HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 278, Issue 50, 50112-50119, December 12, 2003

This Article Abstract Full Text (PDF) All Versions of this Article: 278/50/50112    most recent M307990200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Melzer, N. Articles by Fahlke, C. Search for Related Content PubMed PubMed Citation Articles by Melzer, N. Articles by Fahlke, C. Social Bookmarking                      What's this?

Glutamate Modifies Ion Conduction and Voltage-dependent Gating of Excitatory Amino Acid Transporter-associated Anion Channels^*

Nico Melzer, Alexander Biela, and Christoph Fahlke¶

From the Institute of Physiology, RWTH Aachen, 52057 Aachen, Germany and Centro de Estudios Cientificos, Avenida Prat 514, Valdivia, Chile

Received for publication, July 23, 2003 , and in revised form, September 18, 2003.

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Excitatory amino acid transporters (EAATs) mediate two distinct transport processes, a stoichiometrically coupled transport of glutamate, Na⁺, K⁺, and H⁺, and a pore-mediated anion conductance. We studied the anion conductance associated with two mammalian EAAT isoforms, hEAAT2 and rEAAT4, using whole-cell patch clamp recording on transfected mammalian cells. Both isoforms exhibited constitutively active, multiply occupied anion pores that were functionally modified by various steps of the Glu/Na⁺/H⁺/K⁺ transport cycle. Permeability and conductivity ratios were distinct for cells dialyzed with Na⁺- or K⁺-based internal solution, and application of external glutamate altered anion permeability ratios and the concentration dependence of the anion influx. EAAT4 but not EAAT2 anion channels displayed voltage-dependent gating that was modified by glutamate. These results are incompatible with the notion that glutamate only increases the open probability of the anion pore associated with glutamate transporters and demonstrate unique gating mechanisms of EAAT-associated anion channels.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Excitatory amino acid transporters (EAATs)¹ mediate the removal of glutamate from the synaptic cleft in the central nervous system and the uptake of glutamate in kidney and intestine (1–3). Five structurally distinct subtypes of mammalian glutamate transporters, EAAT1–EAAT5, have been identified in recent years (4–9). Each of these isoforms exhibits two separate transport processes: a stoichiometrically coupled cotransport of one glutamate with three Na⁺ ions and one H⁺, in countertransport to one K⁺ ion (10, 11); and an anion conductance that appears to be pore-mediated. The EAAT-associated anion channel is currently thought to function as a glutamate-gated channel with a tight coupling of channel opening and closing to conformational changes of the corresponding carrier domain. In this model, only certain carrier conformations are associated with conducting anion pores, and the anion channel cycles between conducting and non-conducting states during transitions through various conformational states of the glutamate carrier (8, 9, 12–15).

We investigated anion conduction properties of two EAAT isoforms, human EAAT2 and rat EAAT4, using patch clamp recordings of transfected tsA201 cells under conditions that eliminate the Glu/Na⁺/H⁺/K⁺ current component. Our results demonstrated that opening and closing of the EAAT-associated anion channels as well as the interaction with the glutamate uptake process are more complex than previously assumed. External glutamate modifies intrinsic properties of EAAT2- and EAAT4-associated anion channels such as anion selectivity and the rate constants of anion permeation. Moreover, EAAT4 anion channels exhibit voltage-dependent opening and closing transitions that are modified by glutamate. These results provide novel insights into the function of neurotransmitter transporters and illustrate similarities as well as differences between transporter-associated pores and ion channels.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Expression of hEAAT2 and rEAAT4 in tsA201 and HEK293 Cells—A pcDNA3.1-hEAAT2 construct was generated by subcloning the coding region of hEAAT2 (16) (kindly provided by Dr. M. Hediger, Harvard Medical School, Boston) into a pcDNA3.1 vector (Invitrogen) using flanking NotI restriction sites. The pcDNA3.1-rEAAT4 construct was kindly provided by Dr. J. Rothstein, The Johns Hopkins University, Baltimore (17). Transient transfection of tsA201 cells using the Ca₃(PO₄)₂ technique was performed as described previously (18). To identify cells with a high probability of expressing recombinant transporters, cells were cotransfected with a plasmid encoding the CD8 antigen and incubated 5 min before use with polystyrene microbeads precoated with anti-CD8 antibodies (Dynabeads M-450 CD8, Dynal, Great Neck, NY) (19). Only cells decorated with microbeads were used for electrophysiological recordings. By adjusting the CD8/EAAT cDNA ratio, we optimized our conditions so that almost every cell with beads exhibited a current component with characteristic properties shown in Fig. 1. For each construct, two independent recombinants were examined and exhibited indistinguishable functional properties. Oligoclonal cell lines stably expressing hEAAT2 were obtained by selection for resistance to the aminoglycoside antibiotic geneticin (G418, Roche Applied Science) as described previously (18). The stable cell lines so obtained exhibited electrical properties indistinguishable from transiently transfected tsA201 cells. Untransfected tsA201 cells displayed a negligible endogenous anion current component (current amplitude at –175 mV: Cl^–, –54 ± 6 pA; , –63 ± 13 pA; SCN^–, –56 ± 10 pA; at +125 mV: Cl^–, 40 ± 3 pA; , 31 ± 3 pA; SCN^–, 211 ± 13 pA, n > 5 cells) that was not modified by the addition of glutamate.

