INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Nicotinic acetylcholine receptors are pentamers of homologous subunits. The primordial AChR pentamer likely contained only one type of subunit, exemplified by the 7 homopentamer found in brain. However, evolution led to subunit diversity and in parallel heteropentamers appeared. Presumably, fine tuning of subunit structure, together with combinatorial diversity of heteropentamers, allowed a wide range of physiological demands to be met. Therefore conserved residues in homologous subunits of heteropentamers potentially represent structures that contribute to particular physiological functions. Here we combine mutagenesis and single channel recordings to examine functional contributions of conserved residues in the M4 transmembrane domain of the heteromeric muscle AChR.

AChRs have the composition ₂ in adult muscle and ₂ in embryonic and denervated muscle. Each subunit contains an amino-terminal extracellular domain of ~210 amino acids, four transmembrane domains (M1-M4), and a short extracellular tail. Extracellular domains of four of the five subunits contribute to the two agonist binding sites where the biological response is initiated, whereas the M2 domain of each subunit contributes to the cation-selective channel, the endpoint of the biological response (Unwin, 1995). The locations and functional roles of the M1, M3, and M4 transmembrane domains are not as well understood as those of the M2 domain. However, together with M2 they likely comprise the channel gating apparatus.

The M4 domain is the least conserved among the transmembrane domains, is the most hydrophobic, and has been extensively labeled by hydrophobic probes (Blanton and Cohen, 1992, 1994). Although M4 is likely the outermost of the transmembrane domains, several findings suggest that it is essential for proper activation of the AChR (Bouzat et al., 1994; Lee et al., 1994; Ortiz-Miranda et al., 1997; Bouzat et al., 1998). We recently demonstrated that T422 located at position 14' of the M4 domain of the -subunit contributes through a hydrogen bond to channel gating (Bouzat et al., 2000). Here we combine site-directed mutagenesis with single-channel recordings to analyze the contribution to gating of this highly conserved residue in the M4 domain of the muscle non--subunits. Our studies reveal that the functional contribution of the residue at position 14' is not conserved among the homologous subunits but is restricted to the - and -subunits. Additionally, whereas in the -subunit mutation of T14' affects both channel opening and closing steps, mutations of the corresponding residue in the -subunit affects mainly the channel closing step.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Construction of mutant subunits

Mouse cDNAs were subcloned into the cytomegalovirus-based expression vector pRBG4 (Sine, 1993). Mutant subunits were constructed using the QuikChange Site-Directed mutagenesis kit (Stratagene, La Jolla, CA). Restriction mapping and DNA sequencing confirmed all constructs.

Expression of AChR

HEK293 cells were transfected with -, -, -, and -cDNA subunits (wild-type or mutants) using calcium phosphate precipitation at a subunit ratio of 2:1:1:1 for :::, respectively, essentially as described previously (Bouzat et al., 1994, 1998). For transfections, cells at 40 to 50% confluence were incubated for 8 to 12 h at 37°C with the calcium phosphate precipitate containing the cDNAs in Dulbecco's Modified Eagle Medium (DMEM) plus 10% fetal bovine serum. Cells were used for single-channel measurements 1 or 2 days after transfection.

Patch-clamp recordings

Recordings were obtained in the cell-attached configuration (Hamill et al., 1981) at a membrane potential of 70 mV and at 20°C. The bath and pipette solutions contained 142 mM KCl, 5.4 mM NaCl, 1.8 mM CaCl₂, 1.7 mM MgCl₂, and 10 mM HEPES (pH 7.4). Patch pipettes were pulled from 7052 capillary tubes (Garner Glass Co, Claremont, CA) and coated with Sylgard (Dow Corning, Midland, MI). Acetylcholine (ACh) at final concentrations of 1 to 300 µM or 20 mM choline were added to the pipette solution.

Single-channel currents were recorded using an Axopatch 200B patch-clamp amplifier (Axon Instruments, Inc., Union City, CA), digitized at 5-µs intervals with the PCI-6111E interface (National Instruments, Austin, TX), recorded to the hard disk of a computer using the program Acquire (Bruxton Corporation, Seattle, WA), and detected by the half-amplitude threshold criterion using the program TAC (Bruxton Corporation, Seattle, WA) at a final bandwidth of 10 kHz. Data of AChRs activated by 20 mM choline were analyzed at a bandwidth of 5 kHz to avoid detection of some blockages that can be resolved at 10 kHz. Therefore, channel kinetics can be reduced to that of the closed to open reaction (Grosman and Auerbach, 2000). Open and closed time histograms were plotted using a logarithmic abscissa and a square root ordinate (Sigworth and Sine, 1987) and fitted to the sum of exponential functions by maximal likelihood using the program TACFit (Bruxton Corporation, Seattle, WA).

Open probability within clusters (p_open) was experimentally determined at each ACh concentration by calculating the mean fraction of time the channel is open within a cluster.

Kinetic analysis

Kinetic analysis was restricted to clusters of channel openings that each reflect the activity of a single AChR. Clusters of openings corresponding to a single channel were identified as a series of closely spaced events preceded and followed by closed intervals greater than a critical duration (_crit); this duration was taken as the point of intersection of the predominant closed component and the succeeding one in the closed time histogram. To minimize errors in assigning cluster boundaries, we analyzed only recordings from patches with low channel activity in which both components are clearly differentiated from one another. Because each cluster contains one more opening than closing, to avoid biasing in favor of opening, only clusters containing more than 10 openings were considered for further analysis. In addition, any clusters showing double openings were rejected. The predominant closed duration component, associated with closings within clusters, became shorter with increasing concentrations of agonist. Thus, we assume that this component reflects the set of transitions between unliganded closed and diliganded open states. Examples of the predominant closed component and the corresponding values of _crit for AChRs activated by different concentrations of ACh are: wild type, 10 µM, 6.50 ± 1.00 ms, _crit 40 to 50 ms; 30 µM, 1.10 ± 0.18 ms, _crit 8 to 10 ms; 100 µM, 0.18 ± 0.09 ms, _crit 1 to 1.5 ms; 300 µM, 0.08 ± 0.05 ms, _crit 0.8 ms; S14'A, 10 µM, 4.90 ± 1.00 ms, _crit 40 to 50 ms; 30 µM, 1.20 ± 0.09 ms, _crit 8 to 10 ms; 100 µM, 0.23 ± 0.08 ms, _crit 1 to 1.5 ms; 300 µM, 0.09 ± 0.04 ms, _crit 0.8 ms; T14'A, 30 µM, 4.40 ± 0.50 ms, _crit 40 to 50 ms; 100 µM, 0.86 ± 0.13 ms, _crit 10 to 15 ms; 300 µM, 0.38 ± 0.12 ms, _crit 2 to 3 ms; 600 µM, 0.13 ± 0.02 ms, _crit 1 ms.

For each recording, kinetic homogeneity was determined by selecting clusters on the basis of their distribution of mean open duration and p_open, as described before (Wang et al., 1997; Bouzat et al., 2000). For each cluster within a recording, we calculated the p_open, mean open duration, and mean closed duration and plotted their distributions. Typically, the distributions contained a dominant approximately Gaussian component and minor contributions of clusters with different properties. Clusters showing mean open duration and open probability values within 2 SD of the major component were selected and retained for the kinetic analysis. Typically, more than 80% of the events were selected (see Table 1). As shown in the examples of Table 1, mean values obtained from distributions of open probability, mean open channel duration, and mean closed channel duration of clusters do not change significantly after the selection procedure. In addition, comparison of closed- and open-time histograms before and after selection indicates that the selected clusters are representative of the predominant population of AChR in the patch (Bouzat et al., 2000).

The resulting open and closed intervals from different patches, each at a specified ACh concentration or at 20 mM choline, were analyzed according to kinetic schemes using the program MIL (QuB suite, State University of New York, Buffalo, NY). Briefly, the program allows simultaneous fitting of recordings at different agonist concentrations and estimates the rate constants using a maximal likelihood method that corrects for missed events (Qin et al., 1996). The dead time was typically 30 µs; calculated rate constants were stable over a range of dead times varying from 18 to 40 µs (Bouzat et al., 2000). Probability density functions of open and closed durations were calculated from the fitted rate constants and instrumentation dead time and superimposed on the experimental dwell time histogram as described by Qin et al. (1996). Calculated rates were accepted only if the resulting probability density functions correctly fitted the experimental open and closed duration histograms.

For wild-type and some mutant AChRs activated by ACh, ₂ was constrained to its previously estimated value (Sine et al., 1995; Wang et al., 1997; Salamone et al., 1999) because brief closings due to gating and channel blocking become indistinguishable at high ACh concentrations (Wang et al., 1997; Salamone et al., 1999; Bouzat et al., 2000). When ₂ was allowed to vary freely, MIL failed to converge to a well-defined set of rate constants and approached a value of ~100,000 s¹ (Salamone et al., 1999; Bouzat et al., 2000). Additionally, association and dissociation rate constants were assumed to be equal at both binding sites (Akk and Auerbach, 1996; Wang et al., 1997; Salamone et al., 1999). Rate constants are shown with standard deviations.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Analysis of M4 mutant AChRs

We previously demonstrated that T14' forms a hydrogen bond essential for proper rates of channel opening and closing (Bouzat et al., 2000). To further examine functional contributions of the equivalent residues in the non--subunits, we replaced the conserved threonines in - and -subunits by alanine and serine, as well as the equivalent serine in the -subunit by threonine and alanine. We transfected HEK 293 cells with mutant plus wild-type subunit cDNAs and recorded single channel currents activated by a range of desensitizing concentrations of ACh (3-300 µM). At these ACh concentrations, single channel currents activate in clear clusters of events corresponding to a single channel (Sakmann et al., 1980).

Cotransfection of HEK 293 cells with mutant and wild-type subunit cDNAs could result in the surface expression of subunit-omitted AChRs. However, omitting the - or -subunits eliminates all cell-surface expression. Omitting the -subunit reduces the surface expression to only 16 to 20% of that of ₂ pentamers. In addition, channel activity of ₂₂ pentamers is difficult to detect (Bouzat et al., 1994; Engel et al., 1996). In contrast, channel activity from all mutant AChRs in this study was as easily detected as that from wild-type ₂ AChRs. Furthermore, surface expression of each of the mutant AChRs was similar to that of wild-type AChRs. Thus subunit-omitted receptors do not contaminate recordings from our mutant AChRs.

Wild-type as well as all mutant AChRs open in clusters of well-defined activation episodes at ACh concentrations greater than 3 µM. Single-channel recordings from AChRs containing subunits mutated at position 14' show that only mutations in - and -subunits affect gating kinetics (Fig. 1). We previously reported that the mutant T14'A AChR shows both briefer openings and prolonged intracluster closings compared to wild-type AChRs (Bouzat et al., 2000). By contrast, AChRs containing the mutant S14'A subunit show greatly prolonged openings, opposite to T14'A. AChRs containing the mutant T14'A and T14'A subunits exhibit open and closed intervals similar to wild-type AChRs (Fig. 1 and Table 2).

View larger version (51K): [in this window] [in a new window]   FIGURE 1   Clusters of single-channel currents from wild-type and mutant AChRs. Channels activated by 30 µM ACh were recorded from HEK cells expressing mouse wild-type (2) and AChRs carrying the mutant subunits (T14'A, T14'A, T14'A, S14'A, and T14'A plus S14'A). Currents are displayed at a bandwidth of 9 kHz with channel openings as upward deflections. Membrane potential: 70 mV.

Histograms of open intervals exhibit a main component with a mean of ~900 µs for wild-type AChRs (Table 2). The mean duration of this component decreases by an order of magnitude in the mutant T14'A AChR, whereas it increases by an order of magnitude in the S14'A mutant (Table 2). Maintaining hydrogen bonding capability of the side chain with the T14'S and S14'T mutations retains normal open intervals (Table 2). Histograms of open intervals for AChRs containing the mutations T14'A and T14'A AChRs are indistinguishable from that of wild-type AChR; similar results were obtained with the corresponding mutations to serine in the  and  subunits (Table 2). Thus, substitutions at position 14' of M4 of all subunits that preserve the hydrogen bonding ability do not affect channel open intervals.

For both wild-type and mutant AChRs, histograms of closed intervals within clusters show a main component that becomes progressively briefer with increasing ACh concentration. Fig. 2 clearly illustrates the dependence of the duration of this component on agonist concentration in the T422A AChR. This main component reflects the set of transitions between unliganded closed and diliganded open states. We previously reported that for the T14'A mutant, the mean for this component greatly exceeds that of wild-type AChR at all agonist concentrations (Bouzat et al., 2000; Fig. 1 and Table 2). On the contrary, for the S14'A, T14'A, and T14'A mutants, the mean closed duration at a given agonist concentration is similar to that of wild-type AChR (Fig. 1 and Table 2). Again, substitutions preserving the hydrogen bond ability at position 14' of all subunits do not affect distributions of closed intervals.

View larger version (17K): [in this window] [in a new window]   FIGURE 2   Closed time histograms of T14'A AChRs activated by different ACh concentrations. Mutant AChR channels were activated by a given ACh concentration. The predominant closed duration component, associated with closings within clusters, becomes shorter with increasing concentrations of agonist. This component reflects the set of transitions between unliganded closed and diliganded open states.

Given that open intervals decrease in the T14'A mutant but increase in the S14'A mutant, we investigated the effect of combining both mutant subunits in a single AChR. Open intervals for receptors containing both mutations are similar to those of wild type (Fig. 1 and Table 2). Thus, the opposing effects of T14'A and S14'A offset each other when present in the same AChR. Because the two T14' residues contribute additively to the duration of the open state (Bouzat et al., 1998), the effect of mutating the  subunit is approximately twice that of mutating the  subunit. On the other hand, closed intervals are slightly prolonged in the double mutant compared to wild-type AChR (Table 2).

Kinetic analysis of M4 mutant AChRs

To identify the kinetic step affected by the mutant subunits and to quantify the kinetic changes, we fitted the classical activation scheme (Scheme 1) to the open and closed dwell times.

View larger version (6K): [in this window] [in a new window]   Scheme 1

In Scheme 1, two agonists (A) bind to receptors (R) in the resting state with association rates k₊₁ and k₊₂ and dissociate with rates k₁ and k₂. Receptors occupied by one agonist open with rate ₁ and close with rate ₁, and AChRs occupied by two agonists open with rate ₂ and close with rate _2. At high agonist concentrations (greater than 100 µM ACh) channel blockade is evident and thus the blocked state, A₂B, is included. To estimate the set of rate constants, Scheme 1 was fitted to the data using the program MIL (Qin et al., 1996). We analyzed recordings obtained at multiple ACh concentrations (10-300 µM) simultaneously with the aim of representing sojourns in all the states in Scheme 1 in the analysis.

Rate constant estimates obtained for wild-type AChR (Fig. 3 and Table 3) agree with those previously reported for mouse AChR (Wang et al., 1997; Salamone et al., 1999; Bouzat et al., 2000). Data for all mutants are also well described by Scheme 1 (Fig. 3). The fitted rate constants reveal that the mutation S14'A markedly slows the rate of channel closing (Table 3). In addition, small changes in association and dissociation rate constants are observed in this mutant AChR. In contrast, the corresponding mutations in the - and -subunits do not affect activation rate constants (Fig. 3 and Table 3). Thus, the amino acid located at position 14' of M4 contributes mainly to channel opening and closing rates in the -subunit (Bouzat et al., 2000), to the channel closing rate in the -subunit, but does not contribute to channel gating in the - and -subunits.

View larger version (48K): [in this window] [in a new window]   FIGURE 3   Kinetics of activation of wild-type and mutant AChRs. Open and closed-time histograms of wild-type and mutant AChRs containing the indicated mutant subunit(s) with the fit for Scheme 1 superimposed to the experimental data. Histograms were constructed with the selected clusters. The ordinates correspond to the square root of the fraction of events per bin. The fitting was performed simultaneously with a range of ACh concentrations (3-300 µM) for each mutant. The data shown correspond to 30 µM ACh.

When both mutant - and -subunits are combined into a single AChR, the resulting channels activate with kinetics similar to wild-type AChR. Kinetic analysis of the double mutant AChR reveals normal rate constants except for a slight slowing in the rate of channel opening (Fig. 3 and Table 3). For this mutant, data could be well-fitted allowing ₂ to vary freely (Table 3). Thus, our results show that residue 14' in M4 of the - and -subunits contributes additively to the channel closing rate but contributes nonadditively to the channel opening rate.

Activation of AChRs by choline

Because the channel opening rate constant for wild-type AChRs activated by ACh approaches the time resolution limits of the patch clamp, and because the S14'A mutation increases the probability of channel opening (see below), we used choline, an agonist with a very slow opening rate, to test the possibility that S14'A affects the rate of channel opening (Grosman and Auerbach, 2000). At 20 mM choline, openings occur in easily recognizable clusters for both wild-type and mutant AChRs. Open-channel blockade elicited by the high concentration of choline appears as a reduction in channel amplitude (Grosman and Auerbach, 2000). The mean current of 20 mM choline-activated channels, 2.6 ± 0.3 pA, is ~50% lower than that of AChRs activated by 1 to 100 µM ACh (5.5 ± 0.3 pA) or 100 µM choline (5.6 ± 0.5 pA).

For both wild-type and mutant choline-activated AChRs, open time histograms are well described by a single exponential, but the mean durations differ by approximately an order of magnitude as observed for ACh (0.45 ± 0.07 ms and 3.40 ± 0.70 ms for wild-type and S14'A AChRs, respectively). Closed time histograms for the choline-activated AChRs show a main component that corresponds to intracluster closings (mean durations are 17 ± 3 ms and 11 ± 2 ms for wild-type and S14'A AChRs, respectively). To analyze the kinetics of channel opening and closing in the presence of 20 mM choline, we fitted Scheme 2 to the closed and open intervals of the selected clusters.

View larger version (6K): [in this window] [in a new window]   Scheme 2

Scheme 2 is a subset of Scheme 1, which reduces the kinetics of AChR activity to the closed to open transition because 20 mM choline is a saturating agonist concentration (Grosman and Auerbach, 2000). The resulting estimates for opening and closing rate constants are: wild-type, 110 ± 10 s¹ and 2200 ± 50 s¹ for  and , respectively; S14'A AChR, 125 ± 3 s¹ and 285 ± 7 s¹ for  and , respectively. Thus the changes in gating rate constants using choline as the agonist agree with those using ACh, indicating that S14' selectively affects the channel closing step.

Changes in free energy and double mutant cycles analysis

The channel gating equilibrium constant for ACh-activated receptors, , calculated as ₂/₂, increases approximately sixfold in the S14'A mutant compared to wild type ( = 33), whereas it decreases 20-fold in the T14'A mutant, indicating opposite and unequal contributions of the - and -subunits to the gating equilibrium. Interestingly, when both mutant subunits are combined, the gating equilibrium is close to that of the wild-type AChR (Table 4). Table 4 shows that compared with wild-type AChR, the free energy change of the gating equilibrium is twofold greater in the T14'A than in the S14'A AChR. However, if each -subunit of the AChR contributed equally to the gating equilibrium, the free energy changes would be equal for both - and -subunits.

To determine the degree of coupling between residues at positions 14' of -and -subunits, we used thermodynamic mutant cycles to analyze the gating equilibrium (Horovitz and Fersht, 1990; Hidalgo and MacKinnon, 1995). Our mutational analysis can be described by the following pair of joined mutant cycles:

View larger version (9K): [in this window] [in a new window]   Scheme 3

The cycle begins in the upper left corner containing the wild-type AChR and ends in the lower right corner with mutations in all three subunits. We do not analyze the steps indicating one mutant -subunit because of the problem of separating kinetically different types of AChRs after coexpression of both wild-type and mutant -subunits (Bouzat et al., 1998). Thus, our measured free-energy change upon mutating the -subunit corresponds to the sum of the two horizontal steps in the cycle. Based on this cycle, we calculated a coupling coefficient  according to:
which corresponds to a coupling free energy (RT ln ) of ~0.8 kcal/mol.

We also calculated the contribution of residue 14' to activation free energies for the closed to open transition and for the open to closed transition. Changes of activation free energies due to the mutations could result from changes in ground or transition states or both. Whereas the mutation in the -subunit increases the free energy for channel closing, the mutation in the -subunit decreases it to a similar extent (Table 4). No changes in activation-free energies for channel closing are observed when both mutant subunits are combined (Table 4). As described for the gating equilibrium, assuming that each -subunit contributes equally to closing and opening steps, our results indicate that the -subunit affects the closing transition to a greater extent than the -subunit. In contrast, whereas the activation free energy for channel opening is increased in the -mutant, it is unchanged in the A14' AChR (Table 4).

Open probability of M4 mutant AChRs

To determine the overall consequences of each mutation for receptor activation we determined the open probability as a function of ACh concentration and compared it with that predicted by the kinetically determined rate constants. The predicted dose-response curves superimpose upon the open probability measurements, supporting the estimated rate constants (Fig. 4). For wild-type AChRs, open probability increases with increasing ACh concentration, showing an EC₅₀ of 40 µM. As described before, the dose-response curve of the mutant T14'A AChR is displaced to higher ACh concentrations owing to impaired channel gating (Bouzat et al., 2000). In contrast, the curve for S14'A-containing AChRs is shifted to lower ACh concentrations, decreasing the EC₅₀ to 9 µM. As expected, the profiles for the T14'A (EC₅₀ = 58 µM), T14'A (EC₅₀ = 46 µM), as well as that for the double mutant T14'A-S14'A AChRs (EC₅₀ = 57 µM) are similar to that of the wild-type AChR (Fig. 4).

View larger version (17K): [in this window] [in a new window]   FIGURE 4   Agonist concentration dependence of the channel open probability. The mean fraction of time the channel is open during a cluster (popen) was experimentally determined at the indicated concentrations of ACh. Each point corresponds to the mean ± SD of three patches. The symbols correspond to: , wild type; , S14'A; , T14'A; , T14'A + S14'A; , T14'A; , T14'A. The curves were calculated from Scheme 1, using the fitted rate constants in Table 3, and superimposed on the experimentally determined popen values.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Although M4 is the least conserved of the four transmembrane domains, T14' is highly conserved among subunits and species. We previously demonstrated that in the  subunit a hydrogen bond involving the side chain at position 14' contributes to rapid and efficient gating of the muscle AChR. Here we determine whether these structural and mechanistic contributions to gating are conferred by non--subunits. The results indicate that the functional contribution of the residue at position 14' of M4 is restricted to the - and -subunits, showing that at this position sequence identity among subunits is not accompanied by a common functional contribution.

The present kinetic analysis shows no changes due to mutation at position 14' in - and -subunits. In contrast, mutation of position 14' in the -subunit profoundly slows the step underlying channel closing. Interestingly, this effect is quantitatively opposite to that observed for the T14'A AChR. When mutations in the - and -subunits are combined in a single receptor, the resulting closing rate is similar to that of wild-type AChR. Thus, mutations in - and -subunits contribute additively to the rate of channel closing. Because each AChR contains two T14' residues, and because both -subunits contribute additively and equally to the closing rate (Bouzat et al., 1998), the present results indicate that the residue at 14' of the -subunit has a greater effect on the channel closing rate than that of the -subunit.

The channel opening rate is affected differently by mutations in the - and -subunits. Channel opening is slowed in the T14'A mutant AChR, as revealed by prolonged closings within clusters at high ACh concentrations (Fig. 1) and channel closing is accelerated (Table 3). In addition, data for the T14'A AChR are not well described by Scheme 1. Instead, activation is described by a kinetic scheme containing two consecutive doubly liganded open states (Bouzat et al., 2000). This scheme was previously used to describe activation of other mutant AChRs, which also showed slower rates of channel opening (Wang et al., 1999). Thus, consecutive double liganded open states may be present but not distinguished in wild-type AChRs because the observed rate of channel closing is relatively rapid.

Recordings obtained in the presence of ACh show that the equivalent mutation in the -subunit, S14'A, does not change the rate of channel opening. Because the opening rate constant of wild-type AChRs (₂ in Scheme 1) is at the upper limit of reliable estimation, even a modest increase makes this parameter too fast to be resolved (Grosman and Auerbach, 2000). Therefore, we study the opening rate of S14'A channels activated by saturating concentrations of a slowly opening, low-efficacious agonist as choline. At a saturating concentration of choline, the kinetics of the channel can be reduced to that of the closed to open reaction. Because the opening rate, , is slow in the choline-activated AChRs, this constant can be well measured and thus an increase in such constant can be easily detected. The calculated values for closing and opening rates of wild-type AChRs activated by choline generally agree with those previously reported (Grosman and Auerbach, 2000). The calculated opening rate constant for S14'A mutant is similar to that of wild-type AChR. Thus, in contrast to mutations in the  subunit, which affect both the opening and closing steps, the mutation in the  subunit selectively affects the closing step.

The extent of coupling between different mutant residues can be quantified using double mutant cycles analysis (Horovitz and Fersht, 1990). Applied to the channel gating equilibrium, mutations of residue 14' in the - and -mutants show a low coupling free energy of 0.8 kcal/mol, indicating nearly independent contributions of the mutant subunits. However, to obtain purely independent contributions of both mutant subunits to the gating equilibrium, the opening rate of the S14'A AChR should be ~230,000 s¹ rather than the observed 50,000 s¹. Alternatively, the opening rate for the double mutant, which was 36,000 s¹, should approach the 12,000 s¹ observed for the T14'A AChR. Either of these expected opening rates should have been easily detected in our experiments. Thus the - and -subunits appear to be weakly coupled in contributing to the channel gating equilibrium. Because the - and -subunits contribute additively to the closing rate, and because the closing rate of the double mutant is similar to that of the wild-type AChR, coupling between subunits appears to originate in the channel opening step.

The apparent coupling between - and -subunits may owe to a selective effect of the mutations on ground versus transition state free energies. The double mutant AChR (T14'A plus S14'A) shows a slight decrease in the opening rate constant compared to wild type. However, the rate constant for the double mutant is significantly greater than that of the T14'A AChR. Also, in contrast to the -mutant AChR, activation of the double mutant receptor can be well described by Scheme 1. These results indicate that the S14'A subunit by itself does not affect the opening rate, but when combined with the T14'A mutant subunit, it counteracts the profound decrease in opening rate due to the T14'A mutation. A possible explanation is that the S14'A mutation affects free energy of the closed ground state and the transition state to similar extents, which would not affect the rate constant for opening. The T14'A mutant, on the other hand, may stabilize the closed ground state, leaving the transition state unaffected, which would slow the rate of opening. However, when the mutant  subunit is combined with the mutant -subunit, the combination of effects on ground and transition states leads to an intermediate rate of channel opening.

Our results also reveal the structural basis of the contribution of this lipid-exposed residue in M4 to channel gating. Substitution to threonine in the -subunit, which preserves the hydrogen bonding ability of the side chain, does not affect gating kinetics. In addition, the free energy changes for channel closing are in the range expected for hydrogen bonds between uncharged residues in an aqueous solution (0.5-1.5 kcal/mol; Fersht et al., 1985). Together with our previous analysis of T14' (Bouzat et al., 2000), the present results suggest that although the functional contributions to channel gating of the S14' and T14' are different, the structural bases are common, in both cases mediated by a hydrogen bond.

One explanation for the subunit-selective function of the side chain at position 14' of M4 is that the orientation of this highly conserved residue differs among the subunits. In this regard, T14' in M4 has been labeled by 3-trifluoromethyl-3-(m-iodophenyl)diazirine (TID) in the -subunit of Torpedo, suggesting that it is situated at the lipid-protein interface of the AChR. However, the equivalent residues in -, -, and -subunits were not labeled by TID (Blanton and Cohen, 1994). An alternative explanation is based on the key role that the -subunit plays in governing gating kinetics. In this respect, it is responsible for the fast kinetics typical of the adult AChR. Moreover, we previously showed that residues in the HA region, located within the M3-M4 intracellular domain, together with residues in M4 underlie the different open durations between - and -containing AChRs (Bouzat et al., 1994). Thus, the unique functional roles of the - and -subunits could also explain the subunit-specific contributions to gating of residue at 14' position of the M4 domain.