Subunit-specific contribution to agonist binding and channel gating re
http://www.100md.com
《大脑学杂志》
1 Department of Neurology and Neuromuscular Research Laboratory
2 Department of Physiology and Biophysics and Receptor Biology Laboratory, Mayo Clinic, Rochester, MN, USA
Summary
We trace the cause of congenital myasthenic syndromes in two patients to mutations in the subunit of the muscle acetylcholine receptor (AChR). Both patients harbour deletion of an asparagine residue in the subunit (N436del) at the C-terminus of the cytoplasmic loop linking the third (M3) and fourth (M4) transmembrane domains. The presence of a null mutation in the second allele of the subunit shows that N346del determines the phenotype. Endplate studies show markedly reduced expression of the N346del-AChR and compensatory accumulation of fetal -AChR. Expression studies in HEK cells reveal decreased expression of N436del-AChR and abnormally brief channel openings. Thus, neuromuscular transmission is compromised by AChR deficiency, fast channel kinetics of the N346del-AChR and incomplete phenotypic rescue by -AChR. Single-channel kinetic analysis shows that the N436del shortens channel openings by reducing stability of the diliganded receptor: rates of channel closing and of ACh dissociation are increased and the rate of channel opening is decreased. In addition to shortening the M3–M4 loop, N436del shifts a negatively charged aspartic acid residue adjacent to M4; the effects of N436del are shown to result from shortening of the M3–M4 loop and not from juxtaposition of a negative charge to M4. To determine whether the consequences of N346del are subunit-specific, we deleted residues that align with N436 in , and subunits. Each deletion mutant reduces AChR expression, but whereas the and mutants curtail channel open duration, the mutant strikingly prolongs open duration. Kinetic analysis reveals that the mutant increases the stability of the diliganded receptor: rates of channel closing and of ACh dissociation are decreased and the rate of channel opening is increased. The overall studies reveal subunit asymmetry in the contributions of the M3–M4 loops in optimizing AChR activation through allosteric links to the channel and the agonist binding site.
Key Words: acetylcholine receptor; congenital myasthenic syndrome; M3–M4 loop; mutagenesis; single-channel patch-clamp recordings
Abbreviations: ACh = acetylcholine; AChR = acetylcholine receptor; -bgt = -bungarotoxin; CMS = congenital myasthenic syndrome; EP = endplate; HEK = human embryonic kidney; M = transmembrane domain; MEPC = miniature endplate current
Introduction
The nicotinic acetylcholine receptor (AChR) at the motor endplate (EP) is a heteropentamer of homologous subunits with stoichiometries 2 for the adult-type receptor and 2 for the fetal type. Each subunit contains four transmembrane domains and short (M1–M2) and long (M3–M4) cytoplasmic loops. The M3–M4 loops of the subunits constitute most of the cytoplasmic mass of AChR (Popot and Changeux, 1984) and harbour three predicted amphipathic helices (Le Novere et al., 1999), one of which borders M4. The M3–M4 loops regulate the flow of ions through the channel (Miyazawa et al., 1999) and affect the rate of channel closing; structural differences between the loops of the and subunits are major determinants of the change from fetal to adult AChR kinetics (Bouzat et al., 1994). The M3–M4 loops also interact with rapsyn to cluster AChR at the EP, but the residues that bind the receptor to rapsyn have not been determined (Gensler et al., 2001; Huebsch and Maimone, 2003; Maimone and Merlie, 1993; Yu and Hall, 1994).
Congenital myasthenic syndromes (CMS) are heterogeneous disorders caused by defects in presynaptic, synaptic basal lamina or postsynaptic gene products (Engel et al., 2003). Most CMS are postsynaptic, and most postsynaptic CMS are caused by mutations in AChR subunits (Engel et al., 2003). To date, six missense mutations have been observed in the M3–M4 loops. An in-frame duplication of residues 413–418 (1254ins18) (Milone et al., 1998) and a missense mutation (A411P) (Wang et al., 2000), both in the amphipathic helix of the subunit, corrupt the fidelity of gating and result in irregular channel kinetics. A three-codon deletion of residues 426–428 of the subunit (426EQEdel) disrupts a specific interaction between the and subunits and impairs AChR assembly (Quiram et al., 1999). Three missense mutations [R311W (Ohno et al., 1997), P331L (Croxen et al., 2001) and V402F (Milone et al., 1999)] reduce surface expression of AChR. Additionally, R311W mildly shortens and V402F modestly prolongs channel opening events.
Here we trace the cause of a myasthenic syndrome in two patients to two heteroallelic mutations in the acetylcholine receptor (AChR) subunit: deletion of the C-terminal residue of the M3–M4 cytoplasmic loop (N436del) plus a null mutation in the second allele. When N436del-AChR and corresponding deletion mutants of other AChR subunits are expressed in human embryonic kidney (HEK) cells, each mutant reduces AChR expression, but whereas the , and deletion mutants decrease, the deletion mutant markedly increases the duration of channel opening episodes. Kinetic analysis reveals that the deletion mutant has decreased ACh affinity for the diliganded closed state and impaired gating efficiency, whereas the deletion mutant markedly enhances ACh affinity and gating efficiency. Thus, the presence of the C-terminal residue of each M3–M4 loop is essential for normal expression, and loops from the different subunits contribute in an asymmetrical manner to optimize activation of AChR.
Methods
Muscle specimens
Intercostal muscle specimens were obtained intact from origin to insertion from patients and control subjects without muscle disease undergoing thoracic surgery. All human studies were in accord with the guidelines of the Institutional Review Board of the Mayo Clinic.
AChR and acetylcholinesterase were detected in cryostat sections by two-colour fluorescence (Hutchinson et al., 1993). Endplates (EPs) were localized for electron microscopy and analysed by the established methods (Engel 1994a,b). Peroxidase-labelled -bungarotoxin (-bgt) was used for the ultrastructural localization of AChR (Engel et al., 1977). The number of AChRs per EP was measured with [125I]-bgt (Engel et al., 1993).
Electrophysiology of muscle specimens
Recordings of miniature EP currents and estimates of the number of transmitter quanta released by nerve impulse were carried out as described elsewhere (Engel et al., 1993; Uchitel et al., 1993). Single-channel patch-clamp recordings from EP AChR were performed in the cell-attached mode as previously described (Milone et al., 1994).
Mutation analysis
We directly sequenced the AChR subunit gene using genomic DNA (Ohno et al., 1996). For family analysis, we traced the IVS9-1GC mutation by EcoNI and the 911delT mutation by Eco72I restriction analysis of PCR products. The N436del mutations was traced with allele-specific PCR in family members and in 200 normal alleles of 100 unrelated controls.
Construction and expression of wild-type and mutant AChRs
Sources of human , , and subunit cDNAs were as previously described (Luther et al., 1989; Ohno et al., 1996; Schoepfer et al., 1988). All four cDNAs were subcloned into the CMV-based expression vector pRBG4 (Sine, 1993) for expression in human embryonic kidney fibroblast (293 HEK) cells. The artificial mutations were engineered into wild-type AChR subunit cDNAs in pRBG4 using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). The presence of each mutation and the absence of unwanted mutations was confirmed by sequencing the entire inserts. HEK cells were transfected with a total of 7.2 μg of plasmids, comprising pRBG4-, -, -, - and pEGFP-N1 in a ratio of 2 : 1 : 1 : 1 : 1 per 35 mm dish using the calcium phosphate precipitation method (Bouzat et al., 1994; Ohno et al., 1996), or a total of 2 μg of plasmids using 6 μl of the FuGene6 transfection reagent (Roche). For patch-clamp studies, we enhanced expression of DN435del-, H408del-, R446del-, R450del-, -omitted and -omitted AChRs by adding 1 μg of the pAdVAntage plasmid (Promega) per 10 μg of total AChR subunit cDNAs.
Bungarotoxin binding measurements
The total number of [125I]-bgt sites on the surface of transfected human embryonic kidney (HEK) cells and ACh competition against the initial rate of [125I]-bgt binding were determined as described elsewhere (Ohno et al., 1996). ACh competition measurements were analysed using the monophasic Hill equation for wild-type and N436del-AChR (Equation 1) or the two-binding-site equation for -omitted AChR (Equation 2):
where Y is fractional occupancy by ACh, n is the Hill coefficient, KOV is an overall dissociation constant, KA and KB are the dissociation constants for the two binding sites, and fractA is the fraction of sites with dissociation constant KA.
Patch-clamp recordings from AChRs expressed in HEK cells
Recordings were obtained in the cell-attached configuration at a membrane potential of –80 mV, at 22°C, and with bath and pipette solutions containing (mM): KCl 142, NaCl 5.4, CaCl2 1.8, MgCl2 1.7, HEPES 10, pH 7.4 (Bouzat et al., 1994; Ohno et al., 1996). Single-channel currents were recorded using an Axopatch 200A amplifier (Axon Instruments) at a bandwidth of 50 kHz, digitized at 5 μs intervals using a Digidata 1200A (Axon Instruments), and recorded to hard disk using the program Clampex 8 (Axon Instruments). Records were analysed at a uniform bandwidth of 11.7 kHz with TACx4.0.9 software (Bruxton). Dwell-time histograms were plotted on a logarithmic abscissa and fitted to the sum of exponentials by maximum likelihood (Sigworth and Sine, 1987).
To estimate rate constants underlying AChR activation, we employed desensitizing concentrations of ACh that cause events from a single channel to cluster into identifiable activation episodes (Qin et al., 1996). Clusters were identified as a series of closely spaced openings preceded and followed by closed intervals greater than a defined critical time. The critical time was determined by a method that misclassifies an equal number of events between two adjacent closed-time components (Colquhoun and Sakmann, 1985). For each receptor, the critical time that provided the best fit for the closed time histogram was chosen for the final analysis. Clusters with fewer than five openings were excluded from analysis. Individual clusters were examined for homogeneity by determining the mean open probability and open duration for each cluster, and clusters within two standard deviations of the means were accepted for further analysis (Qin et al., 1996; Shen et al., 2003). The resulting open and closed intervals were analysed according to kinetic schemes of receptor activation using the program MIL, which uses an interval-based maximum likelihood method that also corrects for missed events (Qin et al., 1996). A dead time of 23 μs was imposed on all recordings. For each type of AChR, single-channel dwell times obtained at a range of ACh concentrations were fitted simultaneously. Data for wild-type AChR were obtained at 10, 20, 30, 50, 70, 100, 200 and 300 μM ACh, for N436del-AChR at 20, 30, 50, 70, 100, 200 and 300 μM ACh, and for H408del-AChR at 0.3, 1, 3 and 10 μM ACh. An average of 8900 events were analysed for each ACh concentration with the range from 1867 to 15 500. The final set of rate constants were checked by superimposing probability density functions calculated from the rate constants on the experimental dwell time histograms, and by their ability to predict burst length at low ACh concentrations (Colquhoun and Sigworth, 1995; Colquhoun and Hawkes, 1995).
Results
Characteristics of CMS patients
Patient 1, a 25-year-old woman (Fig. 1A), and patient 2, a 12-year-old-girl, had severe myasthenic symptoms since birth, a decremental compound muscle action potential response on repetitive stimulation of motor nerves at 2 Hz, and negative tests for anti-AChR antibodies. Both patients responded partially to pyridostigmine and improved further with the additional use of 3,4-diaminopyridine. The parents are not consanguineous, and there are no similarly affected relatives.
Endplate studies
The configuration of the EPs, evaluated from the cytochemical reaction for acetylcholinesterase on teased muscle fibres, was abnormal, with an increased number of small EP regions (1–10 in patient 1 and 2–8 in patient 2) distributed over an increased span of the muscle fibre surface. The number of [125I]-bgt binding sites per EP was 10% of normal (Table 1). Electron microscopy examination of 16 EPs in patient 1 (Fig. 1B and C) and 24 EPs in patient 2 showed a decreased density and restricted distribution of AChR on the junctional folds. The integrity of the junctional folds and nerve terminals was preserved but some postsynaptic regions were simplified (Fig. 1B). Quantal release by nerve impulse was higher than normal (Table 1), probably as an adaptive response to decreased postsynaptic sensitivity to ACh (Plomp et al., 1992, 1995). The miniature EP current (MEPC) amplitude was reduced to 34% of normal in patient 1 (Fig. 1A) and to 44% of normal in patient 2. In patient 1, most MEPCs observed at 17 EPs decayed abnormally slowly and were best fitted by a single exponential, but a small proportion of the EPs (11%) at 15 of the 17 EPs was best fitted by two exponentials, with one component shorter and one component three times longer than normal (Table 1 and Fig. 2A). In patient 2, MEPCs recorded from all 13 EPs decayed biexponentially, with one component shorter and the other twice longer than normal (Table 1).
Single-channel recordings from EPs of patient 2 showed that most channels opened to a low conductance (40 pS) and had long burst open durations characteristic of fetal-type -AChRs (Fig. 2B), but a small proportion (6%) of channels opened to the 60 pS conductance of adult-type -AChRs and had shorter than normal burst durations (Table 1). To summarize, EP studies revealed AChR deficiency, expression of fetal -AChR, and abnormally brief activation episodes of the expressed adult -AChR.
Mutation analysis
To examine the genetic basis of the observed morphological and physiological abnormalities, we directly sequenced the AChR subunit gene and detected three mutations in the two patients (Fig. 3D). Both patients have a 3-bp deletion (1306delAAC) in exon 12 that predicts deletion of an asparagine residue at the C-terminus of the M3–M4 loop of the subunit (N436del). The deleted asparagine at codon 436 is conserved in the mouse and rat but not in the Xenopus or bovine subunit or in other human subunits (Fig. 3A). N436del was not observed in 200 normal alleles.
The second mutation in patient 1 is a previously reported splice site mutation at the 3' end of intron 9 (IVS9-1GC) that alters the canonical AG to AC at the splice acceptor site (Fig. 3D). The mutation causes retention of intron 9, predicting 67 missense amino acids followed by a stop codon. The genetically engineered aberrantly spliced transcript is not expressed on the surface of HEK cells (Ohno et al., 2003). The second mutation in patient 2 is a previously reported frameshifting null mutation in the M3 domain of the subunit (911delT) (Brengman et al., 2000; Sieb et al., 2000) (Fig. 3D).
Family analysis in both patients indicates that the observed mutations are heteroallelic and recessive (Fig. 3E and F). Because IVS9-1GC and 911delT are null mutations, N436del determines the phenotype of both patients.
Expression studies of N436del-AChR expressed in HEK cells
To determine whether N436del hinders the amount of AChR expressed on the cell surface, we engineered N436del into the human subunit and coexpressed it with complementary wild-type , and subunits in HEK cells. As a control, we coexpressed , and subunits in the absence of the subunit. Measurement of [125I]-bgt binding revealed N436del-AChR reduced surface expression to 50% of wild-type, while that of -omitted 22-AChR was 38% of wild-type (Fig. 4A).
To distinguish between lack of incorporation and reduced expression of the mutant subunit, we measured ACh binding at steady state by competition against the initial rate of [125I]-bgt binding (Sine and Taylor, 1979). Wild-type 2 pentamers bind ACh in a monophasic manner, whereas -omitted 22 pentamers bind ACh in a biphasic manner (Fig. 4B) (Ohno et al., 1996). N436del-AChR binds ACh in a monophasic manner like wild-type AChR (Fig. 4B), indicating that the mutant subunit incorporates into most if not all cell surface pentamers. The apparent dissociation constant of N436del-AChR for ACh was very similar to that of wild-type AChR (Fig. 4B).
To examine kinetic effects of N436del, we recorded single-channel currents from human embryonic kidney (HEK) cells expressing N436del-AChR or wild-type AChR activated by a low concentration of ACh (50 nM). Open interval and burst duration histograms of both wild-type and mutant AChRs showed three components, presumably corresponding to two brief mono-liganded open states and one long diliganded open state. Mean durations of diliganded openings are reduced to 46% of wild type by N436del, and those of the corresponding bursts are reduced to 38% (Table 2 and Fig. 5A and B).
The effects of N436del are caused by shortening of the M3–M4 loop
The N436del mutation shortens the M3–M4 loop and at the same time shifts a negatively charged aspartic acid residue to the N-terminal end of M4. To determine whether either or both effects reduce expression or alter activation of AChR, we engineered four site-directed mutants (Fig. 3B) and expressed each together with complementary wild-type subunits in HEK cells.
The first engineered mutant, D435del, shortens the M3–M4 loop by a single residue, without shifting a negative charge against M4 (Fig. 3B). This mutation reduces surface expression of AChR to 56% of wild-type (Fig. 4A) and decreases the mean durations of the major components of channel openings and bursts (Table 2 and Fig. 5C) to 65 and 50% of wild-type, respectively. The second engineered mutant, DN435del, which removes two residues from the M3–M4 loop (Fig. 3B), reduces AChR expression to 10% of wild-type, and decreases the mean durations of the longest components of channel openings and bursts to 65 and 55% of wild-type, respectively (Table 2 and Fig. 5F). The third and fourth engineered mutants, N436R and N436D, do not shorten the M3–M4 loop but position a positive or negative charge next to M4 (Fig. 3B). Neither mutant has an appreciable effect on AChR surface expression (Fig. 4A) or channel kinetics (Table 2 and Fig. 5D and E). Thus, the effects of N436del on AChR expression and channel kinetics can be attributed to shortening of the M3–M4 loop and not to shift of a negative charge adjacent to M4.
Deletion of equivalent residues in non- subunits
To determine whether the effects of deleting a C-terminal residue of the M3–M4 loop are subunit specific, we constructed corresponding deletion mutants of the (H408del), (R446del) and (R450del) subunits (Fig. 3C). The and deletion mutants reduce surface expression of AChR to 5% of wild-type (Fig. 4A) and shorten channel opening events by about the same amount as N436del (Table 2 and Fig. 5H and I). The H408del mutation reduces expression of AChR to 20% of wild-type (Fig. 4A) and, in striking contrast to the and deletion mutants, prolongs the dominant component of open intervals 3.5-fold and that of bursts 13.5-fold (Table 2 and Fig. 5G).
Because HEK cells bound similar low levels of [125I]-bgt after transfection with R450del-AChR, 22-AChR, R446del-AChR and 22-AChR (Fig. 4A), the channel events recorded from the transfected HEK cells could have arisen from -omitted or -omitted receptors rather than from R450del-AChR or R446del-AChR. To test this possibility, we expressed either - or -omitted AChRs in HEK cells in each of three different experiments and searched for channel openings in 30 and 31 patches, respectively, but detected no channel openings. Therefore, it is unlikely that either - or -omitted AChRs, if present, are functional.
Activation kinetics of N436del-AChR and H408del-AChR
To determine the mechanism by which the N436del receptor shortens and the 408Hdel receptor prolongs channel opening events, we examined their kinetics of activation at desensitizing concentrations of ACh (see Methods). Wild-type and N436del-AChRs generated well-defined clusters of openings at ACh concentrations as low as 10 and 20 μM, respectively, but H408del-AChR produced clusters of openings even at 0.3 μM ACh, indicating an enhanced propensity of receptors containing the mutant subunit to become desensitized (Fig. 6, left column).
For both wild-type and N436del-AChRs, the longest closed-time component shifted to shorter durations with increasing ACh concentration (Fig. 6, central column). The open times of the N436del receptors were briefer and those of the H408del receptor longer compared with wild-type (Fig. 6, right column). Dwell times for wild-type and mutant receptors show typical dependence on ACh concentration: closings become briefer with increasing ACh concentration, and the major component of openings changes little across ACh concentrations.
To determine the consequences of the mutations on rate constants underlying receptor activation, we analysed the global set of open and closed dwell times according to scheme 1:
In this scheme, two agonists (A) bind to the receptor (R) with association rate constants k+1 and k+2, and dissociate with rate constants k–1 and k–2. Receptors occupied by one agonist open with rate 1 and close with rate 1, while receptors occupied by two agonists open with rate 2 and close with rate 2. Asterisks indicate open states and RB indicates the blocked state of the receptor. At high ACh concentrations, ACh blocks the open channel with rate k+b, and the channel unblocks with rate k–b. The fitted rate constants allow calculation of the equilibrium dissociation constants (Kn=k–n/k+n, KB=k–b/k+b) and the channel gating equilibrium constants (n=n/n). This scheme allows for only two open states, whereas low-concentration recordings (Fig. 5 and Table 2) had revealed three open states. However, at the high concentrations of ACh required to elicit clusters of channel events due to a single channel, only two components of openings are observed, presumably because the briefest class of monoliganded openings occurs too infrequently to be identified.
Scheme 1 provided a good description of the closed and open intervals for wild-type and N436del receptors but did not provide a satisfactory description of the closed intervals for the H408del mutant. The chief problem was complexity of the closed duration distribution, which exhibited at least one more component than accounted for by scheme 1. We therefore examined alternative scheme 2, which incorporated an additional closed state branching from the diliganded open state:
An analogous brief closed state was suggested by Elenes and Auerbach (2002) for wild-type mouse muscle AChR, and by Hatton and colleagues in describing the activation of the L221F slow-channel mutation (Hatton et al., 2003), both groups suggesting that it corresponded to a short-lived desensitized state. Schemes 1 and 2 provided equally satisfactory fits and similar rate constants for the human wild-type and N436del receptors. To compare activation kinetics of wild type and mutant receptors, we analysed our data according to scheme 2 (Table 3).
The fitted rate constants in Table 3 indicate the following functional consequences of the N346del mutant: by decreasing k+2 and increasing k–2, it increases K2 2.2-fold, thus decreasing the affinity of the diliganded closed receptor for ACh; and by decreasing 2 and increasing 2, it decreases the gating efficiency of the diliganded receptor (2) 2.3-fold. The fitted rate constants were used to predict burst duration at very low concentrations of agonist, which agreed very well with the independently determined burst duration obtained at 50 nM ACh (Table 3). Overall, the N346del burst duration is reduced to 40% of wild-type, and the mutant receptor has mild fast-channel properties (Engel et al., 2003).
The H408del mutation, on the other hand, has opposite effects to N346del: by increasing k+1 and k+2 and decreasing k–1 and k–2, it decreases K1 3-fold and K2 17-fold, thus markedly enhancing both monoliganded and diliganded closed state affinities; by increasing 2 1.5-fold and slowing 2 3.0-fold, it increases the gating efficiency of the diliganded receptor 5-fold. Again, burst duration predicted by the fitted rate constants agreed well with the independently determined burst duration obtained at 50 nM ACh (Table 3). Burst duration increases 14-fold, and the H408del receptor has pronounced slow-channel properties (Engel et al., 2003).
Figure 7 shows plots of open probabilities (Popen) within defined clusters of channel events at different concentrations of ACh for the wild-type, N436del-AChR and H408del receptors. The plotted points are well described by the theoretical Popen curve computed from the rate constants determined by fitting scheme 2 to the dwell times (Fig. 7). The Popen curve for N436del receptor is right-shifted with respect to wild-type, indicating a 3.5-fold decreased probability of opening at the EC50. In contrast, the Popen curve for the H408del receptor is left-shifted with respect to wild-type, indicating a 31-fold increased probability of opening at EC50.
Discussion
Phenotypic consequences of N436del
Both CMS patients carry the N436del mutation plus a null mutation in the second allele; therefore, N346del determines the phenotype. EP studies demonstrate severe AChR deficiency, compensatory expression of fetal-type -AChR, and short opening events of the expressed adult -AChRs. Expression of -AChR at the EP has been documented with low-expressor or null mutations of the subunit, where it probably serves as a means of phenotypic rescue (Engel et al., 2003). Because the number of AChRs at both patients' EPs is only 10% of normal and most expressed AChRs harbour the fetal subunit, neuromuscular transmission is primarily compromised by the AChR deficiency, and this is compounded by the fast-channel kinetics of the N436del-AChRs.
Expression of N436del-AChR in HEK cells was 50% of wild-type. Higher AChR expression in HEK cells than at the EP has been observed previously with other low-expressor missense mutations of AChR (Milone et al., 1998; Ohno et al., 1997; Shen et al., 2002; Wang et al., 1999). This may be due to differences in the rate of synthesis or destruction of the mutant receptor in muscle fibres versus HEK cells. Interaction with the cytoplasmic anchoring protein rapsyn, absent in HEK cells, may also be compromised by the mutant receptors at the EP.
Expression of AChRs harbouring C-terminal deletion mutants of M3–M4 loops
All three C-terminal deletion mutants of the M3–M4 loop (N436del, D435del and DN435del) as well as the corresponding deletion mutants of the (408Hdel), (R446del) and (R450del) subunits curtail surface expression of AChR, whereas the two C-terminal charge mutants of the M3–M4 loop (N436D or N436R) do not; therefore the decreased AChR expression probably stems from shortening of the M3–M4 loop. Decreased AChR expression, in turn, may stem from abnormal folding and accelerated destruction of nascent peptides in the endoplasmic reticulum, or because C-terminal residues of the M3–M4 loops are required for efficient subunit assembly, or both.
Activation kinetics of C-terminal deletion mutants of the M3–M4 loops
Scheme 1 described well the kinetics of activation the wild-type and N436del receptors, but did not account for all closed states of the H408del receptor. We therefore explored an alternative scheme (scheme 2) that allowed an additional closed state connected to the open state of the receptor (A2RG); scheme 2 described both closed and open times of the H408del receptor. Interestingly, applying scheme 2 to the wild-type and the N436del receptors did not significantly alter the rate constants of activation or improve the likelihood of the fitted parameters compared to scheme 1. This finding differs from that in the mouse receptor, where allowing a non-conducting gap to arise from the diliganded open state improved the quality of the fit (Elenes and Auerbach, 2002). The functional significance of the A2RG state is not known; it may represent a brief desensitized state of the receptor (Salamone et al., 1999), but that the dwell in the A2RG state is 13 times shorter for the desensitization prone H408del receptor than for the wild-type or the N436del receptor appears inconsistent with this interpretation.
The results shed new light on the structure–function relationships of the AChR. In particular, the M3–M4 linker has been shown to be a primary determinant of the fetal-to-adult kinetic switch, conferred by change from to subunits (Bouzat et al., 1994), as well as a determinant of the fidelity of channel gating (Milone et al., 1998; Wang et al., 2000). The collective studies point to and as subunits that specifically contribute to activation kinetics, but the present work is the first to show a contribution of the M3–M4 loop of the subunit. The single residue deletion from the subunit prolongs channel activation episodes and enhances desensitization to extents comparable to severe slow channel mutations. The enhanced activation is attributed to enhanced affinity of ACh for the resting closed state of the receptor as well as to enhanced gating. Both the ACh binding site (Celie et al., 2004) and the channel gate (Miyazawa et al., 2003) are distant from the M3–M4 linker, indicating an allosteric contribution of the M3–M4 loop of the subunit to receptor function. Thus, combined clinical, morphological, electrophysiological and genetic studies of our two patients unravelled the pathophysiological basis of a CMS. Further experimentation revealed subunit-specific contributions of C-terminal residues of the M3–M4 loops of AChR that allosterically affect gating of the ion channel and ACh occupancy of the more distant binding site.
Acknowledgements
This work was supported by NIH grants to A. G. E. (NS-6277) and to S. M. S. (NS-31744) and by a Muscular Dystrophy Association Grant to A. G. E.
References
Bouzat C, Bren N, Sine SM. Structural basis of different gating kinetics of fetal and adult acetylcholine receptors. Neuron 1994; 13: 1395–402.
Brengman JM, Ohno K, Milone M, Friedman RL, Feldman RG, Engel AG. Identification and functional characterization of eight novel acetylcholine receptor mutations in six congenital myasthenic syndrome kinships. Neurology 2000; 54 (Suppl. 3): A182–3.
Celie PH, van Rossum-Fikkert SE, van Dijk WJ, Brejc K, Smit AB, Sixma TK. Nicotine and carbamycholine binding to nicotinic acetylcholine receptors as studied in AChBP crystal structures. Neuron 2004; 41: 907–14.
Colquhoun D, Hawkes AG. A Q-matrix cookbook: how to write only one program to calculate the single-channel and macroscopic predictions for any kinetic mechanism. In: Sakmann B, Neher E, editors. Single-channel recording. New York: Plenum Press; 1995. p. 589–633.
Colquhoun D, Sakmann B. Fast events in single channel currents activated by acetylcholine and its analogues at the frog muscle end-plate. J Physiol (Lond) 1985; 369: 501–57.
Colquhoun D, Sigworth FJ. Fitting and statistical analysis of single-channel records. In: Sakmann B, Neher E, editors. Single-channel recording. New York: Plenum Press; 1995. p. 483–587.
Croxen R, Young C, Slater C, Haslam S, Brydson M, Vincent A, et al. Endplate - and subunit mRNA levels in AChR deficiency syndrome due to subunit null mutations. Brain 2001; 124: 1362–72.
Elenes S, Auerbach A. Desensitization of diliganded mouse muscle nicotinic acetylcholine receptor channels. J Physiol (Lond) 2002; 541: 367–83.
Engel AG. Quantitative morphological studies of muscle. In: Engel AG, Franzini-Armstrong C, editors. Myology. New York: McGraw-Hill; 1994a. p. 1018–45.
Engel AG. The muscle biopsy. In: Engel AG, Franzini-Armstrong C, editors. Myology. New York: McGraw-Hill; 1994b. p. 822–31.
Engel AG, Nagel A, Walls TJ, Harper CM, Waisburg HA. Congenital myasthenic syndromes. I. Deficiency and short open-time of the acetylcholine receptor. Muscle Nerve 1993; 16: 1284–92.
Engel AG, Lindstrom JM, Lambert EH, Lennon VA. Ultrastructural localization of the acetylcholine receptor in myasthenia gravis and in its experimental autoimmune model. Neurology 1977; 27: 307–15.
Engel AG, Ohno K, Sine SM. Sleuthing molecular targets for neurological diseases at the neuromuscular junction. Nat Rev Neurosci 2003; 4: 339–52.
Gensler S, Sander A, Korngreen A, Traina G, Giese G, Witzemann V. Assembly and clustering of acetylcholine receptors containing GFP-tagged or subunits. Selective targeting to the neuromuscular junction in vivo. Eur J Biochem 2001; 268: 2209–17.
Hatton CJ, Shelley C, Brydson M, Beeson D, Colquhoun D. Properties of the human muscle nicotinic receptor, and of the slow-channel myasthenic syndrome mutant L221F, inferred from maximum likelihood fits. J Physiol (Lond) 2003; 547: 729–60.
Huebsch KA, Maimone MM. Rapsyn-mediated clustering of acetylcholine receptor subunits requires the major cytoplasmic loop of the receptor subunits. J Neurobiol 2003; 54: 486–501.
Hutchinson DO, Walls TJ, Nakano S, Camp S, Taylor P, Harper CM, et al. Congenital endplate acetylcholinesterase deficiency. Brain 1993; 116: 633–53.
Le Novere N, Corringer PJ, Changeux J-P. Improved secondary structure predictions for a nicotinic receptor subunit: incorporation of solvent accessibility and experimental data into a two-dimensional representation. Biophys J 1999; 76: 2329–45.
Luther MA, Schoepfer R, Whiting P, Casey B, Blatt Y, Montal MS, et al. A muscle acetylcholine receptor is expressed in the human cerebellar medulloblastoma cell line TE671. J Neurosci 1989; 9: 1082–96.
Maimone MM, Merlie JP. Interaction of the 43 kd postsynaptic protein with all subunits of the muscle nicotinic acetylcholine receptor. Neuron 1993; 11: 53–66.
Milone M, Hutchinson DO, Engel AG. Patch-clamp analysis of the properties of acetylcholine receptor channels at the normal human endplate. Muscle Nerve 1994; 17: 1364–9.
Milone M, Wang H-L, Ohno K, Prince RJ, Shen X-M, Brengman JM, et al. Mode switching kinetics produced by a naturally occurring mutation in the cytoplasmic loop of the human acetylcholine receptor subunit. Neuron 1998; 20: 575–88.
Milone M, Shen X-M, Ohno K, Harper CM, Fukudome T, Stilling G, et al. Unusual congenital myasthenic syndrome with endplate AChR deficiency caused by alpha subunit mutations and a remitting-relapsing course. Neurology 1999; 52 Suppl. 2: 185–6.
Miyazawa A, Fujiyoshi Y, Unwin N. Nicotinic acetylcholine receptor at 4.6 resolution: transverse tunnels in the channel wall. J Mol Biol 1999; 288: 765–86.
Miyazawa A, Fujiyoshi Y, Unwin N. Structure and gating mechanism of the acetylcholine receptor pore. Nature 2003; 423: 949–55.
Ohno K, Wang H-L, Milone M, Bren N, Brengman JM, Nakano S, et al. Congenital myasthenic syndrome caused by decreased agonist binding affinity due to a mutation in the acetylcholine receptor subunit. Neuron 1996; 17: 157–70.
Ohno K, Quiram P, Milone M, Wang H-L, Harper CM, Pruitt JN, et al. Congenital myasthenic syndromes due to heteroallelic nonsense/missense mutations in the acetylcholine receptor subunit gene: identification and functional characterization of six new mutations. Hum Mol Genet 1997; 6: 753–66.
Ohno K, Milone M, Shen X-M, Engel AG. A frameshifting mutation in CHRNE unmasks skipping of the preceding exon. Hum Mol Genet 2003; 12: 3055–66.
Plomp JJ, van Kempen GTH, Molenaar PC. Adaptation of quantal content to decreased postsynaptic sensitivity at single endplates in -bungarotoxin treated rats. J Physiol (Lond) 1992; 458: 487–99.
Plomp JJ, van Kempen GTH, De Baets MB, Graus YMF, Kuks JBM, Molenaar PC. Acetylcholine release in myasthenia gravis: regulation at single end-plate level. Ann Neurol 1995; 37: 627–36.
Popot JL, Changeux J-P. Nicotinic receptor of acetylcholine: structure of an oligomeric integral membrane protein. Physiol Rev 1984; 64: 1162–239.
Qin F, Auerbach A, Sachs F. Estimating single-channel kinetic parameters from idealized patch-clamp data containing missed events. Biophys J 1996; 70: 264–80.
Quiram P, Ohno K, Milone M, Patterson MC, Pruitt NJ, Brengman JM, et al. Mutation causing congenital myasthenia reveals acetylcholine receptor / subunit interaction essential for assembly. J Clin Invest 1999; 104: 1403–10.
Salamone FN, Zhou M, Auerbach A. A re-examination of adult mouse nicotinic acetylcholine receptor channel activation kinetics. J Physiol (Lond) 1999; 516: 315–30.
Schoepfer R, Luther MA, Lindstrom J. The human medulloblastoma cell line TE671 expresses a muscle-like acetylcholine receptor. Cloning of the alpha-subunit cDNA. FEBS Lett 1988; 226: 235–40.
Shen X-M, Ohno K, Fukudome T, Tsujino A, Brengman JM, De Vivo DC, et al. Congenital myasthenic syndrome caused by low-expressor fast-channel AChR subunit mutation. Neurology 2002; 59: 1881–8.
Shen X-M, Ohno K, Tsujino A, Brengman JM, Gingold M, Sine SM, et al. Mutation causing severe myasthenia reveals functional asymmetry of AChR signature Cys-loops in agonist binding and gating. J Clin Invest 2003; 111: 497–505.
Sieb JP, Kraner S, Rauch M, Steinlein OK. Immature end-plates and utrophin deficiency in congenital myasthenic syndrome caused by epsilon-AChR subunit truncating mutations. Hum Genet 2000; 107: 160–4.
Sigworth FJ, Sine SM. Data transformation for improved display and fitting of single-channel dwell time histograms. Biophys J 1987; 52: 1047–54.
Sine SM. Molecular dissection of subunit interfaces in the acetylcholine receptor: identification of residues that determine curare selectivity. Proc Natl Acad Sci USA 1993; 90: 9436–40.
Sine SM, Taylor P. Functional consequences of agonist mediated state transitions in the cholinergic receptor. Studies in cultured muscle cells. J Biol Chem 1979; 254: 3315–25.
Uchitel O, Engel AG, Walls TJ, Nagel A, Atassi ZM, Bril V. Congenital myasthenic syndromes. II. A syndrome attributed to abnormal interaction of acetylcholine with its receptor. Muscle Nerve 1993; 16: 1293–301.
Wang H-L, Milone M, Ohno K, Shen X-M, Tsujino A, Batocchi AP, et al. Acetylcholine receptor M3 domain: stereochemical and volume contributions to channel gating. Nat Neurosci 1999; 2: 226–33.
Wang H-L, Ohno K, Milone M, Brengman JM, Evoli A, Batocchi AP, et al. Fundamental gating mechanism of nicotinic receptor channel revealed by mutation causing a congenital myasthenic syndrome. J Gen Physiol 2000; 116: 449–60.
Yu X-M, Hall ZW. The role of the cytoplasmic domains of individual subunits of the acetylcholine receptor in 43 kDa protein-induced clustering in COS cells. J Neurosci 1994; 14: 785–95.(Xin-Ming Shen, Kinji Ohno, Steven M. Sin)
2 Department of Physiology and Biophysics and Receptor Biology Laboratory, Mayo Clinic, Rochester, MN, USA
Summary
We trace the cause of congenital myasthenic syndromes in two patients to mutations in the subunit of the muscle acetylcholine receptor (AChR). Both patients harbour deletion of an asparagine residue in the subunit (N436del) at the C-terminus of the cytoplasmic loop linking the third (M3) and fourth (M4) transmembrane domains. The presence of a null mutation in the second allele of the subunit shows that N346del determines the phenotype. Endplate studies show markedly reduced expression of the N346del-AChR and compensatory accumulation of fetal -AChR. Expression studies in HEK cells reveal decreased expression of N436del-AChR and abnormally brief channel openings. Thus, neuromuscular transmission is compromised by AChR deficiency, fast channel kinetics of the N346del-AChR and incomplete phenotypic rescue by -AChR. Single-channel kinetic analysis shows that the N436del shortens channel openings by reducing stability of the diliganded receptor: rates of channel closing and of ACh dissociation are increased and the rate of channel opening is decreased. In addition to shortening the M3–M4 loop, N436del shifts a negatively charged aspartic acid residue adjacent to M4; the effects of N436del are shown to result from shortening of the M3–M4 loop and not from juxtaposition of a negative charge to M4. To determine whether the consequences of N346del are subunit-specific, we deleted residues that align with N436 in , and subunits. Each deletion mutant reduces AChR expression, but whereas the and mutants curtail channel open duration, the mutant strikingly prolongs open duration. Kinetic analysis reveals that the mutant increases the stability of the diliganded receptor: rates of channel closing and of ACh dissociation are decreased and the rate of channel opening is increased. The overall studies reveal subunit asymmetry in the contributions of the M3–M4 loops in optimizing AChR activation through allosteric links to the channel and the agonist binding site.
Key Words: acetylcholine receptor; congenital myasthenic syndrome; M3–M4 loop; mutagenesis; single-channel patch-clamp recordings
Abbreviations: ACh = acetylcholine; AChR = acetylcholine receptor; -bgt = -bungarotoxin; CMS = congenital myasthenic syndrome; EP = endplate; HEK = human embryonic kidney; M = transmembrane domain; MEPC = miniature endplate current
Introduction
The nicotinic acetylcholine receptor (AChR) at the motor endplate (EP) is a heteropentamer of homologous subunits with stoichiometries 2 for the adult-type receptor and 2 for the fetal type. Each subunit contains four transmembrane domains and short (M1–M2) and long (M3–M4) cytoplasmic loops. The M3–M4 loops of the subunits constitute most of the cytoplasmic mass of AChR (Popot and Changeux, 1984) and harbour three predicted amphipathic helices (Le Novere et al., 1999), one of which borders M4. The M3–M4 loops regulate the flow of ions through the channel (Miyazawa et al., 1999) and affect the rate of channel closing; structural differences between the loops of the and subunits are major determinants of the change from fetal to adult AChR kinetics (Bouzat et al., 1994). The M3–M4 loops also interact with rapsyn to cluster AChR at the EP, but the residues that bind the receptor to rapsyn have not been determined (Gensler et al., 2001; Huebsch and Maimone, 2003; Maimone and Merlie, 1993; Yu and Hall, 1994).
Congenital myasthenic syndromes (CMS) are heterogeneous disorders caused by defects in presynaptic, synaptic basal lamina or postsynaptic gene products (Engel et al., 2003). Most CMS are postsynaptic, and most postsynaptic CMS are caused by mutations in AChR subunits (Engel et al., 2003). To date, six missense mutations have been observed in the M3–M4 loops. An in-frame duplication of residues 413–418 (1254ins18) (Milone et al., 1998) and a missense mutation (A411P) (Wang et al., 2000), both in the amphipathic helix of the subunit, corrupt the fidelity of gating and result in irregular channel kinetics. A three-codon deletion of residues 426–428 of the subunit (426EQEdel) disrupts a specific interaction between the and subunits and impairs AChR assembly (Quiram et al., 1999). Three missense mutations [R311W (Ohno et al., 1997), P331L (Croxen et al., 2001) and V402F (Milone et al., 1999)] reduce surface expression of AChR. Additionally, R311W mildly shortens and V402F modestly prolongs channel opening events.
Here we trace the cause of a myasthenic syndrome in two patients to two heteroallelic mutations in the acetylcholine receptor (AChR) subunit: deletion of the C-terminal residue of the M3–M4 cytoplasmic loop (N436del) plus a null mutation in the second allele. When N436del-AChR and corresponding deletion mutants of other AChR subunits are expressed in human embryonic kidney (HEK) cells, each mutant reduces AChR expression, but whereas the , and deletion mutants decrease, the deletion mutant markedly increases the duration of channel opening episodes. Kinetic analysis reveals that the deletion mutant has decreased ACh affinity for the diliganded closed state and impaired gating efficiency, whereas the deletion mutant markedly enhances ACh affinity and gating efficiency. Thus, the presence of the C-terminal residue of each M3–M4 loop is essential for normal expression, and loops from the different subunits contribute in an asymmetrical manner to optimize activation of AChR.
Methods
Muscle specimens
Intercostal muscle specimens were obtained intact from origin to insertion from patients and control subjects without muscle disease undergoing thoracic surgery. All human studies were in accord with the guidelines of the Institutional Review Board of the Mayo Clinic.
AChR and acetylcholinesterase were detected in cryostat sections by two-colour fluorescence (Hutchinson et al., 1993). Endplates (EPs) were localized for electron microscopy and analysed by the established methods (Engel 1994a,b). Peroxidase-labelled -bungarotoxin (-bgt) was used for the ultrastructural localization of AChR (Engel et al., 1977). The number of AChRs per EP was measured with [125I]-bgt (Engel et al., 1993).
Electrophysiology of muscle specimens
Recordings of miniature EP currents and estimates of the number of transmitter quanta released by nerve impulse were carried out as described elsewhere (Engel et al., 1993; Uchitel et al., 1993). Single-channel patch-clamp recordings from EP AChR were performed in the cell-attached mode as previously described (Milone et al., 1994).
Mutation analysis
We directly sequenced the AChR subunit gene using genomic DNA (Ohno et al., 1996). For family analysis, we traced the IVS9-1GC mutation by EcoNI and the 911delT mutation by Eco72I restriction analysis of PCR products. The N436del mutations was traced with allele-specific PCR in family members and in 200 normal alleles of 100 unrelated controls.
Construction and expression of wild-type and mutant AChRs
Sources of human , , and subunit cDNAs were as previously described (Luther et al., 1989; Ohno et al., 1996; Schoepfer et al., 1988). All four cDNAs were subcloned into the CMV-based expression vector pRBG4 (Sine, 1993) for expression in human embryonic kidney fibroblast (293 HEK) cells. The artificial mutations were engineered into wild-type AChR subunit cDNAs in pRBG4 using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). The presence of each mutation and the absence of unwanted mutations was confirmed by sequencing the entire inserts. HEK cells were transfected with a total of 7.2 μg of plasmids, comprising pRBG4-, -, -, - and pEGFP-N1 in a ratio of 2 : 1 : 1 : 1 : 1 per 35 mm dish using the calcium phosphate precipitation method (Bouzat et al., 1994; Ohno et al., 1996), or a total of 2 μg of plasmids using 6 μl of the FuGene6 transfection reagent (Roche). For patch-clamp studies, we enhanced expression of DN435del-, H408del-, R446del-, R450del-, -omitted and -omitted AChRs by adding 1 μg of the pAdVAntage plasmid (Promega) per 10 μg of total AChR subunit cDNAs.
Bungarotoxin binding measurements
The total number of [125I]-bgt sites on the surface of transfected human embryonic kidney (HEK) cells and ACh competition against the initial rate of [125I]-bgt binding were determined as described elsewhere (Ohno et al., 1996). ACh competition measurements were analysed using the monophasic Hill equation for wild-type and N436del-AChR (Equation 1) or the two-binding-site equation for -omitted AChR (Equation 2):
where Y is fractional occupancy by ACh, n is the Hill coefficient, KOV is an overall dissociation constant, KA and KB are the dissociation constants for the two binding sites, and fractA is the fraction of sites with dissociation constant KA.
Patch-clamp recordings from AChRs expressed in HEK cells
Recordings were obtained in the cell-attached configuration at a membrane potential of –80 mV, at 22°C, and with bath and pipette solutions containing (mM): KCl 142, NaCl 5.4, CaCl2 1.8, MgCl2 1.7, HEPES 10, pH 7.4 (Bouzat et al., 1994; Ohno et al., 1996). Single-channel currents were recorded using an Axopatch 200A amplifier (Axon Instruments) at a bandwidth of 50 kHz, digitized at 5 μs intervals using a Digidata 1200A (Axon Instruments), and recorded to hard disk using the program Clampex 8 (Axon Instruments). Records were analysed at a uniform bandwidth of 11.7 kHz with TACx4.0.9 software (Bruxton). Dwell-time histograms were plotted on a logarithmic abscissa and fitted to the sum of exponentials by maximum likelihood (Sigworth and Sine, 1987).
To estimate rate constants underlying AChR activation, we employed desensitizing concentrations of ACh that cause events from a single channel to cluster into identifiable activation episodes (Qin et al., 1996). Clusters were identified as a series of closely spaced openings preceded and followed by closed intervals greater than a defined critical time. The critical time was determined by a method that misclassifies an equal number of events between two adjacent closed-time components (Colquhoun and Sakmann, 1985). For each receptor, the critical time that provided the best fit for the closed time histogram was chosen for the final analysis. Clusters with fewer than five openings were excluded from analysis. Individual clusters were examined for homogeneity by determining the mean open probability and open duration for each cluster, and clusters within two standard deviations of the means were accepted for further analysis (Qin et al., 1996; Shen et al., 2003). The resulting open and closed intervals were analysed according to kinetic schemes of receptor activation using the program MIL, which uses an interval-based maximum likelihood method that also corrects for missed events (Qin et al., 1996). A dead time of 23 μs was imposed on all recordings. For each type of AChR, single-channel dwell times obtained at a range of ACh concentrations were fitted simultaneously. Data for wild-type AChR were obtained at 10, 20, 30, 50, 70, 100, 200 and 300 μM ACh, for N436del-AChR at 20, 30, 50, 70, 100, 200 and 300 μM ACh, and for H408del-AChR at 0.3, 1, 3 and 10 μM ACh. An average of 8900 events were analysed for each ACh concentration with the range from 1867 to 15 500. The final set of rate constants were checked by superimposing probability density functions calculated from the rate constants on the experimental dwell time histograms, and by their ability to predict burst length at low ACh concentrations (Colquhoun and Sigworth, 1995; Colquhoun and Hawkes, 1995).
Results
Characteristics of CMS patients
Patient 1, a 25-year-old woman (Fig. 1A), and patient 2, a 12-year-old-girl, had severe myasthenic symptoms since birth, a decremental compound muscle action potential response on repetitive stimulation of motor nerves at 2 Hz, and negative tests for anti-AChR antibodies. Both patients responded partially to pyridostigmine and improved further with the additional use of 3,4-diaminopyridine. The parents are not consanguineous, and there are no similarly affected relatives.
Endplate studies
The configuration of the EPs, evaluated from the cytochemical reaction for acetylcholinesterase on teased muscle fibres, was abnormal, with an increased number of small EP regions (1–10 in patient 1 and 2–8 in patient 2) distributed over an increased span of the muscle fibre surface. The number of [125I]-bgt binding sites per EP was 10% of normal (Table 1). Electron microscopy examination of 16 EPs in patient 1 (Fig. 1B and C) and 24 EPs in patient 2 showed a decreased density and restricted distribution of AChR on the junctional folds. The integrity of the junctional folds and nerve terminals was preserved but some postsynaptic regions were simplified (Fig. 1B). Quantal release by nerve impulse was higher than normal (Table 1), probably as an adaptive response to decreased postsynaptic sensitivity to ACh (Plomp et al., 1992, 1995). The miniature EP current (MEPC) amplitude was reduced to 34% of normal in patient 1 (Fig. 1A) and to 44% of normal in patient 2. In patient 1, most MEPCs observed at 17 EPs decayed abnormally slowly and were best fitted by a single exponential, but a small proportion of the EPs (11%) at 15 of the 17 EPs was best fitted by two exponentials, with one component shorter and one component three times longer than normal (Table 1 and Fig. 2A). In patient 2, MEPCs recorded from all 13 EPs decayed biexponentially, with one component shorter and the other twice longer than normal (Table 1).
Single-channel recordings from EPs of patient 2 showed that most channels opened to a low conductance (40 pS) and had long burst open durations characteristic of fetal-type -AChRs (Fig. 2B), but a small proportion (6%) of channels opened to the 60 pS conductance of adult-type -AChRs and had shorter than normal burst durations (Table 1). To summarize, EP studies revealed AChR deficiency, expression of fetal -AChR, and abnormally brief activation episodes of the expressed adult -AChR.
Mutation analysis
To examine the genetic basis of the observed morphological and physiological abnormalities, we directly sequenced the AChR subunit gene and detected three mutations in the two patients (Fig. 3D). Both patients have a 3-bp deletion (1306delAAC) in exon 12 that predicts deletion of an asparagine residue at the C-terminus of the M3–M4 loop of the subunit (N436del). The deleted asparagine at codon 436 is conserved in the mouse and rat but not in the Xenopus or bovine subunit or in other human subunits (Fig. 3A). N436del was not observed in 200 normal alleles.
The second mutation in patient 1 is a previously reported splice site mutation at the 3' end of intron 9 (IVS9-1GC) that alters the canonical AG to AC at the splice acceptor site (Fig. 3D). The mutation causes retention of intron 9, predicting 67 missense amino acids followed by a stop codon. The genetically engineered aberrantly spliced transcript is not expressed on the surface of HEK cells (Ohno et al., 2003). The second mutation in patient 2 is a previously reported frameshifting null mutation in the M3 domain of the subunit (911delT) (Brengman et al., 2000; Sieb et al., 2000) (Fig. 3D).
Family analysis in both patients indicates that the observed mutations are heteroallelic and recessive (Fig. 3E and F). Because IVS9-1GC and 911delT are null mutations, N436del determines the phenotype of both patients.
Expression studies of N436del-AChR expressed in HEK cells
To determine whether N436del hinders the amount of AChR expressed on the cell surface, we engineered N436del into the human subunit and coexpressed it with complementary wild-type , and subunits in HEK cells. As a control, we coexpressed , and subunits in the absence of the subunit. Measurement of [125I]-bgt binding revealed N436del-AChR reduced surface expression to 50% of wild-type, while that of -omitted 22-AChR was 38% of wild-type (Fig. 4A).
To distinguish between lack of incorporation and reduced expression of the mutant subunit, we measured ACh binding at steady state by competition against the initial rate of [125I]-bgt binding (Sine and Taylor, 1979). Wild-type 2 pentamers bind ACh in a monophasic manner, whereas -omitted 22 pentamers bind ACh in a biphasic manner (Fig. 4B) (Ohno et al., 1996). N436del-AChR binds ACh in a monophasic manner like wild-type AChR (Fig. 4B), indicating that the mutant subunit incorporates into most if not all cell surface pentamers. The apparent dissociation constant of N436del-AChR for ACh was very similar to that of wild-type AChR (Fig. 4B).
To examine kinetic effects of N436del, we recorded single-channel currents from human embryonic kidney (HEK) cells expressing N436del-AChR or wild-type AChR activated by a low concentration of ACh (50 nM). Open interval and burst duration histograms of both wild-type and mutant AChRs showed three components, presumably corresponding to two brief mono-liganded open states and one long diliganded open state. Mean durations of diliganded openings are reduced to 46% of wild type by N436del, and those of the corresponding bursts are reduced to 38% (Table 2 and Fig. 5A and B).
The effects of N436del are caused by shortening of the M3–M4 loop
The N436del mutation shortens the M3–M4 loop and at the same time shifts a negatively charged aspartic acid residue to the N-terminal end of M4. To determine whether either or both effects reduce expression or alter activation of AChR, we engineered four site-directed mutants (Fig. 3B) and expressed each together with complementary wild-type subunits in HEK cells.
The first engineered mutant, D435del, shortens the M3–M4 loop by a single residue, without shifting a negative charge against M4 (Fig. 3B). This mutation reduces surface expression of AChR to 56% of wild-type (Fig. 4A) and decreases the mean durations of the major components of channel openings and bursts (Table 2 and Fig. 5C) to 65 and 50% of wild-type, respectively. The second engineered mutant, DN435del, which removes two residues from the M3–M4 loop (Fig. 3B), reduces AChR expression to 10% of wild-type, and decreases the mean durations of the longest components of channel openings and bursts to 65 and 55% of wild-type, respectively (Table 2 and Fig. 5F). The third and fourth engineered mutants, N436R and N436D, do not shorten the M3–M4 loop but position a positive or negative charge next to M4 (Fig. 3B). Neither mutant has an appreciable effect on AChR surface expression (Fig. 4A) or channel kinetics (Table 2 and Fig. 5D and E). Thus, the effects of N436del on AChR expression and channel kinetics can be attributed to shortening of the M3–M4 loop and not to shift of a negative charge adjacent to M4.
Deletion of equivalent residues in non- subunits
To determine whether the effects of deleting a C-terminal residue of the M3–M4 loop are subunit specific, we constructed corresponding deletion mutants of the (H408del), (R446del) and (R450del) subunits (Fig. 3C). The and deletion mutants reduce surface expression of AChR to 5% of wild-type (Fig. 4A) and shorten channel opening events by about the same amount as N436del (Table 2 and Fig. 5H and I). The H408del mutation reduces expression of AChR to 20% of wild-type (Fig. 4A) and, in striking contrast to the and deletion mutants, prolongs the dominant component of open intervals 3.5-fold and that of bursts 13.5-fold (Table 2 and Fig. 5G).
Because HEK cells bound similar low levels of [125I]-bgt after transfection with R450del-AChR, 22-AChR, R446del-AChR and 22-AChR (Fig. 4A), the channel events recorded from the transfected HEK cells could have arisen from -omitted or -omitted receptors rather than from R450del-AChR or R446del-AChR. To test this possibility, we expressed either - or -omitted AChRs in HEK cells in each of three different experiments and searched for channel openings in 30 and 31 patches, respectively, but detected no channel openings. Therefore, it is unlikely that either - or -omitted AChRs, if present, are functional.
Activation kinetics of N436del-AChR and H408del-AChR
To determine the mechanism by which the N436del receptor shortens and the 408Hdel receptor prolongs channel opening events, we examined their kinetics of activation at desensitizing concentrations of ACh (see Methods). Wild-type and N436del-AChRs generated well-defined clusters of openings at ACh concentrations as low as 10 and 20 μM, respectively, but H408del-AChR produced clusters of openings even at 0.3 μM ACh, indicating an enhanced propensity of receptors containing the mutant subunit to become desensitized (Fig. 6, left column).
For both wild-type and N436del-AChRs, the longest closed-time component shifted to shorter durations with increasing ACh concentration (Fig. 6, central column). The open times of the N436del receptors were briefer and those of the H408del receptor longer compared with wild-type (Fig. 6, right column). Dwell times for wild-type and mutant receptors show typical dependence on ACh concentration: closings become briefer with increasing ACh concentration, and the major component of openings changes little across ACh concentrations.
To determine the consequences of the mutations on rate constants underlying receptor activation, we analysed the global set of open and closed dwell times according to scheme 1:
In this scheme, two agonists (A) bind to the receptor (R) with association rate constants k+1 and k+2, and dissociate with rate constants k–1 and k–2. Receptors occupied by one agonist open with rate 1 and close with rate 1, while receptors occupied by two agonists open with rate 2 and close with rate 2. Asterisks indicate open states and RB indicates the blocked state of the receptor. At high ACh concentrations, ACh blocks the open channel with rate k+b, and the channel unblocks with rate k–b. The fitted rate constants allow calculation of the equilibrium dissociation constants (Kn=k–n/k+n, KB=k–b/k+b) and the channel gating equilibrium constants (n=n/n). This scheme allows for only two open states, whereas low-concentration recordings (Fig. 5 and Table 2) had revealed three open states. However, at the high concentrations of ACh required to elicit clusters of channel events due to a single channel, only two components of openings are observed, presumably because the briefest class of monoliganded openings occurs too infrequently to be identified.
Scheme 1 provided a good description of the closed and open intervals for wild-type and N436del receptors but did not provide a satisfactory description of the closed intervals for the H408del mutant. The chief problem was complexity of the closed duration distribution, which exhibited at least one more component than accounted for by scheme 1. We therefore examined alternative scheme 2, which incorporated an additional closed state branching from the diliganded open state:
An analogous brief closed state was suggested by Elenes and Auerbach (2002) for wild-type mouse muscle AChR, and by Hatton and colleagues in describing the activation of the L221F slow-channel mutation (Hatton et al., 2003), both groups suggesting that it corresponded to a short-lived desensitized state. Schemes 1 and 2 provided equally satisfactory fits and similar rate constants for the human wild-type and N436del receptors. To compare activation kinetics of wild type and mutant receptors, we analysed our data according to scheme 2 (Table 3).
The fitted rate constants in Table 3 indicate the following functional consequences of the N346del mutant: by decreasing k+2 and increasing k–2, it increases K2 2.2-fold, thus decreasing the affinity of the diliganded closed receptor for ACh; and by decreasing 2 and increasing 2, it decreases the gating efficiency of the diliganded receptor (2) 2.3-fold. The fitted rate constants were used to predict burst duration at very low concentrations of agonist, which agreed very well with the independently determined burst duration obtained at 50 nM ACh (Table 3). Overall, the N346del burst duration is reduced to 40% of wild-type, and the mutant receptor has mild fast-channel properties (Engel et al., 2003).
The H408del mutation, on the other hand, has opposite effects to N346del: by increasing k+1 and k+2 and decreasing k–1 and k–2, it decreases K1 3-fold and K2 17-fold, thus markedly enhancing both monoliganded and diliganded closed state affinities; by increasing 2 1.5-fold and slowing 2 3.0-fold, it increases the gating efficiency of the diliganded receptor 5-fold. Again, burst duration predicted by the fitted rate constants agreed well with the independently determined burst duration obtained at 50 nM ACh (Table 3). Burst duration increases 14-fold, and the H408del receptor has pronounced slow-channel properties (Engel et al., 2003).
Figure 7 shows plots of open probabilities (Popen) within defined clusters of channel events at different concentrations of ACh for the wild-type, N436del-AChR and H408del receptors. The plotted points are well described by the theoretical Popen curve computed from the rate constants determined by fitting scheme 2 to the dwell times (Fig. 7). The Popen curve for N436del receptor is right-shifted with respect to wild-type, indicating a 3.5-fold decreased probability of opening at the EC50. In contrast, the Popen curve for the H408del receptor is left-shifted with respect to wild-type, indicating a 31-fold increased probability of opening at EC50.
Discussion
Phenotypic consequences of N436del
Both CMS patients carry the N436del mutation plus a null mutation in the second allele; therefore, N346del determines the phenotype. EP studies demonstrate severe AChR deficiency, compensatory expression of fetal-type -AChR, and short opening events of the expressed adult -AChRs. Expression of -AChR at the EP has been documented with low-expressor or null mutations of the subunit, where it probably serves as a means of phenotypic rescue (Engel et al., 2003). Because the number of AChRs at both patients' EPs is only 10% of normal and most expressed AChRs harbour the fetal subunit, neuromuscular transmission is primarily compromised by the AChR deficiency, and this is compounded by the fast-channel kinetics of the N436del-AChRs.
Expression of N436del-AChR in HEK cells was 50% of wild-type. Higher AChR expression in HEK cells than at the EP has been observed previously with other low-expressor missense mutations of AChR (Milone et al., 1998; Ohno et al., 1997; Shen et al., 2002; Wang et al., 1999). This may be due to differences in the rate of synthesis or destruction of the mutant receptor in muscle fibres versus HEK cells. Interaction with the cytoplasmic anchoring protein rapsyn, absent in HEK cells, may also be compromised by the mutant receptors at the EP.
Expression of AChRs harbouring C-terminal deletion mutants of M3–M4 loops
All three C-terminal deletion mutants of the M3–M4 loop (N436del, D435del and DN435del) as well as the corresponding deletion mutants of the (408Hdel), (R446del) and (R450del) subunits curtail surface expression of AChR, whereas the two C-terminal charge mutants of the M3–M4 loop (N436D or N436R) do not; therefore the decreased AChR expression probably stems from shortening of the M3–M4 loop. Decreased AChR expression, in turn, may stem from abnormal folding and accelerated destruction of nascent peptides in the endoplasmic reticulum, or because C-terminal residues of the M3–M4 loops are required for efficient subunit assembly, or both.
Activation kinetics of C-terminal deletion mutants of the M3–M4 loops
Scheme 1 described well the kinetics of activation the wild-type and N436del receptors, but did not account for all closed states of the H408del receptor. We therefore explored an alternative scheme (scheme 2) that allowed an additional closed state connected to the open state of the receptor (A2RG); scheme 2 described both closed and open times of the H408del receptor. Interestingly, applying scheme 2 to the wild-type and the N436del receptors did not significantly alter the rate constants of activation or improve the likelihood of the fitted parameters compared to scheme 1. This finding differs from that in the mouse receptor, where allowing a non-conducting gap to arise from the diliganded open state improved the quality of the fit (Elenes and Auerbach, 2002). The functional significance of the A2RG state is not known; it may represent a brief desensitized state of the receptor (Salamone et al., 1999), but that the dwell in the A2RG state is 13 times shorter for the desensitization prone H408del receptor than for the wild-type or the N436del receptor appears inconsistent with this interpretation.
The results shed new light on the structure–function relationships of the AChR. In particular, the M3–M4 linker has been shown to be a primary determinant of the fetal-to-adult kinetic switch, conferred by change from to subunits (Bouzat et al., 1994), as well as a determinant of the fidelity of channel gating (Milone et al., 1998; Wang et al., 2000). The collective studies point to and as subunits that specifically contribute to activation kinetics, but the present work is the first to show a contribution of the M3–M4 loop of the subunit. The single residue deletion from the subunit prolongs channel activation episodes and enhances desensitization to extents comparable to severe slow channel mutations. The enhanced activation is attributed to enhanced affinity of ACh for the resting closed state of the receptor as well as to enhanced gating. Both the ACh binding site (Celie et al., 2004) and the channel gate (Miyazawa et al., 2003) are distant from the M3–M4 linker, indicating an allosteric contribution of the M3–M4 loop of the subunit to receptor function. Thus, combined clinical, morphological, electrophysiological and genetic studies of our two patients unravelled the pathophysiological basis of a CMS. Further experimentation revealed subunit-specific contributions of C-terminal residues of the M3–M4 loops of AChR that allosterically affect gating of the ion channel and ACh occupancy of the more distant binding site.
Acknowledgements
This work was supported by NIH grants to A. G. E. (NS-6277) and to S. M. S. (NS-31744) and by a Muscular Dystrophy Association Grant to A. G. E.
References
Bouzat C, Bren N, Sine SM. Structural basis of different gating kinetics of fetal and adult acetylcholine receptors. Neuron 1994; 13: 1395–402.
Brengman JM, Ohno K, Milone M, Friedman RL, Feldman RG, Engel AG. Identification and functional characterization of eight novel acetylcholine receptor mutations in six congenital myasthenic syndrome kinships. Neurology 2000; 54 (Suppl. 3): A182–3.
Celie PH, van Rossum-Fikkert SE, van Dijk WJ, Brejc K, Smit AB, Sixma TK. Nicotine and carbamycholine binding to nicotinic acetylcholine receptors as studied in AChBP crystal structures. Neuron 2004; 41: 907–14.
Colquhoun D, Hawkes AG. A Q-matrix cookbook: how to write only one program to calculate the single-channel and macroscopic predictions for any kinetic mechanism. In: Sakmann B, Neher E, editors. Single-channel recording. New York: Plenum Press; 1995. p. 589–633.
Colquhoun D, Sakmann B. Fast events in single channel currents activated by acetylcholine and its analogues at the frog muscle end-plate. J Physiol (Lond) 1985; 369: 501–57.
Colquhoun D, Sigworth FJ. Fitting and statistical analysis of single-channel records. In: Sakmann B, Neher E, editors. Single-channel recording. New York: Plenum Press; 1995. p. 483–587.
Croxen R, Young C, Slater C, Haslam S, Brydson M, Vincent A, et al. Endplate - and subunit mRNA levels in AChR deficiency syndrome due to subunit null mutations. Brain 2001; 124: 1362–72.
Elenes S, Auerbach A. Desensitization of diliganded mouse muscle nicotinic acetylcholine receptor channels. J Physiol (Lond) 2002; 541: 367–83.
Engel AG. Quantitative morphological studies of muscle. In: Engel AG, Franzini-Armstrong C, editors. Myology. New York: McGraw-Hill; 1994a. p. 1018–45.
Engel AG. The muscle biopsy. In: Engel AG, Franzini-Armstrong C, editors. Myology. New York: McGraw-Hill; 1994b. p. 822–31.
Engel AG, Nagel A, Walls TJ, Harper CM, Waisburg HA. Congenital myasthenic syndromes. I. Deficiency and short open-time of the acetylcholine receptor. Muscle Nerve 1993; 16: 1284–92.
Engel AG, Lindstrom JM, Lambert EH, Lennon VA. Ultrastructural localization of the acetylcholine receptor in myasthenia gravis and in its experimental autoimmune model. Neurology 1977; 27: 307–15.
Engel AG, Ohno K, Sine SM. Sleuthing molecular targets for neurological diseases at the neuromuscular junction. Nat Rev Neurosci 2003; 4: 339–52.
Gensler S, Sander A, Korngreen A, Traina G, Giese G, Witzemann V. Assembly and clustering of acetylcholine receptors containing GFP-tagged or subunits. Selective targeting to the neuromuscular junction in vivo. Eur J Biochem 2001; 268: 2209–17.
Hatton CJ, Shelley C, Brydson M, Beeson D, Colquhoun D. Properties of the human muscle nicotinic receptor, and of the slow-channel myasthenic syndrome mutant L221F, inferred from maximum likelihood fits. J Physiol (Lond) 2003; 547: 729–60.
Huebsch KA, Maimone MM. Rapsyn-mediated clustering of acetylcholine receptor subunits requires the major cytoplasmic loop of the receptor subunits. J Neurobiol 2003; 54: 486–501.
Hutchinson DO, Walls TJ, Nakano S, Camp S, Taylor P, Harper CM, et al. Congenital endplate acetylcholinesterase deficiency. Brain 1993; 116: 633–53.
Le Novere N, Corringer PJ, Changeux J-P. Improved secondary structure predictions for a nicotinic receptor subunit: incorporation of solvent accessibility and experimental data into a two-dimensional representation. Biophys J 1999; 76: 2329–45.
Luther MA, Schoepfer R, Whiting P, Casey B, Blatt Y, Montal MS, et al. A muscle acetylcholine receptor is expressed in the human cerebellar medulloblastoma cell line TE671. J Neurosci 1989; 9: 1082–96.
Maimone MM, Merlie JP. Interaction of the 43 kd postsynaptic protein with all subunits of the muscle nicotinic acetylcholine receptor. Neuron 1993; 11: 53–66.
Milone M, Hutchinson DO, Engel AG. Patch-clamp analysis of the properties of acetylcholine receptor channels at the normal human endplate. Muscle Nerve 1994; 17: 1364–9.
Milone M, Wang H-L, Ohno K, Prince RJ, Shen X-M, Brengman JM, et al. Mode switching kinetics produced by a naturally occurring mutation in the cytoplasmic loop of the human acetylcholine receptor subunit. Neuron 1998; 20: 575–88.
Milone M, Shen X-M, Ohno K, Harper CM, Fukudome T, Stilling G, et al. Unusual congenital myasthenic syndrome with endplate AChR deficiency caused by alpha subunit mutations and a remitting-relapsing course. Neurology 1999; 52 Suppl. 2: 185–6.
Miyazawa A, Fujiyoshi Y, Unwin N. Nicotinic acetylcholine receptor at 4.6 resolution: transverse tunnels in the channel wall. J Mol Biol 1999; 288: 765–86.
Miyazawa A, Fujiyoshi Y, Unwin N. Structure and gating mechanism of the acetylcholine receptor pore. Nature 2003; 423: 949–55.
Ohno K, Wang H-L, Milone M, Bren N, Brengman JM, Nakano S, et al. Congenital myasthenic syndrome caused by decreased agonist binding affinity due to a mutation in the acetylcholine receptor subunit. Neuron 1996; 17: 157–70.
Ohno K, Quiram P, Milone M, Wang H-L, Harper CM, Pruitt JN, et al. Congenital myasthenic syndromes due to heteroallelic nonsense/missense mutations in the acetylcholine receptor subunit gene: identification and functional characterization of six new mutations. Hum Mol Genet 1997; 6: 753–66.
Ohno K, Milone M, Shen X-M, Engel AG. A frameshifting mutation in CHRNE unmasks skipping of the preceding exon. Hum Mol Genet 2003; 12: 3055–66.
Plomp JJ, van Kempen GTH, Molenaar PC. Adaptation of quantal content to decreased postsynaptic sensitivity at single endplates in -bungarotoxin treated rats. J Physiol (Lond) 1992; 458: 487–99.
Plomp JJ, van Kempen GTH, De Baets MB, Graus YMF, Kuks JBM, Molenaar PC. Acetylcholine release in myasthenia gravis: regulation at single end-plate level. Ann Neurol 1995; 37: 627–36.
Popot JL, Changeux J-P. Nicotinic receptor of acetylcholine: structure of an oligomeric integral membrane protein. Physiol Rev 1984; 64: 1162–239.
Qin F, Auerbach A, Sachs F. Estimating single-channel kinetic parameters from idealized patch-clamp data containing missed events. Biophys J 1996; 70: 264–80.
Quiram P, Ohno K, Milone M, Patterson MC, Pruitt NJ, Brengman JM, et al. Mutation causing congenital myasthenia reveals acetylcholine receptor / subunit interaction essential for assembly. J Clin Invest 1999; 104: 1403–10.
Salamone FN, Zhou M, Auerbach A. A re-examination of adult mouse nicotinic acetylcholine receptor channel activation kinetics. J Physiol (Lond) 1999; 516: 315–30.
Schoepfer R, Luther MA, Lindstrom J. The human medulloblastoma cell line TE671 expresses a muscle-like acetylcholine receptor. Cloning of the alpha-subunit cDNA. FEBS Lett 1988; 226: 235–40.
Shen X-M, Ohno K, Fukudome T, Tsujino A, Brengman JM, De Vivo DC, et al. Congenital myasthenic syndrome caused by low-expressor fast-channel AChR subunit mutation. Neurology 2002; 59: 1881–8.
Shen X-M, Ohno K, Tsujino A, Brengman JM, Gingold M, Sine SM, et al. Mutation causing severe myasthenia reveals functional asymmetry of AChR signature Cys-loops in agonist binding and gating. J Clin Invest 2003; 111: 497–505.
Sieb JP, Kraner S, Rauch M, Steinlein OK. Immature end-plates and utrophin deficiency in congenital myasthenic syndrome caused by epsilon-AChR subunit truncating mutations. Hum Genet 2000; 107: 160–4.
Sigworth FJ, Sine SM. Data transformation for improved display and fitting of single-channel dwell time histograms. Biophys J 1987; 52: 1047–54.
Sine SM. Molecular dissection of subunit interfaces in the acetylcholine receptor: identification of residues that determine curare selectivity. Proc Natl Acad Sci USA 1993; 90: 9436–40.
Sine SM, Taylor P. Functional consequences of agonist mediated state transitions in the cholinergic receptor. Studies in cultured muscle cells. J Biol Chem 1979; 254: 3315–25.
Uchitel O, Engel AG, Walls TJ, Nagel A, Atassi ZM, Bril V. Congenital myasthenic syndromes. II. A syndrome attributed to abnormal interaction of acetylcholine with its receptor. Muscle Nerve 1993; 16: 1293–301.
Wang H-L, Milone M, Ohno K, Shen X-M, Tsujino A, Batocchi AP, et al. Acetylcholine receptor M3 domain: stereochemical and volume contributions to channel gating. Nat Neurosci 1999; 2: 226–33.
Wang H-L, Ohno K, Milone M, Brengman JM, Evoli A, Batocchi AP, et al. Fundamental gating mechanism of nicotinic receptor channel revealed by mutation causing a congenital myasthenic syndrome. J Gen Physiol 2000; 116: 449–60.
Yu X-M, Hall ZW. The role of the cytoplasmic domains of individual subunits of the acetylcholine receptor in 43 kDa protein-induced clustering in COS cells. J Neurosci 1994; 14: 785–95.(Xin-Ming Shen, Kinji Ohno, Steven M. Sin)