Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2002 Sep 13;544(Pt 3):695–705. doi: 10.1113/jphysiol.2002.029413

Contributions of the non-α subunit residues (loop D) to agonist binding and channel gating in the muscle nicotinic acetylcholine receptor

Gustav Akk 1
PMCID: PMC2290637  PMID: 12411516

Abstract

The agonist binding site of the nicotinic acetylcholine receptor has a loop-based structure, and is formed by residues located remotely to each other in terms of primary structure. Amino acid residues in sites δ57 and δ59, and the equivalent residues in the ε subunit, have been identified as part of the agonist binding site and designated as loop D. The effects of point mutations in sites δ57, δ59, ε55 and ε57 on agonist binding and channel gating were studied. The mutated receptors were expressed transiently in HEK 293 cells and the currents were recorded using the cell-attached single-channel patch clamp technique. The results demonstrate that the mutations mainly affect channel gating with the major portion of the effect due to a reduction in the channel opening rate constant. For both the δ57/ε55 and the δ59/ε57 site, a mutation in the ε subunit had a greater effect on channel gating than a mutation in the δ subunit. In all instances, agonist binding was affected to a lesser degree than channel gating. Previous data have placed the δ57 and δ59 residues in or near the agonist binding pocket. The data presented here suggest that these two residues (and the homologous sites in the ε subunit) are not involved in specific interactions with the nicotinic agonist and that they affect the activation of the nicotinic receptor by shaping the overall structure of the agonist binding site.

The agonist binding sites in the muscle nicotinic acetylcholine receptor (nAChR) are located at the interfaces of the α-δ and α-γ (or α-ɛ) subunits. Both the α and non-α sides of the binding sites are formed by residues widely spaced from each other but joined together through loop-like structures. The residues forming the loops have been identified using affinity labelling and site-directed mutagenesis, and from a more recent X-ray crystallographic study of the snail acetylcholine-binding protein (Arias, 2000; Corringer et al. 2000; Brejc et al. 2001).

On the side of the α-subunit, loop A is formed by residues W86 and Y93, loop B contains residues W149, Y152 and G153, and loop C contains Y190, C192, C193 and Y198. Loops D, E and F are formed by residues of the δ and γ (or ɛ) subunits. Residues on the non-α-subunit that contribute to the binding site are (numbering from the γ-subunit): W55 and E57 (loop D), L109, S111, C115, I116 and Y117 (loop E), and F172, D174 and E183 (loop F). The numerous aromatic residues of the putative binding site have been proposed to line the entrance or path to the final docking site, similar to the aromatic gorge in the acetylcholinesterase molecule (Sussman et al. 1991). Once in the docking site, the quaternary ammonium group of the ACh molecule may interact with the αW149 or the γD174 residue (Czajkowski & Karlin, 1995; Zhong et al. 1998).

Site-directed mutagenesis has been used to explore the functional effects of amino acid substitutions at a number of these sites. The results of such experiments have been described in a number of reviews (see, for example, Arias, 2000). In general, mutations of most putative binding site residues lead to a rightward shift in the channel dose-response curve. A notable exception is the αG153S mutation which impairs agonist dissociation from its binding site leading to a leftward shift in the dose-response curve and a congenital myasthenic syndrome in the patients having this mutation (Sine et al. 1995).

A reduction in the channel current and a rightward shift in the whole-cell dose-response curve, such as that observed for most binding site mutations, can be caused by an effect on agonist binding, channel gating or the combination of the two. In principle, mutagenesis of residues interacting with the agonist during what is construed as the binding process in kinetic analysis, should affect the rate constants of agonist association and dissociation. On the other hand, substitution of residues located in the binding pocket but not interacting directly with the ligand might affect more strongly the coupling of the binding site to the channel gate, i.e. the channel gating rate constants.

In the present work, the contributions of residues forming loop D (δW57 and δD59, and the equivalent residues in the ɛ-subunit) to agonist binding and channel gating were studied. The receptors were activated by acetylcholine (ACh) or tetramethylammonium (TMA), and the currents were measured using the single-channel patch clamp technique. The results demonstrate that mutations to the residues forming loop D mainly affect channel gating with little or no effect on ligand binding.

Methods

Mouse cDNA encoding the wild-type α, β, δ and ɛ subunits were subcloned into a cytomegalovirus (CMV) promoter-based expression vector, pcDNAIII (Invitrogen, San Diego, CA, USA). The δW57F, δD59N, ɛW55F and ɛG57S mutant clones were made using QuikChange (Stratagene, San Diego, CA, USA). The α subunit differed from the sequence in the GenBank database (accession X03986) by having an alanine rather than a valine at position 433 (Salamone et al. 1999).

The receptors were expressed in HEK 293 cells using transient transfection based on calcium phosphate precipitation. In brief, a total of 3.5 μg of cDNA per 35 mm culture dish in the ratio of 2:1:1:1 (α:β:δ:ɛ) was mixed with 12.5 μl of 2.5 m CaCl2 and dH2O to a final volume of 125 μl. The mixture was then added slowly, without mixing to an equal volume of 2 × Bes-buffered solution. The combined solution was incubated at room temperature for 10 min followed by mixing the contents and another incubation of 15 min. The precipitate was then added to the cells. The cells were incubated at 37 °C with 5 % CO2 for 16-20 h at which time the medium was replaced. The electrophysiological experiments were held at 40-72 h after the start of transfection.

The electrophysiological experiments were performed using the patch clamp technique in the cell-attached configuration (Hamill et al. 1981). The bath solution was Dulbecco's phosphate-buffered saline containing (mm): 137 NaCl, 0.9 CaCl2, 2.7 KCl, 1.5 KH2PO4, 0.5 MgCl2, 6.6 Na2HPO4, pH 7.3. The pipette solution contained (mm): 142 KCl, 1.8 CaCl2, 1.7 MgCl2, 5.4 NaCl, 10 Hepes, pH 7.4. In addition, the pipette solution contained ACh, TMA or carbamylcholine. The patch membrane potential was held at -50 mV based on the combination of the pipette potential and the cell membrane potential estimated from the reversal potential of ionic currents through the nAChR. All experiments were performed at room temperature.

