Pathogenic point mutations in a transmembrane domain of the ε subunit increase the Ca²⁺ permeability of the human endplate ACh receptor

Abstract

The ε subunit of the human endplate ACh receptor (AChR) is a key determinant of the large fraction of the ACh-evoked current carried by Ca²⁺ ions (P_f). Consequently, missense mutations in the ε subunit are potential targets for altering the P_f of human AChR. In this paper we investigate the effects of two pathogenic point mutations in the M2 transmembrane segment AChR ε subunit, εT264P and εV259F, that cause slow-channel syndromes (SCS). When expressed in GH4C1 cells, the mutant receptors subunits raise Ca²⁺ permeability of the receptors ∼1.5 and ∼2-fold above that of wild-type, to attain P_f values of 11.8% (εT264P) and 15.4% (εV259F). The latter value exceeds most P_f values reported to date for ligand-gated ion channels. Consistent with these findings, the biionic Ca²⁺ permeability ratio (P_Ca/P_Cs) of the mutant AChRs is also increased. Upon repetitive stimulation with ACh, the mutant receptors show an enhanced current run-down compared with wild-type, leading to a strong reduction of their function. We propose that the enhanced Ca²⁺ permeability of the mutant receptors overrides the protective effect of desensitization and, together with the prolonged opening events of the AChR channel, is an important determinant of the excitotoxic endplate damage in the SCS.

Slow-channel syndromes (SCS) are caused by point mutations in the human endplate acetylcholine receptor (AChR) that greatly prolong the openings of the AChR channel. In a peculiar form of the SCS, tubulofilamentous inclusions, similar to those found in inclusion body myositis (IBM), accumulate in junctional regions of the muscle fibres (Fidzianska et al. 2005). This SCS is caused by a valine-to-phenylalanine substitution at position 259 in the M2 transmembrane segment of the AChR ε subunit (εV259F) that lines the channel pore.

Hyperphosphorylation of tau protein leads to formation of tubulofilamentous structures in neuronal diseases as well as in IBM, and is associated with, and possibly caused by, a disrupted Ca²⁺ homeostasis (LaFerla, 2002; Pierrot et al. 2004). We recently showed that the fractional Ca²⁺ current (P_f) of human endplate AChR, defined as the percentage of the total ACh-evoked current carried by Ca²⁺ ions, is twice as large than that of the homologous mouse AChR (Fucile et al. 2006). The high Ca²⁺ permeability of human AChR by itself enhances vulnerability of the human endplate to excitotoxic damage (endplate myopathy) in the SCS. An increased fractional Ca²⁺ current would further contribute to Ca²⁺ overloading of the postsynaptic region and to the Ca²⁺-dependent pathological effects. Since the human ε subunit turned out to be a key determinant of the high P_f at the human endplate (Fucile et al. 2006), we postulated that strategically positioned amino acid substitutions in the ε subunit can alter the Ca²⁺ permeability of the AChR channel.

To test this notion, we measured the P_f of the εV259F-AChR and found it almost twice as high as that of wild-type (WT) AChR. To determine whether this effect is mutation specific, we also examined the P_f for another well-characterized SCS mutation in the M2 segment of the ε subunit, εT264P (Ohno et al. 1995). We found that the P_f of the εT264P-AChR is also markedly increased compared with that of the WT receptor. Our results provide the first direct evidence that mutations associated with SCS alter Ca²⁺ permeability as well as channel kinetics of the human endplate AChR.

Methods

Cell cultures and transfection

Rat pituitary GH4C1 cells were grown (5% CO₂, 37°C) in HAM F10 medium plus 10% fetal calf serum and 1% penicillin/streptomycin. Cells were plated onto 35 mm Petri dishes (3 × 10⁵ cells per dish) 24 h prior to transfection. The cDNAs encoding human wild-type α, β, δ, ε subunits, or ε subunits harbouring the εT264P or εV259F mutations were transiently transfected using Lipofectamine 2000, using 1 μg of cDNA for each subunit per Petri dish. Medium was changed after overnight incubation with cDNAs, and experiments carried out after a further 24–48 h. All media were purchased from Invitrogen.

Solutions and chemicals

Cells were superfused with a standard external medium containing (mm): 140 NaCl, 2.8 KCl, 2 CaCl₂, 2 MgCl₂, 10 Hepes-NaOH, 10 glucose, pH 7.3. For cell-attached experiments, patch pipettes were filled with the same solution plus ACh (200 nm).

For whole-cell recordings, patch pipettes were filled with a solution containing (mm): 120 KCl, 5 BAPTA, 10 Hepes-KOH, 2 Mg-ATP, 2 MgCl₂, pH 7.3.

For P_f determinations, intracellular solution contained (mm): 140 N-methyl-d-glucamine (NMDG), 10 Hepes-HCl, 0.25 Fura-2 pentapotassium salt, 0.001 thapsigargin, pH 7.3. Calibration measurements were performed using an extracellular solution made of (mm): 153 NMDG, 10 CaCl₂, 10 Hepes-HCl, pH 7.3.

The shift of the reversal potential of ACh-evoked current was measured using an intracellular solution containing (mm): 140 CsCl, 10 Hepes-CsOH, 20 BAPTA, pH 7.3. The external solution contained (mm): 150 CsCl, 10 Hepes-CsOH, 10 glucose, pH 7.3 plus 1 or 10 CaCl₂.

All chemicals were purchased from Sigma (USA), except for Fura-2 pentapotassium salt (Molecular Probes).

Electrophysiology

Currents were recorded using borosilicate glass patch pipettes (2–5 MΩ tip resistance, Sylgard-coated for single-channel recordings) and an Axopatch 200B amplifier (Molecular Devices, Union City, CA, USA). Data were recorded and analysed using pCLAMP 9 (Molecular Devices). All recordings were performed at 25–27°C.

For whole-cell recordings, series resistance was compensated by 80–90%. Cells were voltage clamped at a holding potential of −60 mV and continuously superfused using a gravity-driven fast exchanger perfusion system (RSC-200, BioLogique, France). Current decay and run-down were fitted with single exponential curves [i(t) =i_∞+i₀e^−t/τ], using Origin 7 (OriginLab Corporation, Northampton, MA, USA).

The relative Ca²⁺ permeability in biionic conditions (P_Ca/P_Cs) was estimated from the shift of the reversal potential (V_rev) of whole-cell ACh-induced current using two different extracellular Ca²⁺ concentrations (Ca_o1 of 1 mm, Ca_o10 of 10 mm). In each cell, V_rev was calculated at both Ca²⁺ concentrations (V_r1 and V_r10, respectively), by linear interpolation of the ACh-evoked currents peak amplitude plotted versus the test potential. The use of voltage ramp yielded inaccurate results because of the rapid desensitization of εV259F-AChR. P_Ca/P_Cs ratios were obtained from the extended Lewis equation adapted to the chosen ion concentrations (Lewis, 1979; Castro & Albuquerque, 1995):

(1)

where F, R, T are the usual thermodynamic constants and V_r= V_r10− V_r1, Ion activity as given by Castro & Albuquerque (1995), was used in calculations. Membrane potential was stepped to the test potential 1 s before ACh application.

ACh dose–current curves were constructed applying to each cell different doses of transmitter (0.1–300 μm, applied at 30–60 s interval) and normalizing the current response to the plateau value. The data were best-fitted using Hill equation by Origin 7 software (Origin Laboratory).

Single-channel data were filtered at 5 kHz and sampled at 25 kHz. Analysis was performed with a 50% threshold criterion, omitting events shorter than 0.12 ms. Slope conductance was calculated by linear fitting of the unitary amplitudes recorded at different pipette potentials (at least three for each cell). The fit was extrapolated to estimate cell resting potential, which was summed to the pipette potential to obtain membrane potential. The critical time used to identify a burst, determined for each cell from the closed time distribution (Colquhoun & Sakmann, 1985), ranged between 1 and 4 ms.

All results are given as mean ±s.e.m. Two data sets were considered statistically different when P < 0.05 (ANOVA test).

P_f determination

The methods to measure P_f have been fully reported previously (Fucile et al. 2000, 2006). Fluorescence determinations were made using a fluorescence upright microscope (Axioskop 2, Zeiss, Germany), a digital 12-bit cooled camera (SensiCam, PCO, Germany) and a monochromator (Cairn, UK). The system was driven by Axon Imaging Workbench 2 software (Molecular Devices), which also triggered the start of electrophysiological recording. All optical parameters and the setting of the digital camera (50 ms exposure time, 4 × 4 binning) were maintained throughout all measurements. The changes in intracellular calcium were monitored at a single excitation wavelength (380 nm), to achieve a higher time resolution, and expressed as the ratio of time-resolved fluorescence variation over the basal fluorescence (ΔF/F₀).

Only isolated cells with low basal intracellular Ca²⁺ (F₃₄₀/F₃₈₀ ratio values < 2) were considered for recordings. Cells were filled with Fura-2 pentapotassium salt through the patch pipette and measurements performed after obtaining a stable value of basal fluorescence, with F₃₈₀ > 200 arbitrary units (a.u), indicating satisfactory loading. The ratio F/Q (expressed in nC⁻¹) was defined as:

where I_Nic is the nicotine-evoked whole-cell current. The charge entered into the cell was calculated at each fluorescence acquisition time. The F/Q ratio was estimated for each cell as the slope of the linear fit of the F versus Q plot. P_f was determined normalizing the F/Q ratio to the calibration value, obtained in a calibration medium containing only Ca²⁺ as permeant ions:

The F/Q_calibration value obtained in this work was 6.6 ± 1.9 nC⁻¹ (mean ±s.e.m., n = 4 cells).

Results

Characterization of the εV259F and εT264P AChRs in GH4C1 cells

We first determined the main functional characteristics of wild-type (WT) and mutant AChRs expressed in GH4C1 cells. We confirmed the slow-channel properties of both εV259F- and εT264P-AChR at the single-channel level by cell-attached recordings (Fig. 1A). For the WT-AChR, the mean single channel slope conductance was about 50 pS and the mean burst duration was 1.9 ms (Table 1). As shown in Fig. 1B (left), the distribution of burst duration was fitted by two exponential components, with the mean time constants (τ_b1 and τ_b2) and weights given in Table 1. The duration of the bursts was greatly enhanced by the εV259F as well as the εT264P mutation, while the conductance of unitary events was not significantly affected (Table 1). The distribution of burst durations presented a third exponential component (time constant, τ_b3) for both mutant AChRs. Typical examples are shown in Fig. 1B (middle and right panels) and mean values of the τ_bs reported in Table 1. These data confirm that both mutant receptors retain their main functional properties when expressed in GH4C1 cells.

Figure 1.

Functional characterization of WT and mutant AChRs in GH4C1 cells A, typical cell-attached recordings of channel openings evoked by ACh (200 nm) in cells transfected with WT or mutant AChRs, as indicated, revealing prolonged burst duration in εV259F and εT264P mutant AChRs. Traces were filtered at 1 kHz for display (5 kHz for analysis). Unitary conductance was: 49.8 pS (WT), 46.0 pS (εV259F) and 46.6 pS (εT264P). B, distribution of burst durations for the same recordings as in A, best fitted with the sum of two or three exponential components (dotted lines). Time constants (weight) of the fitting distributions in these three cells (at −100 mV membrane potential) were: for WT-AChR, τ_b1= 0.46 ms (42%), τ_b2= 2.9 ms (58%), critical τ= 4 ms; for εV259F, τ_b1= 0.17 ms (41%), τ_b2= 2.7 ms (14%), τ_b3= 78 ms (45%), critical τ= 4 ms; for εT264P, τ_b1= 0.24 ms (34%), τ_b2= 7.9 ms (18%), τ_b3= 63 ms (48%), critical τ= 1 ms. C, typical whole-cell responses to repetitive ACh (100 μm for 0.5 s) applications (black bars) in cells transfected as indicated. Dotted lines represent the exponential fit of the current peaks with the parameters given on each panel, showing the enhanced run-down of mutant AChRs. Holding potential, −60 mV. Open bars indicate superfusion with standard solution.

Table 1.

Single channel properties of WT and mutant AChRs

	WT-AChR	εV259F-AChR	εT264P-AChR
Conductance (pS)	48.6 ± 2.4 (7)	49.7 ± 1.7 (7)	51.1 ± 2.6 (4)
Burst duration (ms)	1.88 ± 0.19 (5)	47 ± 11 (4)	42.4 ± 7.6 (4)
Critical time (ms)	3.4 ± 0.6 (5)	5 ± 1 (4)	1.8 ± 0.7 (4)
τb1 (ms)/weight (%)	3.2 ± 0.3/49 ± 5 (5)	3.2 ± 0.3/32 ± 3 (4)	3.2 ± 0.3/33 ± 4 (4)
τb2 (ms)/weight (%)	3.2 ± 0.3/51 ± 5 (5)	3.7 ± 2/18 ± 8 (4)	3.2 ± 0.3/17 ± 9 (4)
τb3 (ms)/weight (%)	not present	73 ± 10/50 ± 6 (4)	74 ± 14/50 ± 6 (4)

Data are given as mean ±s.e.m. (number of cells). Burst analysis was performed on cell-attached recordings obtained with bandwidth of 5 kHz; T = 25–27°C, with an estimated membrane potential of −90 to −105 mV (as determined a posteriori by the i–V relation). τ_b1, τ_b2, τ_b3: time constants of the exponential components best fitting the distribution of burst durations.

At the macroscopic level, the amplitude of the ACh-evoked currents showed large cell-to-cell variation, ranging from −0.2 nA to over −10 nA at −60 mV (ACh, 100 μm) for each type of AChR, suggesting that expression of the mutant AChRs was not appreciably impaired compared with that of WT-AChR. Both mutant receptors had a higher apparent affinity for ACh, as dose–response curves showed a decrease of the EC₅₀ from 33.5 ± 1.4 μm (n = 4) in the WT-AChR, to 1.2 ± 0.2 μm (n = 4) and 0.9 ± 0.2 μm (n = 5) for εV259F-AChR and εT264P-AChR, respectively (data not shown).

During sustained ACh application, whole-cell currents decayed exponentially, reflecting receptor desensitization. At the plateau concentration of 100 μm, the rate of decay was similar for the WT-AChR (τ_decay= 207 ± 28 ms, n = 14) and the εT264P-AChR (246 ± 37 ms, n = 7; P = 0.4), whereas the εV259F-AChR showed a faster decay (128 ± 16 ms, n = 10; P = 0.04). Repetitive ACh applications produced currents with decreasing peak amplitude (current run-down) for all three AChRs examined (Fig. 1C). An exponential fit of the consecutive current peaks showed that WT-AChR (Fig. 1C, left) had a relatively slow and limited run-down (τ_run-down= 0.78 ± 0.08 s, n = 14), with an asymptotic amplitude (i_∞) of 31 ± 4% of the first response. For the εV259F mutant (Fig. 1C, middle), the decay was both faster (τ_run-down= 0.38 ± 0.06 s, n = 9, P = 0.002) and more pronounced (i_∞= 4.5 ± 0.9%) than for WT-AChR. The εT264P mutant AChR (Fig. 1C, right) showed an intermediate behaviour, with a τ_run-down= 0.68 ± 0.05 s (n = 7) and an asymptote at i_∞= 9 ± 2%, significantly less than that of WT-AChR (P = 0.001). Current run-down was not accompanied by an increased rate of desensitization, as, in each cell, the τ_decay of the third current response was not significantly different from that of the first response (P > 0.5). Together, these data indicate that the mutant receptors become less responsive than WT, or unresponsive, to ACh at physiological rates of stimulation.

P_f and P_Ca/P_Cs measurements

For P_f determinations we used nicotine because ACh might activate muscarinic receptors, leading to Ca²⁺ release from internal stores. Currents evoked by nicotine (100 μm) were comparable to those elicited by ACh, except for a slower current decay observed for the WT-AChR (τ_decay= 649 ± 87 ms, n = 14) (e.g. Fig. 2A).

Figure 2.

Calcium fractional current of WT and mutant AchRs A, typical fluorescence (top trace) and current (bottom) responses to nicotine (100 μm, horizontal bar) simultaneously recorded in cells expressing the indicated AChR. Notice the large Ca²⁺ response recorded in the εV259 cell in spite of a relatively small current. B, linear fit of the ΔF/F₀versus Q in the same three cells as in A. The dotted line represents the averaged slope of the calibration experiments. ^, εV259F-AChR; ▵, εT264P-AChR; □, WT-AChR, as indicated.

Nicotine-evoked Ca²⁺ and current (I_Nic) responses were simultaneously recorded (Fig. 2A), and the F/Q ratios calculated (Fig. 2B). In cells expressing WT-AChR (e.g. Fig. 2A, left), normalization to calibration values yielded P_f values of 7.83 ± 0.69% (n = 11), in excellent agreement with previously published values (Fucile et al. 2006). The P_f value measured for εV259F-AChR (Fig. 2A, middle) was 15.4 ± 1.7% (n = 12, P = 0.0007), showing that this mutation doubles the Ca²⁺ permeability of the AChR-channel. For comparison, we measured the P_f of the εT264P-AChR (Fig. 2A, right) and again found a significantly enhanced Ca²⁺ permeability (P_f= 11.76 ± 0.91%, n = 13, P = 0.003). Thus, both slow-channel mutations in the M2 segment of the ε subunit enhance the Ca²⁺ permeability of human AChR.

To test by a different approach that the Ca²⁺ permeability of the mutant AChRs was enhanced compared with that of WT-AChR, we also assessed the relative Ca²⁺ permeability of each isoform by measuring its biionic permeability ratio P_Ca/P_Cs. The currents evoked by ACh (10 μm) were measured at five or more test potentials between −10 and +10 mV, bracketing the current reversal potential, at two different extracellular Ca²⁺ concentrations (1 mm and 10 mm) (Fig. 3A). With 1 mm extracellular Ca²⁺, the average values of the reversal potentials were not statistically different for the εV259F-AChR (0.2 ± 0.6 mV, n = 7) or the εT264P-AChR (−0.5 ± 0.3 mV, n = 10) versus the WT-AChR (0.1 ± 0.3 mV, n = 8; P > 0.2). The higher extracellular Ca²⁺ concentration induced a right-shift of the reversal potential of each isoform (Fig. 3B). Using eqn (1) (see Methods), we obtained P_Ca/P_Cs= 0.73 ± 0.07 (n = 7) for the WT-AChR. The value of P_Ca/P_Cs obtained for the εV259F-AChR (1.20 ± 0.20, n = 5) was significantly larger than that of WT-AChR (P = 0.038). A further increase over the WT-AChR value was observed in the P_Ca/P_Cs ratio determined for the εT264P-AChR (1.35 ± 0.10, n = 7, P = 0.0007).

Figure 3.

Relative Ca²⁺ permeability (P_Ca/P_Cs) of WT and mutant AchRs A, ensemble whole-cell currents recorded in response to ACh (10 μm) applied at various test potentials with low extracellular Ca²⁺ (1 mm, open symbols) or high extracellular Ca²⁺ (10 mm, filled symbols), in transfected cells expressing the indicated AChRs. In each panel, numbers refer to the most positive and negative test potentials shown. Note the reduction of current amplitude in high Ca²⁺ for all the AChR isoforms. B, current peak amplitude plotted versus membrane test potential in the same cells as in A. Lines represent the linear fit of each data group. Note positive shifts of current reversal potentials (0.7 mV for WT-AChR; 1.5 mV for εV259F-AChR; 1.6 mV for εT264P-AChR) at 10 mm Ca²⁺ (filled symbols).

Discussion

This paper provides the first quantitative evidence that the Ca²⁺ permeability of the human endplate AChR can be modified by point mutations in the membrane-spanning M2 segment of the ε subunit. The mutations examined, εV259F and εT264P, gave rise to a fractional Ca²⁺ current of ∼15% and ∼12%, respectively. These P_f values are among the highest known, being equivalent to, or higher than, those reported for the NMDA receptor (up to 11%, Burnashev et al. 1995; 13.5%, Jatzke et al. 2002), or the homomeric α7-AChR (11.4%, Fucile et al. 2003).

The enhanced Ca²⁺ permeability of the ε SCS mutants additionally increases the Ca²⁺ ingress into the junctional sarcoplasm that stems from the prolonged duration of the synaptic current and the intrinsically high Ca²⁺ permeability of human AChR and is thus directly relevant to the pathogenesis of the endplate myopathy associated with the observed mutations. An increased permeability to Ca²⁺ may also occur with SCS mutations in other AChR subunits, as an increase of P_Ca/P_Cs was observed in HEK 293 cells expressing mouse βV299F-AChR, a mutant receptor designed after a recently identified mutation in a SCS patient (Navedo et al. 2006). Both εV259F and εT264P mutations yield a complex effect on receptor function. The affinity of the mutant receptors for ACh is increased by about one order of magnitude, but this increase is accompanied by (or possibly results in) a loss of receptor function upon repetitive stimulation at high ACh concentration, as well as an enhanced rate of desensitization in the case of the εV259F-AChR. That both mutant AChRs show an increased run-down during repetitive exposure to ACh probably contributes to the abnormal fatigability on exertion observed in vivo. Interestingly, the propensity of the mutant receptors to become unresponsive could partially mitigate the adverse effects of the prolonged open time and increased Ca²⁺ permeability of the mutant channels, but not enough to protect the endplate from excitotoxic injury.

In the only patient carrying the εV259F mutation identified to date (Fidzianska et al. 2005), the endplate myopathy was accompanied by the presence of tubulofilamentous inclusion bodies. The formation of these structures might be precipitated by the severe disruption of Ca²⁺ homeostasis caused by the concurrent prolongation of synaptic events and doubled P_f, consistent with the notion that an altered Ca²⁺ homeostasis enhances phosphorylation of tau protein and formation of fibrillary structures (LaFerla, 2002).

Several lines of evidence indicate that the ε subunit is a key determinant of the Ca²⁺ permeability of adult muscle AChR. The P_f is almost 2-fold higher in AChRs containing the ε rather than the γ subunit (Villarroel & Sakmann, 1996; Ragozzino et al. 1998). Moreover, the ε subunit is the only determinant of the high Ca²⁺ permeability of the human adult AChR (Fucile et al. 2006). That a valine-to-phenylalanine mutation in the pore-forming M2 segment of the ε subunit (εV259F) markedly enhances P_f whereas mutation of the corresponding residue in the α subunit (αV249F) has no effect on P_f (Fucile et al. 2006), is also consistent with the important role of the ε subunit in governing the Ca²⁺ permeability of the human receptor. Another mutation in the ε M2 segment, εT264P, also increases the P_f of the mutant receptor. Thus, the data presented here confirm that the ε subunit has a strong effect on the Ca²⁺ permeability of human endplate AChR.

In our experiments, the alterations in P_Ca/P_Cs values only partially mirrored those in the P_f. The rapid run-down of current generated by εV259F-AChR during multiple applications of ACh limits the accuracy of the P_Ca/P_Cs measurements. Furthermore, constant field assumptions leading to eqn (1) may not hold under the actual experimental conditions, as previously reported (Vernino et al. 1994; Burnashev et al. 1995). For instance, studies that compare muscle and neuronal AChRs reveal a 7-fold difference in the P_Ca/P_Cs ration (Vernino et al. 1994), but only a 2-fold difference for the P_f values (Vernino et al. 1993). Thus, biionic permeability ratios often provide qualitative rather than quantitative estimates of the relative Ca²⁺ permeability of a given channel whereas P_f determinations require no a priori assumptions (Zhou & Neher, 1994) and thus yield a more reliable measure of the Ca²⁺ permeability of the AChR channel (Burnashev et al. 1995).

In conclusion, our studies provide further insight into the pathogenic mechanisms whereby single amino acid substitutions in AChR subunits cause disease, which results from the delicate balance between the deleterious effects of the increased Ca²⁺ permeability and the dampening of the synaptic response by run-down of the investigated SCS mutant receptors at physiological rates of stimulation, by loss of AChR from the degenerating folds, and by the altered endplate geometry.