Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2014 Jul 16;592(Pt 16):3687–3696. doi: 10.1113/jphysiol.2014.276493

Complete reversal of Lambert–Eaton myasthenic syndrome synaptic impairment by the combined use of a K+ channel blocker and a Ca2+ channel agonist

Tyler B Tarr 1,2, David Lacomis 3,4, Stephen W Reddel 5, Mary Liang 6,7, Guillermo Valdomir 6,7, Michael Frasso 6,7, Peter Wipf 6,7, Stephen D Meriney 1,2,✉
PMCID: PMC4229355  PMID: 25015919

Abstract

Lambert–Eaton myasthenic syndrome (LEMS) is an autoimmune disorder in which a significant fraction of the presynaptic P/Q-type Ca2+ channels critical to the triggering of neurotransmitter release at the neuromuscular junction (NMJ) are thought to be removed. There is no cure for LEMS, and the current most commonly used symptomatic treatment option is a potassium channel blocker [3,4-diaminopyridine (3,4-DAP)] that does not completely reverse symptoms and can have dose-limiting side-effects. We previously reported the development of a novel Ca2+ channel agonist, GV-58, as a possible alternative treatment strategy for LEMS. In this study, we tested the hypothesis that the combination of GV-58 and 3,4-DAP will elicit a supra-additive increase in neurotransmitter release at LEMS model NMJs. First, we tested GV-58 in a cell survival assay to assess potential effects on cyclin-dependent kinases (Cdks) and showed that GV-58 did not affect cell survival at the relevant concentrations for Ca2+ channel effects. Then, we examined the voltage dependence of GV-58 effects on Ca2+ channels using patch clamp techniques; this showed the effects of GV-58 to be dependent upon Ca2+ channel opening. Based on this mechanism, we predicted an interaction between 3,4-DAP and GV-58. We tested this hypothesis using a mouse passive transfer model of LEMS. Using intracellular electrophysiological ex vivo recordings, we demonstrated that a combined application of 3,4-DAP plus GV-58 had a supra-additive effect that completely reversed the deficit in neurotransmitter release magnitude at LEMS model NMJs. This reversal contrasts with the less significant improvement observed with either compound alone. Our data indicate that a combination of 3,4-DAP and GV-58 represents a promising treatment option for LEMS and potentially for other disorders of the NMJ.

Introduction

Lambert–Eaton myasthenic syndrome (LEMS) is an autoimmune disorder characterized by a presumed loss of a fraction of the presynaptic P/Q-type Ca2+ channels required for action potential-evoked acetylcholine release at the neuromuscular junction (NMJ) (Lambert et al. 1956; Nagel et al. 1988; Vincent et al. 1989; Smith et al. 1995; Meriney et al. 1996). A reduction in the number of presynaptic P/Q-type Ca2+ channels at the NMJ in LEMS is assumed on the basis of multiple lines of evidence, including the presence of antibodies to the P/Q-type Ca2+ channel in sera of LEMS patients (Lennon et al. 1995; Motomura et al. 1997), a reduction in the number of active zone particles following passive transfer of LEMS to mice (Fukunaga et al. 1983) and a reduction in Ca2+ current in various model systems following application of LEMS IgG (Lang et al. 1989; Meriney et al. 1996). A reduction in P/Q-type Ca2+ channels results in a disruption of neuromuscular transmission that predominately causes limb weakness and areflexia that improve after a few seconds of sustained contraction of the weak muscle (Lambert et al. 1956; Titulaer et al. 2008). The electrophysiological correlates are a reduced quantal content [QC (magnitude of acetylcholine release)] and a reduction in the resulting compound muscle action potential (CMAP) (Lambert et al. 1956; Vincent et al. 1989; Titulaer et al. 2008). LEMS can occur as a primary autoimmune disease or as a paraneoplastic autoimmune disorder (most commonly triggered by small cell lung cancer in smokers) (Titulaer et al. 2011a). The most common treatment for LEMS is the potassium channel blocker 3,4-diaminopyridine (3,4-DAP), although pyridostigmine (which increases the persistence of acetylcholine in the synaptic cleft) or immunotherapies such as i.v. immunoglobulin, plasma exchange and immunosuppression are also used in some cases (Lindquist & Stangel, 2011; Titulaer et al. 2011b). These therapies can often control the disease, but generally do not result in a return to complete normality (Sedehizadeh et al. 2012).

We previously reported the initial evaluation of our novel compound GV-58 as a Ca2+ channel agonist that selectively affects P/Q- and N-type, but not L-type, Ca2+ channels (Liang et al. 2012; Tarr et al. 2013). We developed GV-58 as a synthetic analogue of (R)-roscovitine, a trisubstituted purine derivative that is best known as an inhibitor of cyclin-dependent kinases (Cdks) (De Azevedo et al. 1997; Meijer et al. 1997). However, a previous study found that (R)-roscovitine also has unexpected Ca2+ channel agonist activity (Yan et al. 2002). In order to develop a compound to selectively increase chemical transmitter release at diseased synapses, we sought to modify (R)-roscovitine with the goal of decreasing Cdk antagonist activity and enhancing Ca2+ channel agonist activity (Liang et al. 2012). We selected GV-58 as our primary lead structure for further evaluation as it showed the most desirable characteristics as a potential treatment for neuromuscular disease, including a 22-fold reduction in Cdk inhibitory activity and a four-fold increase in Ca2+ channel agonist effect (Liang et al. 2012; Tarr et al. 2013).

In the present study, we evaluated the effects of GV-58 on cell toxicity and survival in a cultured cell assay designed to reveal potential off-target effects on Cdks. Next, we characterized the voltage-dependent properties of the GV-58 effect on Ca2+ channels, the results of which predicted that a combination treatment with 3,4-DAP would be supra-additive. We then tested the hypothesis that GV-58 would work in combination with 3,4-DAP to create a supra-additive effect at LEMS model mouse NMJs that would completely reverse the deficit in neurotransmitter release at the NMJ caused by the passive transfer of LEMS to mice. Lastly, we characterized the effects of this novel treatment approach on short-term synaptic plasticity at NMJs in a LEMS model mouse.

Methods

Ethical approval

Serum from LEMS patients was collected in line with the guidelines set forth by the University of Pittsburgh Institutional Review Board. Animal studies were performed according to protocols approved by the University of Pittsburgh Institutional Animal Care and Use Committee (IACUC).

Cell lines

For the evaluation of effects of GV-58 on P/Q-type channels, tsA-201 cells were transiently transfected with Cav2.1 in combination with Cavβ3 and Cavα2δ1 (Addgene, Inc., Cambridge, MA, USA) using FuGENE 6 (Promega Corp., Madison, WI, USA). SH-SY5Y cells (a kind gift from Dr Susan G. Amara) were used to evaluate Cdk antagonist effects in the cell survival assay. All cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (tsA-201) or 15% (SH-SY5Y) fetal bovine serum.

Cell survival assay

An MTS-based cell survival assay (Ribas & Boix, 2004) using SH-SY5Y cells was used to test Cdk antagonist effects in the presence of physiological levels of ATP. SH-SY5Y is a cell line originating from a human neuroblastoma and has been used in previous studies that examine the effect of Cdk inhibition on apoptosis (Ribas & Boix, 2004; Ribas et al. 2006). Briefly, SH-SY5Y cells were first plated into 96 well clear-bottom plates 1 day prior to drug application. After 24 h of drug treatment, an MTS reagent (CellTiter 96 Kit; Promega Corp.) was added and absorbance at 490 nm was determined using an Infinite 200 PRO microplate reader (Tecan Trading AG, Männedorf, Switzerland). The absorbance values in the drug-treated wells were normalized to the absorbance values in wells containing the vehicle (0.05% DMSO). Background absorbance was determined in wells containing no cells and was subtracted from all values.

Whole-cell perforated patch clamp recordings

To assess the Ca2+ channel agonist effects of GV-58, whole-cell currents through Ca2+ channels were recorded using perforated patch methods as previously described (Tarr et al. 2013). The pipette solution consisted of 70 mm Cs2SO4, 60 mm CsCl, 1 mm MgCl2 and 10 mm Hepes, at pH 7.4. The extracellular saline contained 130 mm choline chloride, 10 mm tetraethylammonium chloride (TEA-Cl), 2 mm CaCl2, 1 mm MgCl2 and 10 mm Hepes, at pH 7.4. All standard chemicals were obtained from Sigma-Aldrich Corp. (St Louis, MO, USA). Patch pipettes were fabricated from borosilicate glass and pulled to a resistance of ∼1 MΩ. Capacitive currents and passive membrane responses to voltage commands were subtracted from the data. A liquid junction potential of −11.7 mV was subtracted during recordings. Currents were activated by a depolarizing step from −100 mV to +20 mV, amplified by an Axopatch 200B amplifier, filtered at 5 kHz, and digitized at 10 kHz for subsequent analysis using pClamp 10 software (Molecular Devices, LLC, Sunnyvale, CA, USA). The tail current integral was measured before and after the application of a compound and the integral of each trace was normalized to its peak. All experiments were carried out at room temperature (22°C). GV-58 was bath-applied via a glass pipette in a ∼0.5 ml static bath chamber during whole-cell recordings of Ca2+ current.

LEMS passive transfer

To test GV-58 in a LEMS model NMJ, we utilized an established LEMS passive transfer mouse model (Fukunaga et al. 1983; Lang et al. 1984; Fukuoka et al. 1987; Smith et al. 1995; Xu et al. 1998; Flink & Atchison, 2002). Serum from patient aBC2 [Ca2+ channel antibody titre: 3.2 fmol L−1 (Tarr et al. 2013)] was used for all LEMS passive transfer models reported here and was collected using plasmapheresis. Control serum was obtained from the University of Pittsburgh Medical Center blood bank. The serum was filtered with a 0.22 μm filter prior to the injection protocol. Adult female CFW mice (16 mice aged 2–3 months at the beginning of passive transfer and weighing 25–32 g; Charles River Laboratories, Inc., Wilmington, MA, USA) received one i.p. injection of LEMS or control serum on day 1, and then an injection of 300 mg kg−1 cyclophosphamide on day 2 to suppress the specific immune response to human IgG. This was followed by an i.p. injection of 1.5 ml LEMS or control serum once per day for 15–30 consecutive days. In all cases, experimenters were blinded to the injection conditions.

Intracellular recordings at mouse NMJs

Following the passive transfer protocol, mice were killed by CO2 inhalation followed by thoracotomy in accordance with procedures approved by the University of Pittsburgh IACUC, and intracellular recordings to assess the LEMS-mediated deficit in transmitter release were made in the mouse epitrochleoanconeus (ETA) ex vivo nerve–muscle preparation as previously described (Tarr et al. 2013). The extracellular saline contained 150 mm NaCl, 5 mm KCl, 11 mm dextrose, 10 mm Hepes, 1 mm MgCl2 and 2 mm CaCl2 (pH 7.3–7.4). The nerve was stimulated with a suction electrode and muscle contractions were blocked by exposure to 1 μm μ-conotoxin GIIIB (Alomone Labs Ltd, Jerusalem, Israel) (Hong & Chang, 1989). Microelectrode recordings were performed using ∼40–60 MΩ borosilicate electrodes filled with 3M potassium acetate. Spontaneous miniature synaptic events [miniature endplate potentials (mEPPs)] were collected for 1–2 min in each muscle fibre, and then 10–30 nerve-evoked synaptic events [endplate potentials (EPPs)] were collected with an inter-stimulus interval of 5 s. Each digitized point in each trace was corrected for non-linear summation (McLachlan & Martin, 1981). To calculate QC, the integral of signal under the average EPP waveform was divided by the integral of signal under the average mEPP waveform recorded from each NMJ. This ratio calculates the average number of quanta that are released following each presynaptic action potential. To evaluate effects on short-term synaptic plasticity, a train of 10 EPPs with an inter-stimulus interval of 20 ms (50 Hz) was collected in each muscle fibre. In some recordings the protocol involved first performing vehicle (0.05% v/v DMSO) control recordings, then recording in the same muscle fibres after a 30–60 min incubation in either 50 μm GV-58 or 1.5 μm 3,4-DAP, and finally recording in the same muscle fibres again after a 30–60 min incubation in a combination of 50 μm GV-58 plus 1.5 μm 3,4-DAP. In other cases, we recorded only vehicle controls in a group of muscle fibres prior to recording in the same muscle fibres following a 30–60 min incubation in a combination of 50 μm GV-58 plus 1.5 μm 3,4-DAP. There was no significant difference in QC between a sequential (vehicle→GV-58 or 3,4-DAP→GV-58 and 3,4-DAP; 104.60 ± 5.45, n = 41) and a non-sequential (vehicle→GV-58 plus 3,4-DAP; 106.00 ± 5.53, n = 22) application of drugs (P = 0.86, Student's t test). Similarly, there was no significant difference in short-term plasticity between a sequential (normalized 10th EPP: 0.725 ± 0.025, n = 45) and a non-sequential (normalized 10th EPP: 0.737 ± 0.024, n = 30) application of drugs (P = 0.72, Student's t test). Therefore, the data from both the sequential and non-sequential application protocols were pooled. Recordings made in 1.5 μm 3,4-DAP also contained the vehicle (0.05% v/v DMSO). A 3,4-DAP concentration of 1.5 μm was chosen because previous studies have reported that oral administration of 3,4-DAP to patients leads to peak serum levels of ∼70–150 ng ml−1 (Aisen et al. 1995; Wirtz et al. 2009), which corresponds to a concentration of ∼0.5–1.5 μm. We chose a GV-58 concentration of 50 μm because this concentration showed maximal agonist effect in patch clamp studies of Ca2+ current (Liang et al. 2012; Tarr et al. 2013). Data were collected using an Axoclamp 900A and digitized at 10 kHz for subsequent analysis using pClamp 10 software.

Statistical analysis

Statistical analysis was performed using either GraphPad Prism Version 5.0 (GraphPad Software, Inc., La Jolla, CA, USA) or Origin 7 (OriginLab Corp., Northampton, MA, USA). Data are presented as the mean ± s.e.m. unless otherwise noted. The significance level was set at P < 0.05 for all statistical tests.

Results

Reduced Cdk activity of GV-58

Our initial screens of GV-58 for Cdk antagonist activity were performed in low ATP levels (10 μm) (Liang et al. 2012; Tarr et al. 2013). However, because (R)-roscovitine and GV-58 inhibit Cdks by competing with ATP at the ATP binding site on Cdks (De Azevedo et al. 1997), we compared the Cdk inhibitory effects of (R)-roscovitine and GV-58 under physiological ATP conditions in a cell survival assay (Fig.1) (Ribas & Boix, 2004). Using this assay, we found an IC50 of 52.6 μm for (R)-roscovitine (n = 6 at each concentration), which is similar to previous results obtained in mammalian cell lines (Meijer et al. 1997; Ribas & Boix, 2004). Interestingly, GV-58 did not reduce cell survival at concentrations up to 50 μm, and caused only an 8% reduction in cell survival at 100 μm (n = 12 at each concentration; P < 0.05; one-sample t test), indicating that, in the presence of physiological levels of ATP in an in vitro assay, GV-58 does not inhibit Cdks in the concentration range used for Ca2+ channel agonist effects (≤50 μm) (Fig.1).

Figure 1. GV-58 does not inhibit cell survival at concentrations required for Ca2+ channel agonist effects.

Figure 1

Open in a new tab

Dose–response curve for inhibition of SH-SY5Y cell survival with (R)-roscovitine and GV-58 in an in vitro assay with physiological levels of ATP. Cell survival was measured at six different concentrations of (R)-roscovitine (n = 6 at each concentration) or GV-58 (n = 12 at each concentration). Data are normalized to values recorded in control cells and expressed as the mean ± s.e.m. for each concentration. The dotted line represents no change in cell survival. The (R)-roscovitine data are fit using the Hill equation for dose–response relationships.

Voltage-dependent effects of GV-58 on Ca2+ channels

Previous studies with (R)-roscovitine have shown that it preferentially affects to Ca2+ channels in the open conformation in order to induce agonist effects (Buraei et al. 2005). To test if GV-58 has a similar mechanism of action, we performed whole-cell patch clamp recordings on cell lines expressing P/Q-type Ca2+ channels and gave square-step depolarizations to +20 mV of varying durations in the presence of 50 μm GV-58 (Fig.2A and B). A long, 100 ms square-step resulted in a strong GV-58 agonist effect on the Ca2+ channel tail current; therefore, in our analysis, all other square-step durations were normalized to this value. As the duration of the square-step was decreased, the magnitude of the GV-58 agonist effect also decreased. Nevertheless, GV-58 still elicited an approximately six-fold increase in the Ca2+ channel tail current during very short duration square-steps of 1 ms (Fig.2B), which is similar to the half-width of a presynaptic action potential. Not only do these data suggest that GV-58 preferentially affects to Ca2+ channels that are in open conformation, but they also support the notion that GV-58 can significantly increase Ca2+ entry during physiological stimulus conditions.

Figure 2. GV-58 elicits Ca2+ channel agonist effects by preferentially affecting to the open conformation of the channel.

Figure 2

Open in a new tab

A, representative traces for the voltage step to +20 mV for 1 ms, 2 ms and 100 ms (top) and the resultant Ca2+ current traces (bottom) in the presence of 50 μm GV-58 (tail currents are shown starting at the end of the depolarizing step). A control trace in response to a 100 ms step prior to application of GV-58 is shown for comparison. The amplitude of each current trace is scaled to the amplitude of the control current trace to allow the comparison of decay kinetics. The dashed line indicates the portion of voltage traces (top) that correspond to the sample current traces (bottom). B, the Ca2+ channel agonist effect of GV-58 at voltage steps to +20 mV of 1 ms, 2 ms, 3 ms, 5 ms, 10 ms and 100 ms duration (n = 3–9 for each duration). Each GV-58-modified tail current integral was first normalized to its peak and then normalized to the control current integral prior to GV-58 application. Data are expressed as the mean ± s.e.m. and fit using the Hill equation for dose–response relationships.

Supra-additive effects of GV-58 plus 3,4-DAP on LEMS model mouse NMJs

The patch clamp data presented in Fig.2 show that GV-58 has a greater effect when more Ca2+ channels are open, which would occur when the depolarizing stimulus is longer in duration. Because 3,4-DAP induces the opening of more Ca2+ channels by prolonging the duration of the presynaptic action potential (Kirsch & Narahashi, 1978), we sought to investigate the intriguing possibility that 3,4-DAP and GV-58 would interact to cause a supra-additive effect. To test this hypothesis, we performed intracellular microelectrode recordings on ex vivo nerve–muscle preparations taken from LEMS passive transfer model mice and measured the magnitude of acetylcholine released at the NMJ. The most sensitive method of quantifying the magnitude of acetylcholine released is to determine the QC. We determined the QC by first measuring the area under the average action potential-evoked EPP waveform and then dividing this value by the area under the average single vesicle release event (mEPP) waveform. Using this approach, we compared the QC among five experimental conditions: control serum NMJs; LEMS NMJs in the vehicle (0.05% DMSO); LEMS NMJs exposed to 50 μm GV-58; LEMS NMJs exposed to 1.5 μm 3,4-DAP, and LEMS NMJs exposed to 50 μm GV-58 plus 1.5 μm 3,4-DAP (Fig.3). We chose to use a concentration of 1.5 μm 3,4-DAP because previous studies have reported the oral administration of 3,4-DAP in patients to lead to peak serum levels of ∼70–150 ng ml−1 (Aisen et al. 1995; Wirtz et al. 2009), which corresponds to a concentration of ∼0.5–1.5 μm. NMJs taken from LEMS model mice showed significantly reduced QC (n = 63 terminals; QC 26.7 ± 1.4; EPP amplitude 10.18 ± 0.62 mV) compared with control serum-treated mouse NMJs (n = 41 terminals; QC 107.5 ± 3.6; EPP amplitude 34.62 ± 1.37 mV) (P < 0.05, one-way ANOVA with Tukey's post hoc test) (Tarr et al. 2013). After exposure to 50 μm GV-58, the QC in LEMS NMJs was significantly larger (n = 20 terminals; QC 48.4 ± 2.7; EPP amplitude 13.75 ± 1.244 mV) than the QC in vehicle controls (P < 0.05, one-way ANOVA with Tukey's post hoc test). In fact, this GV-58-mediated enhancement was very similar to the significant enhancement observed after exposing LEMS model NMJs to 1.5 μm 3,4-DAP (n = 21 terminals; QC 49.0 ± 4.4; EPP amplitude 17.94 ± 1.381 mV) (P < 0.05, one-way ANOVA with Tukey's post hoc test). Interestingly, when LEMS model NMJs were exposed to a combination of 50 μm GV-58 plus 1.5 μm 3,4-DAP, QC increased (n = 63 terminals; QC 105.1 ± 4.0; EPP amplitude 32.30 ± 1.85 mV) (P < 0.05, one-way ANOVA with Tukey's post hoc test) such that it was not significantly different from the QC measured from NMJs taken from control serum-treated mice (P ≥ 0.05, one-way ANOVA with Tukey's post hoc test). These data indicate that the magnitude of transmitter release in LEMS model NMJs was completely restored to control levels under conditions of exposure to both GV-58 and 3,4-DAP (Fig.3B). The combination of these two drugs also slightly, but significantly, prolonged the duration of the EPP [full width at half-maximum 3.08 ± 0.05 ms and 3.84 ± 0.07 ms for control serum (n = 41 terminals) and 50 μm GV-58 plus 1.5 μm 3,4-DAP (n = 63 terminals), respectively; P < 0.05, Student's independent-samples t test]. The potential impact of this small change in EPP duration on neuromuscular function in vivo is unknown.

Figure 3. The supra-additive effect of GV-58 plus 3,4-DAP completely reverses the deficit in the magnitude of neurotransmitter release.

Figure 3

Open in a new tab

A, sample traces showing average endplate potential (EPP) amplitudes following exposure to the indicated conditions. Left: sample average EPP recorded from a neuromuscular junction (NMJ) in a control serum-treated mouse. Middle: sample average EPPs of a Lambert–Eaton myasthenic syndrome (LEMS) model NMJ before and after application of 50 μm GV-58. Right: sample average EPPs recorded from a LEMS model NMJ before drug application (LEMS), following application of 1.5 μm 3,4-DAP, and following application of 50 μm GV-58 plus 1.5 μm 3,4-DAP. B, quantal content for NMJs in each of the five conditions: NMJs from control serum-treated mice (n = 41); LEMS model NMJs in the presence of vehicle (n = 63); LEMS model NMJs following application of 50 μm GV-58 (n = 20); LEMS model NMJs following application of 1.5 μm 3,4-DAP (n = 21), and LEMS model NMJs following application of 50 μm GV-58 plus 1.5 μm 3,4-DAP (n = 63). Data are represented as the mean ± s.e.m.

Effects of GV-58 plus 3,4-DAP on short-term synaptic plasticity

In addition to measuring the properties of individual action potential-evoked events, we also measured and compared short-term synaptic plasticity characteristics among all conditions by eliciting a train of 10 stimuli at 50 Hz (Fig.4). In NMJs taken from mice injected with control patient serum, the magnitude of transmitter release did not change much during the first few stimuli in the train, and the 10th EPP in the train depressed to 66% of the first EPP (n = 41 terminals) (Tarr et al. 2013). By contrast, LEMS model NMJs showed strong facilitation throughout the train, with the 10th EPP showing facilitation to ∼148% of the first EPP (n = 75 terminals). Both the 50 μm GV-58 condition (10th EPP at ∼123% of the first EPP; n = 24 terminals) and the 1.5 μm 3,4-DAP condition (10th EPP at ∼132% of the first EPP; n = 21 terminals) showed only a partial restoration of normal short-term plasticity characteristics. However, when the combination of 50 μm GV-58 plus 1.5 μm 3,4-DAP was applied, there was a near complete restoration of normal short-term plasticity characteristics, with a small amount of facilitation during the first few EPPs of the train and a depression at the 10th EPP to ∼73% of the first EPP (n = 75 terminals) (Fig.4).

Figure 4. The supra-additive effect of GV-58 plus 3,4-DAP elicits a near complete restoration of short-term synaptic plasticity characteristics.

Figure 4

Open in a new tab

A, sample endplate potentials (EPPs) recorded from neuromuscular junctions (NMJs) in each of the five conditions during a 50 Hz train of 10 stimuli. Dashed lines indicate the amplitude of the first EPP in each train. B, average change in EPP amplitude during a 50 Hz stimulus train for NMJs from control serum-treated mice (n = 41), Lambert–Eaton myasthenic syndrome (LEMS) model NMJs in vehicle (n = 75), LEMS model NMJs after application of 50 μm GV-58 (n = 24), LEMS model NMJs following application of 1.5 μm 3,4-DAP (n = 21), and LEMS model NMJs following application of 50 μm GV-58 plus 1.5 μm 3,4-DAP (n = 75). Each EPP in the train is first normalized to the first EPP in the train before responses from many trials are averaged. The average normalized values are then plotted for each treatment condition. Data are represented as the mean ± s.e.m. The dashed line represents no change from the amplitude of the first EPP in the train.

Discussion

The present study further evaluates the potential of GV-58 as a novel treatment for LEMS. In particular, we show that 50 μm GV-58 is not toxic to dividing cells in culture, suggesting that GV-58 is not an effective Cdk inhibitor under physiological ATP conditions at concentrations that show maximal effects on Ca2+ channels. This is in contrast with findings of inhibitory effects of GV-58 at lower concentrations using a Cdk kinase assay with very low concentrations of ATP (Liang et al. 2012; Tarr et al. 2013). These data indicate that GV-58 competes with ATP at the ATP binding site on Cdks; therefore, given the high physiological levels of intracellular ATP (Maechler et al. 1998; Kennedy et al. 1999), these data mitigate the potential concern that Cdk inhibition will impede the development of optimized (R)-roscovitine analogues, such as GV-58, as potential therapeutics in LEMS.

Whereas either GV-58 or 3,4-DAP in isolation caused an increase of about 80% in the magnitude of transmitter release from LEMS model NMJs, the combination of GV-58 plus 3,4-DAP elicited an increase of approximately 300% in transmitter release magnitude, which is beyond that expected of a simple additive effect. This supra-additive effect is not surprising given the mechanistic actions of the two compounds. By blocking potassium channels, 3,4-DAP causes a broadening of the presynaptic action potential (Kirsch & Narahashi, 1978), which, in turn, leads to an increase in the number of open Ca2+ channels per action potential. GV-58 is a Ca2+ channel agonist that increases the amount of Ca2+ influx by prolonging the duration for which Ca2+ channels will remain open only after they have been induced to open by voltage depolarization. When both compounds are applied simultaneously, more Ca2+ channels are opened as a result of the action of 3,4-DAP, and GV-58 can therefore elicit a greater increase in Ca2+ influx than it can when applied alone because more open channels are available. The interaction between these two compounds leads to a complete restoration of neurotransmitter release magnitude in LEMS model mouse NMJs (Fig.3). In addition, the combination of GV-58 and 3,4-DAP caused a small increase (∼0.76 ms) in the duration of the EPP waveform. Although we cannot know how this might affect neuromuscular synaptic transmission, we do not believe it to be a major issue because the change is small.

The slight differences in short-term synaptic plasticity that persist in LEMS model NMJs treated with GV-58 plus 3,4-DAP may reflect several factors. Firstly, the combined effects of GV-58 and 3,4-DAP can completely restore transmitter release magnitude in LEMS model NMJs because both the probability of opening (3,4-DAP) and the flux of Ca2+ (GV-58) through the reduced number of Ca2+ channels that remain in the active zones of these LEMS model NMJs are enhanced. The enhanced Ca2+ flux at the reduced number of Ca2+ entry sites in these nerve terminals would be predicted to create a different spatial and temporal profile of presynaptic Ca2+ concentration following each action potential, which may enhance the residual Ca2+ effects that critically influence short-term synaptic plasticity. Secondly, previous freeze-fracture electron microscopic studies of LEMS active zones have revealed a disruption in the organization of presynaptic proteins (presumed to include Ca2+ channels) (Fukunaga et al. 1983; Fukuoka et al. 1987; Nagel et al. 1988). If this disruption changes the spatial distance between the remaining presynaptic Ca2+ channels and docked synaptic vesicles that are ready for release, this may also affect short-term synaptic plasticity at these synapses. The potential impact of both of these issues at LEMS model synapses will require further study.

Overall, our data show that exposure of LEMS model mouse NMJs to a combination of 3,4-DAP plus our novel Ca2+ channel agonist (GV-58) completely reverses the deficit in neurotransmitter release magnitude that underlies the neuromuscular weakness that is characteristic of LEMS NMJs. This effect of the two compounds is not simply additive, but is supra-additive based on the mechanism of action of each compound. These data lead us to propose a new combination treatment strategy for LEMS that should be explored further in pre-clinical evaluation, including in vivo animal studies and studies to determine factors such as off-target effects, toxicity and blood–brain barrier penetrance. Lastly, this treatment strategy may also prove beneficial for those with disorders that respond positively to 3,4-DAP treatment (Sedehizadeh et al. 2012), such as muscle-specific kinase (MuSK) myasthenia (Mori et al. 2012; Morsch et al. 2013), congenital myasthenic syndromes (Schara et al. 2012), botulism (Adler et al. 2012), and amyotrophic lateral sclerosis (Bertorini et al. 2011).

Acknowledgments

We thank S. G. Amara (National Institute of Mental Health) and C. B. Divito for providing the SH-SY5Y cell line and for helping with the cell viability experiments, the LEMS patients for providing serum and plasma exchange samples with consent, C. Meriney for help with data analysis, and A. E. Homan for help with data analysis and for critically reading the manuscript.

Glossary

Cdk

cyclin-dependent kinase

CMAP

compound muscle action potential

3,4-DAP

3,4-diaminopyridine

EPP

endplate potential

ETA

epitrochleoanconeus

LEMS

Lambert–Eaton myasthenic syndrome

mEPP

miniature endplate potential

MuSK

muscle-specific kinase

NMJ

neuromuscular junction

QC

quantal content

Key points

  • Lambert–Eaton myasthenic syndrome (LEMS) is characterized by an autoimmune-mediated attack on presynaptic P/Q-type Ca2+ channels at the neuromuscular junction (NMJ).

  • The current common symptomatic treatment option is 3,4-diaminopyridine (3,4-DAP), a potassium channel blocker that widens the presynaptic action potential, causing an increase in the amount of neurotransmitter release. This approach, however, does not completely reverse symptoms and can have dose-limiting side-effects. Thus, there is a need for additional treatment options.

  • We show that GV-58, a Ca2+ channel agonist developed from the cyclin-dependent kinase inhibitor (R)-roscovitine, does not significantly inhibit cell division at physiological levels of ATP.

  • We further show that GV-58 has a greater agonist effect when more Ca2+ channels are open, and combining GV-58 and 3,4-DAP elicits a supra-additive effect that completely restores the magnitude of neurotransmitter release in LEMS model NMJs.

  • These results suggest that a combination of GV-58 and 3,4-DAP is promising as a possible alternative treatment approach to LEMS and other neuromuscular diseases.

Additional information

Competing interests

None declared.

Author contributions

This work was performed in the laboratories of P.W. and S.D.M. at the University of Pittsburgh. T.B.T., D.L., S.W.R., P.W. and S.D.M. designed the research. T.B.T., M.L., G.V., M.F. and S.D.M. performed the research. T.B.T., P.W. and S.D.M. analysed the data. T.B.T., S.W.R., P.W. and S.D.M. wrote the manuscript. All authors approved the final version of the manuscript.

Funding

This work was supported by an Achievement Rewards for College Scientists (ARCS) Foundation scholarship (to T.B.T.) and grants from the National Science Foundation (0844604 and 1249546 to S.D.M.), the National Institutes of Health (GM067082 to P.W.), the Muscular Dystrophy Association (295271 to S.D.M.), the Myasthenia Associations of Australia and the Beeren Foundation (to S.W.R.), and the University of Pittsburgh Central Research Development Fund (to S.D.M).

References

  1. Adler M, Deshpande SS, Apland JP, Murray B, Borrell A. Reversal of BoNT/A-mediated inhibition of muscle paralysis by 3,4-diaminopyridine and roscovitine in mouse phrenic nerve–hemidiaphragm preparations. Neurochem Int. 2012;61:866–873. doi: 10.1016/j.neuint.2012.07.015. [DOI] [PubMed] [Google Scholar]
  2. Aisen ML, Sevilla D, Gibson G, Kutt H, Blau A, Edelstein L, Hatch J, Blass J. 3,4-diaminopyridine as a treatment for amyotrophic lateral sclerosis. J Neurol Sci. 1995;129:21–24. doi: 10.1016/0022-510x(94)00225-d. [DOI] [PubMed] [Google Scholar]
  3. Bertorini TE, Rashed H, Zeno M, Tolley EA, Igarashi M, Li YD. Effects of 3,4-diaminopyridine (DAP) in motor neuron diseases. J Clin Neuromuscul Dis. 2011;12:129–137. doi: 10.1097/CND.0b013e3182095b2b. [DOI] [PubMed] [Google Scholar]
  4. Buraei Z, Anghelescu M, Elmslie KS. Slowed N-type calcium channel (CaV2.2) deactivation by the cyclin-dependent kinase inhibitor roscovitine. Biophys J. 2005;89:1681–1691. doi: 10.1529/biophysj.104.052837. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. De Azevedo WF, Leclerc S, Meijer L, Havlicek L, Strnad M, Kim SH. Inhibition of cyclin-dependent kinases by purine analogues: crystal structure of human Cdk2 complexed with roscovitine. Eur J Biochem. 1997;243:518–526. doi: 10.1111/j.1432-1033.1997.0518a.x. [DOI] [PubMed] [Google Scholar]
  6. Flink MT, Atchison WD. Passive transfer of Lambert–Eaton syndrome to mice induces dihydropyridine sensitivity of neuromuscular transmission. J Physiol. 2002;543:567–576. doi: 10.1113/jphysiol.2002.021048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Fukunaga H, Engel AG, Lang B, Newsom-Davis J, Vincent A. Passive transfer of Lambert–Eaton myasthenic syndrome with IgG from man to mouse depletes the presynaptic membrane active zones. Proc Natl Acad Sci U S A. 1983;80:7636–7640. doi: 10.1073/pnas.80.24.7636. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fukuoka T, Engel AG, Lang B, Newsom-Davis J, Vincent A. Lambert–Eaton myasthenic syndrome: II. Immunoelectron microscopy localization of IgG at the mouse motor end-plate. Ann Neurol. 1987;22:200–211. doi: 10.1002/ana.410220204. [DOI] [PubMed] [Google Scholar]
  9. Hong SJ, Chang CC. Use of geographutoxin II (μ-conotoxin) for the study of neuromuscular transmission in mouse. Br J Pharmacol. 1989;97:934–940. doi: 10.1111/j.1476-5381.1989.tb12034.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kennedy HJ, Pouli AE, Ainscow EK, Jouaville LS, Rizzuto R, Rutter GA. Glucose generates sub-plasma membrane ATP microdomains in single islet β-cells. Potential role for strategically located mitochondria. J Biol Chem. 1999;274:13281–13291. doi: 10.1074/jbc.274.19.13281. [DOI] [PubMed] [Google Scholar]
  11. Kirsch GE, Narahashi T. 3,4-diaminopyridine. A potent new potassium channel blocker. Biophys J. 1978;22:507–512. doi: 10.1016/S0006-3495(78)85503-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lambert EH, Eaton LM, Rooke ED. Defect of neuromuscular conduction associated with malignant neoplasms. Am J Physiol. 1956;187:612–613. [Google Scholar]
  13. Lang B, Molenaar PC, Newsom-Davis J, Vincent A. Passive transfer of Lambert–Eaton myasthenic syndrome in mice: decreased rates of resting and evoked release of acetylcholine from skeletal muscle. J Neurochem. 1984;42:658–662. doi: 10.1111/j.1471-4159.1984.tb02733.x. [DOI] [PubMed] [Google Scholar]
  14. Lang B, Vincent A, Murray NM, Newsom-Davis J. Lambert–Eaton myasthenic syndrome: immunoglobulin G inhibition of Ca2+ flux in tumor cells correlates with disease severity. Ann Neurol. 1989;25:265–271. doi: 10.1002/ana.410250310. [DOI] [PubMed] [Google Scholar]
  15. Lennon VA, Kryzer TJ, Griesmann GE, O'Suilleabhain PE, Windebank AJ, Woppmann A, Miljanich GP, Lambert EH. Calcium-channel antibodies in the Lambert–Eaton syndrome and other paraneoplastic syndromes. N Engl J Med. 1995;332:1467–1474. doi: 10.1056/NEJM199506013322203. [DOI] [PubMed] [Google Scholar]
  16. Liang M, Tarr TB, Bravo-Altamirano K, Valdomir G, Rensch G, Swanson L, DeStefino NR, Mazzarisi CM, Olszewski RA, Wilson GM, Meriney SD, Wipf P. Synthesis and biological evaluation of a selective N- and P/Q-type calcium channel agonist. ACS Med Chem Lett. 2012;3:985–990. doi: 10.1021/ml3002083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lindquist S, Stangel M. Update on treatment options for Lambert–Eaton myasthenic syndrome: focus on use of amifampridine. Neuropsychiatr Dis Treat. 2011;7:341–349. doi: 10.2147/NDT.S10464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Maechler P, Wang H, Wollheim CB. Continuous monitoring of ATP levels in living insulin secreting cells expressing cytosolic firefly luciferase. FEBS Lett. 1998;422:328–332. doi: 10.1016/s0014-5793(97)01618-9. [DOI] [PubMed] [Google Scholar]
  19. McLachlan EM, Martin AR. Non-linear summation of end-plate potentials in the frog and mouse. J Physiol. 1981;311:307–324. doi: 10.1113/jphysiol.1981.sp013586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Meijer L, Borgne A, Mulner O, Chong JP, Blow JJ, Inagaki N, Inagaki M, Delcros JG, Moulinoux JP. Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5. Eur J Biochem. 1997;243:527–536. doi: 10.1111/j.1432-1033.1997.t01-2-00527.x. [DOI] [PubMed] [Google Scholar]
  21. Meriney SD, Hulsizer SC, Lennon VA, Grinnell AD. Lambert–Eaton myasthenic syndrome immunoglobulins react with multiple types of calcium channels in small-cell lung carcinoma. Ann Neurol. 1996;40:739–749. doi: 10.1002/ana.410400510. [DOI] [PubMed] [Google Scholar]
  22. Mori S, Kishi M, Kubo S, Akiyoshi T, Yamada S, Miyazaki T, Konishi T, Maruyama N, Shigemoto K. 3,4-Diaminopyridine improves neuromuscular transmission in a MuSK antibody-induced mouse model of myasthenia gravis. J Neuroimmunol. 2012;245:75–78. doi: 10.1016/j.jneuroim.2012.02.010. [DOI] [PubMed] [Google Scholar]
  23. Morsch M, Reddel SW, Ghazanfari N, Toyka KV, Phillips WD. Pyridostigmine but not 3,4-diaminopyridine exacerbates ACh receptor loss and myasthenia induced in mice by muscle-specific kinase autoantibody. J Physiol. 2013;591:2747–2762. doi: 10.1113/jphysiol.2013.251827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Motomura M, Lang B, Johnston I, Palace J, Vincent A, Newsom-Davis J. Incidence of serum anti-P/O-type and anti-N-type calcium channel autoantibodies in the Lambert–Eaton myasthenic syndrome. J Neurol Sci. 1997;147:35–42. doi: 10.1016/s0022-510x(96)05303-8. [DOI] [PubMed] [Google Scholar]
  25. Nagel A, Engel AG, Lang B, Newsom-Davis J, Fukuoka T. Lambert–Eaton myasthenic syndrome IgG depletes presynaptic membrane active zone particles by antigenic modulation. Ann Neurol. 1988;24:552–558. doi: 10.1002/ana.410240412. [DOI] [PubMed] [Google Scholar]
  26. Ribas J, Boix J. Cell differentiation, caspase inhibition, and macromolecular synthesis blockage, but not BCL-2 or BCL-XL proteins, protect SH-SY5Y cells from apoptosis triggered by two CDK inhibitory drugs. Exp Cell Res. 2004;295:9–24. doi: 10.1016/j.yexcr.2003.12.019. [DOI] [PubMed] [Google Scholar]
  27. Ribas J, Boix J, Meijer L. R)-roscovitine (CYC202, Seliciclib) sensitizes SH-SY5Y neuroblastoma cells to nutlin-3-induced apoptosis. Exp Cell Res. 2006;312:2394–2400. doi: 10.1016/j.yexcr.2006.04.021. [DOI] [PubMed] [Google Scholar]
  28. Schara U, Della Marina A, Abicht A. Congenital myasthenic syndromes: current diagnostic and therapeutic approaches. Neuropediatrics. 2012;43:184–193. doi: 10.1055/s-0032-1323850. [DOI] [PubMed] [Google Scholar]
  29. Sedehizadeh S, Keogh M, Maddison P. The use of aminopyridines in neurological disorders. Clin Neuropharmacol. 2012;35:191–200. doi: 10.1097/WNF.0b013e31825a68c5. [DOI] [PubMed] [Google Scholar]
  30. Smith DO, Conklin MW, Jensen PJ, Atchison WD. Decreased calcium currents in motor nerve terminals of mice with Lambert–Eaton myasthenic syndrome. J Physiol. 1995;487:115–123. doi: 10.1113/jphysiol.1995.sp020865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tarr TB, Malick W, Liang M, Valdomir G, Frasso M, Lacomis D, Reddel SW, Garcia-Ocano A, Wipf P, Meriney SD. Evaluation of a novel calcium channel agonist for therapeutic potential in Lambert–Eaton myasthenic syndrome. J Neurosci. 2013;33:10559–10567. doi: 10.1523/JNEUROSCI.4629-12.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Titulaer MJ, Maddison P, Sont JK, Wirtz PW, Hilton-Jones D, Klooster R, Willcox N, Potman M, Sillevis Smitt PA, Kuks JB, Roep BO, Vincent A, van der Maarel SM, van Dijk JG, Lang B, Verschuuren JJ. Clinical Dutch–English Lambert–Eaton myasthenic syndrome (LEMS) tumor association prediction score accurately predicts small-cell lung cancer in the LEMS. J Clin Oncol. 2011a;29:902–908. doi: 10.1200/JCO.2010.32.0440. [DOI] [PubMed] [Google Scholar]
  33. Titulaer MJ, Lang B, Verschuuren JJ. Lambert–Eaton myasthenic syndrome: from clinical characteristics to therapeutic strategies. Lancet Neurol. 2011b;10:1098–1107. doi: 10.1016/S1474-4422(11)70245-9. [DOI] [PubMed] [Google Scholar]
  34. Titulaer MJ, Wirtz PW, Kuks JB, Schelhaas HJ, van der Kooi AJ, Faber CG, van der Pol WL, de Visser M, Sillevis Smitt PA, Verschuuren JJ. The Lambert–Eaton myasthenic syndrome 1988–2008: a clinical picture in 97 patients. J Neuroimmunol. 2008;201–202:153–158. doi: 10.1016/j.jneuroim.2008.05.025. [DOI] [PubMed] [Google Scholar]
  35. Vincent A, Lang B, Newsom-Davis J. Autoimmunity to the voltage-gated calcium channel underlies the Lambert–Eaton myasthenic syndrome, a paraneoplastic disorder. Trends Neurosci. 1989;12:496–502. doi: 10.1016/0166-2236(89)90109-4. [DOI] [PubMed] [Google Scholar]
  36. Wirtz PW, Verschuuren JJ, van Dijk JG, de Kam ML, Schoemaker RC, van Hasselt JG, Titulaer MJ, Tjaden UR, den Hartigh J, van Gerven JM. Efficacy of 3,4-diaminopyridine and pyridostigmine in the treatment of Lambert–Eaton myasthenic syndrome: a randomized, double-blind, placebo-controlled, crossover study. Clin Pharmacol Ther. 2009;86:44–48. doi: 10.1038/clpt.2009.35. [DOI] [PubMed] [Google Scholar]
  37. Xu YF, Hewett SJ, Atchison WD. Passive transfer of Lambert–Eaton myasthenic syndrome induces dihydropyridine sensitivity of ICa in mouse motor nerve terminals. J Neurophysiol. 1998;80:1056–1069. doi: 10.1152/jn.1998.80.3.1056. [DOI] [PubMed] [Google Scholar]
  38. Yan Z, Chi P, Bibb JA, Ryan TA, Greengard P. Roscovitine: a novel regulator of P/Q-type calcium channels and transmitter release in central neurons. J Physiol. 2002;540:761–770. doi: 10.1113/jphysiol.2001.013376. [DOI] [PMC free article] [PubMed] [Google Scholar]

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

RESOURCES