Single-channel currents were amplified with an Axopatch 200B amplifier (Axon Instruments, Union City, CA, USA), digitized at 500 kHz, and saved on a PC hard disk using a Digidata 1322 Series interface (Axon Instruments). For event detection, the data were low-pass filtered at 5-7 kHz and idealized using the program SKM (http://www.qub.buffalo.edu). With the exception of wild-type receptors in the presence of ACh, in all cases the channel closing rate constant was estimated as an inverse of the mean open time duration in the presence of low agonist concentrations (50 μm ACh or 100 μm TMA) when the error due to unresolved channel block is minimal. Otherwise, the analysis was restricted to clusters of channel openings that each reflects the activity of a single AChR (Sakmann et al. 1980).

Clusters were defined as series of openings separated by closed intervals shorter than some critical duration (τcrit). The value of τcrit was established as described previously (Salamone et al. 1999; Akk, 2001). The duration of τcrit varied for data obtained at different agonist concentrations, but was at least five times the duration of the main component in the closed time histograms with the minimal value of 30 ms. For single-channel kinetic analysis, the rate constants for agonist association and dissociation were determined from the analysis of idealized intracluster interval durations using Q-matrix methods. A maximum-likelihood method was employed that incorporated a correction for missed events (Program MIL, http://www.qub.buffalo.edu). Error limits were estimated from the curvature of the likelihood surface at its maximum using the approximation of parabolic shape (Qin et al. 1996).

The following kinetic scheme was used to describe the current interval durations for the wild-type and mutant receptors. A closed, unoccupied receptor (C) binds two agonist molecules (A) to become a doubly liganded, open receptor (A2O): where k+1 and k+2 are the agonist association rate constants, k-1 and k-2 are the agonist dissociation rate constants, β is the channel opening rate constant, and α is the channel closing rate constant. When the agonist binding sites were postulated to have equivalent affinities, then k+1 = 2k+2 and k-2 = 2k-1. In general, the relaxation of the constraint of binding site equivalency did not lead to a statistically significant improvement of the fit. For example, in the ɛW57F mutant receptor, the log-likelihood of the fit was 37 679 when the binding sites were constrained to have equal affinities to ACh, and 37 684 without such constraint. With the homologous mutation in the δ subunit, the log-likelihood was 123 313 with the binding site equivalency constraint, and 123 323 without it. Hence, throughout the analysis, the agonist binding sites were constrained to identical affinity.

In the case of some receptor-agonist combinations, the A2O state was further connected to a closed state corresponding to the short-lived desensitized state observed previously by other investigators (Colquhoun & Sakmann, 1985; Salamone et al. 1999). In the present study its presence was seen in some of the recordings (e.g. wild-type in Fig. 1) while it was absent or masked by the slower activation component in others (e.g. ɛW55F in Fig. 1). Such a closed time component was added to Model 1 in the analysis of wild-type (ACh or TMA), δW57F (ACh or TMA), δD59N (ACh), ɛG57S (ACh) and δD59N + ɛG57S (ACh) receptors. Its mean duration was 2-12 ms for different constructs. No further analysis of this agonist-independent closed time component was performed.

Figure 1. Currents from the wild-type, and δW57F, ɛW55F and δW57F + ɛW55F mutant receptors activated by 200 μm ACh.

Figure 1

Open in a new tab

Single-channel clusters and the first 200 ms at a higher time resolution are shown. Open events are shown downward. Membrane potential is -50 mV. Intracluster closed and open time histograms are given for the patch from which the representative cluster is shown. The continuous line in the histograms is calculated according to the activation rate constants given in Table 2 and Model 1.

Model 1.

Model 1

Open in a new tab

Figure 5. Currents from the δD59N and ɛG57S mutant receptors activated by 5 mm TMA.

Figure 5

Open in a new tab

Single-channel clusters and the first 200 ms at a higher time resolution are shown. Open events are shown downward. Membrane potential is -50 mV. Intracluster closed and open time histograms are given for the patch from which the representative cluster is shown. The continuous line in the histograms is calculated according to the activation rate constants given in Table 2 and Model 1.

The activation rate constants were normally determined from a simultaneous fit of data from several patches obtained at different agonist concentrations. The data were chosen to cover the steep rising phase of the dose-response curve. Data obtained in the presence of saturating concentrations of agonist were not included in the analysis because of its diminished information value as at high agonist concentrations, the receptor rarely enters the unliganded closed state. The channel opening rate constant was usually fixed at a value determined from an independent line of analysis. Two approaches to determine β independently were used.

In most cases, the channel opening rate constant was obtained from the saturation of the effective opening rate (β‘) curve. The effective opening rate is an inverse of a component in the intracluster closed time histograms that scales with agonist concentration. As the agonist concentration is increased, the closed intervals within clusters become briefer, and β’ increases. At saturating agonist concentrations the value for β’ approaches a value set by the channel opening rate constant, β. The curve for β’ vs. [agonist] was fitted using an empirical equation:

graphic file with name tjp0544-0695-m1.jpg (1)

yielding the value for β which was used in the estimation of activation rate constants as described above.

In another approach, the opening rate constant was estimated from the ‘glitch analysis’ of bursts (Sakmann et al. 1980). In such analysis, β, α and k- are estimated by studying the reopening of the ion channel once it closes using program MIL (see above). A simple C ™ AC ™ AO model is used, where β and α are the rate constants governing the transition between the liganded closed and liganded open state, and k- is the rate for dissociation of the agonist. When β >> k-, the receptor keeps reopening, and the bursts contain several openings separated by short glitches whose lifetime is determined by an inverse of the sum of β and k-. If β << k-, once the receptor closes, the ligand molecule dissociates in most cases without the reopening of the channel. The durations of interburst closed times are determined by the affinity of the receptor to the agonist and the concentration of agonist, but also by the number of receptors in the patch. So, to avoid the contamination of intraburst closed events with interburst closed events, only data obtained at relatively low agonist concentrations, and those containing few overlaps (low number of receptors in the patch), were used. In addition, the low agonist concentration helps to reduce channel block, which might also interfere with the analysis.

Double mutant cycle analysis was employed to evaluate additivity of changes in the structure of the receptor or the ligand. The method compares the effects observed upon changes in one parameter (a point mutation in the receptor or a substitution of the ligand) with changes seen upon a pairwise switch (two point mutations in the receptor or a point mutation and ligand substitution). First, a coupling coefficient (Ω) is calculated according to:

graphic file with name tjp0544-0695-m2.jpg (2)

where K represents an equilibrium constant for agonist binding or channel gating, and wild-type and mutant refer to the wild-type and mutant forms of the receptor, respectively. The change in interaction free energy (ΔG) is calculated as ΔG = RT lnΩ. The strength of interaction between sites A and B is estimated based on the value for ΔG. Usually, it is assumed that ΔG values lower than 1-1.5 kcal mol−1 are caused by long-range interactions and should not be interpreted as specific interaction between the two sites (LiCata & Ackers, 1995).

Results

Effects of mutations to sites δ57 and ɛ55 on receptor activation by ACh

The activation rate constants for the mouse adult wild-type receptor have been estimated previously (Akk & Auerbach, 1996; Wang et al. 1997; Salamone et al. 1999). The rate constant estimates used in the present work are given in Table 2. The agonist association rate constant (k+) was estimated as 242 μm s−1, the agonist dissociation rate (k-) was 30 632 s−1 and the channel closing rate constant was 2043 s−1. These estimates were obtained from a simultaneous fit of Model 1 to data from three patches obtained in the presence of 10, 20 and 50 μm ACh (total of 25 699 events). The channel opening rate constant was estimated from fitting the effective opening rate curve as 41 090 s−1. This value is comparable to estimates obtained elsewhere (48 900 s−1, Wang et al. 1997; 48 950 s−1, Salamone et al. 1999).

Table 2.

Activation rate constants of wild type and mutant receptors in the presence of Ach or TMA

Receptor Agonist K+ ( μm−1 s−1) K- (s−1) β(s−1) α(s−1)
Wild-type Ach 242 ± 4 30 632 ± 604 41090 2043 ± 25
δW57F Ach 135 ± 7 30 593 ± 1794 13670 2602 ± 42
ɛW55F Ach 85 ± 12 28 713 ± 4028 3035 1818 ± 179
δW57F + ɛW55F Ach 24 ± 9 10 235 ± 3732 845 2465 ± 124
Wild-type TMA 7.3 ± 0.5 22 877 ± 1439 6089 1850 ± 50
δW57F TMA 1.3 ± 0.2 9752 ± 1392 2258 ± 108 2660 ± 98
ɛW55F TMA — — 609 2189 ± 74
δD59N Ach 89 ± 6 15 308 ± 1308 21200 1852 ± 52
ɛG57S Ach 288 ± 14 34 883 ± 1686 7673 1834 ± 43
δD59N + ɛG57S Ach 188 ± 11 40029 ± 2378 14 500 2622 ± 90
δD59N TMA 13 ± 6 26 576 ± 11 782 1207 1958 ± 233
ɛG57S TMA 4.4 ± 2.8 28498 ± 18 006 2116 2430 ± 395
Open in a new tab

The rate constants and error limits were determined from single-channel kinetic analysis based on Model 1. The opening rate constants were estimated from fitting the effective opening rate curve (see Table 1) with the exception of δW57F and ɛG57S receptors in the presence of TMA. For the δW57F, the β was estimated from the rate constant analysis of two patches obtained in the presence of 5 and 15 mm TMA (total of 14 928 events). For the ɛG57S receptor, the β was estimated from the glitch analysis of data obtained in the presence of 100 μm TMA. The activation rate constants were not estimated for the ɛW55F receptor in the presence of TMA because of the absence of single-channel clusters in the recordings.

The δ subunit has a tryptophan residue in position 57. In the present work, this residue was mutated to phenylalanine and the effect of the mutation on single-channel currents was investigated. The single-channel currents from the mutated receptor are shown in Fig. 1. The presence of a mutation leads to an increase in the intracluster closed time duration and a decrease in the channel open duration. The channel opening rate constant (β) was determined from the saturation with respect to concentration of the effective opening rate (β‘) curve which is presented in Fig. 3. A fit to eqn (1) gave 13 670 s−1 for β (Table 1). This represents an almost 3-fold reduction compared to the wild-type receptor. The estimate for β is confirmed with glitch analysis which yielded 14 391 ± 5657 s−1 for the channel opening rate constant. The effect of the mutation on the agonist association and dissociation rates was determined using Model 1. With the binding sites constrained to equal affinities, the analysis gave: k+ = 135μm−1 s−1 and k- = 30 593 s−1 (KD = 227 μm). Hence, the mutation has a less than 2-fold effect on the ACh binding rate constant with essentially no effect on the dissociation of ACh.

Figure 3. The effective opening rate curves for the wild-type, δW57F, ɛW55F and δW57F+ɛW55F mutant receptors.

Figure 3

Open in a new tab

The effective opening rate curves for the wild-type receptor in the presence of ACh (open circles) or TMA (filled circles), δW57F mutant receptor in the presence of ACh (open squares), TMA (filled squares) or CCh (crossed squares), ɛW55F mutant receptor in the presence of ACh (open triangles) or TMA (filled triangles), and δW57F+ɛW55F double mutant receptor in the presence of ACh (open diamonds). The curves were fitted using eqn (1). For the δW57F receptor activated by TMA, the maximal value for effective opening rate curve was constrained at 2258 s−1. For the ɛW55F receptor activated by TMA, the maximal value for an effective opening rate curve was constrained at 609 s−1. The results of the fit are given in Table 1.

Table 1.

