The atypical cation-conduction and gating properties of ELIC underscore the marked functional versatility of the pentameric ligand-gated ion-channel fold

Abstract

The superfamily of pentameric ligand-gated ion channels (pLGICs) is unique among ionotropic receptors in that the same overall structure has evolved to generate multiple members with different combinations of agonist specificities and permeant-ion charge selectivities. However, aside from these differences, pLGICs have been typically regarded as having several invariant functional properties. These include pore blockade by extracellular quaternary-ammonium cations in the micromolar-to-millimolar concentration range (in the case of the cation-selective members), and a gain-of-function phenotype, which manifests as a slower deactivation time course, as a result of mutations that reduce the hydrophobicity of the transmembrane pore lining. Here, we tested this notion on three distantly related cation-selective members of the pLGIC superfamily: the mouse muscle nicotinic acetylcholine receptor (nAChR), and the bacterial GLIC and ELIC channels. Remarkably, we found that, whereas low millimolar concentrations of TMA(+) and TEA(+) block the nAChR and GLIC, neither of these two quaternary-ammonium cations blocks ELIC at such concentrations; instead, both carry measurable inward currents when present as the only cations on the extracellular side. Also, we found that, whereas lidocaine binding speeds up the current-decay time courses of the nAChR and GLIC in the presence of saturating concentrations of agonists, the binding of lidocaine to ELIC slows this time course down. Furthermore, whereas mutations that reduce the hydrophobicity of the side chains at position 9' of the M2 α-helices greatly slowed the deactivation time course of the nAChR and GLIC, these mutations had little effect--or even sped up deactivation--when engineered in ELIC. Our data indicate that caution should be exercised when generalizing results obtained with ELIC to the rest of the pLGICs, but more intriguingly, they hint at the possibility that ELIC is a representative of a novel branch of the superfamily with markedly divergent pore properties despite a well-conserved three-dimensional architecture.

Figure 1.

Block of the muscle nAChR by TEA⁺. (A) Macroscopic current responses to 2-s pulses of 100 µM ACh recorded at –80 mV from the mouse muscle adult-type nAChR in the outside-out configuration. The response of each patch of membrane was recorded sequentially in the absence of TEA⁺ (left), in the presence of 0.5 mM TEA⁺ flowing through both barrels of the theta-type perfusion tubing (middle), and back again, without TEA⁺ in either barrel (right). The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4 with or without ACh and with or without TEA⁺. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TEA⁺. Note the expanded current scale used for the middle panel; the trace is the average of 25 consecutive responses recorded from a representative patch. (B) Single-channel inward currents elicited by 100 µM ACh recorded in the presence or absence of 0.5 mM TEA⁺ at approximately –80 mV in the cell-attached configuration; openings are downward deflections. The pipette solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, 0.1 ACh, and 10 HEPES/KOH, pH 7.4 with or without TEA⁺. (C) Peak-current amplitudes recorded in response to the application of ACh in the presence of external 0.5 mM or 5.0 mM TEA⁺ and upon TEA⁺ removal normalized to the peak value observed in the initial TEA⁺-free recording. (D) Time constant of desensitization during the application of ACh in the presence of external 0.5 or 5.0 mM TEA⁺ and upon TEA⁺ removal normalized to the time constant fitted to the initial TEA⁺-free recording. The values plotted in C and D are means obtained from six patches each for the two concentrations of TEA⁺; error bars are the corresponding standard errors.

Figure 2.

Block of GLIC by TEA⁺. (A) Macroscopic current responses to 5-s pulses of pH 4.5 solution (pH_holding 7.4) recorded at –80 mV in the outside-out configuration. The response of each patch of membrane was recorded sequentially in the absence of TEA⁺ (left), in the presence of 5 mM TEA⁺ flowing through both barrels of the theta-type perfusion tubing (middle), and back again, without TEA⁺ in either barrel (right). The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂ with or without TEA⁺, and pH buffered with 10 mM HEPES/KOH, pH 7.4, or with 10 mM acetic acid/KOH, pH 4.5. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TEA⁺. Note the expanded current scale used for the middle panel; the trace is the average of 25 consecutive responses recorded from a representative patch. (B) Peak-current amplitudes recorded in response to the application of pH 4.5 in the presence of external 5 mM TEA⁺ and upon TEA⁺ removal normalized to the peak value observed in the initial TEA⁺-free recording. (C) Time constants of desensitization during the application of pH 4.5 in the presence of external 5 mM TEA⁺ and upon TEA⁺ removal normalized to the time constant fitted to the initial TEA⁺-free recording. The values plotted in B and C are means obtained from 6 patches; error bars are the corresponding standard errors.

Figure 3.

Block of the muscle nAChR by lidocaine. (A) Macroscopic current responses to 2-s pulses of 100 µM ACh recorded at –80 mV from the mouse muscle adult-type nAChR in the outside-out configuration. The response of each patch of membrane was recorded sequentially in the absence of lidocaine (left), in the presence of 0.5 mM lidocaine flowing through both barrels of the theta-type perfusion tubing (middle), and back again, without lidocaine in either barrel (right). The insets show the much lower open probability of the channel in the presence of 0.5 mM external lidocaine; openings are downward deflections. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4 with or without ACh and with or without lidocaine. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of lidocaine. Note the expanded current scale used for the middle panel; the trace is the average of 25 consecutive responses recorded from a representative patch. (B) Peak-current amplitudes recorded in response to the application of ACh in the presence of external 0.5 mM lidocaine and upon lidocaine removal normalized to the peak value observed in the initial lidocaine-free recording. (C) Current-decay time constants during the application of ACh in the presence of external 0.5 mM lidocaine and upon lidocaine removal normalized to the time constant fitted to the initial lidocaine-free recording. The values plotted in B and C are means obtained from 4 patches; error bars are the corresponding standard errors. (D) Comparison of the responses of the nAChR to 5-ms pulses of ACh alone and ACh plus lidocaine applied sequentially to the same patch of membrane. The vertical dashed line emphasizes the different current values observed upon fast washout.

Figure 4.

Block of GLIC by lidocaine. (A) Macroscopic current responses to 5-s pulses of pH-4.5 solution (pH_holding 7.4) recorded at –80 mV in the outside-out configuration. The response of each patch of membrane was recorded sequentially in the absence of lidocaine (left), in the presence of 0.5 mM lidocaine flowing through both barrels of the theta-type perfusion tubing (middle), and back again, without lidocaine in either barrel (right). The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂ with or without lidocaine, and pH-buffered with 10 mM HEPES/KOH, pH 7.4, or with 10 mM acetic-acid/KOH, pH 4.5. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of lidocaine. Note the expanded current scale used for the middle panel; the trace is the average of 25 consecutive responses recorded from a representative patch. (B) Peak-current amplitudes recorded in response to the application of pH 4.5 in the presence of external 0.5 mM lidocaine and upon lidocaine removal normalized to the peak value observed in the initial lidocaine-free recording. (C) Current-decay time constants during the application of pH 4.5 in the presence of external 0.5 mM lidocaine and upon lidocaine removal normalized to the time constant fitted to the initial lidocaine-free recording. The values plotted in B and C are means obtained from 12 patches; error bars are the corresponding standard errors.

Figure 5.

50 mM TEA⁺ fails to block ELIC. (A) Macroscopic current responses to 5-s pulses of 10 mM cysteamine recorded at –80 mV in the outside-out configuration. The responses were recorded without TEA⁺ in either barrel (left) or in the presence of 50 mM TEA⁺ flowing through both barrels of the theta-type perfusion tubing (middle). Because ELIC displayed marked rundown in outside-out patches, these traces were not recorded sequentially. Instead, individual patches of membrane were exposed to cysteamine pulses either in the absence or the presence of TEA⁺. In separate recordings, TEA⁺ was also applied to the intracellular side of outside-out patches (right). The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4 with or without cysteamine and with or without TEA⁺. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of external TEA⁺. (B) Peak-current amplitudes recorded in response to the application of cysteamine in the presence of external 50 mM TEA⁺ or internal 5 mM TEA⁺ normalized to the peak value observed in the absence of external or internal TEA⁺. (C) Time constants of desensitization during the application of cysteamine in the presence of external 50 mM TEA⁺ or internal 5 mM TEA⁺ normalized to the time constant fitted to TEA⁺-free recordings. The values plotted in B and C are means obtained from 8 (external 50 mM TEA⁺) and 5 (internal 5 mM TEA⁺) patches; error bars are the corresponding standard errors. Note the lack of effect of millimolar TEA⁺ applied to either side of the membrane.

Figure 6.

Lidocaine effects on ELIC. (A) Macroscopic current response to 5-s pulses of 10 mM cysteamine in the continuous presence of 5 mM lidocaine recorded at –80 mV in the outside-out configuration. The trace shown is the average of 25 consecutive responses recorded from a representative patch. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, 5 lidocaine, and 10 HEPES/KOH, pH 7.4 with or without cysteamine. In the schematic representation of the theta-tubing perfusion, arrows indicate the application of lidocaine. (B) Single-channel inward currents elicited by 10 mM cysteamine recorded at –80 mV in the absence (top) or presence (bottom) of 5 mM lidocaine in the outside-out configuration; openings are downward deflections. The external solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, 10 cysteamine, and 10 HEPES/KOH, pH 7.4, with or without lidocaine.

Figure 7.

X-ray crystal structure of wild-type ELIC bound to cysteamine and either lidocaine or a brominated analogue. (A) Pentameric architecture of ELIC. The inset is a magnified view of the cysteamine-binding site with surrounding side chains shown in ball-and-stick representation. The carbon atoms of the amino acids forming the primary and secondary interfaces are represented in teal and gray, respectively. The yellow mesh represents the 2Fo–Fc electron-density map of cysteamine (carbon atoms, in green) contoured at the level of 1σ. The atoms of oxygen are shown in red; those of nitrogen, in blue; and those of sulfur, in yellow. (B) Structural alignment of wild-type ELIC cocrystalized with cysteamine and lidocaine (teal) with the previously solved model of unliganded wild-type ELIC (PDB code: 2VL0; orange); no change is apparent. For clarity, only one subunit is shown. The inset is a magnified view of the M2 α-helix and the M2–M3 loop. (C) Lateral and top views of the binding site for the bromo derivative N-1-(4-bromophenyl)-N-2,N-2 diethyl glycinamide (shadowed in pink). (D) A magnified view (from the extracellular side) of the bromo-derivative binding site; only two subunits are shown. Protein and brominated-analogue atoms are colored as in A, with the exception of the carbon atoms of the latter, which are shown in orange; the bromine atom is shown in purple. The anomalous-difference map corresponding to the bromo derivative (calculated at 4.8 Å and contoured at 4.5σ) is also displayed in purple. The amino acids in the M2 α-helices are denoted using the prime notation.

Figure 8.

Effects of mutations to the lidocaine-binding site of ELIC. (A–C) Macroscopic current responses to 5-s pulses of 10 mM cysteamine recorded at –80 mV from 3 M2 α-helix mutants in the outside-out configuration. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4, with or without cysteamine. The trace shown in C is the average of 27 responses recorded from 7 patches. (D and E) Peak-current amplitudes and time constants of desensitization normalized to wild-type values. The plotted values are means obtained from 3 (Y14′A), 4 (Y17′A), and 7 (Y14′A + Y17′A) patches; error bars are the corresponding standard errors. The desensitization time courses recorded from the Y14′A + Y17′A double mutant had low peak-current amplitudes, and hence, a total of 27 traces from 7 patches were averaged and fitted (Table 3); a proper fit required two (rather than one) exponential-decay components. (F–H) Single-channel inward currents elicited by 10 mM cysteamine recorded at –80 mV in the outside-out configuration; openings are downward deflections. The external solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, 10 cysteamine, and 10 HEPES/KOH, pH 7.4. For comparison, single-channel traces recorded from the wild-type channel under identical experimental conditions are shown in Fig. 6 B, top. (I and J) Macroscopic current responses to 5-s pulses of 10 mM cysteamine in the continuous presence of 5 mM lidocaine recorded at –80 mV from 2 mutants (in the M3 and M2 α-helices, respectively) in the outside-out configuration. The solutions flowing through the two barrels of the perfusion tubing were as in A–C, with the addition of 5 mM lidocaine. In the schematic representation of the theta-tubing perfusion, arrows indicate the application of lidocaine.

Figure 9.

50 mM TEA⁺ or TMA⁺ fail to block wild-type ELIC and a wider-pore mutant. (A) Macroscopic current responses to 5-s pulses of 10 mM cysteamine recorded from the L9′A + F16′L wider-pore mutant at –80 mV in the outside-out configuration. The responses were recorded without TEA⁺ in either barrel or in the presence of 50 mM TEA⁺ flowing through both barrels of the theta-type perfusion tubing. Because ELIC displayed marked rundown in outside-out patches, these traces were not recorded sequentially. Instead, individual patches of membrane were exposed to cysteamine pulses either in the absence or the presence of TEA⁺. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4, with or without cysteamine and with or without TEA⁺. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TEA⁺. (B) Peak-current amplitudes recorded in response to the application of cysteamine in the presence of external 5 or 50 mM TEA⁺ normalized to the peak value observed in the absence of TEA⁺. The plotted values are means obtained from 4 (5 mM TEA⁺) and 7 (50 mM TEA⁺) patches; error bars are the corresponding standard errors. (C and D) Macroscopic current responses to 5-s pulses of 10 mM cysteamine in the presence of external 50 mM TMA⁺ recorded from the wild type and the L9′A + F16′L mutant at –80 mV in the outside-out configuration. All other conditions are as in A, replacing TEA⁺ with TMA⁺. The corresponding responses in the absence of TMA⁺ are shown in Fig. 5 A (for the wild type) and in panel A of this figure (for the mutant). (E) Peak-current amplitudes recorded in response to the application of cysteamine in the presence of external 50 mM TMA⁺ normalized to the peak value observed in the absence of TMA⁺. The plotted values are means obtained from 9 (wild-type ELIC) and 8 (L9′A + F16′L mutant) patches; error bars are the corresponding standard errors.

Figure 10.

Block of the muscle nAChR by TMA⁺. (A) Macroscopic current response to a 2-s pulse of 50 mM TMA⁺ recorded at –80 mV from the mouse muscle adult-type nAChR in the outside-out configuration. Because TMA⁺ is both a pore blocker and a (desensitizing) agonist of the nAChR, TMA⁺ was applied only transiently rather than continuously (compare with Fig. 9, C and D). The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4, with or without TMA⁺. In the schematic representation of the theta-tubing perfusion, the arrow indicates the application of TMA⁺. (B) Macroscopic current responses to a 1-ms pulse of 100-µM ACh or 50 mM TMA⁺ recorded at –80 mV in the outside-out configuration. The solutions flowing through the two barrels of the perfusion tubing were as described in A with or without ACh and with or without TMA⁺. The two traces are averages of 10 consecutive responses recorded from two separate representative patches and were normalized for displaying purposes in such a way that the current values attained upon ligand removal are the same; this emphasizes the block by TMA⁺ (see red asterisk). In the schematic representation of the theta-tubing perfusion, arrows indicate the application of ACh or TMA⁺. (C) Single-channel inward currents elicited by 0.5 mM or 50 mM TMA⁺ recorded at approximately –100 mV in the cell-attached configuration; openings are downward deflections. The pipette solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, 0.5 or 50 TMA⁺, and 10 HEPES/KOH, pH 7.4. The extent of block caused by TMA⁺ seems to be larger in the single-channel recordings than in the macroscopic recordings. Undoubtedly, this is due to the fast deactivation time constant of the nAChR, which leads to the underestimation of the true peak value of the macroscopic transient upon TMA⁺ removal.

Figure 11.

TMA⁺ conduction through the muscle nAChR and GLIC. (A and B) Macroscopic current responses to 1-ms pulses of 142 mM TMA⁺ recorded at –80 mV from the wild-type (A) and the εT12′P slowly deactivating mutant (B) of the mouse muscle adult-type nAChR in the outside-out configuration. The TMA⁺-containing solution was (in mM) 142 TMACl and 10 HEPES/TMAOH, pH 7.4, whereas the TMA⁺-free solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TMA⁺. (C) Macroscopic current response to a 100-ms pulse of pH-4.5 solution (pH_holding 7.4) containing 142 mM TMA⁺ recorded at –80 mV from GLIC in the outside-out configuration. The TMA⁺-containing solution was (in mM) 142 TMACl and 10 acetic acid/TMAOH, pH 4.5, whereas the TMA⁺-free solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4. The short duration of the pH-4.5-solution application ensured that entry into desensitization within the pulse was negligible. All other conditions are as in A and B. (D) Intra–pulse-to-tail current ratios. The plotted values are means obtained from 17 (wild-type nAChR), 11 (εT12′P nAChR), and 6 (GLIC) patches; error bars are the corresponding standard errors. All current traces shown in this figure are the averages of 10 consecutive responses recorded from representative patches.

Figure 12.

TMA⁺ and TEA⁺ conduction through ELIC. (A and B) Macroscopic current responses to 500-ms pulses of 10 mM cysteamine plus 142 mM TMA⁺ recorded at –80 mV from the wild type (A) and the L9′A + F16′L wider-pore mutant (B) in the outside-out configuration. The TMA⁺-containing solution was (in mM) 142 TMACl, 10 cysteamine, and 10 HEPES/TMAOH, pH 7.4, whereas the TMA⁺-free solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TMA⁺. (C) Intra–pulse-to-tail current ratios. The plotted values are means obtained from 5 (wild type) and 6 (L9′A + F16′L mutant) patches; error bars are the corresponding standard errors. (D) Single-channel inward currents elicited by 10 mM cysteamine recorded at –80 mV in the outside-out configuration (openings are downward deflections) and corresponding all-point amplitude histograms. For the panel on the left, the external solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, 10 cysteamine, and 10 HEPES/KOH, pH 7.4, whereas for the panel on the right, this solution was (in mM) 142 TMACl, 10 cysteamine, and 10 HEPES/TMAOH, pH 7.4. (E–G) Macroscopic current responses and intra–pulse-to-tail current ratios. All conditions are as in (A–C), replacing TMA⁺ with TEA⁺. The values plotted in G are means obtained from 9 (wild type) and 7 (L9′A + F16′L mutant) patches; error bars are the corresponding standard errors. All macroscopic current traces shown in this figure are the averages of 10 consecutive responses recorded from representative patches.

Figure 13.

TEA⁺–K⁺ competition in ELIC. (A and B) Macroscopic current responses to 500-ms pulses of 10 mM cysteamine plus 142 mM TEA⁺ recorded at –80 mV from the wild type (A) and the L9′A + F16′L wider-pore mutant (B) in the outside-out configuration. The TEA⁺-containing solution was (in mM) 142 KCl, 142 TEACl, 10 cysteamine, and 10 HEPES/KOH, pH 7.4, whereas the TEA⁺-free solution was (in mM) 142 KCl and 10 HEPES/KOH, pH 7.4. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TEA⁺. (C) Intra–pulse-to-tail current ratios. The plotted values are means obtained from 9 (wild type) and 3 (L9′A + F16′L mutant) patches; error bars are the corresponding standard errors. Both current traces shown in this figure are the averages of 10 consecutive responses recorded from two representative patches.

Figure 14.

TPA⁺–K⁺ competition and TPA⁺ conduction through ELIC. (A and B) Macroscopic current responses to 500-ms pulses of 10 mM cysteamine plus 142 mM TPA⁺ recorded at –80 mV from the wild type in the outside-out configuration. In (A), the TPA⁺-containing solution was (in mM) 142 KCl, 142 TPACl, 10 cysteamine, and 10 HEPES/KOH, pH 7.4, whereas the TPA⁺-free solution was (in mM) 142 KCl and 10 HEPES/KOH, pH 7.4. In (B), the TPA⁺-containing solution was (in mM) 142 TPACl, 10 cysteamine, and 10 HEPES/TPAOH, pH 7.4, whereas the TPA⁺-free solution was (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4. In the schematic representations of the theta-tubing perfusion, arrows indicate the application of TPA⁺. (C) Intra–pulse-to-tail current ratios. The plotted values are means obtained from 5 patches for each condition; error bars are the corresponding standard errors. Both current traces shown in this figure are the averages of 10 consecutive responses recorded from two representative patches.

Figure 15.

Effect of mutations at position 9′ of M2 on deactivation of the muscle nAChR and GLIC. (A and B) Macroscopic current responses to 1-ms pulses of 100-µM ACh recorded at –80 mV from the wild-type mouse muscle adult-type nAChR (A) and the αL9′A mutant (B) in the outside-out configuration. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4, with or without ACh. (C and D) Macroscopic current responses to 50-ms pulses of pH-4.5 solution recorded at –80 mV from wild-type GLIC (C) and the I9′A mutant (D) in the outside-out configuration. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and the pH was buffered with 10 HEPES/KOH (pH 7.4), 10 TABS/KOH (pH 9.0) or 10 acetic-acid/KOH (pH 4.5). In the case of the gain-of-function I9′A mutant, the (proton-gated) activity of the channel at pH 7.4 was relatively high, and hence, the pH of the solution applied during the “no-agonist” intervals was increased to 9.0. For all panels, the insets emphasize the time courses of deactivation; lines are fits to monoexponential-decay functions. (E) Mutant-construct deactivation time constants obtained from monoexponential fits normalized to the corresponding wild-type time constants. The plotted values are means obtained from 4 (L9′A nAChR) and 2 (I9′A GLIC) patches. An error bar (standard error) is only displayed for the nAChR; for GLIC, the two averaged values were 155 ms and 169 ms. All current traces shown in this figure are the averages of 10–25 consecutive responses recorded from representative patches.

Figure 16.

Effect of mutations at position 9′ of M2 on deactivation of ELIC. (A–E) Macroscopic current responses to 500-ms pulses of 10 mM cysteamine recorded at –80 mV from wild-type ELIC and the indicated mutants in the outside-out configuration. The solutions flowing through the two barrels of the perfusion tubing were (in mM) 142 KCl, 5.4 NaCl, 1.8 CaCl₂, 1.7 MgCl₂, and 10 HEPES/KOH, pH 7.4 with or without cysteamine. The insets emphasize the time courses of deactivation; lines are fits to monoexponential-decay functions. (F) Mutant-construct deactivation time constants obtained from monoexponential fits normalized to the wild-type time constant. The plotted values are means obtained from 6 patches for each mutant; error bars are the corresponding standard errors. All current traces shown in this figure are the averages of 10 consecutive responses recorded from representative patches.