View larger version (19K): [in this window] [in a new window]   FIG. 1. Constitutive anion currents in cells expressing EAAT2 or EAAT4. Representative whole-cell current traces recorded from tsA201 cells expressing EAAT2 (A–C) and EAAT4 (D–F) transporters. Cells were dialyzed with a NaCl-based internal solution and perfused with a glutamate-free external solution containing Cl– (A and D), (B and E), or SCN– (C and F) as main anions. Cells were held at 0 mV, and voltage steps between –115 and +125 mV in 40-mV intervals were applied.

View larger version (16K): [in this window] [in a new window]   FIG. 4. Glutamate alters anion conduction properties of EAAT transporters. A, current ratios of isochronal current amplitudes before and after external application of 0.5 mM glutamate at +125 mV obtained from cells expressing EAAT2 (n = 3–14) or EAAT4 (n = 4–7). Cells were dialyzed with a standard NaCl-based internal solution and perfused with solutions containing either Cl–, I–, , or SCN– as principal anions. * and ** indicate statistically significant differences based on a Student's t test (*, p < 0.05; **, p < 0.01). B, glutamate-induced changes of the anion permeability ratio for EAAT-2 and EAAT-4 in the absence of internal potassium. C, the EAAT-associated anion conductance is mediated by a multiply occupied pore. Dependence of the current reversal potential on the absolute anion concentration in cells expressing EAAT2 (, n = 9) or EAAT4 (, n = 3). The concentration of []i and [SCN]e was altered in a way that the concentration ratio stayed constant. The measured reversal potentials were different at p < 0.01.

View larger version (25K): [in this window] [in a new window]   FIG. 5. Effect of changes of [SCN–]o on isochronal current amplitudes in tsA201 cells expressing EAAT2. A, mean current amplitudes for various external [SCN–] in the absence of external glutamate. SCN– was substituted equimolarly with Cl–. B, mean current amplitudes for various external [SCN–] in the presence of 0.5 mM external glutamate. C and D, voltage dependence of the KD for SCN– in the absence (•) and in the presence () of external glutamate. Means ± S.E. from 7 cells.

View larger version (28K): [in this window] [in a new window]   FIG. 2. Anion selectivity of EAAT2- and EAAT4-associated anion channels. A and C, voltage dependence of isochronal current amplitudes determined for different external anion compositions. Current amplitudes were measured in an external solution containing Cl–, I–, , or SCN– as the principal anions, in cells expressing EAAT2 (A) and EAAT4 (C). Cells were dialyzed with a standard NaCl-based internal solution. B and D, voltage dependence of isochronal current amplitudes for different internal anion compositions. Current amplitudes were measured in cells dialyzed with a solution containing Cl–, , or SCN– as the principal anions, in cells expressing EAAT2 (B) and EAAT4 (D). The external solution had the standard composition. Means ± S.E. of absolute current data from at least 5 cells for all figures.

To obtain current-voltage relationships in various internal or external anions, isochronal current amplitudes were determined 1 ms after the voltage step. Current ratios given in Fig. 4 were determined by dividing isochronal current amplitudes obtained from the same cell after application of 0.5 mM L-glutamate by the corresponding value before application of external substrate. For the determination of normalized current amplitudes as a mean of channel selectivity (Tables I and II), we divided current amplitudes in various external anions by the corresponding value determined from the same cell in a -based external solution, rather than in a standard external solution to minimize errors introduced by dividing by the rather small current amplitudes of cells expressing EAAT2 in external Cl^–. Reversal potentials were only used for further analysis when the cell exhibited a whole-cell conductance that was at least three times larger than the mean value for untransfected cells under identical conditions. Permeability ratios were calculated from reversal potential measurements under biionic conditions using the Goldman-Hodgkin-Katz equation as described (18) (Tables I and II). Affinity constants (K_D) (Fig. 5) were calculated by plotting isochronal current amplitudes versus [SCN^–]_e after a Lineweaver-Burk transformation and fitting a linear function with a least square routine with Excel (Microsoft) and Sigma Plot (Jandel Scientific). K_D values were determined from individual cells, in the absence and in the presence of L-glutamate, and all K_D values provided are means ± S.E. of 7 individual cells. The voltage dependence of the relative open probability (Fig. 7A) was determined by plotting the normalized instantaneous current amplitude at + 90 mV after 0.2-s prepulses to different voltages versus the preceding potential. Activation curves obtained in this manner were then fit with a single Boltzmann term plus a voltage-independent value (I(V) = I₀/(1 + e^V–V0.5/kV) + P_min). To describe the time course of current inactivation, a sum of two exponentials and a time-independent value were fit to data recorded during a series of voltage steps from a holding potential of 0 mV (Fig. 7B). To describe the recovery from inactivation, we used a pulse protocol consisting of a 150-ms conditioning pulse to +160 mV, a hyperpolarizing prepulse (to potentials between –180 and –120 mV) of variable duration, and a fixed test step to +160 mV (Fig. 7C). The dependence of the peak current amplitude on the prepulse duration was fit with a single exponential giving the time constant of recovery from inactivation (Fig. 7D).

