant GABA_A receptor desensitization: the role of 2 subunit and its physiological significance. J. Physiol. (Lond.) 497: 145-159, 1996.[Medline]
• DRAGUHUN, A., VERDOORN, T. A., EWERT, M., SEEBURG, P. H., SAKMANN, B. Functional and molecular distinction between recombinant rat GABA_A receptor subtypes by Zn²⁺. Neuron 5: 781-788, 1990.[Medline]
• EDWARDS, F. A., KONNERTH, A., SAKMANN, B. Quantal analysis of inhibitory synaptic transmission in the dentate gyrus of rat hippocampal slices: a patch-clamp study. J. Physiol. (Lond.) 430: 213-249, 1990.[Abstract/Free Full Text]
• FROSCH, M. P., LIPTON, S. A., DICHTER, M. A. Desensitization of GABA-activated currents and channels in cultured cortical neurons. J. Neurosci. 12: 3042-3053, 1992.[Abstract]
• GYENES, M., WANG, Q., GIBBS, T. T., FARB, D. H. Phosphorylation factors control neurotransmitter and neuromodulator actions at the -aminobutyric acid type A receptor. Mol. Pharmacol. 46: 542-549, 1994.[Abstract]
• HARATA, N., WU, J., ISHIBASHI, H., ONO, K., AKAIKE, N. Run-down of the GABA_A response under experimental ischaemia in acutely dissociated CA1 pyramidal neurones of the rat. J. Physiol. (Lond.) 500: 673-688, 1997.[Medline]
• HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, E., SIGWORTH, F. J. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Arch. 391: 85-100, 1981.[Medline]
• HUGUENARD, J. R., ALGER, B. E. Whole cell voltage-clamp study of the fading of GABA-activated currents in acutely dissociated hippocampal neurons. J. Neurophysiol. 56: 1-18, 1986.[Abstract/Free Full Text]
• INOUE, M., OOMURA, Y., YAKUSHIJI, T., AKAIKE, N. Intracellular calcium ions decrease the affinity of the GABA receptor. Nature 324: 156-158, 1986.[Medline]
• JONES, M. V., WESTBROOK, G. L. Desensitized states prolong GABA_A channel responses to brief agonist pulses. Neuron 15: 181-191, 1995.[Medline]
• KAILA, K. Ionic basis of GABA_A receptor channel function in the nervous system. Prog. Neurobiol. 42: 489-537, 1994.[Medline]
• KAY, A. R., WONG, R.K.S. Isolation of neurons suitable for patch-clamping from adult mammalian central nervous system. J. Neurosci. Methods 16: 227-238, 1986.[Medline]
• LLANO, I., LERESCHE, N., MARTY, A. Calcium entry increases the sensitivity of cerebellar Purkinje cells to applied GABA and decreases inhibitory synaptic currents. Neuron 6: 565-574, 1991.[Medline]
• MARTINA, M., KILIC, G., CHERUBINI, E. The effects of intracellular Ca²⁺ on GABA-activated currents in cerebellar granule cells in culture. J. Membr. Biol. 142: 209-216, 1994.[Medline]
• MARTINA, M., MOZRZYMAS, J. W., BODDEKE, H.W.G.M., CHERUBINI, E. The calcineurin inhibitor cyclosporin A - cyclophilin A complex reduces desensitization of GABA_A-mediated responses in acutely dissociated rat hippocampal neurons. Neurosci. Lett. 215: 95-98, 1996a.[Medline]
• MARTINA, M., MOZRZYMAS, J. W., STRATA, F., CHERUBINI, E. Zinc modulation of bicuculline-sensitive and insensitive GABA receptors in the developing rat hippocampus. Eur. J. Neurosci. 8: 2168-2176, 1996b.[Medline]
• MISGELD, U., BIJAK, M., JAROLIMEK, W. A physiological role for GABA_B receptors and the effects of baclofen in the mammalian nervous system. Prog. Neurobiol. 46: 423-462, 1995.[Medline]
• MOSS, S. J., GORRIE, G. H., AMATO, A., SMART, T. G. Modulation of GABA_A receptors by tyrosine phosphorylation. Nature 377: 344-348, 1995.[Medline]
• MOUGINOT, D., FELTZ, P., SCHLICHTER, R. Modulation of GABA-gated chloride currents by intracellular Ca²⁺ in cultured porcine melanotrophs. J. Physiol. (Lond.) 437: 109-132, 1991.[Abstract/Free Full Text]
• OH, D. J., DICHTER, M. A. Desensitization of GABA-induced currents in cultured rat hippocampal neurons. Neuroscience 49: 571-576, 1992.[Medline]
• PUIA, G., COSTA, E., VICINI, S. Functional diversity of GABA-activated Cl^- currents in Purkinje versus granule neurons in rat cerebellar slices. Neuron 12: 117-126, 1994.[Medline]
• SCHÖNROCK, B., BORMANN, J. Functional heterogeneity of hippocampal GABA_A receptors. Eur. J. Neurosci. 5: 1042-1049, 1993.[Medline]
• SIVILOTTI, L., NISTRI, A. GABA receptor mechanism in the central nervous system. Prog. Neurobiol. 36: 35-92, 1991.[Medline]
• SMART, T. G., XIE, X., KRISHEK, B. J. Modulation of inhibitory and excitatory amino acid receptor ion channels by zinc. Prog. Neurobiol. 42: 393-441, 1994.[Medline]
• STELZER, A. Intracellular regulation of GABA_A-receptor function. In: Ion Channels, edited by and T. Narahashi . New York: Plenum, 1992, vol. 3, p. 83-136
• STELZER, A., KAY, A. R., WONG, R.K.S. GABA_A-receptor function in hippocampal cells is maintained by phosphorylation factors. Science 241: 339-341, 1988.[Abstract/Free Full Text]
• STELZER, A., SHI, H. Impairment of GABA_A receptor function by N-methyl-D-aspartate-mediated calcium influx in isolated CA1 pyramidal cells. Neuroscience 62: 813-828, 1994.[Medline]
• TAUCK, D. L., FROSCH, M. P., LIPTON, S. A. Characterization of GABA- and glycine-induced currents of solitary rodent retina ganglion cells in culture. Neuroscience 27: 193-203, 1988.[Medline]
• WANG, R. A., CHENG, G., KOLAJ, M., RANDIC, M. -Subunit of calcium/calmodulin-dependent protein kinase II enhances -aminobutyric acid in inhibitory synaptic responses of rat neurons in vitro. J. Neurophysiol. 73: 2099-2106, 1995.[Abstract/Free Full Text]
• WEISS, D. S., BARNES, E. M., HABLITZ, J. J. Whole cell and single-channel recordings of GABA-gated currents in cultured chick cerebral neurons. J. Neurophysiol. 59: 495-513, 1988.[Abstract/Free Full Text]
• YOON, K. W. Voltage-dependent modulation of GABA_A receptor channel desensitization in rat hippocampal neurons. J. Neurophysiol. 71: 2151-2160, 1994.[Abstract/Free Full Text]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society