Desensitization contributes to the synaptic response of gain-of-function mutants of the muscle nicotinic receptor

Abstract

Although the muscle nicotinic receptor (AChR) desensitizes almost completely in the steady presence of high concentrations of acetylcholine (ACh), it is well established that AChRs do not accumulate in desensitized states under normal physiological conditions of neurotransmitter release and clearance. Quantitative considerations in the framework of plausible kinetic schemes, however, lead us to predict that mutations that speed up channel opening, slow down channel closure, and/or slow down the dissociation of neurotransmitter (i.e., gain-of-function mutations) increase the extent to which AChRs desensitize upon ACh removal. In this paper, we confirm this prediction by applying high-frequency trains of brief ( approximately 1 ms) ACh pulses to outside-out membrane patches expressing either lab-engineered or naturally occurring (disease-causing) gain-of-function mutants. Entry into desensitization was evident in our experiments as a frequency-dependent depression in the peak value of succesive macroscopic current responses, in a manner that is remarkably consistent with the theoretical expectation. We conclude that the comparatively small depression of the macroscopic currents observed upon repetitive stimulation of the wild-type AChR is due, not to desensitization being exceedingly slow but, rather, to the particular balance between gating, entry into desensitization, and ACh dissociation rate constants. Disruption of this fine balance by, for example, mutations can lead to enhanced desensitization even if the kinetics of entry into, and recovery from, desensitization themselves are not affected. It follows that accounting for the (usually overlooked) desensitization phenomenon is essential for the correct interpretation of mutagenesis-driven structure-function relationships and for the understanding of pathological synaptic transmission at the vertebrate neuromuscular junction.

Figure 1.

An allosteric reaction mechanism for the muscle AChR. C, O, and D denote the closed, open, and desensitized conformations of the channel, whereas A denotes a molecule of ACh. For simplicity, only one of the (probably several) desensitized conformations is included in the model. Also, the two neurotransmitter binding sites are assumed to be functionally equivalent and independent and, therefore, the two possible monoliganded configurations are considered to be functionally indistinguishable. These simplifications have no consequences on the interpretation of our data. The red arrows indicate the rate constants that, according to Eq. 1, determine the kinetics of the macroscopic current decay upon stepping the concentration of ACh from saturating to zero (i.e., during channel deactivation).

Figure 2.

AChR deactivation. (A) The kinetics of deactivation were estimated by exposing outside-out patches (−80 mV) to ∼1-ms, 100 μM ACh pulses. Each plotted trace is the average response of a patch to 10 such pulses applied as a ≤1-Hz train. For clarity, only the responses of some of the studied constructs are shown. All traces were aligned and normalized for easier comparison, and their decaying phases were fitted (least-squares method) from the peak of the current until the end of the transient with single exponential components. The deactivation time constants, averaged over a number of patches per construct, are listed in Table I. (B) Correlation between the (macroscopic) deactivation time constants and the (single-channel) time constants of the slowest component of burst length distributions. It can be shown that for a kinetic scheme like that in Fig. 1 and rate constants like those of the constructs studied here, these two quantities coincide (Colquhoun et al., 1997; Wyllie et al., 1998). For all constructs, single-channel measurements were performed in the cell-attached configuration. In addition, for the Asp, Asn, and Gln mutants, these measurements were also done in the outside-out configuration, and the results were indistinguishable from those obtained in cell-attached experiments. Hence, the reason for the deviation of these mutants' data points from the expected straight line of unity slope (dashed line) remains unclear. Error bars are standard errors.

Figure 3.

Burst length distributions. Example bursts of single-channel openings, and burst length duration histograms corresponding to recordings obtained from individual patches. Currents were recorded at −80 mV in the steady presence of 100 nM ACh. Openings are downward deflections. Display f _c ≅ 6 kHz. The zero-current level is indicated, on each trace, with a dotted line. Bursts were defined using the criterion of Jackson et al. (1983) as outlined in Materials and methods. The time constant of the slowest component of the burst length distribution, the t _crit value, the total number of analyzed events (i.e., openings plus shuttings), and the fraction of misclassified shuttings in the particular cases shown are, respectively: (A) 1.29 ms, 0.27 ms, 5,691, 0.0022; (B) 1.76 ms, 0.22 ms, 25,505, 0.054; (C) 19.5 ms, 1.48 ms, 11,493, 0.047; (D) 19.6 ms, 0.85 ms, 10,183, 0.028. The plotted bursts, especially in the cases of the His and Val mutants, are among the longest bursts in their respective distributions. The “length” of a burst includes the durations of all openings and shuttings within it.

Figure 4.

Calibration of the solution-switching system. The different parameters of the solution-switching system (i.e., diameter of the theta tube openings, relative positioning of the theta tube and patch pipette, flow rate of solutions, bandwidth of the computer-generated waveform) were adjusted so as to optimize the time course of the solution exchange. The latter was estimated by measuring the liquid junction potential by alternatively exposing the tip of an open pipette to 1 M KCl (for ∼1 ms) and 140 mM KCl (for ∼20 ms) solutions. In this particular recording, 48 1-M pulses were applied. The trace was segmented in 12 groups of four pulses, and these were aligned and averaged (red trace). The 10–90% risetime during both the onset and the offset, as well as the duration of exposure to the 1 M KCl solution, are indicated for this representative recording.

Figure 5.

Gain-of-function AChR mutants desensitize during deactivation. The hypothesis that mutant AChRs desensitize upon neurotransmitter removal was tested in the outside-out configuration with the application of high-frequency trains of brief ACh pulses. (A–L) Example current traces from individual patches. Each panel is the response of a different construct to the application of a 50-Hz train of 1-ms, 100 μM ACh pulses. One such trains is indicated in A above the current trace. The zero-current level is indicated, on each trace, with a dotted line. The plots are presented in increasing order of deactivation time constant (Table I). The prediction of Eq. 2, namely that (everything else being equal) the slower the deactivation time course, the more pronounced the depression, is borne out by these recordings. Of course, because the kinetics of entry into and recovery from desensitization are not completely unaffected by the mutations (i.e., “everything else” is not exactly equal; Table I), this relationship cannot be perfect. However, the trend is, undoubtedly, clear (see also Fig. 6).

Figure 6.

Depression of ACh-evoked currents upon repetitive stimulation. (A–L) Peak current values in response to 50-Hz trains of 1-ms, 100 μM ACh pulses were normalized with respect to the first peak in each series and averaged (black circles). The number of averaged responses (n) is indicated. Vertical error bars are standard errors. Example current traces for each construct are given in Fig. 5. The red circles correspond to the fits with the reaction scheme in Fig. 9 A (see Results). For each construct, the fitted parameter was the value of the recovery rate constant (Desensitized→Activatable; expressed as their reciprocals in Table I). The values of the other variables needed for these fits were taken from the experimental data, as follows. The sum of the rate constants leading away from the activated states was calculated as the reciprocal of the deactivation time constant of each construct (Table I). The value of the (single) entry-into-desensitization rate constant (note that desensitization of activated receptors occurs as a single-step isomerization in this scheme) was calculated from the half-time values in Table I, as (ln2)/t _1/2. In turn, these half-times were calculated as explained in Materials and methods and in the legend to Table I. For those mutants displaying double-exponential desensitization time courses, the use of a single desensitization rate constant is, of course, an oversimplification. However, it can be shown that the monoexponential decays calculated using these half-times approximate very well the experimentally obtained double-exponential desensitization time courses during the first tens of milliseconds (see Materials and methods).

Figure 7.

AChR desensitization. The kinetics of entry into desensitization were estimated by exposing outside-out patches (−80 mV) to 2-s, 100 μM ACh pulses. As shown by the example current traces, opening is a transient event. For clarity, only the responses of some of the studied constructs are shown. All traces were aligned and normalized, and their decaying phases were fitted (least-squares method) from the peak of the currents until the end of the 2-s ACh applications with one or two exponential components (only the first second is shown). The parameters of these fits were used to calculate desensitization half-times (see Materials and methods). The corresponding averages, over a number of responses per construct, are listed in Table I.

Figure 8.

AChR recovery from desensitization. The kinetics of recovery from desensitization were estimated using pairs of conditioning and test, 100 μM ACh pulses (1 s and 100 ms in duration, respectively) with intervening intervals of variable length. Desensitization was nearly complete at the end of each conditioning pulse; hence, recovered-fraction values were estimated as the ratio between the peak current elicited by the test pulse and that elicited by the conditioning pulse in each pair. The interval between any two consecutive pairs of pulses was ≥6 s to ensure complete recovery from desensitization. (A) A representative recording from the δS268C mutant. (B) Plots of recovered fraction as a function of the duration of the interpulse interval were well fitted with monoexponential rise functions in spite of the multiple steps that must be involved in the recovery of ACh-diliganded desensitized receptors (i.e., DA₂→DA→D→C or DA₂→DA→CA→C); the corresponding time constants are listed in Table I. Vertical error bars are standard errors. The kinetics of recovery within trains of pulses were estimated as indicated in Fig. 6.

Figure 9.

A mechanistic view of AChR deactivation. (A) A simplified version of the reaction scheme in Fig. 1. This scheme was used to interpret the response of the AChR during zero-ACh, interpulse intervals. The correspondence between states in Fig. 1 and sets of states in this reduced model is explained in Results. Upon ACh removal, receptors in activated states either lose ACh (becoming activatable) or desensitize. Desensitized receptors, in turn, can recover and become activatable. All three steps in this kinetic scheme can be considered to be unidirectional because, in the absence of ACh (a) ACh rebinding cannot occur, (b) diliganded desensitized receptors are much morelikely to recover than to reopen, and (c) activatable receptors do not desensitize much; hence, the single arrows. In this model, the deactivation time constant is the reciprocal of the sum of the two rate constants leading away from the activated states. Consistent with the experimental observations, this model predicts monoexponential time courses of deactivation and recovery from desensitization. Also, consistent with the behavior of wild-type AChRs and of some of the mutants, the predicted time course of entry into desensitization is monoexponential, as well. (B) Kinetics of interconversion among the different sets of states in A upon stepping the concentration of ACh to zero. The time course of each set of states in this “triangular” kinetic scheme was solved analytically (see Materials and methods), and is plotted for a hypothetical gain-of-function mutant (deactivation time constant = 15 ms; entry-into-desensitization rate constant = 50 s⁻¹; recovery rate constant = 10 s⁻¹). The concentration of ACh is stepped to 0 at time zero. All receptors are assumed to be activated before this concentration jump.

Figure 10.

Effect of different variables on the macroscopic response to trains of brief neurotransmitter pulses. (A–D) Normalized peak current values under different hypothetical situations. These plots were generated by analytically solving for the fraction of channels that become activated upon each successive pulse, using the kinetic scheme in Fig. 9 A. Although these calculations do not distinguish between activated closed (CA₂) and activated open (OA₂) receptors, the fraction of activated channels that are open is expected to be large in the case of wild-type and gain-of-function mutant AChRs (∼0.95 in the case of the adult wild-type receptor). Thus, the calculated fraction of activated channels is a good approximation to the fraction of open channels. Unless otherwise indicated, τ_deactivation = 15 ms, desensitization rate constant = 50 s⁻¹, recovery rate constant = 10 s⁻¹, and train frequency = 50 Hz. Each arriving pulse of ACh is considered to be too short for desensitization within a pulse to be appreciable, and is assumed to convert all activatable receptors into activated ones. The latter is a very good approximation because, using the full model in Fig. 1 and adult wild-type rate constant values, it can be calculated (using Q-matrix methods; Colquhoun and Hawkes, 1995) that ∼90% of all activatable AChRs would be activated by the end of a 1-ms, 100 μM ACh pulse. (E and F) Experimental data from two of the gain-of-function mutants exposed to trains of different frequencies. Vertical error bars are standard errors. The y axis value corresponding to the first pulse in each train (black circles) is the same for all trains.

Figure 11.

A naturally occurring gain-of-function mutant desensitizes during deactivation. (A) The kinetics of deactivation were estimated and analyzed as described in Fig. 2 A. The plotted trace is the average response of a patch expressing the ɛT264P mutant to 10 1-ms, 100 μM ACh pulses applied as a 0.33-Hz train. The response of the adult wild-type AChR is also shown for comparison. The deactivation time constant of this mutant, averaged over a number of such trains, is 28 ± 2 ms (mean ± SEM, n = 5 trains), whereas that of the adult wild-type receptor is 0.99 ± 0.08 ms (mean ± SEM, n = 13 trains). (B) The kinetics of desensitization were estimated and analyzed as described in Fig. 7. The response of the adult wild-type AChR is also shown for comparison. In five out of six patches expressing this mutant, the best fit to the decaying phase was obtained with two exponential components. The desensitization half-time of the mutant, averaged over these six patches, is 106 ± 17 ms (mean ± SEM), whereas that of the adult wild-type receptor is 26 ± 4 ms (mean ± SEM, n = 8 patches). (C) The kinetics of recovery from desensitization were estimated and analyzed as described in Fig. 8. Vertical error bars are standard errors. The recovery time constant of the mutant is 742 ± 37 ms (data points from two patches), whereas that of the adult wild-type receptor is 306 ± 19 ms (data points from five patches). (D) Example current trace in response to a 50-Hz train of 1-ms, 100 μM ACh pulses. The zero-current level is indicated with a dotted line.