View larger version (32K): [in this window] [in a new window]   FIG. 7. Time- and voltage-dependent changes of EAAT4-associated anion currents. A, voltage dependence of the relative open probability of the EAAT4 anion channels in the absence (•, n = 7) and in the presence of 0.5 mM glutamate (, n = 7). B, voltage dependence of time constants of inactivation in the absence (•, n = 31) and in the presence (, n = 31) of 0.5 mM glutamate. C, current response of a tsA201 cell expressing EAAT4 to a pulse protocol consisting of a pulse to +160 mV followed by a voltage step –120 mV of variable duration and a fixed test step to +160 mV. D, voltage dependence of the time constants of the recovery from inactivation in the absence (•, n = 7) and in the presence of 0.5 mM glutamate (, n = 5). All data were obtained from cells dialyzed with a standard intracellular solution and extracellularly perfused with NaNO3-based solution.

(Eq. 1)

To determine the single channel amplitude at a certain potential, the background variance () measured at the current reversal potential was subtracted from variances determined at various time periods after a voltage step to positive potentials (+120 to +160 mV). The subtracted variances were sorted into current bins (25) and plotted versus the corresponding mean current amplitude (Fig. 6). Under these experimental conditions the absolute open probability changes with time as given by Equation 2,

(Eq. 2)

yielding Equation 3.

(Eq. 3)

View larger version (22K): [in this window] [in a new window]   FIG. 6. Variance analysis for EAAT4 anion channels. A, mean current trace obtained from 300 current responses to a test step to +120 mV from a holding potential of –53 mV. B, time course of the variance. C, variance versus current plot fitted to the function 2 = iI(t) – I(t)2/N (solid line). D, voltage dependence of the unitary current amplitudes for three different test potentials (•). The dashed line represents the current-voltage relationship of the instantaneous macroscopic current from 6 cells measured under identical conditions scaled down to superimpose on the single channel currents. All data were obtained from cells dialyzed with a standard intracellular solution and extracellularly perfused with NaNO3-based solution.

(Eq. 4)

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
EAAT-associated Anion Currents Are Active in the Absence of Transporter Substrate—Fig. 1 shows representative recordings from tsA201 cells expressing EAAT2 and EAAT4. Cells were intracellularly dialyzed with a potassium-free solution and extracellularly perfused with glutamate-free solutions with either Cl^– (Fig. 1, A and D), (Fig. 1, B and E), or SCN^– (Fig. 1, C and F) as main anions. These conditions eliminated the Glu/Na⁺/H⁺/K⁺ current component (20) and allowed us to record anion currents in isolation. For cells expressing EAAT2, currents in external Cl^– were small (current amplitude at +125 mV: 0.11 ± 0.02 nA, n = 16) (Fig. 1A). Substitution of Cl^– by (Fig. 1B) or SCN^– (Fig. 1C) increased the anion influx at positive potentials and also the chloride efflux at negative potentials. Cl^– currents in cells expressing EAAT4 were larger than those in cells expressing EAAT2 (current amplitude at +125 mV: 0.27 ± 0.05 nA, n = 12) (Fig. 1D). Again, external (Fig. 1E) and SCN^– (Fig. 1F) enhanced the current amplitudes at negative and positive potentials as compared with recordings in external Cl^–. Under all external anion compositions EAAT2- and EAAT4-associated anion currents rose instantaneously upon voltage steps to negative and positive potentials. The two isoforms differed in the time course of current responses to depolarizing voltage steps. Whereas EAAT4-associated currents decayed in a voltage- and time-dependent manner, such current relaxations could not be observed in cells expressing EAAT2 (Fig. 1).

Fig. 2A shows the voltage dependence of the absolute EAAT2-associated current amplitudes for external Cl^–, I^–, , and SCN^– in the absence of external glutamate. EAAT2 displayed a SCN > NO₃ > I > Cl conductivity sequence (Table I). Fig. 2B presents averaged isochronal current amplitudes for cells intracellularly dialyzed with solutions containing Cl^–, , and SCN^– illustrating a higher conductivity also for intracellular polyatomic anions compared with Cl^– (Table II). Permeability ratios from the reversal potentials obtained in these experiments revealed for EAAT2 a relative anion permeability sequence of P_I > P_SCN > P_NO3 > P_Cl for external anions (Table I), and a P_SCN > P_NO3 > P_Cl sequence for internal anions (Table II). Similar experiments with EAAT4 (Fig. 2, C and D) demonstrated an SCN > NO₃ > I > Cl conductivity and permeability sequence for external as well as for internal anions (Tables I and II).

EAAT2- and EAAT4-associated anion currents differ in the relative anion conductivities as well as in the relative anion permeabilities (Tables I and II), in the absolute current amplitude (Fig. 2) and in the time and voltage dependence of the current amplitude. These differences dismiss the possibility that anion currents measured in the absence of glutamate are conducted by endogenous anion channels up-regulated by overexpression of EAAT transporters and provide evidence that anion pores are formed by EAAT proteins either in homomeric or in heteromeric complexes together with endogenous proteins.

Transitions in the Glutamate Transport Cycle Modify EAAT-associated Anion Pathways—Application of external glutamate enhanced EAAT-associated anion currents and modified the underlying anion channels. Fig. 3 shows EAAT4-associated current amplitudes in the absence and in the presence of glutamate under two ionic compositions: NaCl-based internal and NaNO₃-based external solutions (Fig. 3, A and C) and NaNO₃-based internal and NaCl-based external solutions (Fig. 3, B and D). For both conditions, application of external glutamate caused about a 2-fold increase of the anion current amplitude (Fig. 3, C and D, and Fig. 4A). Similar increases of the anion current amplitude by glutamate are observed in cells expressing EAAT2 (Fig. 4A). This demonstrates that there is a substantial constitutive activity of EAAT anion channels in the absence of glutamate.

View larger version (35K): [in this window] [in a new window]   FIG. 3. Application of external glutamate increases macroscopic current amplitudes and modifies functional properties of EAAT anion channels. A and B, representative whole-cell current traces recorded from tsA201 cells expressing EAAT4 transporters intracellularly dialyzed with NaCl-based solution but perfused with a -based solution (A), or from cells intracellularly dialyzed with a -based solution in NaCl-based external solution (B) before and after external application of 0.5 mM glutamate. C and D, corresponding mean current amplitudes before (•) and after () external application of 0.5 mM glutamate from cells intracellularly dialyzed with NaCl-based solution, but perfused with a -based solution (C) and from cells intracellularly dialyzed with a -based solution in standard external solution (D). Insets, glutamate causes a shift of the anion current reversal potential. Mean ± S.E. from 5 cells.

For EAAT2 and EAAT4, the glutamate-induced current increase depended on the anion composition. For cells intracellularly dialyzed with a NaCl-based internal solution, these values were highest for and smaller for SCN^–, I^–, and Cl^– (Fig. 4A). Moreover, for various ionic conditions application of glutamate caused a significant shift of the reversal potential (Fig. 3, C and D, insets, Tables I and II). These shifts were not due to glutamate permeation through the anion pore, as glutamate caused shifts to more negative under certain conditions (Fig. 3C, inset) and shifts to more positive potentials under others (Fig. 3D, inset). They rather indicate an alteration of the relative anion permeability of EAAT2- and EAAT4-associated anion pathways by glutamate (Fig. 4B).

Other steps in the glutamate transport cycle (20, 26–28) also altered the selectivity of the anion pores. Internal solutions without K⁺ prevent the occupation of K⁺-bound states and might therefore result in a distinct distribution of transporter states than under standard conditions (20). If all anion conducting states displayed the same anion permeability, the reversal potentials determined with the two distinct internal solutions would be identical. In contrast, differences in measured reversal potentials indicate that distinct conformations of EAAT transporters are associated with different anion permeabilities. For EAAT2 and EAAT4, the anion permeability ratios for certain external anions measured in the absence of glutamate depended on the intracellular cation. The relative permeability of EAAT2 for external I^– and for internal SCN^– was significantly different for cells dialyzed with Na⁺ or K⁺ (Table I). For EAAT4, distinct anion permeabilities for external SCN^– and but for none of the tested internal anions were observed (Table I). These results indicate that various conformational states of EAAT2 and EAAT4 are associated with distinct anion selectivities.

EAAT2 and EAAT4 Exhibit Multiply Occupied Anion-selective Pores—The EAAT-associated anion conductance exhibited several properties that appear unusual for a pore-mediated process: an increase of Cl^– efflux upon application of more permeant external anions (Fig. 2) comparable with a transacceleration phenomenon, a change of permeability and conductivity upon addition of glutamate (Figs. 3 and 4, Tables I and II), and the dependence of anion selectivity on the intracellular cation (Tables I and II). These observations could cast some doubt on the notion that the EAAT-anion conductance is pore-mediated. The concentration dependence of the reversal potential under biionic conditions is a very sensitive test for ion conduction by a multiply occupied ion pore (29). By using two permeant anions, SCN^– and , we recorded reversal potentials under biionic conditions with a fixed concentration ratio for various absolute concentrations. As Glu/Na⁺/H⁺/K⁺ currents are absent and anion current amplitudes are significantly larger than endogenous currents, reversal potentials reflect the relative anion permeability of the EAAT-associated anion conductance. In both isoforms, reversal potentials shifted upon reduction of the absolute anion concentration, although the ratio of the anion concentrations at both sides of the membrane stayed constant (Fig. 4C). This demonstrated that anion transport does not occur in a fixed stoichiometry and excluded a carrier-mediated anion transport. Moreover, it indicated that the anion pores of EAAT2 and EAAT4 contain more than one anion.

External Glutamate Modifies the Transfer of Anions through EAAT-associated Pores—The observed changes of anion conduction (Fig. 4) suggest that certain steps in anion permeation through the EAAT anion pores are altered by external substrate. To test which of these steps are primarily affected by external glutamate, we determined the effects of varying the external [SCN^–] composition on the EAAT-associated anion current. Experiments were performed in the absence (Fig. 5A) and in the presence of external glutamate (Fig. 5B). For both conditions, the current amplitude-[SCN^–] plot could be well fit with a Michaelis-Menten relationship with a Hill coefficient of 1 revealing SCN^– dissociation constants (K_D) for various voltages (Fig. 5, C and D). External substrate increased the K_D values about 2-fold at positive potentials (Fig. 5C) but left them unaltered at negative voltages (Fig. 5D).

Our analysis was performed at voltages more than 80 mV away from the current reversal potential. Under these conditions, anion influx through a multiply occupied pore greatly exceeds anion efflux at positive potentials and is negligible at negative potentials ((j_eff/j_in) = e^–n(VF/RT), where n is the number of simultaneously binding ions and V the transmembrane voltage (30)). The concentration dependence of the current amplitude thus gives the concentration dependence of the SCN^– influx at positive potentials or of the Cl^– efflux at negative potentials. Saturation of the ion flux with increasing external [SCN^–] can occur when the binding-unbinding step to a first binding site within the ion conduction pathway becomes rate-limiting. This happens when the rate of entry approaches the maximum rate of the unbinding step. As unbinding can be followed either by diffusing into the external solution or by translocation through the pore, binding of external SCN^– can be described by three distinct rate constants: the binding constant (k_on), the rate constant of unbinding and passing through the pore (k_trans), and the rate constant of unbinding and diffusing into the external solution (k_off)(K_D = (k_off + k_trans)/k_on). The observed glutamate and voltage dependence of the SCN^– dissociation constants can be nicely explained by assuming that external glutamate accelerates the translocation step for external SCN^–. This assumption correctly predicts that glutamate increases the SCN^– influx (Fig. 4A) as well as the measured K_D values at positive potentials (Fig. 5C). It is also in agreement with the unchanged dissociation constants at negative potentials (Fig. 5D). Most likely, at negative potentials the k_trans rate constant (describing a transition against the electrostatic gradient) is substantially smaller than at positive potentials. Its contribution to the dissociation constants at negative membrane potentials will therefore be negligible. Thus, a glutamate-induced change of k_trans will only little affect the K_D values at negative potentials, in agreement with the experimental results (Fig. 5D). The glutamate-induced acceleration of the k_trans rate constant predicts an increase of the unitary current amplitude, providing a novel explanation for the increase of the macroscopic current amplitude in the presence of substrate.

Unitary Properties of EAAT4-associated Anion Channels—In external I^–, , and SCN^–, EAAT4-associated anion currents displayed prominent time- and voltage-dependent current relaxations in both the absence and the presence of external glutamate (Figs. 1 and 3). These transient currents are not capacitive charge movements; they are changes of ionic current amplitudes as illustrated by the following three pieces of evidence: the amplitude differed for distinct anions (Figs. 1 and 2), the reversal potential shifted upon alterations of the anion composition (Fig. 3), and integrals of the transient currents elicited by steps to positive potentials were always distinct from the integral determined after stepping back to the holding potential. The observed current relaxations therefore can be caused by two distinct mechanisms: time-dependent changes of the open probability of the anion channel or decreases of the unitary current amplitude. We employed non-stationary noise analysis to distinguish between the two possibilities.

Fig. 6 shows variance analysis from a tsA201 cell expressing EAAT4 dialyzed with NaCl-based internal solution in a NaNO₃-based external solution. Fig. 6, A and B, illustrates the time course of the average current for a voltage step from the current reversal potential (–53 mV) to a test potential of +120 mV and the corresponding time course of the variance. Although the current amplitude decreased with time after the voltage step, the noise increased. Such a time dependence of the current amplitude and the noise (² = Ni² P(1 – P) = (I²/N)(1/P – 1)) indicates that the EAAT4-associated current relaxations are due to changes of the number of open channels. Gating of EAAT4 occurs by the same mechanism as gating of ion channels.

To obtain the unitary current amplitude and the absolute open probability of the underlying channels, the variance-mean current plot in Fig. 6C was fitted to Equation 3. From this fit, the unitary current amplitude, 57 ± 7 fA (n = 9) at +120 mV, and the number of channels in the cell were obtained. By dividing the instantaneous current amplitude by the product of the unitary current amplitude and the number of channels, we then calculated the absolute open probability at the holding potential (0.84 ± 0.06, n = 9)). Measurements were repeated in several cells at three different test potentials (Fig. 6D).

Because the open probability of EAAT4 is distinct from zero at the end of the test pulse, the variance-mean current plot is not a complete parabola. However, as the fitted parabola assumed its maximum within the experimentally determined variance-mean current plot, the fits were well defined. We obtained reproducible results for the unitary current amplitude and the open probabilities at the beginning and the end of the voltage step in several cells. All these results demonstrated that our non-stationary noise analysis reliably determined the unitary current amplitude of EAAT4. The high absolute open probability of EAAT4 anion channels in glutamate-free solutions demonstrates that a mere increase of the open probability cannot explain the glutamate-induced increase of the macroscopic current amplitude. Obviously, glutamate also increases the unitary current amplitude of EAAT4 anion channels.

Gating of EAAT4-associated Anion Channels—We next studied time- and voltage-dependent opening and closing transitions of EAAT4-associated anion channels in cells dialyzed with NaCl-based internal solution in a NaNO₃-based external solution. In the absence of glutamate, steady-state inactivation can be described with a single Boltzmann distribution with a midpoint of inactivation at 75.4 ± 9.2 mV (n = 7) and a voltage-dependent minimum open probability of 0.19 ± 0.02 (n = 7) (Fig. 7A). The significant minimum open probability can be assigned to an incomplete inactivation of EAAT4 anion channels by noise analysis (Fig. 6C). Current amplitudes decayed in a biexponential time course with two time constants of about 5 and 60 ms (Fig. 7B). Stepping back to negative potentials after a positive test step caused a monoexponential increase of the amplitude of the currents elicited by a following step to +160 mV (Fig. 7, C and D).

Voltage-dependent gating of EAAT4 anion channels is modified by external glutamate. Addition of external glutamate increased the minimum open probability to 0.60 ± 0.05 (n = 7) and shifted the midpoint of inactivation to 106.3 ± 3.2 mV (n = 7) (Fig. 7A). Before application of glutamate, currents decayed in a biexponential time course at positive potentials (Fig. 7B). After application of 0.5 mM glutamate, the current decay became monoexponential with a time constant of about 5 ms and a significantly increased relative late current amplitude, i.e. a significant increase of the ratio of the late by the instantaneous current amplitude at positive potentials (Fig. 7, A and B). Furthermore, glutamate decreased the time constants of recovery from inactivation (Fig. 7D). The glutamate-induced gating alterations will result in an increase of the EAAT4 anion conductance under various conditions. In the absence as well as in the presence of external glutamate, membrane depolarization might cause an inactivation of EAAT-associated anion channels. In the presence of glutamate, this inactivation will be less complete and recovered faster after returning to the resting potential causing a decreased input resistance and membrane length constant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
EAAT-associated Anion Currents—The small background anion permeability of our expression system enabled us to measure directly EAAT-associated anion currents in the absence as well as in the presence of external substrate. For both isoforms, anion currents were significantly different from currents observed in untransfected cells. Application of external glutamate caused a 2-fold increase of the anion current amplitude (Fig. 4A) indicating that EAAT-associated anion channels are active in the absence of substrate and that this tonically active anion conductance is a substantial component of the activated anion current (Figs. 3 and 4). The existence of a glutamate-independent anion conductance has been reported (15, 31, 32); however, previous estimates of the relative amplitude of this current amplitude for EAAT1 and EAAT3 were much lower (between 0.05 and 0.2 (15, 32)). This difference likely arises from the methods used to measure the glutamate-independent anion currents. Earlier experiments employed blocking agents to estimate the glutamate-independent anion conductance. It appears possible that block of the anion conductance by these agents is incomplete resulting in an under-estimation of the substrate-independent conductance. In contrast, we determined current amplitudes without subtraction procedures. As anion currents are potentially contaminated by currents conducted by endogenous anion channels, our estimate of the tonically active EAAT-associated anion currents might be above the real value. For measurements with Cl^– as the only permeant anion, currents associated with EAAT2 are quite small, and indeed the glutamate-induced current increase determined under these conditions is smaller than the values for other anions (Fig. 4A). However, for other experimental conditions, the contribution of endogenous anions currents appears to be negligible for the following experimental results. The substrate-independent anion currents in EAAT2 and EAAT4-expressing cells differ in selectivity (Table I) and display a distinct time and voltage dependence of the current amplitude (Fig. 1), demonstrating that the anion currents are conducted by distinct heterologously expressed anion channels. Moreover, if a significant component of the substrate-independent anion current were conducted by endogenous channels, one would expect that the relative glutamate-induced anion current increases differ for varying expression levels of the EAAT protein, which was not observed in our experiments with I^–, , or SCN^– as permeant anions.

Anion Selectivity of EAAT-associated Pores—For EAAT2 and EAAT4, larger and polyatomic anions display a higher permeability and conductivity than Cl^–. The mechanisms underlying this particular anion selectivity of EAAT-associated anion channels are currently not very well understood. However, a comparison with a different class of anion-selective channels, the ClC family of voltage-gated chloride channels, allows us insights into the processes that are involved in anion permeation and selection between different anions. ClC channels, similar to every anion channel tested so far, exhibit pores with anion-binding sites that prefer large and polyatomic anions (34). There are two classes of ClC channels that differ in the selectivity between anions, isoforms with a Cl^– > and isoforms or mutant channels exhibiting an > Cl^– permeability and conductivity sequence. These two classes of ClC channels do not differ in the selectivity of binding but in the absolute binding interaction energy for anions (34). For example, in ClC-1, a ClC isoform with a Cl^– > anion permeability and conductivity sequence, binding of anions is tight. Therefore, larger and polyatomic anions dwell longer than small anions at the anion-binding sites. This reduces the permeability of these anions and causes the block of chloride current (35, 36). In contrast, in ClC-4, exhibiting a > Cl^– permeability sequence, the interaction energy is substantially lower. Association steps to the binding sites are rate-limiting and thus permeability follows the binding selectivity of the anion-binding sites (18). For EAAT2 and EAAT4, we found no indication of block by permeating anions. Apparently, for both anion pores, anion binding is less tight than in ClC channels resulting in the observed SCN^– > > I^– > Cl^– conductivity sequence.

EAAT4-associated Anion Channels Have a Small Unitary Conductance and a High Open Probability in Substrate-free Solution—By using non-stationary noise analysis, we determined unitary current amplitudes of EAAT4 anion channels in the fA range (Fig. 6D). By taking the distinct voltages and anion composition into account, these results are significantly larger than published unitary current amplitudes for EAAT1 (15) and smaller than those reported for a native glutamate transporter in photoreceptors from the tiger salamander (12) and heterologously expressed EAAT5 (33). We do not know the reasons for these differences. They might be caused by isoform-specific differences in anion conductance, but they could also be due to the distinct experimental procedures employed. The earlier results were obtained by applying various concentrations of glutamate assuming that the open probability in the absence of substrate is close to zero. If the studied glutamate transporter-associated anion channels exhibited comparably large open probabilities in the absence of substrate as EAAT4, it could be that EAAT current amplitudes and variances have been subtracted by these procedures.

A Substrate-modified Anion Pore in EAAT2 and EAAT4 — For EAAT2 and EAAT4, glutamate application alters the current reversal potentials under various asymmetric anion conditions (Figs. 3 and 4 and Tables I and II) indicating changes of the selectivity of the EAAT-associated anion channels. Experiments on cells dialyzed with either K⁺-containing or K⁺-free solution demonstrated that not only binding of glutamate but also other steps in the glutamate uptake cycle modify the anion selectivity of the EAAT anion pore. The finding that conformational changes of the glutamate carrier domain cause alterations of the anion selectivity filter suggests a close spatial proximity of glutamate transport and anion pore.

Comparing dissociation constants for SCN^– in the absence and in the presence of glutamate revealed that the K_D values at positive voltages were increased by application of substrate but unaffected in the negative range (Fig. 4), in agreement with the notion that glutamate increases the rate constant for transversing the pore. These results suggest that glutamate causes an augmentation of the unitary current amplitude. This idea is supported by the high absolute open probability of EAAT4 anion channels in the absence of glutamate (0.84 ± 0.06, n = 9 at –53 mV). An increase of the open probability alone cannot explain the 3-fold increase of the macroscopic current amplitude by glutamate under these conditions (Figs. 3A and 4A).

A Glutamate-modified Voltage-gated Anion Channel—So far, EAAT-associated anion channels have been thought to be gated by a tight coupling to the glutamate carrier domain. Opening and closing were assumed to be governed by transitions in the glutamate uptake cycle resulting in gating features that are very similar to ligand-gated channels. Our results demonstrate certain discrepancies to this concept. EAAT-associated anion channels exhibit a substantial open probability in the absence of glutamate transport. Moreover, voltage-dependent gating of EAAT4 illustrates opening and closing transitions that are independent of transitions in the glutamate uptake cycle. These gating transitions are modified by glutamate in a way favoring the conducting state, i.e. by increasing the voltage-independent open probability (Fig. 5) and accelerating the recovery from inactivation (Fig. 5). In the presence as well as in the absence of substrate, gating of EAAT4 depends on the anion composition (Fig. 3, A and B), quite similar to other anion channels, such as ClC-type (37, 38) or volume-activated anion channels (39).

Our results suggest a new view of EAAT-associated anion channels. They function as voltage-dependent anion channels much like conventional anion channels, but the functional properties are altered by the glutamate carrier domain. This modification not only alters transitions between conducting and non-conducting states but also affects pore properties, in clear contrast to almost all known examples of a modification of an ion channel by a ligand-bound receptor.

	FOOTNOTES

 
^* This work was supported by the Deutsche Forschungsgemeinschaft Grant FOR450/1 (to C. F.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Institute of Physiology, RWTH Aachen, Pauwelsstr. 30, 52057 Aachen, Germany. Tel.: 49 241 80 88810; Fax: 49 241 80 82434; E-mail: chfahlke{at}physiology.rwth-aachen.de.

¹ The abbreviation used is: EAATs, excitatory amino acid transporters.

	ACKNOWLEDGMENTS

 
We thank Dr. M. Hediger and Dr. J. Rothstein for providing expression constructs for hEAAT2 and rEAAT4; Dr. Patricia Hidalgo and Dr. J. P. Johnson for critical review of the manuscript and helpful discussions; Dr. Louis DeFelice for helpful discussions; and Hannelore Heidtmann and Barbara Poser for excellent technical assistance.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Takahashi, M., Billups, B., Rossi, D., Sarantis, M., Hamann, M., and Attwell, D. (1997) J. Exp. Biol. 200, 401–409[Abstract]
2. Seal, R. P., and Amara, S. G. (1999) Annu. Rev. Pharmacol. Toxicol. 39, 431–456[CrossRef][Medline] [Order article via Infotrieve]
3. Kanner, B. I. (1994) J. Exp. Biol. 196, 237–249[Abstract/Free Full Text]
4. Kanai, Y., and Hediger, M. A. (1992) Nature 360, 467–471[CrossRef][Medline] [Order article via Infotrieve]
5. Pines, G., Danbolt, N. C., Bjoras, M., Zhang, Y., Bendahan, A., Eide, L., Koepsell, H., Storm-Mathisen, J., Seeberg, E., and Kanner, B. I. (1992) Nature 360, 464–467[CrossRef][Medline] [Order article via Infotrieve]
6. Storck, T., Schulte, S., Hofmann, K., and Stoffel, W. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 10955–10959[Abstract/Free Full Text]
7. Arriza, J. L., Fairman, W. A., Wadiche, J. I., Murdoch, G. H., Kavanaugh, M. P., and Amara, S. G. (1994) J. Neurosci. 14, 5559–5569[Abstract]
8. Arriza, J. L., Eliasof, S., Kavanaugh, M. P., and Amara, S. G. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 4155–4160[Abstract/Free Full Text]
9. Fairman, W. A., Vandenberg, R. J., Arriza, J. L., Kavanaugh, M. P., and Amara, S. G. (1995) Nature 375, 599–603[CrossRef][Medline] [Order article via Infotrieve]
10. Zerangue, N., and Kavanaugh, M. P. (1996) Nature 383, 634–637[CrossRef][Medline] [Order article via Infotrieve]
11. Levy, L. M., Warr, O., and Attwell, D. (1998) J. Neurosci. 18, 9620–9628[Abstract/Free Full Text]
12. Larsson, H. P., Picaud, S. A., Werblin, F. S., and Lecar, H. (1996) Biophys. J. 70, 733–742[Abstract/Free Full Text]
13. Wadiche, J. I., Amara, S. G., and Kavanaugh, M. P. (1995) Neuron 15, 721–728[CrossRef][Medline] [Order article via Infotrieve]
14. Billups, B., Rossi, D., and Attwell, D. (1996) J. Neurosci. 16, 6722–6731[Abstract/Free Full Text]
15. Wadiche, J. I., and Kavanaugh, M. P. (1998) J. Neurosci. 18, 7650–7661[Abstract/Free Full Text]
16. Trotti, D., Rolfs, A., Danbolt, N. C., Brown, R. H., Jr., and Hediger, M. A. (1999) Nat. Neurosci. 2, 427–433[CrossRef][Medline] [Order article via Infotrieve]
17. Lin, C. L., Tzingounis, A. V., Jin, L., Furuta, A., Kavanaugh, M. P., and Rothstein, J. D. (1998) Brain Res. 63, 174–179[CrossRef]
18. Hebeisen, S., Heidtmann, L., Cosmelli, D., Gonzalez, C., Latorre, R., Alvarez, O., and Fahlke, Ch. (2003) Biophys. J. 84, 2306–2318[Abstract/Free Full Text]
19. Jurman, M. E., Boland, L. M., Liu, Y., and Yellen, G. (1994) BioTechniques 17, 876–881[Medline] [Order article via Infotrieve]
20. Bergles, D. E., Tzingounis, A. V., and Jahr, C. E. (2002) J. Neurosci. 22, 10153–10162[Abstract/Free Full Text]
21. Barry, P. H. (1994) J. Neurosci. Methods 51, 107–116[CrossRef][Medline] [Order article via Infotrieve]
22. Neher, E. (1992) Methods Enzymol. 207, 123–131.[Medline] [Order article via Infotrieve]
23. DeFelice, L. J. (1981) in Introduction to Membrane Noise (DeFelice, L. J., ed) pp. 291–323, Plenum Publishing Corp., New York
24. Sigworth, F. J. (1980) J. Physiol. (Lond.) 307, 97–129[Abstract/Free Full Text]
25. Heinemann, S. H., and Conti, F. (1992) Methods Enzymol. 207, 131–148[Medline] [Order article via Infotrieve]
26. Otis, T. S., and Jahr, C. E. (1998) J. Neurosci. 18, 7099–7110[Abstract/Free Full Text]
27. Watzke, N., Bamberg, E., and Grewer, C. (2001) J. Gen. Physiol. 117, 547–562[Abstract/Free Full Text]
28. Auger, C., and Attwell, D. (2000) Neuron 28, 547–558[CrossRef][Medline] [Order article via Infotrieve]
29. Dani, J. A. (1989) J. Neurosci. 9, 884–892[Abstract]
30. Hodgkin, A. L., and Keynes, R. D. (1955) J. Physiol. (Lond.) 128, 61–88
31. Eliasof, S., and Jahr, C. E. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 4153–4158[Abstract/Free Full Text]
32. Watzke, N., and Grewer, C. (2001) FEBS Lett. 503, 121–125[CrossRef][Medline] [Order article via Infotrieve]
33. Palmer, M. J., Taschenberger, H., Hull, C., Tremere, L., and von Gersdorff, H. (2003) J. Neurosci. 23, 4831–4841[Abstract/Free Full Text]
34. Fahlke, Ch. (2001) Am. J. Physiol. 280, F748–F757
35. Fahlke, Ch., Dürr, C., and George, A. L., Jr. (1997) J. Gen. Physiol. 110, 551–564[Abstract/Free Full Text]
36. Rychkov, G. Y., Pusch, M., Roberts, M. L., Jentsch, T. J., and Bretag, A. H. (1998) J. Gen. Physiol. 111, 653–665[Abstract/Free Full Text]
37. Richard, E. A., and Miller, C. (1990) Science 247, 1208–1210[Abstract/Free Full Text]
38. Pusch, M., Ludewig, U., Rehfeldt, A., and Jentsch, T. J. (1995) Nature 373, 527–530[CrossRef][Medline] [Order article via Infotrieve]
39. Voets, T., Droogmans, G., and Nilius, B. (1997) J. Gen. Physiol. 110, 313–325[Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati