- Email: [email protected]
Desensitization characteristics of rat recombinant GABAA receptors consisting of α1β2γ2S and α1β2 subunits expressed in HEK293 cells
Neuroscience Letters 278 (2000) 21±24 www.elsevier.com/locate/neulet
Desensitization characteristics of rat recombinant GABAA receptors consisting of a1b2g2S and a1b2 subunits expressed in HEK293 cells Klaus Kramp¯ a, Johannes Bu¯er a, Alexandra Lepier b, Josef Dudel b, Helmuth Adelsberger b,* a
Department of Neurology of the Medical Department, Medizinische Hochschule Hannover, D-30623 Hannover, Germany b Institute of Physiology, Technical University Munich, Biedersteinerstrasse 29, D-80802 Munich, Germany Received 22 September 1999; received in revised form 22 October 1999; accepted 25 October 1999
Abstract Desensitization kinetics of rat recombinant typeA GABAergic receptors consisting of the subunits a1b2g2S or a1b2 was investigated on application of 10±0.001 mM GABA to whole cell patches using a piezo driven liquid ®lament switch for fast application and deapplication. At high GABA concentrations desensitization was triphasic showing increasing time constants and a decreasing extent of desensitization on lowering the GABA concentration. Below agonist concentrations of 1 mM for the trimeric receptor and 0.1 mM for the dimeric one desensitization was biphasic switching to monophasic kinetics at GABA concentrations #0.01 mM for the a1b2g2S-type and #0.003 mM for the a1b2-type, respectively. Comparison with former studies performed with GABAergic receptors consisting of different subunits revealed differences in the desensitization kinetics. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: g-Aminobutyric acid; GABAA-receptor; Rat; desensitization; HEK293; Patch-clamp; Whole cell; Subunit composition
Inhibition in the vertebrate central nervous system is mediated mainly by typeA GABAergic receptors. A great diversity of subtypes with distinct kinetic and pharmacological properties is generated by different combinations of subtypes [3,5,11]. Based on their homologies the cloned subtypes are grouped into six families (a1±6, b1±3, g1±3, d, 1, p), forming channels consisting of ®ve subunits [10]. Heterologous expression systems like Xenopus oocytes or HEK293 cells give the chance to study molecularly de®ned receptor subtypes with distinct subunit compositions. While in such expression systems dimeric functional channels consisting of only ab or ag subunits can be generated, the majority of GABAA receptors in vivo seems to be trimeric, consisting of abg or abd subunits [9]. In previous studies it has been demonstrated that activation and desensitization kinetics of these receptors mimic the time course of inhibitory synaptic currents (IPSCs) [7,8]. Therefore, by variation of subunit composition, organisms can modulate synaptic currents. For the GABAergic receptor consisting of the subunits * Corresponding author. Tel.: 149-89-4140-3376; fax: 149-894140-3377. E-mail address: [email protected] (H. Adelsberger)
a1b2g2L it was shown that the maximal slope of the dose response relationship was .2, indicating at least three binding steps for GABA [6]. The current rise times were dose dependent and shorter than 1 ms with saturating GABA concentrations of 10 mM. In a very detailed investigation a kinetic reaction scheme was proposed and simulated describing the kinetic properties of GABAergic receptors containing b3 and g2L subunits [3]. In the present study we describe the desensitization kinetics of a receptor containing b2 and g2S subunits over a wide GABA concentration range. Comparison with other types of GABAergic channels revealed the in¯uence of subunit composition on desensitization kinetics. cDNAs for the a1, b2 and g2S subunits were cloned into the vector pCDM8 (Invitrogen) and the S65T-green ¯uorescent protein (GFP)-cDNA [4] into the vector pcDNA3 (Invitrogen) for mammalian expression. For electroporation human embryonic kidney (HEK)-293 cells (ATCC CRL1573) were resuspended in electroporation buffer (50 mM K2HPO4, 20 mM K-acetate, pH 7.35) to a density of approx. 8 £ 105 cells/ml. MgSO4 was added to a ®nal concentration of 25 mM. 0.4 ml of the cell suspension was then added to puri®ed plasmid DNA (5 mg for each GABA receptor subunit and 10 mg for S65T-GFP). Electroporation was performed in 0.4-cm cuvettes at 225 V, 750 mF and
0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 88 8- 5
22
K. Kramp¯ et al. / Neuroscience Letters 278 (2000) 21±24
