Activation and block of mouse muscle-type nicotinic receptors by tetraethylammonium

Abstract

We have studied the activation and inhibition of the mouse muscle adult-type nicotinic acetylcholine receptor by tetraethylammonium (TEA) and related quaternary ammonium derivatives. The data show that TEA is a weak agonist of the nicotinic receptor. No single-channel clusters were observed at concentrations as high as 5 mM TEA or in the presence of a mutation which selectively increases the efficacy of the receptor. When coapplied with 1 mM carbamylcholine (CCh), TEA decreased the effective opening rate demonstrating that it acts as a competitive antagonist of CCh-mediated activation. Kinetic analysis of currents elicited by CCh and TEA allowed an estimate of receptor affinity for TEA of about 1 mM, while an upper limit of 10 s-1 could be set for the wild-type channel-opening rate constant for receptors activated by TEA alone. At millimolar concentrations, TEA inhibited nicotinic receptor currents by depressing the single-channel amplitude. The effect had an IC50 of 2-3 mM, depending on the conditions of the experiment, and resembled a standard open-channel block. However, the decrease in channel amplitudes was not accompanied by an increase in the mean burst duration, indicating that a linear open-channel blocking mechanism is not applicable. Upon studying block by other nicotinic receptor ligands it was found that block by CCh, tetramethylammonium and phenyltrimethylammonium can be accounted for by the sequential blocking mechanism while block in the presence of methyltriethylammonium, ethyltrimethylammonium or choline was inconsistent with such a mechanism. A mechanism in which receptors blocked by TEA can close would account for the experimental findings.

Figure 2. Receptor activation in the presence of CCh and TEA

A, single-channel clusters recorded in the presence of 1 mM CCh, 1 mM CCh plus 5 mM TEA or 100 μM CCh plus 5 mM TEA. Openings are shown downward. The presence of TEA leads to a reduction in the cluster open probability and the amplitude of channel openings. No clusters are seen in the presence of 100 μM CCh and 5 mM TEA. Burst and closed duration histograms are shown next to the traces. The results of the fits are: 1 mM CCh: burst, 0.66 ms; closed, 0.32 ms (99 %), 5.8 ms (1 %); 1 mM CCh + 5 mM TEA: burst: 0.54 ms; closed, 2.4 ms (67 %), 0.07 ms (33 %); 100 m M CCh + 5 mM TEA: burst, 0.42 ms; closed, 9.3 ms (91 %), 0.07 ms (9 %). B, effective opening rate of single-channel clusters obtained in the presence of 1 mM CCh and different concentrations of TEA. The presence of TEA reduces the effective opening rate. C, the apparent receptor affinity to CCh in the presence of TEA. The K_D was determined according to model 1. The presence of TEA reduces the apparent affinity of the receptor to CCh.

Model 1

Model 2

Figure 1. Single-channel currents from wild-type (A) and αS269I (B) mutant receptors

The currents were elicited by 100 or 2000 μM TEA. Openings are shown downward. No clusters of single-channel activity were observed with any concentrations of TEA tested. Burst duration histograms are shown next to the traces. The data were best characterized by one- (at 2000 μM TEA) or two-exponential fits (at 100 μM TEA). The mean event durations were: wild-type, 100 μM TEA: 0.17 ms (93 %) and 0.77 ms (7 %); wild-type, 2000 μM TEA: 0.16 ms; αS269I, 100 μM TEA: 0.63 ms (81 %), 6.0 ms (19 %); αS269I, 2000 μM TEA: 7.0 ms.

Figure 3. The observed values for β‘, with values obtained from simulations of single-channel activity using model 2

A, the values obtained when 1 mM CCh is used as agonist, in the presence of various concentrtions of TEA (○). The small filled circles, connected by a line, show the values obtained when single-channel records were simulated using model 2 with the following values for CCh: k₊₁ = 44 μM⁻¹ s⁻¹, k-₁ = 26000 s⁻¹, β = 7630 s⁻¹ and α = 1000 s⁻¹, and for TEA k₊₁ = 4 μM⁻¹ s⁻¹, k-₁ = 6300 s⁻¹. B, values obtained for ACh at different concentrations in the absence of TEA (○) and in the presence of 1 mM TEA (▵). The lines and small symbols show values obtained from simulated data generated using model 2 with the same parameters for TEA and the following values for ACh: k₊₁ = 110 μM⁻¹ s⁻¹, k_-1 = 18 000 s⁻¹, β = 18 000 s⁻¹ and α = 1000 s⁻¹. Approximately 2500 events were synthesized in each condition, then analysed using SKM and MIL programs as for experimental data.

Figure 4. TEA affects channel event amplitudes and the current-voltage relationship

A, the amplitudes of bursts for the wild-type receptor in the presence of TEA only (○) or 1 mM CCh plus TEA (•), and for the αS269I mutant receptor in the presence of TEA only (□). The currents were recorded at -50 mV. Increased concentrations of TEA result in a reduction of the single-channel current. The results of the fit are given in the text. B, I-V plots from the αS269I mutant receptor activated by TEA. TEA affected the slope of inward current. C, normalized conductance at various membrane potentials as a function of TEA concentration. The curves shown are the results of fits of the equation i([TEA])/i_max = ([TEA]ⁿ/([TEA]ⁿ + (K_D)ⁿ). The fit values for K_D and n are: 906 μM, 0.84 (-100 mV); 1280 μM, 0.83 (-75 mV); 2921 μM, 0.83 (-50 mV); and 5949 μM, 1.35 (-25 mV).

Figure 5. The channel burst durations are minimally affected by TEA

The inverse of the mean burst durations in the wild-type receptor (○) activated by 1 mM CCh plus TEA, and in the αS269I mutant receptor (•) activated by TEA. Channel burst durations were determined on data filtered at 2 kHz. The lines show predicted inverse durations based on the observed reduction in channel amplitudes, assuming simple sequential block.

Figure 6. Relationship between the inverse burst duration and membrane potential

A, inverse burst duration vs. membrane potential in the αS269I mutant receptor activated by TEA. The voltage dependence of the inverse duration decreases, and eventually becomes negative, as the concentration of TEA is increased. B, inverse mean burst duration vs. membrane potential in the wild-type receptor activated by 1 mM CCh in the presence of various concentrations of TEA. The voltage dependence of the inverse duration decreases as the concentration of TEA is increased. C, the normalized inverse burst duration for αS269I receptors activated by TEA is plotted against the fraction of time spent blocked. The inverse burst duration at a given voltage and [TEA] was divided by the predicted burst duration at a very low concentration of TEA, at that voltage, to normalize data at potentials between -25 mV and -75 mV (see A). The fraction of time blocked was estimated from the reduction in burst amplitude. Note that in some cases the mean burst duration was longer than control (ordinal value less than 1) or the burst amplitude was greater than control (abscissal value less than 0). There are three lines plotted on the figure. The dashed line with a slope of -1 shows the prediction for linear channel block. The continuous line through the points is the prediction of a model in which blocked receptors close with a rate φ = 3652 s⁻¹ while unblocked receptors close with a rate α = 1826 s⁻¹. The dotted line shows the linear regression line for the data, with ordinal intercept 0.95 and slope 0.45 (probability that the parameter differs from 0 is P < 10⁻⁸ for intercept, P = 0.01 for slope). D, the normalized inverse burst duration for wild-type receptors activated by 1 mM CCh in the presence of different concentrations of TEA is plotted against the fraction of time spent blocked. The data are presented as described for panel C. The continuous line through the points is the prediction of a model in which blocked receptors close with a rate α = 3652 s⁻¹ while unblocked receptors close with a rate α = 2740 s⁻¹. The dotted line shows the linear regression line for the data, with ordinal intercept 1.15 and slope 0.84 (probability that the parameter differs from 0 is P < 10⁻¹⁸ for intercept, P < 0.0001 for slope).

Figure 7. Single-channel currents from various receptor-ligand combinations in the presence of low and high agonist concentrations

Low agonist concentrations were: 20 μM (wild type-CCh) or 100 μM (all remaining combinations). High agonist concentrations were: 1 mM (wild type-PTMA), 2 mM (wild type-MTEA), 5 mM (wild type-ETMA, wild type-choline), 10 mM (wild type-CCh, αY93F-CCh, wild type-TMA) or 20 mM (αY198F-TMA). In all cases, higher agonist concentrations resulted in reduced apparent single-channel amplitudes.

Figure 8. Reduction in channel amplitudes leads to an increase in channel burst durations for some but not all ligands

The data shown are the relative change in inverse burst duration divided by the relative change in single-channel current. For a simple linear blocking model with rapid block kinetics, the apparent burst duration increases as the single-channel current is reduced and the ratio is one. The mean burst duration and mean single-channel current were estimated at two concentrations of agonist (termed low and high, see legend to Fig. 7), and the ratio of changes computed. The ratio was computed as: z_low - z_high)/(z_low)/i_low - i_high)/(i_low) Some data were obtained with receptors containing mutated subunits (CCh*: αY93F, TMA*: αY198F, TEA*: αS269I), while all other data were obtained using wild-type receptors.