Transporters buffer synaptically released glutamate on a submillisecond time scale

Abstract

The role of transporters in clearing free glutamate from the synaptic cleft was studied in rat CA1 hippocampal neurons cultured on glial microislands. The time course of free glutamate in the cleft during a synaptic event was estimated by measuring the extent to which the rapidly dissociating AMPA receptor antagonist kynurenate (KYN) was replaced by glutamate during a synaptic response. Dose inhibition of the AMPA receptor EPSC by KYN was less than predicted by the equilibrium affinity of the antagonist, and the rise time of AMPA receptor miniature EPSCs (mEPSCs) was slowed by KYN. Both results indicated that KYN dissociated from AMPA receptors and was replaced by synaptically released transmitter. When transporters were blocked by D,L-threo-beta-hydroxyaspartic acid (THA) or Li+, the mEPSC rise time in the presence of KYN was slowed further, indicating that transporters affect the glutamate concentration in the first few hundred microseconds of the synaptic response. The glutamate transient necessary to cause these effects was determined by developing a detailed kinetic model of the AMPA receptor. The model replicated the effects of KYN on the amplitude and rise time of the synaptic responses when driven by glutamate transients that were similar to previous estimates (; ). The effects of THA were replicated by slowing and enlarging the slower phase of the dual component transient by about 20% or by prolonging the single component by almost 40%. Because transport is too slow to account for these effects, it is concluded that transporters buffer glutamate in the synaptic cleft.

Fig. 4.

KYN, but not NBQX, is replaced by glutamate during a synaptic response. Solid lines indicate the expected block of the EPSC without antagonist unbinding, according to the assumptions described in the text. Circles indicate normalized EPSC charge transfer recorded in different concentrations of NBQX (filled circles; n = 11) or KYN (open circles; n = 10; only four different KYN concentrations were tested in any one cell). Synaptic currents were integrated to determine the charge transfer and normalized to the response in the absence of antagonist. Similar results were obtained by measuring EPSC amplitude directly. EPSCs were recorded under conditions of low release probability (1 mmCa/1 mm Mg or 3 mm Ca/1 mm Mg/5 μm Cd; no differences, other than peak amplitude, were observed between these two conditions) to reduce the possibility of multivesicular release (Tong and Jahr, 1994a). Left inset, EPSCs recorded in the presence of 0, 30, 100, 300, and 1000 nm NBQX. Right inset, EPSCs recorded from a different cell in the presence of 0, 100, 200, 1000, and 3000 μm KYN.

Fig. 8.

A kinetic model to mimic AMPA receptor behavior.A, Autaptically evoked AMPA receptor EPSC from a cultured CA1 hippocampal neuron (see Materials and Methods). EPSC amplitudes in this cell were unusually small. B, Markov model used to reproduce AMPA receptor kinetics observed in patch experiments. Two binding sites were configured to be equal and independent. Rates were as follows [units are μm⁻¹ msec⁻¹ (fork_a and k_b) or msec⁻¹]: k_a, 0.0133; k_−a, 6.24;k_b, 0.03325;k_−b, 5.985;k₋₁, 0.020;k₂, 0.65;k₋₂, 0.018;k₃, 1;k₋₃, 3; α, 1.1; and β, 5.7.k₁ was set (0.361) to satisfy microscopic reversibility. C, l-Glu (10 mm)-evoked current in an outside-out patch excised from the same neuron as in A. The top trace shows the junction potential change caused by the solution exchange across the open tip of the electrode after patch breakdown (see Materials and Methods). In the patch response, artifacts from the voltage pulse to the piezo have been blanked. D, Simulated response to a brief 10 mm pulse of glutamate, similar to the experiment shown in C. E, l-Glu-evoked responses in an outside-out patch. Five traces are superimposed, each with a different interval (20, 50, 100, 150, and 200 msec) between the end of an initial long (70 msec) application of 10 mml-glu and the beginning of a subsequent brief (4 msec) application.F, Simulated responses to a long (50 msec) pulse ofl-glu, followed at varying intervals by brief (4 msec) pulses, similar to the experiment shown in E.

Fig. 1.

Focally evoked mEPSCs. A, Response to a single puff of Ba/K solution. The large inward current, which persists for several seconds after the puff, was attributable primarily to summation of AMPA receptor-mediated synaptic events, as it was greatly reduced by 5 μm NBQX (data not shown).Inset, Portions of the larger trace, from the areas indicated by letters, displayed on a faster time scale.B, Cumulative probability histogram of rise times of mEPSCs evoked at the soma (solid line;n = 341) or ∼150 μm out along a dendrite (dotted line; n = 363).Inset, Averages of events in both conditions.

Fig. 2.

Schild analysis of NBQX and KYN. A, Whole-cell recording of AMPA receptor-mediated responses to different concentrations of KA in the presence of 300 μm KYN. KA (150 μm, approximately an EC₅₀ dose) was applied in the absence of KYN to determine the half-maximal response.B, Schild plot showing the shift in the KA EC₅₀ (the dose ratio) caused by different concentrations of NBQX (filled circles) or KYN (open circles).

Fig. 3.

Measuring mEPSC rise times. A, Part of a simulated file containing 1000 mEPSCs, each with randomly chosen time courses (sum of rising and decaying exponentials, each with randomly selected time constants) and amplitudes (30 ± 15 pA). The bottom panel shows the same events as the top panel, but with gaussian noise added. B, Test of how well the algorithm measured the rise times of the simulated mEPSCs. The rise time of each noisy event, as measured by the algorithm, was divided by the rise time measured directly in the absence of noise. This ratio was then plotted versus the amplitude of the event, measured in the absence of noise (see Materials and Methods). C, mEPSCs recorded from a cultured CA1 hippocampal neuron.D, Cumulative probability histogram of amplitudes of mEPSCs recorded from the same cell as in C, at −50 mV (dashed line; n = 739) and −100 mV (solid line; n = 781).E, Cumulative probability histogram of 20–80% rise times of the same events analyzed in D.

Fig. 5.

KYN, but not NBQX, slows the mEPSC rise time.A, Cumulative probability histogram from a single cell. mEPSC rise times in control (solid line;n = 403), 200 μm KYN (dashed line; n = 230), and 60 nm NBQX (dotted line; n = 242) are compared.Inset, Unscaled averages of the rising phase of the events in 200 μm KYN (dashed line) and 60 nm NBQX (dotted line). B, mEPSCs from six cells, combined, in control (solid line;n = 2071) and 60 nm NBQX (dotted line; n = 1294). C, mEPSCs from the same six cells as in B, combined, in control (solid line; n = 2071) and 200 μm KYN (dashed line; n= 1158).

Fig. 6.

THA slows the mEPSC rise in the presence of KYN.A, Cumulative probability histogram of mEPSC rise times from a single cell in 60 nm NBQX (solid line) and NBQX plus 300 μm THA (dotted line). B, mEPSC rise times from five cells, combined, in NBQX alone (solid line;n = 1094) and in NBQX and THA (dotted line; n = 890). C, Cumulative probability histogram of mEPSC rise times from the same cell as in A in 200 μm KYN (solid line) and KYN plus 300 μm THA (dotted line). D, mEPSCs from the same six cells as inB, combined, in KYN alone (solid line;n = 1158) and in KYN and THA (dotted line; n = 1059).

Fig. 7.

Li⁺ slows the mEPSC rise in the presence of KYN. A, Cumulative probability histogram of mEPSC rise times from a single cell in 60 nm NBQX in Na⁺ (solid line) and Li⁺ (dotted line). B, mEPSC rise times from four cells, combined, in NBQX with Na⁺ (solid line;n = 1017) and Li⁺ (dotted line; n = 866). C, Cumulative probability histogram of mEPSC rise times from the same cell as in A in 200 μm KYN with Na⁺ (solid line) and Li⁺ (dotted line). D, mEPSC rise times from the same six cells as in B, combined, in KYN with Na⁺ (solid line; n = 1150) and Li⁺(dotted line; n = 806).

Fig. 9.

KYN binds AMPA receptors faster thanl-glu. A, Patch responses to 10 mml-glu alone (large response) and 10 mml-glu together with 10 mm KYN (smaller response). Inset, Earliest phase of both responses, displayed on a faster time scale. B, Simulation of AMPA receptor responses under the same conditions as inA. Magnification of the inset is the same as in A.

Fig. 10.

KYN dissociates rapidly from the AMPA receptor.A, Patch responses to 10 mml-glu alone (thin solid line) or in the presence of 1 mm KYN (thick solid line). In the second case, KYN was present in both the control andl-glu solutions. The dotted line indicates the response in KYN scaled to the control trace at the conclusion of the glutamate application. B, The thick solid line indicates the arithmetic difference between the control trace and the scaled KYN trace in A. This difference current was integrated, normalized (dashed line), and fitted by a single exponential function (thin line; see Results). C, D, Simulated AMPA receptor responses under the same conditions as in A,B.

Fig. 11.

Modeled effects of KYN and THA on AMPA receptor mEPSCs. A, Simulated mEPSCs (bottom panel) evoked by single-component glutamate transients (top panel). All simulated transients reached their peak values in 10 μsec. See Results for amplitudes and decay time constants of transients and for rise times of the simulated mEPSCs. B, Same as A, but for the two-component transients described in Results. C, Simulated dose inhibition by KYN when the model was driven by the transient in A (filled circles) orB (open circles) compared with the experimentally observed dose inhibition (the shaded region represents mean ± SD from Fig. 4) and the equilibrium prediction (solid line, from Fig. 4).D, Mean EPSC amplitudes (±SD; n = 6), normalized, in control, 300 μm THA, 250 μm KYN, and 250 μm KYN plus 300 μm THA (gray bars). Open circles show data from individual cells. Responses in KYN are from same cells as in control; connections between points in THA and points in KYN have been omitted for clarity. Inset, Averaged EPSCs from one cell in each of the four conditions. The smallest response is in the presence of KYN alone.