329 V, resulting in a mean pulse time of 25 ms. Successfully electroporated cells were detected by their green ¯uorescence. Cells were superfused with an extracellular solution containing (in mM): 162 NaCl, 5.3 KCl, 0.67 Na2PO4, 0.22 KH2PO4, 15 HEPES and 5.6 glucose, pH 7.4 adjusted with NaOH at room temperature. The patch pipettes were pulled from borosilicate glass with an outer diameter of 1.5 mm (Clark Electromedical Instruments, Pangbourne, England) to a resistance of 8±12 MV. For recordings in the whole-cell con®guration, the pipettes were ®lled with an intracellular solution containing (in mM): 140 KCl, 2 MgCl2, 10 EGTA and 10 HEPES, pH 7.4 adjusted with KOH. Agonists were applied in 10 s intervals to cells on the pipette using a piezo-driven liquid ®lament switch allowing application and deapplication times of well de®ned agonist concentrations of less than 0.3 ms [2]. The patches were polarised at 230 to 250 mV. Currents were ®ltered with 1 kHz. The rise time of averaged current traces was de®ned as the time from 10 to 90% of the peak current. The decay time constants of averaged currents after rapid washout of agonists were determined after logarithmation. All data are given as means ^ SEM. GABA was applied at different concentrations to whole cell patches with GABAergic receptors consisting of the subunits a1b2g2S or a1b2 (Fig. 1). In both cases saturation was achieved with 10 mM GABA, the currents rising within 1 ms to maximum amplitudes. On lowering the agonist concentration the current amplitudes decreased while the current rise times increased in a dose dependent manner. After reaching the peak amplitude with high or intermediate GABA concentrations currents declined due to desensitization (Fig. 1). The number as well as the relative proportions of the different time constants with which desensitization could be ®tted was concentration dependent. The desensitization time constants were named tfast for the shortest time constant and tintermediate and tslow for the two longer components. High agonist concentrations of $1 mM GABA resulted in desensitization that could be well ®tted with three time constants for the a1b2g2s receptor, while the a1b2 receptor showed three desensitization time constants at GABA concentrations of $0.1 mM (Fig. 2). On lowering the GABA concentration the relative proportion of the shortest time constant t on the whole current amplitude declined. At GABA concentrations ,1 mM for the abgtype and ,0.1 mM for the ab-type of receptor the relative proportion of tfast was less than 1%. Therefore, the number of time constants in the ®ts was reduced from three to two below these GABA concentrations. Current traces obtained with 1 mM GABA for the abg-type and 0.1 mM GABA for the ab-type of receptor were alternatively ®tted with three and two desensitization time constants, respectively. The deviation of the ®ts from the recorded data calculated using the `Minimum-x 2 Method' [1] was signi®cantly (Ftest, P 0:01) lower with three time constants in these cases. At lower agonist concentrations the values for x 2
Fig. 1. Averaged current traces (10±20 each) elicited on application of different GABA concentrations at polarization values of 250 mV to patches in the whole cell con®guration. (A) Application of 10, 1, 0.1, 0.01 and 0.003 mM GABA to a cell containing GABAergic receptors consisting of the subunits a1b2g2S. (B) Application of 10, 1, 0.1, 0.01, 0.003 and 0.001 mM GABA to a cell containing GABAergic receptors consisting of the subunits a1b2.
were signi®cantly lower with two desensitization time constants. This fact was taken as a second objective criterion for the reduction of the number of time constants for GABA concentrations with which the amplitude of the shortest time constant was below 1%. Consequently, at GABA concentrations of #0.01 for the abg-type and #0.003 for the ab-type desensitization was ®tted with only one time constant, because at these concentrations the relative amplitude of tintermediate was lower than 1%. The proportion of the relative current amplitudes of the different time constants on the maximum current amplitude was very similar at all GABA concentrations when a1b2 GABAA receptor channels were used (Table 1). In contrast, each of the two faster time constants of desensitization, tintermediate and tfast, reduced the current amplitude by about 10% when experiments were performed with a1b2g2S GABAA receptor channels. At 0.1 mM GABA tfast disappeared. However, more than 70% of the whole current amplitude at different concentrations of GABA decayed with the slowest time constant tslow of desensitization
K. Kramp¯ et al. / Neuroscience Letters 278 (2000) 21±24
23
Table 1 Relative proportions of the three desensitization time constants of the GABAergic receptors consisting of the subunits a1b2g2S and a1b2 a a1b2g2S
a1b2
10 mM GABA Percentage tfast Percentage tintermediate Percentage tslow
9.8 ^ 1.5 10.1 ^ 3.2 80.1 ^ 4.1
41.9 ^ 8.4 21.9 ^ 0.5 36.2 ^ 7.9
1 mM GABA Percentage tfast Percentage tintermediate Percentage tslow
8.2 ^ 2.1 14 ^ 5.1 77.8 ^ 7
42 ^ 2.8 30.6 ^ 1.6 27.4 ^ 3.8
0.1 mM GABA Percentage tfast Percentage tintermediate Percentage tslow
2 28.2 ^ 4.5 71.8 ^ 4.5
42.2 ^ 5.3 22.3 ^ 4.2 35.5 ^ 9.5
a The data (mean ^ SEM) represent the amplitude proportions of the different desensitization time constants relative to the maximal current amplitudes on activation with 10, 1 and 0.1 mM GABA.