Concentration–response parameters of wild type and mutant receptors in the presence of ACh or TMA

Receptor Agonist β(s−1) EC50 ( μm) n
Wild-type Ach 41090 ± 5580 383 ± 54 1.6 ± 0.03
δW57F Ach 13670 ± 1630 493 ± 95 1.5 ± 0.1
ɛW55F Ach 3035 ± 994 640 ± 322 1.7 ± 0.5
δW57F + ɛW55F Ach 845 ± 345 531 ± 297 2.2 ± 1.0
Wild-type TMA 6089 ± 2954 6759 ± 4605 1.4 ± 0.3
δW57F TMA 2258 27 280 ± 5271 1.2 ± 0.1
ɛW55F TMA 609 68 370 ± 7376 1.2 ± 0.1
δD59N Ach 21 200 ± 2246 445 ± 62 1.7 ± 0.1
δG57S Ach 7673 ± 791 233 ± 43 1.6 ± 0.1
δD59N + ɛG57S Ach 14 500 ± 3645 644 ± 229 1.4 ± 0.1
δD59N TMA 1207 ± 82 3052 ± 295 2.0 ± 0.1
ɛG57S TMA 2116 27 370 ± 7617 0.9 ± 0.1
Open in a new tab

The effective opening rate curves were fitted using eqn (1) on data from Figs 2 and 4. The values are best-fit parameters ± standard deviation estimated from the fit. For the δW57F receptor in the presence of TMA, the channel opening rate constant was estimated directly from the rate constant analysis (Table 2) and was constrained to that value in fitting the effective opening rate curve. For the ɛW55F receptor in the presence of TMA, the channel opening rate constant was calculated by assuming that the mutation affects the β similarly in the presence of ACh or TMA. For the ɛG57S receptor in the presence of TMA, the channel opening rate constant presented in the table was estimated from the glitch analysis of data obtained in the presence of 100 μm TMA.

The equivalent mutation in the ɛ subunit also affects single-channel currents. A sample cluster from the ɛW55F receptor obtained in the presence of 200 μm ACh is shown in Fig. 1. From the saturation of the effective opening rate curve, the channel opening rate constant was estimated to be 3035 s−1 (Fig. 3 and Table 1). This is similar to a β estimate obtained using glitch analysis (2772 ± 2256 s−1). The ACh association and dissociation rates were 85 μm−1 s−1 and 28 713 s−1 (KD = 338 μm), respectively. Thus, the mutation leads to a 3-fold reduction in the agonist binding rate with almost no effect on the dissociation of ACh.

Both mutations affect mainly the channel opening rate constant with only minor effects on receptor affinity to ACh and the channel closing rate constant. To study whether the effects of the two mutations on channel gating are additive, currents from a double mutant, δW57F + ɛW55F, were measured. As might be expected, the double mutant had a channel opening rate constant which was less than that for either of the single mutants. The effective opening rate curve for the double mutant is presented in Fig. 3. The curve saturates at 845 s−1, which corresponds to the double mutant channel opening rate constant. A channel opening rate constant estimated from glitch analysis was 818 ± 859 s−1. The receptor affinity to ACh was affected to a lesser degree, the agonist association rate constant was 24 μm−1 s−1 and dissociation rate constant, 10235 s−1 (KD = 426 μm).

Effects of mutations to sites δ57 and ɛ55 on receptor activation by TMA

Tetramethylammonium is a nicotinic receptor agonist. Compared to ACh, in the presence of TMA, the dose-response curve of the wild-type receptor is shifted toward higher agonist concentrations. Such a shift is caused by a reduction in the channel opening rate constant and a lower affinity of TMA for the receptor. The rate constants for the mouse adult wild-type receptor activated by TMA have been estimated previously for a NaCl-based pipette medium (Akk & Auerbach, 1996). The agonist association rate is dependent on the nature of the medium to which the receptors are exposed (ibid.). So, in the present work, the activation rate constants for TMA-activated wild-type receptors were re-estimated for a KCl-based pipette solution. A simultaneous analysis of three patches obtained at 1, 2 and 5 mm TMA (total of 46210 events) was carried out. The results of such analysis are given in Table 2. The association rate constant for TMA is 7.3 μm−1 s−1 and the dissociation rate is 22 877 s−1. The receptors open with the rate of 6089 s−1 and close at 1850 s−1. The channel opening rate constant was verified using glitch analysis of bursts obtained in the presence of 100 μm TMA. Such approach gave 5359 ± 4022 s−1 for β.

The activation of the δW57F and ɛW55F mutant receptors in the presence of TMA was studied here. Single-channel currents of the wild-type and mutant receptors activated by 5 mm TMA are shown in Fig. 2. For the δW57F receptor, single-channel clusters were observed in the presence of 2-15 mm TMA. However, the effective opening rate curve did not saturate within this agonist concentration range (see Fig. 3). Thus, the channel opening rate constant could not be determined from the effective opening rate curve. However, even without the constraint for β, Model 1 fitted to currents from the δW57F receptor yielded: k+ = 1.1 μm−1 s−1, k- = 8917 s−1 (KD = 8106 μm), β = 2258 s−1 and α = 2713 s−1.

Figure 2. Currents from the wild-type, δW57F and ɛW55F mutant receptors activated by 5 mm TMA.

Figure 2

Open in a new tab

Single-channel clusters and the first 200 ms at a higher time resolution are shown. Open events are shown downward. Membrane potential is -50 mV. Intracluster closed and open time histograms are given for the patch from which the representative cluster is shown. The continuous line in the histograms is calculated according to the activation rate constants given in Table 2 and Model 1. However, for the ɛW55F receptor, the continuous lines are fits to a single exponential with the time constants of 42 ms (closed times) and 0.5 ms (open times).

In the wild-type receptor, carbamylcholine (CCh) and TMA activate the receptor with similar efficacy (Akk & Auerbach, 1996, 1999). However, the dose-response curve for CCh is shifted toward lower agonist concentrations because the receptor has higher affinity to CCh than TMA. Thus, the channel opening rate constant can be more readily determined from the saturation of the effective opening rate for CCh. The effective opening rate curve for the δW57F in the presence of CCh is given in Fig. 3. A fit to eqn (1) gives: β=2418 ± 437 s−1 and EC50 = 1335 ± 314 μm. The value for β is similar to one obtained in the presence of TMA using an unconstrained fit to Model 1 (see above) suggesting that the mutation does not affect the relative gating efficiency of CCh and TMA.