Fig. 2. Concentration dependence of desensitization time constants of GABAergic receptors consisting of the subunits a1b2g2S or a1b2. (A) Desensitization of a1b2g2S type receptor ®tted with two time constants in the range of 0.01±1 mM GABA (dashed) and three time constants in the range of 1±10 mM GABA (solid). (B) Desensitization of a1b2 type receptor ®tted with two time constants in the range of 0.01±0.1 mM GABA (dashed) and three time constants in the range of 0.1±10 mM GABA (solid). For each point four to seven recordings were performed.
when a1b2g2S subunits were used, whereas with the a1b2 receptor, tslow reduced the current amplitude only by about 30%. The concentration dependence of steady state currents of the two receptor types is shown in Table 2. In case of the a1b2g2S type this amplitude declined from 83% at 0.001 mM GABA to 20% at 10 mM GABA. For the a1b2 type, desensitization was closer to completion, resulting in residual amplitudes of 8% at 10 mM GABA and 52% on application of 0.001 mM GABA. The values of the time constants of desensitization were also concentration dependent (Fig. 2). With the a1b2g2S receptor the shortest time constant tfast increased from 25.3 ^ 2.5 ms at 10 mM GABA to 31.1 ^ 4.2 ms at 1
mM GABA. With the a1b2 receptor the values for tfast were in the same range as for the a1b2g2S with values of 28.2 ^ 3.3 ms at 10 mM GABA and 26.5 ^ 3.7 ms at 1 mM GABA. For the second time constant tintermediate, signi®cant differences between the two receptor types were obvious. Whereas with the a1b2g2S type the values for tintermediate increased from 209.6 ^ 15.3 ms at 10 mM to 259.9 ^ 45.5 ms at 1 mM GABA, for the dimeric type respective time constants of 114.7 ^ 3.8 ms at 10 mM GABA and 139.2 ^ 13.3 ms at 1 mM GABA were determined. Also, the third time constants tslow with values ranging from 1653 ^ 349.3 ms at 10 mM GABA to 2668 ^ 578 ms at
Table 2 Extent of desensitization of GABAergic receptors consisting of the subunits a1b2g2S and a1b2 a GABA concentration [mM]
a1b2g2S
a1b2
Steady state amplitude [%] 10 3 1 0.3 0.1 0.03 0.01 0.003 0.001
20 ^ 4 22 ^ 6 26 ^ 4 22 ^ 3 33 ^ 5 35 ^ 5 44 ^ 5 68 ^ 11 83 ^ 1
8^1 2 8^1 2 12 ^ 2 17 ^ 4 17 ^ 2 29 ^ 1 52 ^ 5
a Data (mean ^ SEM) are given as the current amplitudes at the steady state of desensitization divided by the maximal current amplitudes.
24
K. Kramp¯ et al. / Neuroscience Letters 278 (2000) 21±24
1 mM GABA ®tted for the a1b2g2S type were signi®cantly longer than the ones determined for the a1b2 type. For the latter one, the values were 660 ^ 6.1 ms at 10 mM GABA and 647 ^ 34.3 ms at 1 mM GABA. At GABA concentrations where the relative amplitudes of tfast were smaller than 1% desensitization was ®tted with the two time constants tintermediate and tslow. The values for these constants were again concentration dependent and increased on decreasing the GABA concentration. If ®tted with two time constants the values of the shorter time constant tintermediate increased for the channel consisting of the subunits a1b2g2S from 68 ^ 6.4 ms at 1 mM GABA to 515.7 ^ 0.01 ms on application of 0.01 mM GABA. For the channel consisting of the subunits a1b2, a short time constant of 53.1 ^ 7.7 ms at 0.1 mM GABA increasing to 96.2 ^ 18.1 ms at 0.01 mM GABA was determined. A larger difference between the two types of receptors was obvious for the second longer time constant tslow. For the trimeric receptor type the values were 1291 ^ 210 ms at 1 mM GABA and 4113 ^ 500 ms at 0.01 mM GABA, in contrast to 572 ^ 71.8 ms at 0.1 mM GABA and 606.9 ^ 46.9 ms at 0.01 mM GABA for the dimeric type. Comparison of the present results with a study performed with GABAergic receptors consisting of the subunits a1b2g2L [6] revealed an in¯uence of the type of g-subunit. The long type of g-subunit used by Jahn et al. [6] is characterised by an eight amino acid insertion between the transmembrane domains three and four [13]. This insertion was demonstrated to in¯uence ethanol modulation of the GABAergic receptor complex [12]. The main difference was a lower extent of desensitization with a steady state amplitude of about 60% on application of 3 mM GABA in contrast to 22% found here. Additionally, the values of the time constants were slightly longer in the former study. Desensitization kinetics at 1 mM GABA studied with receptors consisting of the subunits a1b2g2L and a1b2 also showed differences to the results of the present study. In this study [3] shorter values of the desensitization time constants and a larger extent of desensitization were found compared to the present results. Although an in¯uence of a different expression system that was used there cannot be excluded, the difference in subunit composition seemed to in¯uence the desensitization kinetics.
We thank Prof. H. LuÈddens, Mainz for kindly providing us with the cloned subunits. This work was supported by the Deutsche Forschungsgemeinschaft (SFB 391/A2).
[1] Colquhoun, D. and Sigworth, F.J., The Minimum-x2 Method. In B. Sakmann and E. Neher (Eds.), Single-Channel Recording, Plenum Press, New York, 1995, pp. 551±552. [2] Franke, C., Hatt, H. and Dudel, J., Liquid ®lament switch for ultra-fast exchanges of solutions at excised patches of synaptic membrane of cray®sh muscle. Neurosci. Lett., 77 (1987) 199±204. [3] Haas, K.F. and Macdonald, R.L., GABAA receptor subunit gamma2 and delta subtypes confer unique kinetic properties on recombinant GABAA receptor currents in mouse ®broblasts. J. Physiol., 514.1 (1999) 27±45. [4] Heim, R., Cubitt, A.B. and Tsien, R.Y., Improved green ¯uorescence. Nature, 373 (1995) 663±664. [5] Hevers, W. and LuÈddens, H., The diversity of GABAA receptors. Mol. Neurobiol., 18 (1998) 35±86. [6] Jahn, K., Hertle, I., Bu¯er, J., Adelsberger, H., Pestel, E., ZieglgaÈnsberger, W., Dudel, J. and Franke, C., Activation kinetics and single channel properties of recombinant a1b2g2L GABAA receptor channels. NeuroReport, 8 (1997) 3443±3446. [7] Jones, M.W. and Westbrook, G.L., Desensitized states prolong GABAA channel responses to brief agonist pulses. Neuron, 15 (1995) 181±191. [8] Jones, M.W. and Westbrook, G.L., The impact of receptor desensitization on fast synaptic transmission. Trends Neurosci., 19 (1996) 96±101. [9] McKernan, R.M. and Whiting, P.J., Which GABAA receptor subtypes really occur in the brain? Trends Neurosci, 19 (1996) 139±143. [10] Nayeem, N., Green, T.P., Martin, I.L. and Barnard, E.A., Quaternary structure of the native GABAA receptor determined by electron microscopic image analysis. J. Neurochem., 62 (1994) 815±818. [11] Sieghart, W., Structure and pharmacology of g-aminobutyric acidA receptor subtypes. Pharmacol. Rev., 47 (1995) 182± 234. [12] Wafford, K.A., Burnett, D.M., Leidenheimer, M.J., Burt, D.R., Wang, J.B., Kofuji, P., Dunwiddie, T.V., Harris, R.A. and Sikel, J.M., Ethanol sensitivity of the GABAA receptor expressed in Xenopus oocytes requires 8 amino acids contained in the g2L subunit. Neuron, 7 (1991) 27±33. [13] Whiting, P., McKernan, R.M. and Iversen, L.L., Another mechanism for creating diversity in g-aminobutyrate type A receptors: RNA splicing directs expression of two forms of g2 phosphorylation site. PNAS, 87 (1990) 9966±9970.
Desensitization characteristics of rat recombinant GABAA receptors consisting of a1b2g2S and a1b2 subunits expressed in HEK293 cells Klaus Kramp¯ a, Johannes Bu¯er a, Alexandra Lepier b, Josef Dudel b, Helmuth Adelsberger b,* a
Department of Neurology of the Medical Department, Medizinische Hochschule Hannover, D-30623 Hannover, Germany b Institute of Physiology, Technical University Munich, Biedersteinerstrasse 29, D-80802 Munich, Germany Received 22 September 1999; received in revised form 22 October 1999; accepted 25 October 1999
Abstract Desensitization kinetics of rat recombinant typeA GABAergic receptors consisting of the subunits a1b2g2S or a1b2 was investigated on application of 10±0.001 mM GABA to whole cell patches using a piezo driven liquid ®lament switch for fast application and deapplication. At high GABA concentrations desensitization was triphasic showing increasing time constants and a decreasing extent of desensitization on lowering the GABA concentration. Below agonist concentrations of 1 mM for the trimeric receptor and 0.1 mM for the dimeric one desensitization was biphasic switching to monophasic kinetics at GABA concentrations #0.01 mM for the a1b2g2S-type and #0.003 mM for the a1b2-type, respectively. Comparison with former studies performed with GABAergic receptors consisting of different subunits revealed differences in the desensitization kinetics. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: g-Aminobutyric acid; GABAA-receptor; Rat; desensitization; HEK293; Patch-clamp; Whole cell; Subunit composition
Inhibition in the vertebrate central nervous system is mediated mainly by typeA GABAergic receptors. A great diversity of subtypes with distinct kinetic and pharmacological properties is generated by different combinations of subtypes [3,5,11]. Based on their homologies the cloned subtypes are grouped into six families (a1±6, b1±3, g1±3, d, 1, p), forming channels consisting of ®ve subunits [10]. Heterologous expression systems like Xenopus oocytes or HEK293 cells give the chance to study molecularly de®ned receptor subtypes with distinct subunit compositions. While in such expression systems dimeric functional channels consisting of only ab or ag subunits can be generated, the majority of GABAA receptors in vivo seems to be trimeric, consisting of abg or abd subunits [9]. In previous studies it has been demonstrated that activation and desensitization kinetics of these receptors mimic the time course of inhibitory synaptic currents (IPSCs) [7,8]. Therefore, by variation of subunit composition, organisms can modulate synaptic currents. For the GABAergic receptor consisting of the subunits * Corresponding author. Tel.: 149-89-4140-3376; fax: 149-894140-3377. E-mail address: [email protected] (H. Adelsberger)
a1b2g2L it was shown that the maximal slope of the dose response relationship was .2, indicating at least three binding steps for GABA [6]. The current rise times were dose dependent and shorter than 1 ms with saturating GABA concentrations of 10 mM. In a very detailed investigation a kinetic reaction scheme was proposed and simulated describing the kinetic properties of GABAergic receptors containing b3 and g2L subunits [3]. In the present study we describe the desensitization kinetics of a receptor containing b2 and g2S subunits over a wide GABA concentration range. Comparison with other types of GABAergic channels revealed the in¯uence of subunit composition on desensitization kinetics. cDNAs for the a1, b2 and g2S subunits were cloned into the vector pCDM8 (Invitrogen) and the S65T-green ¯uorescent protein (GFP)-cDNA [4] into the vector pcDNA3 (Invitrogen) for mammalian expression. For electroporation human embryonic kidney (HEK)-293 cells (ATCC CRL1573) were resuspended in electroporation buffer (50 mM K2HPO4, 20 mM K-acetate, pH 7.35) to a density of approx. 8 £ 105 cells/ml. MgSO4 was added to a ®nal concentration of 25 mM. 0.4 ml of the cell suspension was then added to puri®ed plasmid DNA (5 mg for each GABA receptor subunit and 10 mg for S65T-GFP). Electroporation was performed in 0.4-cm cuvettes at 225 V, 750 mF and
0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 88 8- 5
22
K. Kramp¯ et al. / Neuroscience Letters 278 (2000) 21±24
329 V, resulting in a mean pulse time of 25 ms. Successfully electroporated cells were detected by their green ¯uorescence. Cells were superfused with an extracellular solution containing (in mM): 162 NaCl, 5.3 KCl, 0.67 Na2PO4, 0.22 KH2PO4, 15 HEPES and 5.6 glucose, pH 7.4 adjusted with NaOH at room temperature. The patch pipettes were pulled from borosilicate glass with an outer diameter of 1.5 mm (Clark Electromedical Instruments, Pangbourne, England) to a resistance of 8±12 MV. For recordings in the whole-cell con®guration, the pipettes were ®lled with an intracellular solution containing (in mM): 140 KCl, 2 MgCl2, 10 EGTA and 10 HEPES, pH 7.4 adjusted with KOH. Agonists were applied in 10 s intervals to cells on the pipette using a piezo-driven liquid ®lament switch allowing application and deapplication times of well de®ned agonist concentrations of less than 0.3 ms [2]. The patches were polarised at 230 to 250 mV. Currents were ®ltered with 1 kHz. The rise time of averaged current traces was de®ned as the time from 10 to 90% of the peak current. The decay time constants of averaged currents after rapid washout of agonists were determined after logarithmation. All data are given as means ^ SEM. GABA was applied at different concentrations to whole cell patches with GABAergic receptors consisting of the subunits a1b2g2S or a1b2 (Fig. 1). In both cases saturation was achieved with 10 mM GABA, the currents rising within 1 ms to maximum amplitudes. On lowering the agonist concentration the current amplitudes decreased while the current rise times increased in a dose dependent manner. After reaching the peak amplitude with high or intermediate GABA concentrations currents declined due to desensitization (Fig. 1). The number as well as the relative proportions of the different time constants with which desensitization could be ®tted was concentration dependent. The desensitization time constants were named tfast for the shortest time constant and tintermediate and tslow for the two longer components. High agonist concentrations of $1 mM GABA resulted in desensitization that could be well ®tted with three time constants for the a1b2g2s receptor, while the a1b2 receptor showed three desensitization time constants at GABA concentrations of $0.1 mM (Fig. 2). On lowering the GABA concentration the relative proportion of the shortest time constant t on the whole current amplitude declined. At GABA concentrations ,1 mM for the abgtype and ,0.1 mM for the ab-type of receptor the relative proportion of tfast was less than 1%. Therefore, the number of time constants in the ®ts was reduced from three to two below these GABA concentrations. Current traces obtained with 1 mM GABA for the abg-type and 0.1 mM GABA for the ab-type of receptor were alternatively ®tted with three and two desensitization time constants, respectively. The deviation of the ®ts from the recorded data calculated using the `Minimum-x 2 Method' [1] was signi®cantly (Ftest, P 0:01) lower with three time constants in these cases. At lower agonist concentrations the values for x 2
Fig. 1. Averaged current traces (10±20 each) elicited on application of different GABA concentrations at polarization values of 250 mV to patches in the whole cell con®guration. (A) Application of 10, 1, 0.1, 0.01 and 0.003 mM GABA to a cell containing GABAergic receptors consisting of the subunits a1b2g2S. (B) Application of 10, 1, 0.1, 0.01, 0.003 and 0.001 mM GABA to a cell containing GABAergic receptors consisting of the subunits a1b2.
were signi®cantly lower with two desensitization time constants. This fact was taken as a second objective criterion for the reduction of the number of time constants for GABA concentrations with which the amplitude of the shortest time constant was below 1%. Consequently, at GABA concentrations of #0.01 for the abg-type and #0.003 for the ab-type desensitization was ®tted with only one time constant, because at these concentrations the relative amplitude of tintermediate was lower than 1%. The proportion of the relative current amplitudes of the different time constants on the maximum current amplitude was very similar at all GABA concentrations when a1b2 GABAA receptor channels were used (Table 1). In contrast, each of the two faster time constants of desensitization, tintermediate and tfast, reduced the current amplitude by about 10% when experiments were performed with a1b2g2S GABAA receptor channels. At 0.1 mM GABA tfast disappeared. However, more than 70% of the whole current amplitude at different concentrations of GABA decayed with the slowest time constant tslow of desensitization
K. Kramp¯ et al. / Neuroscience Letters 278 (2000) 21±24
23
Table 1 Relative proportions of the three desensitization time constants of the GABAergic receptors consisting of the subunits a1b2g2S and a1b2 a a1b2g2S
a1b2
10 mM GABA Percentage tfast Percentage tintermediate Percentage tslow
9.8 ^ 1.5 10.1 ^ 3.2 80.1 ^ 4.1
41.9 ^ 8.4 21.9 ^ 0.5 36.2 ^ 7.9
1 mM GABA Percentage tfast Percentage tintermediate Percentage tslow
8.2 ^ 2.1 14 ^ 5.1 77.8 ^ 7
42 ^ 2.8 30.6 ^ 1.6 27.4 ^ 3.8
0.1 mM GABA Percentage tfast Percentage tintermediate Percentage tslow
2 28.2 ^ 4.5 71.8 ^ 4.5
42.2 ^ 5.3 22.3 ^ 4.2 35.5 ^ 9.5
a The data (mean ^ SEM) represent the amplitude proportions of the different desensitization time constants relative to the maximal current amplitudes on activation with 10, 1 and 0.1 mM GABA.
Fig. 2. Concentration dependence of desensitization time constants of GABAergic receptors consisting of the subunits a1b2g2S or a1b2. (A) Desensitization of a1b2g2S type receptor ®tted with two time constants in the range of 0.01±1 mM GABA (dashed) and three time constants in the range of 1±10 mM GABA (solid). (B) Desensitization of a1b2 type receptor ®tted with two time constants in the range of 0.01±0.1 mM GABA (dashed) and three time constants in the range of 0.1±10 mM GABA (solid). For each point four to seven recordings were performed.
when a1b2g2S subunits were used, whereas with the a1b2 receptor, tslow reduced the current amplitude only by about 30%. The concentration dependence of steady state currents of the two receptor types is shown in Table 2. In case of the a1b2g2S type this amplitude declined from 83% at 0.001 mM GABA to 20% at 10 mM GABA. For the a1b2 type, desensitization was closer to completion, resulting in residual amplitudes of 8% at 10 mM GABA and 52% on application of 0.001 mM GABA. The values of the time constants of desensitization were also concentration dependent (Fig. 2). With the a1b2g2S receptor the shortest time constant tfast increased from 25.3 ^ 2.5 ms at 10 mM GABA to 31.1 ^ 4.2 ms at 1
mM GABA. With the a1b2 receptor the values for tfast were in the same range as for the a1b2g2S with values of 28.2 ^ 3.3 ms at 10 mM GABA and 26.5 ^ 3.7 ms at 1 mM GABA. For the second time constant tintermediate, signi®cant differences between the two receptor types were obvious. Whereas with the a1b2g2S type the values for tintermediate increased from 209.6 ^ 15.3 ms at 10 mM to 259.9 ^ 45.5 ms at 1 mM GABA, for the dimeric type respective time constants of 114.7 ^ 3.8 ms at 10 mM GABA and 139.2 ^ 13.3 ms at 1 mM GABA were determined. Also, the third time constants tslow with values ranging from 1653 ^ 349.3 ms at 10 mM GABA to 2668 ^ 578 ms at
Table 2 Extent of desensitization of GABAergic receptors consisting of the subunits a1b2g2S and a1b2 a GABA concentration [mM]
a1b2g2S
a1b2
Steady state amplitude [%] 10 3 1 0.3 0.1 0.03 0.01 0.003 0.001
20 ^ 4 22 ^ 6 26 ^ 4 22 ^ 3 33 ^ 5 35 ^ 5 44 ^ 5 68 ^ 11 83 ^ 1
8^1 2 8^1 2 12 ^ 2 17 ^ 4 17 ^ 2 29 ^ 1 52 ^ 5
a Data (mean ^ SEM) are given as the current amplitudes at the steady state of desensitization divided by the maximal current amplitudes.
24
K. Kramp¯ et al. / Neuroscience Letters 278 (2000) 21±24
1 mM GABA ®tted for the a1b2g2S type were signi®cantly longer than the ones determined for the a1b2 type. For the latter one, the values were 660 ^ 6.1 ms at 10 mM GABA and 647 ^ 34.3 ms at 1 mM GABA. At GABA concentrations where the relative amplitudes of tfast were smaller than 1% desensitization was ®tted with the two time constants tintermediate and tslow. The values for these constants were again concentration dependent and increased on decreasing the GABA concentration. If ®tted with two time constants the values of the shorter time constant tintermediate increased for the channel consisting of the subunits a1b2g2S from 68 ^ 6.4 ms at 1 mM GABA to 515.7 ^ 0.01 ms on application of 0.01 mM GABA. For the channel consisting of the subunits a1b2, a short time constant of 53.1 ^ 7.7 ms at 0.1 mM GABA increasing to 96.2 ^ 18.1 ms at 0.01 mM GABA was determined. A larger difference between the two types of receptors was obvious for the second longer time constant tslow. For the trimeric receptor type the values were 1291 ^ 210 ms at 1 mM GABA and 4113 ^ 500 ms at 0.01 mM GABA, in contrast to 572 ^ 71.8 ms at 0.1 mM GABA and 606.9 ^ 46.9 ms at 0.01 mM GABA for the dimeric type. Comparison of the present results with a study performed with GABAergic receptors consisting of the subunits a1b2g2L [6] revealed an in¯uence of the type of g-subunit. The long type of g-subunit used by Jahn et al. [6] is characterised by an eight amino acid insertion between the transmembrane domains three and four [13]. This insertion was demonstrated to in¯uence ethanol modulation of the GABAergic receptor complex [12]. The main difference was a lower extent of desensitization with a steady state amplitude of about 60% on application of 3 mM GABA in contrast to 22% found here. Additionally, the values of the time constants were slightly longer in the former study. Desensitization kinetics at 1 mM GABA studied with receptors consisting of the subunits a1b2g2L and a1b2 also showed differences to the results of the present study. In this study [3] shorter values of the desensitization time constants and a larger extent of desensitization were found compared to the present results. Although an in¯uence of a different expression system that was used there cannot be excluded, the difference in subunit composition seemed to in¯uence the desensitization kinetics.
We thank Prof. H. LuÈddens, Mainz for kindly providing us with the cloned subunits. This work was supported by the Deutsche Forschungsgemeinschaft (SFB 391/A2).
[1] Colquhoun, D. and Sigworth, F.J., The Minimum-x2 Method. In B. Sakmann and E. Neher (Eds.), Single-Channel Recording, Plenum Press, New York, 1995, pp. 551±552. [2] Franke, C., Hatt, H. and Dudel, J., Liquid ®lament switch for ultra-fast exchanges of solutions at excised patches of synaptic membrane of cray®sh muscle. Neurosci. Lett., 77 (1987) 199±204. [3] Haas, K.F. and Macdonald, R.L., GABAA receptor subunit gamma2 and delta subtypes confer unique kinetic properties on recombinant GABAA receptor currents in mouse ®broblasts. J. Physiol., 514.1 (1999) 27±45. [4] Heim, R., Cubitt, A.B. and Tsien, R.Y., Improved green ¯uorescence. Nature, 373 (1995) 663±664. [5] Hevers, W. and LuÈddens, H., The diversity of GABAA receptors. Mol. Neurobiol., 18 (1998) 35±86. [6] Jahn, K., Hertle, I., Bu¯er, J., Adelsberger, H., Pestel, E., ZieglgaÈnsberger, W., Dudel, J. and Franke, C., Activation kinetics and single channel properties of recombinant a1b2g2L GABAA receptor channels. NeuroReport, 8 (1997) 3443±3446. [7] Jones, M.W. and Westbrook, G.L., Desensitized states prolong GABAA channel responses to brief agonist pulses. Neuron, 15 (1995) 181±191. [8] Jones, M.W. and Westbrook, G.L., The impact of receptor desensitization on fast synaptic transmission. Trends Neurosci., 19 (1996) 96±101. [9] McKernan, R.M. and Whiting, P.J., Which GABAA receptor subtypes really occur in the brain? Trends Neurosci, 19 (1996) 139±143. [10] Nayeem, N., Green, T.P., Martin, I.L. and Barnard, E.A., Quaternary structure of the native GABAA receptor determined by electron microscopic image analysis. J. Neurochem., 62 (1994) 815±818. [11] Sieghart, W., Structure and pharmacology of g-aminobutyric acidA receptor subtypes. Pharmacol. Rev., 47 (1995) 182± 234. [12] Wafford, K.A., Burnett, D.M., Leidenheimer, M.J., Burt, D.R., Wang, J.B., Kofuji, P., Dunwiddie, T.V., Harris, R.A. and Sikel, J.M., Ethanol sensitivity of the GABAA receptor expressed in Xenopus oocytes requires 8 amino acids contained in the g2L subunit. Neuron, 7 (1991) 27±33. [13] Whiting, P., McKernan, R.M. and Iversen, L.L., Another mechanism for creating diversity in g-aminobutyrate type A receptors: RNA splicing directs expression of two forms of g2 phosphorylation site. PNAS, 87 (1990) 9966±9970.