No clusters were detected for the ɛW55F mutant receptor at TMA concentrations up to 10 mm (Fig. 3). While it was not possible to determine the agonist association and dissociation rate constants one could still estimate approximate values for the channel opening and closing rate constants. The channel closing rate constant can be determined as an inverse of the channel mean open duration. In the presence of 100 μm TMA, the channel mean open duration is 0.43 ms. Hence, the estimated α is equal to 2317 s−1. The channel opening rate constant can be estimated by assuming that the mutation causes a similar reduction in channel gating for both ACh and TMA. So, in the presence of ACh, the ɛW55F mutations leads to a 1.49 kcal mol−1 increase in the free energy of gating. Assuming a similar change for TMA, the channel opening rate constant can be calculated as 609 s−1. The main component in the total patch closed time histograms was 42 ms at 5 mm and 19 ms at 10 mm TMA. The EC50 of the β’ curve can then be obtained using the major closed interval durations at 5 and 10 mm TMA and a fixed value for the channel opening rate constant (609 s−1). So, a fit to eqn (1) yields an EC50 of ≈68 mm for the ɛW55F receptor in the presence of TMA.

It should be noted that this estimate is based on two assumptions. First, that the mutation similarly affects activation by ACh and TMA. Second, that the patch closed times in the presence of 5 and 10 mm TMA are identical (or, at least, similar) to the intracluster closed times. Since no clusters were observed at these TMA concentrations, the intrapatch closed times only set a higher limit for β‘.

Effects of mutations to sites δ59 and ɛ57 on receptor activation by ACh

The δ subunit has an aspartate residue in position 59. The functional effect of an aspartate-to-asparagine mutation has been examined here. Figure 4 shows the single-channel currents for the δD59N mutant receptor in the presence of 200 μm ACh. Compared to the data obtained for the wild-type receptor, there is a slight increase in the mean intracluster closed time duration. The β’ curve for the mutant receptor is given in Fig. 6. The channel opening rate constant, estimated from the saturation of the β’ curve, is 21 200 s−1. The ACh association and dissociation rate constants are 89 μm−1 s−1 and 15308 s−1, respectively. Thus, the affinity of the mutant receptor to ACh is essentially unchanged (KD = 172 μm) while the actual rate constants were reduced 2- to 3-fold.

Figure 4. Currents from the δD59N, ɛG57S and δD59N + ɛG57S mutant receptors activated by 200 μm ACh.

Figure 4

Open in a new tab

Single-channel clusters and the first 200 ms at a higher time resolution are shown. Open events are shown downward. Membrane potential is -50 mV. Intracluster closed and open time histograms are given for the patch from which the representative cluster is shown. The continuous line in the histograms is calculated according to the activation rate constants given in Table 2 and Model 1.

Figure 6. The effective opening rate curves for the wild-type, δD59N, ɛG57S and δD59N+ɛG57S mutant receptors.

Figure 6

Open in a new tab

The effective opening rate curves for the wild-type receptor in the presence of ACh (open circles) or TMA (filled circles), δD59N mutant receptor in the presence of ACh (open squares) or TMA (filled squares), ɛG57S mutant receptor in the presence of ACh (open triangles) or TMA (filled triangles), and δD59N + ɛG57S double mutant receptor in the presence of ACh (open diamonds). The curves were fitted using eqn (1). For the ɛG57S receptor activated by TMA, the maximal value for effective opening rate curve was constrained at 2116 s−1. The results of the fit are given in Table 1.

The ɛ subunit has a glycine residue in position 57 which was mutated to serine. A sample cluster for the ɛG57S mutant receptor activated by 200 μm ACh is shown in Fig. 4. The β’ curve is given in Fig. 6. A fit of eqn (1) yielded a channel opening rate constant of 7673 s−1. Thus, the mutation reduces β by almost 5-fold. On the other hand, the mutation had essentially no effect on the agonist association and dissociation rate constants. The k+ was 288 μm−1 s−1 and k- = 34883 s−1 (KD = 121 μm).

The kinetic properties of a receptor containing mutations δD59N and ɛG57S were also studied. The effective opening rate curve for the double mutant is shown in Fig. 6. Surprisingly, the curve saturates at a level which is higher than that for the receptor containing one of the single mutations, ɛG57S receptor. The β for the double mutant was 14 500 s−1 compared to 7673 s−1 for the receptor with only the ɛG57S substitution. The β value was confirmed with glitch analysis which yielded 17 422 s−1 for the channel opening rate constant. The rate constant analysis demonstrated that only slight changes in the agonist binding rates take place in the double mutant receptor. The ACh association rate constant is 188 μm−1 s−1 and the dissociation rate constant is 40 029 s−1 (KD = 213 μm).

Effects of mutations to sites δ59 and ɛ57 on receptor activation by TMA

TMA is a much weaker agonist for the δD59N and ɛG57S receptors. For the δD59N receptor, the effective opening rate curve saturated at 1207 s−1. An estimate for β from glitch analysis was 1179 ± 739 s−1. The channel opening rate constant is reduced by 5-fold compared to the wild-type receptor. On the other hand, the equilibrium dissociation constant was actually reduced by the mutation. The k+ is 13.0 μm−1 s−1 and k- is 26576 s−1 (KD = 2044 μm).

For the ɛG57S receptor, single-channel clusters were observed at 2-10 mm TMA. However, direct determination of the channel opening rate constant was not possible because the β’ curve did not saturate within this concentration range. The lower limit for β was estimated as 667 s−1. This corresponds to the β’ value in the presence of the highest TMA concentration used (15 mm).

Another attempt to estimate β was made using a ‘glitch analysis’ of bursts that had been recorded in the presence of 100 μm TMA. Using such an approach, the channel opening rate constant for the ɛG57S receptor was estimated as 2116 s−1. With the β constrained at 2116 s−1, the receptor affinity to TMA was then estimated as 6477 μm (k+ = 4.4 μm−1 s−1, k- = 28498 s−1).

Discussion

The tryptophan residues in the γ (γW55) and δ (δW57) subunits of the Torpedo nicotinic receptor have been long known to contribute to the structure of the agonist binding site. In fact, based on the covalent incorporation of [3H]nicotine, the γW55 residue was the first site on a non-α subunit that had been identified as part of the binding site (Chiara et al. 1998). Both the γW59 and δW57 residues incorporate a competitive antagonist of the nicotinic receptor, [3H]d-tubocurarine (Chiara & Cohen, 1997). The δD59 residue has been implicated in specific interactions with carbamylcholine (Prince & Sine, 1996). When mutated to cysteine, a residue in the equivalent site of the γ subunit is not accessible to alkylating agents following the addition of d-tubocurarine suggesting that the competitive antagonist of the nicotinic receptor comes into close contact with the γE57 residue (Sullivan et al. 2002). In addition, a mutation of a homologous residue in the neuronal α7 receptor (Q56) has been demonstrated to lead to a reduction in the apparent affinity of the receptor to acetylcholine and nicotine (Corringer et al. 1995).

The non-α subunit 57 and 59 residues are likely to be pointed or exposed to different environments. In the Torpedo nicotinic receptor, the γE57 residue (homologous to ɛG57 or δD59) is accessible to modification by several sulfhydryl modifying agents while the γW55 residue (homologous to ɛW55 or δW57) is not (Sullivan & Cohen, 2000), suggesting that the latter might be part of the hydrophobic core of the agonist binding site. Also, according to the crystal structure of the snail ACh-binding protein, a residue homologous to W57 forms part of the lower half of the binding pocket while the site equivalent to the ɛG57/δD59 residues is part of the wall of the agonist binding site (Brejc et al. 2001). In summary, there is plenty of evidence placing the residues forming loop D in the immediate vicinity of the nicotinic agonist binding site.

In this manuscript, the effects of mutations to residues forming loop D of the mouse nicotinic receptor agonist binding site were examined. The data demonstrate that W-to-F mutations in position 57/55 of the δ and ɛ subunits, and a D-to-N and a G-to-S mutation in position 59/57 of the δ and ɛ subunits, respectively, affect predominantly channel opening with little effect on the affinity of the receptor to ACh or the channel closing rate constant. Similar results were obtained when the receptors were activated by TMA.

Comparison of data from the δW57F and ɛW55F receptors reveals that a mutation in the ɛ subunit has an almost 2-fold greater impact on channel gating. In the presence of ACh, there is an increase of 0.79 kcal mol−1 in gating free energy in the δW57F and 1.45 kcal mol−1 in the ɛW55F receptor. In the presence of TMA, the increase in free energy of gating is 0.84 kcal mol−1 in the δW57F and 1.40 kcal mol−1 in the ɛW55F receptor.

Previous work has indicated that in the Torpedo nicotinic receptor, a leucine substitution in the δ57 site affects the receptor apparent affinity to a lesser degree than a mutation in the homologous site of the γ subunit (Xie & Cohen, 2001). The magnitude of shift in the channel half-maximal response was nearly twice as high following the γW55L mutation compared to the δW57L mutation. However, the dissection of the whole-cell response into the binding and gating components was not possible in these studies, so it is not obvious whether the shift in the dose-response curves was entirely due to changes in the gating process.

The comparison of effects of mutations in the δ59 and ɛ57 sites is somewhat more speculative. The residues in the δ and ɛ subunits differ - there is an aspartate residue in the δ subunit, and a glycine in the ɛ subunit. Hence, the difference in effects observed in δD59N and ɛG57S mutant receptors could be explained by the non-equivalent changes in the properties of residues. Nevertheless, when activated by ACh, the increase in the free energy of gating is 0.33 kcal mol−1 in the δD59N and 0.92 kcal mol−1 in the ɛG57S receptor.

The disproportionate sensitivity of ɛ subunit residues to modifications is in agreement with previous data which demonstrated that an αD200N mutation in the α-ɛ binding site affected channel gating by ≈2-fold more than the same mutation located in the α-δ site (Akk et al. 1996). In these experiments it was found that two non-identical modes of activity result when the receptor contains one mutated and one wild-type α subunit (i.e. hybrid receptors). By comparing the two types of activity in hybrid receptors in adult and embryonic configurations it was concluded that an αD200N mutation in the α-γ or α-ɛ binding site leads to a greater reduction in the channel opening rate constant. In contrast, the effects of αY93F and αW149F mutations were deemed identical in the α-ɛ and α-δ binding sites (Akk, 2001). Here, only a single type of receptor activity was observed for receptors containing one mutated and one wild-type binding site. The effect on gating in the receptor with mutations in both binding sites was roughly double of that in the hybrid receptor. So, it was concluded that the activity from the two hybrid receptors was indistinguishable in terms of kinetic properties.

The presence of specific short-range interaction between two sites can be examined using double mutant cycle analysis (Horovitz & Fersht, 1990). This method evaluates the interaction by examining the additivity of effects caused by mutations or changes in the two potentially interacting sites. In the present work, the mutation of the residues forming loop D was evaluated along with the change of agonist from ACh to TMA. Table 3 gives the estimated interaction free energies for sites δ57, δ59 and ɛ57, and the tailgroup of ACh. The interaction energy for the ɛ55 site could not be computed due to the absence of clusters in the presence of TMA. The calculated ΔG values for all sites are below 1 kcal mol−1, a value commonly accepted as the threshold for potential short-range interaction (LiCata & Ackers, 1995). Thus, the data suggest that there is no specific interaction between the tailgroup of ACh and the residues forming loop D. The functional changes caused by the mutations and substitution of ACh by TMA appear additive.

Table 3.

Estimation of the binding and gating interaction

Site Ωbinding ΔGbinding (kcal mol−1) Ωgating ΔGgating (kcal mol−1)
δW57 1.3 0.2 1.0 0
δD59 0.5 −0.4 0.3 −0.7
ɛG57 2.0 0.4 1.3 0.1
Open in a new tab

The coupling coefficient (Ω) and interaction free energy changes (ΔG) between the tailgroup of ACh and the δ57, δ59 and ɛ57 sites. The Ω and ΔG were calculated separately for the agonist binding process and the channel gating. Ω was calculated as (Kwild type- Ach X Kmutant-TMA)/ (Kmutant-ACH X Kwild-type-TMA) where k refers to the binding equilibrium constant KD (k-/k+) or the gating equilibrium constant Θ(β/α). ΔG was calculated as 0.59 In (Ω).

The additivity of the effects of mutations to channel gating in the two agonist binding sites can be used to evaluate the contributions of the two binding sites to the overall channel gating. To do that, the sum of effects observed in the δW57F (or δD59N) and ɛW55F (or ɛG57S) receptors is compared to the effect seen in the double mutant δW57F + ɛW55F (or δD59N + ɛG57S). If the two are equal, the two binding sites contribute to channel gating independently. For the δ57/ɛ55 site, the effects of individual mutations are 0.79 and 1.45 kcal mol−1 in the δ and ɛ subunits, respectively. In the double mutant, there is 2.40 kcal mol−1 increase in the gating free energy. Hence, the sum of the effects seen in the two individual mutants is within 0.16 kcal mol−1 from the effect of the double mutant receptor. For the δ59/ɛ57 residue, the individual mutations lead to a change of 0.33 and 0.92 kcal mol−1 in the free energy of gating. Compared to the effect measured for the double mutant (0.76 kcal mol−1), this is within -0.49 kcal mol−1. Thus, in both cases the double mutant alters gating just slightly more (δ57/ɛ55 site) or less (δ59/ɛ57 site) than what would be predicted assuming fully independent binding sites. This is in agreement with previous data where the independency of the contributions of the binding sites to channel gating was examined using the αD200N, αY93F and αW149F mutations (Akk et al. 1996; Akk, 2001). In all cases the difference beween the ΔΔG for the double mutant vs. the sum of ΔΔG for the individual mutations was within 1 kcal mol−1.

For all modifications to the agonist binding site, the alterations in the agonist binding were relatively small (within a factor of three) while the effect on the channel closing rate constant was negligible. It has been the impression from the experience of the author that mutations of the putative binding site residues rarely result in significant changes in agonist affinity without a concurrent change in the channel opening rate constant. Even modifications of the aromatic residues αY93, αW149, αY190 and αY198, residues which are believed to interact directly with the ACh molecule (Aylwin & White, 1994; Sine et al. 1994; Zhong et al. 1998), affect the coupling of the ligand binding site to the channel gate to a far greater degree than the resting receptor affinity (Chen et al. 1995; Akk et al. 1999; Akk & Steinbach, 2000; Akk, 2001). The two notable exceptions are the αG153S and αE184Q mutations. The former leads to a > 20-fold reduction in the ACh dissociation rate constant (Sine et al. 1995) while the latter speeds the dissociation of ACh from its binding site (Akk et al. 1999). In both cases, the channel opening rate constant is unaffected by the mutation. On the other hand, mutations to the binding site residues have little, if any, influence on the channel closing rate constant. Thus, the data demonstrate that the opening but not the closing of the ion channel is particularly sensitive to changes in the overall shape and configuration of the binding pocket.

Acknowledgments

I thank Joe Henry Steinbach for comments during the course of the work and Jessie Zhang for molecular biology and tissue culture work. This work was supported by the National Science Foundation.

References

  1. Akk G. Aromatics at the murine nicotinic receptor agonist binding site: mutational analysis of the αY93 and αW149 residues. Journal of Physiology. 2001;535:729–740. doi: 10.1111/j.1469-7793.2001.00729.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Akk G, Auerbach A. Inorganic, monovalent cations compete with agonists for the transmitter binding site of nicotinic acetylcholine receptors. Biophysical Journal. 1996;70:2652–2658. doi: 10.1016/S0006-3495(96)79834-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Akk G, Auerbach A. Activation of muscle nicotinic acetylcholine receptor channels by nicotinic and muscarinic agonists. British Journal of Pharmacology. 1999;128:1467–1476. doi: 10.1038/sj.bjp.0702941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Akk G, Sine S. Binding sites contribute unequally to the gating of mouse nicotinic αD200N acetylcholine receptors. Journal of Physiology. 1996;496:185–196. doi: 10.1113/jphysiol.1996.sp021676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Akk G, Steinbach JH. Structural elements near the C-terminus are responsible for changes in nicotinic receptor gating kinetics following patch excision. Journal of Physiology. 2000;527:405–417. doi: 10.1111/j.1469-7793.2000.t01-2-00405.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Akk G, Zhou M. A mutational analysis of the acetylcholine receptor channel transmitter binding site. Biophysical Journal. 1999;76:207–218. doi: 10.1016/S0006-3495(99)77190-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Arias HR. Localization of agonist and competitive antagonist binding sites on nicotinic acetylcholine receptors. Neurochemistry International. 2000;36:595–645. doi: 10.1016/s0197-0186(99)00154-0. [DOI] [PubMed] [Google Scholar]
  8. Aylwin ML, White MM. Ligand-receptor interactions in the nicotinic acetylcholine receptor probed using multiple substitutions at conserved tyrosines on the α subunit. FEBS Letters. 1994;349:99–103. doi: 10.1016/0014-5793(94)00649-0. [DOI] [PubMed] [Google Scholar]
  9. Brejc K, Van Dijk WJ, Klaassen RV, Schuurmans M, van Der Oost J, Smit AB, Sixma TK. Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors. Nature. 2001;411:269–276. doi: 10.1038/35077011. [DOI] [PubMed] [Google Scholar]
  10. Chen J, Zhang Y, Akk G, Sine S, Auerbach A. Activation kinetics of recombinant mouse nicotinic acetylcholine receptors: mutations of α-subunit tyrosine 190 affect both binding and gating. Biophysical Journal. 1995;69:849–859. doi: 10.1016/S0006-3495(95)79959-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chiara DC, Cohen JB. Identification of amino acids contributing to high and low affinity d-tubocurarine sites in the Torpedo nicotinic acetylcholine receptor. Journal of Biological Chemistry. 1997;272:32940–32950. doi: 10.1074/jbc.272.52.32940. [DOI] [PubMed] [Google Scholar]
  12. Chiara DC, Middleton RE, Cohen JB. Identification of tryptophan 55 as the primary site of [3H]nicotine photoincorporation in the γ-subunit of the Torpedo nicotinic acetylcholine receptor. FEBS Letters. 1998;423:223–226. doi: 10.1016/s0014-5793(98)00093-3. [DOI] [PubMed] [Google Scholar]
  13. Colquhoun D, Sakmann B. Fast events in single-channel currents activated by acetylcholine and its analogues at the frog muscle end-plate. Journal of Physiology. 1985;369:501–557. doi: 10.1113/jphysiol.1985.sp015912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Corringer PJ, Le NN, Changeux JP. Nicotinic receptors at the amino acid level. Annual Review of Pharmacology and Toxicology. 2000;40:431–458. doi: 10.1146/annurev.pharmtox.40.1.431. [DOI] [PubMed] [Google Scholar]
  15. Czajkowski C, Karlin A. Structure of the nicotinic receptor acetylcholine-binding site. Identification of acidic residues in the δ subunit within 0.9 nm of the α subunit-binding site disulfide. Journal of Biological Chemistry. 1995;270:3160–3164. doi: 10.1074/jbc.270.7.3160. [DOI] [PubMed] [Google Scholar]
  16. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Archiv. 1981;391:85–100. doi: 10.1007/BF00656997. [DOI] [PubMed] [Google Scholar]
  17. Horovitz A, Fersht AR. Strategy for analysing the co-operativity of intramolecular interactions in peptides and proteins. Journal of Molecular Biology. 1990;214:613–617. doi: 10.1016/0022-2836(90)90275-Q. [DOI] [PubMed] [Google Scholar]
  18. Licata VJ, Ackers GK. Long-range, small magnitude nonadditivity of mutational effects in proteins. Biochemistry. 1995;34:3133–3139. doi: 10.1021/bi00010a001. [DOI] [PubMed] [Google Scholar]
  19. Prince RJ, Sine SM. Molecular dissection of subunit interfaces in the acetylcholine receptor. Identification of residues that determine agonist selectivity. Journal of Biological Chemistry. 1996;271:25770–25777. doi: 10.1074/jbc.271.42.25770. [DOI] [PubMed] [Google Scholar]
  20. Qin F, Auerbach A. Estimating single-channel kinetic parameters from idealized patch-clamp data containing missed events. Biophysical Journal. 1996;70:264–280. doi: 10.1016/S0006-3495(96)79568-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Sakmann B, Patlak J. Single acetylcholine-activated channels show burst-kinetics in presence of desensitizing concentrations of agonist. Nature. 1980;286:71–73. doi: 10.1038/286071a0. [DOI] [PubMed] [Google Scholar]
  22. Salamone FN, Zhou M, Auerbach A. A re-examination of adult mouse nicotinic acetylcholine receptor channel activation kinetics. Journal of Physiology. 1999;516:315–330. doi: 10.1111/j.1469-7793.1999.0315v.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Sine SM, Ohno K, Bouzat C, Auerbach A, Milone M, Pruitt JN, Engel AG. Mutation of the acetylcholine receptor α subunit causes a slow-channel myasthenic syndrome by enhancing agonist binding activity. Neuron. 1995;15:229–239. doi: 10.1016/0896-6273(95)90080-2. [DOI] [PubMed] [Google Scholar]
  24. Sine SM, Quiram P, Papanikolau F, Kreienkamp HJ, Taylor P. Conserved tyrosines in the α subunit of the nicotinic acetylcholine receptor stabilize quaternary ammonium groups of agonists and curariform antagonists. Journal of Biological Chemistry. 1994;269:8808–8816. [PubMed] [Google Scholar]
  25. Sullivan D, Chiara DC, Cohen JB. Mapping the agonist binding site of the nicotinic acetylcholine receptor by cysteine scanning mutagenesis: antagonist footprint and secondary structure prediction. Molecular Pharmacology. 2002;61:463–472. doi: 10.1124/mol.61.2.463. [DOI] [PubMed] [Google Scholar]
  26. Sullivan D, Cohen JB. Mapping the agonist binding site of the nicotinic acetylcholine receptor. Orientation requirements for activation by covalent agonist. Journal of Biological Chemistry. 2000;275:12651–12660. doi: 10.1074/jbc.275.17.12651. [DOI] [PubMed] [Google Scholar]
  27. Sussman JL, Harel M, Frolow F, Oefner C, Goldman A, Toker L, Silman I. Atomic structure of acetylcholinesterase from Torpedo californica: A prototypic acetylcholine-binding protein. Science. 1991;253:872–879. doi: 10.1126/science.1678899. [DOI] [PubMed] [Google Scholar]
  28. Wang HL, Auerbach A, Bren N, Ohno K, Engel AG, Sine SM. Mutation in the M1 domain of the acetylcholine receptor alpha subunit decreases the rate of agonist dissociation. Journal of General Physiology. 1997;109:757–766. doi: 10.1085/jgp.109.6.757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Xie Y, Cohen JB. Contributions of Torpedo nicotinic acetylcholine receptor γTrp-55 and δTrp57 to agonist and competitive antagonist function. Journal of Biological Chemistry. 2001;276:2417–2426. doi: 10.1074/jbc.M009085200. [DOI] [PubMed] [Google Scholar]
  30. Zhong W, Gallivan JP, Zhang Y, Li L, Lester HA, Dougherty DA. From ab initio quantum mechanics to molecular neurobiology: a cation-π binding site in the nicotinic receptor. Proceedings of the National Academy of Sciences of the USA. 1998;95:12088–12093. doi: 10.1073/pnas.95.21.12088. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES