A 5-H-DIBENZ [B,F] AZEPINE-5-C ARBOXAMIDE DERIVATIVE FOR USE IN A METHOD FOR PREVENTING OR TREATING ANDERMANN SYNDROME

FIELD OF THE INVENTION:

The invention relates to a 5H-dibenz[b,fJazepine-5-carboxamide derivative such as carbamazepine (CBZ) for use in a method for preventing or treating Andermann syndrome.

BACKGROUND OF THE INVENTION:

Andermann syndrome is an autosomal recessive neurodevelopmental and neurodegenerative disorder characterized by peripheral neuropathy with variable agenesis of the corpus callosum (ACCPN) ^ ² , hypotonia and amyotrophy ³ and is caused by mutations within the cation-chloride cotransporter KCC3 ⁴ . KCC3-deficient mice reproduce the typical ACCPN sensorimotor histophysiopathology ^4"6 , displaying impaired locomotor and sensorimotor gating capacities and decreased peripheral nerve conduction ^{4 7} . Axonal swelling, neurodegeneration and hypomyelination of the sciatic nerve seem to account for the peripheral neuropathy ^4"6 ' ⁸ . Both enveloping glial cells and demyelinating events were proposed as responsible for reduced nerve conduction and neurodegeneration in ubiquitously-deleted KCC3 mice and patients ⁴ ' ⁷ ' ⁹ . However, sciatic nerve analysis from neuron-specific KCC3-deleted mice reveals that neuronal loss of KCC3 is sufficient to induce peripheral neuropathy ⁶ . These findings thus support a cell-autonomous role for KCC3 in neurons, implying that the phenotype of ubiquitous mutant mice illustrates a primarily neuronal-dependent neuropathy.

KCC3 belongs to the family of cation-chloride cotransporters coded by the Slcl2a genes ¹⁰ . In the nervous system, the Slcl2a gene family comprises five members coding for one inwardly directed Na ⁺ -K ⁺ -2C1 ^" cotransporter (Slcl2a2 for NKCC1), and four outwardly directed K ⁺ -C1 ^" cotransporters, (Slcl2a4-7 for KCCl-4, respectively) ^u . These membrane proteins are responsible for various functions such as cell volume regulation and maintenance of intracellular chloride concentration [Cl ^" ]i. In neurons, only NKCC1, KCC2 and KCC3 have been shown to regulate the [Cl ^" ]i (reviewed in ¹² ). As a regulator of [Cl ^" ]i, KCC3 could participate in the establishment and/or maintenance of spinal cord synaptic networks ⁵ ' ¹³ ' ¹⁴ . Indeed, additional molecular mechanisms linking KCCs to electrical activity and ionic homeostasis have recently emerged. Of interest is the demonstrated interaction between Na ⁺ /K ⁺ -ATPase a2 and KCC2 in synaptic membrane fractions ¹⁵ as well as between the KCC3a isoform and the Na ⁺ /K ⁺ -ATPase al in gastric parietal cells ¹⁶ . The Na ⁺ /K ⁺ -ATPase is an ion pump essential for the maintenance of cellular electrochemical gradients (reviewed in ¹⁷ ) and neuromuscular junction (NMJ) organization ¹⁸ . Moreover, although axonal swelling is a hallmark of Andermann syndrome, it is surprising that the functionally related volume regulators KCC1 and KCC4 do not compensate for the loss of KCC3. In addition, impaired locomotor and sensorimotor gating capacities occur prior to axonal degeneration, suggesting that neuronal loss is in fact a consequence of the primary pathological mechanisms that cause Andermann syndrome ⁴ .

SUMMARY OF THE INVENTION:

In a first aspect, the invention relates to a 5H-dibenz[b,f]azepine-5-carboxamide derivative for use in a method for preventing or treating Andermann syndrome.

DETAILED DESCRIPTION OF THE INVENTION:

The invention relies on the unexpected discovery that KCC3 deletion induces profound structural modifications at the NMJ and skeletal muscle atrophy, prior to any spinal motoneuron degeneration. The inventors show that Andermann syndrome is not related to a chloride imbalance but most likely due to aberrant localization and decreased activity of the Na ⁺ /K ⁺ -ATPase al in the motoneuron soma. Thus, combined with the functional analysis of nerve-evoked muscle contraction in ACCPN patients, the present results highlight motoneuron-dependent defects as central in Andermann syndrome. More importantly, the inventors show that chronic carbamazepine (CBZ) treatment in vivo restores the normal expression of the Na ⁺ /K ⁺ -ATPase l at the plasma membrane of spinal motoneuron of Slcl2a6 ^~!~ mice and prevents, in vitro, the loss of activity-dependent plateau accommodation that occurs in Slcl2a6 ^~ ' ^~ motoneurons at a concentration ineffective on action potential amplitude. Importantly, CBZ partly prevents the denervation at the neuromuscular junction which makes this compound a therapeutically valuable tool for the treatment of Andermann syndrome.

Accordingly, in a first aspect, the invention relates to a 5H-dibenz[b,f]azepine-5- carboxamide derivative for use in a method for preventing or treating Andermann syndrome. In one embodiment of the invention, said derivative is a voltage-gated sodium channel (VGSC) blocker.

In one embodiment of the invention, said derivative is an anticonvulsant or antiepileptic drug.

In one embodiment of the invention, said derivative is a tricyclic antidepressant. In one embodiment of the invention, said derivative is a compound of Formula (I):

wherein Rl and R2 are independently selected from H, alkyl, hydroxyalkyl, amino, alkylamino, nitro, oxo, cyano, acyloxy preferably acetoxy, halogen, trifluoromethyl and hydroxy groups, and pharmaceutically acceptable salts thereof.

5H-dibenz[b,f]azepine-5-carboxamide derivatives and the syntheses for preparing them are known in the art and have been extensively described (see for instance U.S. Pat. No. 3,643,775; U.S. Pat. No. 3,367,667) and U.S. Pat. No. 4,076,812).

In one particular embodiment, the compound of Formula (I) is carbamazepine (5H- dibenz[b,f]azepine-5-carboxamide, CBZ, marketed as Tegretol® by Novartis), which is a widely used anticonvulsant agent having the structural formula:

The 5H-dibenz[b,f]azepine-5-carboxamide derivatives may be substituted or unsubstituted at the 10- or 11 -position. The 10 or 11 -positions may be substituted with, alkyl, hydroxyalkyl, amino, alkylamino, nitro, oxo, cyano, acyloxy such as acetoxy, halogen, trifluoromethyl and hydroxy groups, preferably oxo, acetoxy, or hydroxy groups. When there is an oxo, acetoxy or a hydroxy group at the 10-position, the 11 -position is preferably unsubstituted and vice versa. Accordingly, in addition to carbamazepine (5H-dibenz[b,f]azepine-5-carboxamide), examples of suitable compounds of the general formula I are: 10-oxo-10,l l-dihydro-5H- dibenz[b,f]azepine-5-carboxamide (cf. U.S. Pat. No. 3,643,775) and 10-hydroxy-10,l 1- dihydro-5H-dibenz[b,f]azepine-5-carboxamide (cf. U.S. Pat. No. 3,367,667) as well as 10- fluoro-, 10-chloro- and 10-bromo-5H-dibenz[b,fJazepine-5-carboxamide (cf. U.S. Pat. No. 4,076,812), and 10-cyano-5H-dibenz[b,fJazepine-5-carboxamide.

Preferred compounds are selected from carbamazepine (Tegretol®) (CBZ), oxacarbazepine (Trileptal ®), eslicarbazepine acetate (Zebinix®) (ESL). In one embodiment, the compound of Formula (I) is oxcarbazepine (10-oxo-10,l l- dihydro-5H-dibenz[b,f]azepine-5-carboxamide, OXC, marketed as Trileptal® by Novartis), which is also used as an anticonvulsant agent having the structural formula:

In one embodiment, the compound of Formula (I) is eslicarbazepine acetate ((S)-(-)- 10-Acetoxy-10,ll-dihydro-5H-dibenz[b,fJazepine-5-carboxamide , ESL, marketed as Aptiom® or Zebinix® by Eisai) which is an also used anticonvulsant agent having the structural formula:

Oral forms of 5H-dibenz[b,f]azepine-5-carboxamide derivatives arc well known to the skilled in the art and are suitable for repeated administration over a prolonged period of treatment to ensure a uniform concentration of active agent in the blood. Extended-release formulations (e.g. capsules or tablets) of 5H-dibenz[b.f]azepine-5-carboxamide derivatives such as carbamazepine (Equetro™) and oxcarbazepine (Oxtellar XR™) are also available. Alternatively, parenteral formulations, in particular which are suitable for use intravenously, are also known from the skilled in the art. Another aspect of the invention relates to a method for preventing or treating Andermann syndrome in a patient in need thereof comprising a step of administering an effective amount of a 5H-dibenz[b,f]azepine-5-carboxamide derivative to said patient. As used herein, the term "patient in need thereof denotes a mammal. Preferably, a patient according to the invention is a human (more preferably a newborn or a child) diagnosed with Andermann syndrome or having a risk of developing such syndrome

As used herein, the term "effective amount" is intended for a minimal amount of active agent, which is necessary to impart therapeutic benefit to a patient. For example, an "effective amount of a compound" to a subject is an amount of the compound that induces, ameliorates or causes an improvement in the pathological symptoms, disease progression, or physical conditions associated with the disease affecting the patient. A 5H-dibenz[b,fJazepine-5-carboxamide derivative as above described may be combined with pharmaceutically acceptable excipients, and optionally sustained-release matrices, such as biodegradable polymers, to form pharmaceutical compositions.

As used herein, the terms "Pharmaceutically" or "pharmaceutically acceptable" refer to molecular entities and compositions that do not produce an adverse, allergic or other untoward reaction when administered to a mammal, especially a human, as appropriate. A pharmaceutically acceptable carrier or excipient refers to a non-toxic solid, semi-solid or liquid filler, diluent, encapsulating material or formulation auxiliary of any type. The form of the pharmaceutical compositions, the route of administration, the dosage and the regimen naturally depend upon the condition to be treated, the severity of the illness, the age, weight, and sex of the patient, etc. The pharmaceutical compositions of the invention can be formulated for a topical, oral, intranasal, intraocular, intravenous, intramuscular or subcutaneous administration and the like.

The doses used for the administration can be adapted as a function of various parameters, and in particular as a function of the mode of administration used, of the relevant pathology, or alternatively of the desired duration of treatment. For example, it is well within the skill of the art to start doses of the compound at levels lower than those required to achieve the desired therapeutic effect and to gradually increase the dosage until the desired effect is achieved. However, the daily dosage of the products may be varied over a wide range from 0.01 to 2,000 mg per adult per day. Preferably, the compositions contain 0.01, 0.05, 0.1, 0.5, 1.0, 2.5, 5.0, 10.0, 15.0, 25.0, 50.0, 100, 250 and 500 mg of the active ingredient for the symptomatic adjustment of the dosage to the subject to be treated. A medicament typically contains from about 0.01 mg to about 500 mg of the active ingredient, preferably from 1 mg to about 100 mg of the active ingredient. An effective amount of the drug is ordinarily supplied at a dosage level from 0.0002 mg/kg to about 20 mg/kg of body weight per day, especially from about 0.001 mg/kg to 7 mg/kg of body weight per day.

In a particular embodiment of the invention, the 5H-dibenz[b,fJazepine-5-carboxamide derivative of the invention as described above such as CBZ is administered to the patient in need thereof at a concentration allowing an increase of the expression and the activity of Na ⁺ /K ⁺ - ATPase l isoform but that does not inhibit voltage-gated Na ⁺ channels (in other words, at a concentration ineffective on the action potential) as previously described ³³ and in the section EXAMPLES below.

The invention will be further illustrated by the following figures and examples. However, these examples and figures should not be interpreted in any way as limiting the scope of the present invention.

FIGURES:

Figure 1: Effects of carbamazepine on electrical activity of S/c/2a6-deficient motoneurons. Representative images of the ventral horn area of lumbar spinal cord from Slcl2a6 ^+/+ , Slcl2a6 ^~ ' ^~ vehicle -treated and Slcl2a6 ^~ ' ^~ carbamazepine (CBZ)-treated mice. Spinal cord cross-sections were immuno labeled using the Na ⁺ /K ⁺ ATPase l antibody. In Slcl2a6 ^~ ' ^~ CBZ-treated mice, Na ⁺ /K ⁺ -ATPase al is uniformly expressed along the membrane of motoneuron cell body while in Slcl2a6 ^~ ' ^~ vehicle-treated mice motoneurons display a non- uniform membrane expression of Na ⁺ /K ⁺ -ATPase al . Scale bar = 25 μιη. (B) and quantification (3 mice in each condition) of the standard deviation (SD) of Na ⁺ /K ⁺ -ATPase al staining intensity along motoneuron membrane perimeter shows a significantly lower uneven distribution of Na ⁺ /K ⁺ -ATPase al in Slcl2a6 ^~ ' ^~ CBZ-treated mice. (C) Representative electrophysiological recordings of motoneuron firing elicited by current injection (500 ms duration) showing the recovery of firing-dependent plateau accommodation in Slcl2a6 ^~ ' ^~ CBZ treated motoneurons (dashed line shows the accommodation level). (D) Quantitative analysis of plateau accommodation, expressed as the difference between the first AHP peak amplitude and each spike-induced AHP during a 500 ms depolarization, in Slcl2a6 ^~ ' ^~ CBZ treated motoneurons (n=4) shows no significant differences with Slcl2a6 ^+/+ motoneurons.

Figure 2: Effects of carbamazepine on neuromuscular junction oi Slcl2a6 ^' ' ^' mice.

(A) Quantification of denervation status of whole tibialis anteriors shows significantly less partially and completely denervated neuromuscular junctions in Slcl2a6 ^~ ' ^~ CBZ treated mice compared to Slcl2a6 ^~ ' ^~ vehicle-treated mice, but remains significantly more denervated than control Slcl2a6 ^+/+ mice (nerve is stained with anti-neurofilament and anti-synaptic vesicle antibodies (green) and neuromuscular junction is revealed with tetramethylrhodamine conjugated a-bungarotoxine (red); 3 mice in each condition; two-way ANOVA). (B) Endplates (tetramethylrhodamine conjugated a-bungarotoxine) were morphologically classified as either pretzel (typical normal mature morphology) or disorganized (perforations with bright lines or plaque-like). No differences in the number of disorganized endplates were observed between vehicle- or CBZ-treated Slcl2a6 ^~ ' ^~ mice (same mice than in A and in Fig. 1 A; two-way ANOVA). EXAMPLE 1: Neuromuscular junction and Na ⁺ /K ⁺ pump localization defects in

Andermann syndrome

Material & Methods

Animals: The care and use of mice conformed to institutional policies and guidelines. Slcl2a6 ^+I~ mice ⁴ were maintained on a C57BL/6 background. Mice were housed in cages with a 12 h light/12 h dark cycle, with food and water supplied ad libitum. The neuron- specific Slcl2a6-mA\ mice {nsSlcl2a ^A18/Als ) are described in ⁶ . Mice were euthanized by CO2 inhalation followed by cervical dislocation. Antibodies: The primary antibodies used were as follows: mouse anti-2H3

(Neurofilament 165 kDa) (1 :250; Developmental Studies Hybridoma Bank), mouse anti-SV2 (1 : 100; Developmental Studies Hybridoma Bank), mouse anti-SMI32 (1 : 1000; Stemberger Monoclonals), rabbit antineuro filament AB1987 (1 : 1000; Millipore), rabbit anti-βΙΙΙ tubulin (1 :250; Sigma), rabbit anti-SlOO (1 :500; Sigma) and mouse anti-Na ⁺ K ⁺ -ATPase al (1 : 100; Developmental Studies Hybridoma Bank). The secondary antibodies used were as follows: Alexa Fluor 488 goat anti-mouse (1 :500; Life Technologies), Alexa Fluor 488 goat anti-rabbit (1 :500; Life Technologies), Alexa Fluor 647 goat anti-rabbit (1 :500; Life Technologies), and Alexa Fluor 555 goat anti-mouse (1 :500; Life Technologies). The a-bungarotoxin ( BTX) conjugated to tetramethylrhodamine isothiocyanate was from Life Technologies (1 : 1000).

Genotyping: For experiments, genotyping was carried out by PCR on tail DNA of E12.5 embryos obtained from Slcl2a6 ^+/~ mice breeding. During the course of genotyping, embryos were kept at 4°C in Hibernate-E medium (Invitrogen).

Motoneuron cultures: Motoneuron cultures were prepared from Slcl2a6 ^+/+ and Slcl2a6 ^~ ' ^~ E12.5 mouse embryos as described previously ⁴⁰ . Briefly, cells were dissociated mechanically after trypsin treatment of the dissected spinal cords. The largest cells were isolated using iodixanol density gradient purification. To get highly purified motoneuron cultures, we added an immunopurification step using magnetic cell sorting technology ⁴¹ . In experiments using Hb9::GFP embryos to identify motoneurons, we found that this protocol yielded roughly 80-90% of GFP-positive neurons. Briefly, isolated neurons were incubated at +4°C in 80 μΐ L-15 medium plus 2 μΐ anti-mouse p75 monoclonal antibody (Millipore) for 20 minutes, followed, after washout, by a 15 minute incubation in 80 μΐ L-15 plus 10 μΐ goat anti-mouse IgG microbeads (Miltenyi Biotec). The magnetically labeled cells were applied onto a separation column and retained using a magnet (Miltenyi Biotec). After washout of the negative fraction, cell sorting was achieved by removing the magnet. After a final BSA cushion, motoneurons were plated onto poly-ornithine laminin-coated wells in supplemented Neurobasal (Life Technologies) medium in the presence of a mixture of neurotrophic factors (1 ng/ml BDNF, 100 pg/ml GDNF, and 10 ng/ml CNTF), completed with 2% horse serum, B27 supplement (Life Technologies), 50 μΜ L-glutamine, 25 μΜ L-glutamate, and 25 μΜ 2 β-mercaptoethanol .

Electrophysiological recordings: For chloride reversal potential determination of 1 and 7 days in vitro (DIV) motoneurons, the chloride GABAA current was recorded from 20 to 22°C with the gramicidin-perforated patchclamp technique using an Axopatch 200B amplifier (Dipsi Industrie). The bathing solution contained 140 mM TEA-C1, 3.5 mM MgCk, 10 mM glucose and 10 mM HEPES, adjusted to pH 7.4 with CsOH, and the pipette solution contained 140 mM CsCl, 1.5 mM Mg-ATP, 0.5 mM Na-GTP, 0.1 mM EGTA and 10 mM HEPES, adjusted to pH 7.35. We determined [Cl ^~ ]i, by adding 50 μg/ml gramicidin A (Sigma- Aldrich) to the pipette solution as described in ⁴² . [Cl ^~ ]i was calculated according to the Nernst equation: E _re v = RT/ZF x (log ([Cl ^" ]i / [Cl ^" ] _e ), where RT/ZF = 58 mV at room temperature and [Cl-] _e = 147 mM.

For electrical activity measurements, 7-8 DIV motoneurons from Slcl2a6 ^+I+ and Slcl2a6 ^~ ' ^~ were recorded under whole-cell patch clamp at room temperature in a bathing solution containing 140 mM NaCl, 5 mM KC1, 2 mM CaC12, 1.5 mM MgC12, 10 mM glucose and 10 mM HEPES, adjusted to pH 7.35. The patch pipette contained 10 mM KC1, 135 mM Kmethane-sulfonate, 1.5 mM Mg-ATP, 0.5 mM Na-GTP, 0.1 mM EGTA and 10 mM HEPES, adjusted to pH 7.35. Ouabain (Sigma) was prepared at a 1 mM stock solution and electrical activity was measured 2 minutes after an application of 50 μΜ.

RNA extraction and RT-qPCR: For primary cultures, total mRNA was extracted from roughly 30,000 purified embryonic motoneurons at 1 or 7 DIV with the RNeasy Mini Kit (Qiagen). For El 2.5, PI and P30 lumbar spinal cords, the tissues were harvested in KNAlater stabilization buffer (Qiagen). Lysis buffer was used for pestle tissue crushing and homogenization by passing the lysate through needles. Lysates were then mixed with an equal volume of 70% ethanol, and total mRNA was separated from other cellular components on RNeasy minispin columns. The eluted mRNA was quantified by spectrophotometry (Nanodrop). Following gDNA wipe out, reverse transcription (RT) was performed with 100 ng to 1 μg of mRNA with the Quantitect RT kit (Qiagen). The collected cDNA was diluted to 50 ng (culture) or 100 ng (tissue) with water and stored at -20°C until further use. Primers were designed with Primers 3.0 software. The sequences of the primers used are published in ¹³ . Quantitative PCR (qPCR) was performed on 5 or 10 ng cDNA with SYBR Green (Qiagen) for detection and the LightCycler system (Roche Diagnostics). After initial activation for 15 min at 95°C, 45 cycles of 94°C for 15 s, 60°C for 20 s and 72°C for 35 s were carried out. After PCR amplification, a melting curve analysis was carried out to check that the PCR was specific. Polymerase (RNA) II polypeptide J (Polr2J) levels were used to normalize the amounts of cDNA. AG was calculated as the difference between the G values, determined with the equation 2 ^~Δα .

Immunohistochemistry of spinal cord sections: For immunohistochemistry, 4-week- old mice were anaesthetized and perfused transcardially with 4% paraformaldehyde. Lumbar spinal cords were removed and post-fixed in 4% paraformaldehyde, flash-frozen and cut at a 12-μιη thickness. For Na ⁺ K ⁺ -ATPase immunohistochemistry experiments only, a citrate buffer antigen retrieval step was performed. The sections were then rinsed 5 minutes in PBS and incubated for 2 hours at room temperature in blocking solution (TBLS (10% NaN ₃ ), 20% goat serum, 0.3% Triton X-100). This was followed by an overnight incubation at +4°C with the primary antibodies. Subsequently, sections were incubated 1 hour with the secondary antibodies. All washes were done with PBS. Slides were mounted in Mowiol. Images were taken with a Zeiss confocal microscope, equipped with filters suitable for FITC/Cy3 fluorescence. Quantification of motoneuron area was done with the Image J software.

Neuromuscular junction immunohistochemistry: Tibialis anterior (TA) muscle sections were labeled by immunohistochemistry to allow quantification of neuromuscular innervation and endplate morphology as described previously ²⁰ . For ubiquitous Slcl2a6 ^~ ' ^~ mice, whole TA muscles were dissected from 4-week-old animals and fixed in 4% paraformaldehyde for 15 minutes. Following the removal of connective tissue, the TA muscles were incubated with tetramethylrhodamine-conjugated alpha-a-bungarotoxin for 20 minutes at room temperature. The TA muscles were then incubated in methanol at -20°C for 5 minutes, followed by an overnight incubation at +4°C with the SV2 and neurofilament antibodies. Incubation with the secondary antibodies was performed the following day at room temperature for 1 hour. Finally, two to three thin filets per stained TA were cut and mounted in Mowiol. All filets were single sections of the surface of the TA muscle.

For neuron-specific Slcl2a6 deletion (nsSlcl2a6 ^A18/A18 ^ mice, NMJ analysis was performed on longitudinal frozen sections of 8-week-old animals. TAs were post-fixed in 4% paraformaldehyde, flash- frozen and cut at a 10 μιη thickness. The sections were then rinsed 5 minutes in PBS and incubated for 2 hours at room temperature in blocking solution (TBLS (10%) NaN ₃ ), 20%) goat serum, 0.3%> Triton X-100). This was followed by an overnight incubation at 4°C with the primary antibodies. Subsequently, sections were incubated 1 hour with the secondary antibodies and the tetramethylrhodamine-conjugated alpha-a- bungarotoxin. All washes were done with PBS. Slides were then mounted in Mowiol.

Images were taken with a Zeiss confocal microscope, equipped with filters suitable for FITC/Cy ³ /Cy ⁵ fluorescence. The Image J software was used to calculate the area of NMJs. Electron microscopy: Tibialis anterior (TA) collected from 4-week-old mice were fixed in a 2% paraformaldehyde/0.2% glutaraldehyde (Sigma-Aldrich) solution in PBS for 2 hours at +4°C. Samples were postfixed overnight at 4°C in 2% PFA and rinsed in PBS. Bundles of about 20 muscle fibers were teased out and treated free-floating in solution. Muscle fibers were processed for electron microscopy as follows: junctions were identified by detecting nicotinic AChRs with a-bungarotoxin coupled to biotin (αΒΤΧ-biot) using a pre- embedding immunogold method. Muscle fibers were incubated in aBTX-biot (10 μg/ml PBS) overnight at 20 °C. After washing, muscles were incubated in streptavidin coupled to gold particles (0.8 nm in diameter; Nanoprobes; 1 : 100 in PBS/BSA). The fibers were then washed and post- fixed in 1% glutaraldehyde. After washing in acetate buffer (0.1 M, pH 7), the signal of the gold immunoparticles was increased using a silver enhancement kit (HQ silver; Nanoprobes) for 3 minutes at room temperature in the dark. Finally, after washing in acetate buffer, sections were treated with 1% osmium, dehydrated and embedded in resin. Ultrathin sections were cut, stained with lead citrate and examined in a Philips CM 120 transmission electron microscope (Eindhoven) equipped with a camera (Morada, Soft Imaging System, Olympus).

Hematoxylin and eosin staining of tibialis anterior sections: Whole tibialis anterior (TA) were dissected and placed overnight in 4% paraformaldehyde, then transferred to 70% ethanol solution. Tissues were embedded in paraffin, cut at a thickness of 5 μιη, deparaffinized and stained for hematoxylin and eosin. Images were taken with a Leica DMRB. Muscle fiber area was measured using the Image J software.

Compound muscle action potential recordings: Subjects were studied with a MEDTRONIC electromyography system (Natus Medical Inc.) at CHU de Quebec. For

ACCPN patients, repetitive nerve stimulations were performed for 3 different nerves. The spinal accessory nerve was recorded at the trapezius muscle, the facial nerve was recorded at the nasaiis and the ulnar nerve was recorded at the abductor digiti minimi. Studies were performed using constant-current stimulation and a square-wave pulse o 0.2 ms duration delivered through a hand-held surface-stimulating electrode (13L36; Alpine Biomed) at the frequency of 3Hz. Filters were set at 3 Hz to 5 kHz. A decrement of more than 10% between the first and 4th compound muscle action potential (CMAP) was considered significant for a neuromuscular transmission defect. Statistics: All statistical analyses were done with the Graphpad Prism software. When appropriate, a Student's unpaired two-tail t-test, a one-way or two-way analysis of variance (ANOVA) followed by a Holm-Sidak's post hoc tests were used. Data were considered significantly different at/? < 0.05.

Results

Loss of KCC3 results in aberrant pre-synaptic neuromuscular junction terminals: To investigate a potential motoneuron-specific role for KCC3, we used 4-week- old mice, as this age represents an early symptomatic stage that precedes axonal degeneration within the sciatic nerve ⁵ ' ⁸ . We first evaluated the impact of KCC3 depletion on motoneuron survival within the lumbar spinal cord. Similar to what was previously reported ⁴ , we find that both wild type Slcl2a6 ^+/+ and mutant Slcl2a6 ^~ ' ^~ mice display the same number of motoneurons, suggesting that loss of KCC3 does not lead to motoneuron death at this early time point. Seeing as KCC3 has been ascribed a role in cell volume regulation ⁴ ' ⁸ , we determined the cell body area of Slcl2a6 ^+I+ and Slcl2a6 ^~ ' ^~ spinal motoneurons. This quantification shows similar size distributions between control and mutant mice, implying that KCC3 role in volume regulation and swelling may be restricted to certain cell types and/or cellular compartments.

Seeing as other motoneuron pathologies such as spinal muscular atrophy (SMA) and amyotrophic lateral sclerosis (ALS) are characterized by sequential NMJ denervation, axonal degeneration and motoneuron loss (reviewed in ¹⁹ ), we assessed whether KCC3 depletion could also impact NMJ innervation. To do so, we labeled the innervating axon with neurofilament (NF) and synaptic vesicle 2 (SV2) markers and the post-synaptic apparatus with alpha-bungarotoxin of tibialis anterior (TA) muscle of Slcl2a6 ^+I+ and Slcl2a6 ^~ ' ^~ mice. Our results show that Slcl2a6 ^~ ' ^~ mice display aberrant NMJ innervations compared to Slcl2a6 ^+I+ littermates, with a significant increase in pre-synaptic terminal accumulations, in partially innervated and completely denervated NMJs as well as in the number of terminal and nodal axonal sprouts. We next imaged wildtype and mutant TA by electron microscopy (EM) to get a more in-depth look at the pre-synaptic NMJ terminals of Slcl2a6 ^~ ' ^~ mice. In Slcl2a6 ^+/+ mice, all three cellular partners of the NMJ, the nerve terminal (T), the muscle cell (MC) and the terminal Schwann cell (TSC) were identified. Indeed, the nerve terminal is filled with dense synaptic vesicles and some mitochondria while the TSC adequately caps the nerve terminal. In Slcl2a6-I- mice however, we observed different types of pre-synaptic NMJ terminals. Indeed, while some Slcl2a6 ^~ ' ^~ NMJs normally display the three cellular partners of the NMJ, others present either a very small nerve terminal that in some cases has disappeared from the pre-synaptic site. In the latter event, the TSC processes were found to completely fill the pre-synaptic area. Thus, our analysis reveals that while motoneuron cell bodies are not lost in the spinal cord of Slcl2a6 ^~ ' ^~ mice, there are clear axonal defects at the NMJ that most probably influence the synaptic transmission between the nerve terminal and the muscle and may contribute to the early locomotor and sensorimotor gating defects observed in KCC3- depleted mice ⁴ .

Loss of KCC3 results in aberrant post-synaptic neuromuscular junction terminals and skeletal muscle atrophy: We next evaluated if loss of KCC3 had any effect on the post-synaptic compartment of the NMJ. We used tetramethylrhodamine-conjugated a- bungarotoxin to visualize NMJ endplates of the TA of wild type and Slcl2a6 ^~ ' ^~ mice. We find similar endplate areas between control and mutant mice, suggesting that deletion of Slcl2a6 does not impact endplate growth. As previously described ²⁰ , post-synaptic terminals were categorized based on morphology whereas normal appearing endplates were identified as pretzel-like while perforated-endplates displaying bright lines and plaque-like endplates were identified as disorganized. This analysis shows that Slcl2a6 ^~/~ mice display significantly more morphologically immature endplates than Slcl2a6 ^+/+ littermates. It thus appears that the presynaptic defects observed in Slcl2a6 ^~ ' ^~ NMJs concomitantly occur with abnormalities at the post-synaptic terminal.

Despite locomotor and gait defects having previously been reported in Slcl2a6 ^~ ' ^~ mice ⁴ , the effect of KCC3 depletion on skeletal muscle has never been studied. Further, seeing as our results highlight aberrancies at the interface between the nerve and skeletal muscle, we set out to determine the impact of loss of KCC3 on skeletal muscle parameters. Analysis of TA cross-sections reveals a reduction in myofiber area in Slcl2a6-I- mice compared to Slcl2a6 ^+I+ littermate. This decrease in myofiber area does not result from reduced overall or muscle weight as both Slcl2a6 ^+I+ and Slcl2a6 ^~ ' ^~ display similar body weights and TA weight/body weight ratios. We have thus uncovered that skeletal muscle atrophy occurs in Slcl2a6 ^~/~ mice in the absence of spinal motoneuron degeneration, suggesting that the observed pre- and postsynaptic NMJ defects are sufficient to influence muscle pathology. Neuronal-specific KCC3 deletion results in muscle atrophy and aberrant postsynaptic neuromuscular junction terminals: It is possible that the abnormalities observed in Slcl2a6 ^~ ' ^~ NMJs and skeletal muscles are due to KCC3 -dependent intrinsic defects within muscle. To address this point, we examined a mouse model where KCC3 is solely deleted in neuronal cells, termed nsSlcl2a6 ^{A18/A18 6} . We firstly compared TA cross-sections of 8-week- old Slcl2a6 ^+I+ and nsSlcl2a6 ^A18/A18 mice. Quantification of myofiber area shows that mutant mice display smaller myofibers than control mice, suggesting that skeletal muscle atrophy also characterizes the neuronal-specific nsSlcl2a6 ^A18/A18 mice. We next investigated the phenotype of nsSlcl2a6 ^A18/A18 NMJs. Similarly to ubiquitous iS7ci2a(5-deleted mice, we observed a variety of pre-synaptic terminal aberrancies in nsSlcl2a6 ^A18/A18 mice such as presynaptic accumulation, partial or complete denervation as well as nodal and terminal sprouting. Importantly, when morphologically categorizing endplates as either pretzel-like (normal) or disorganized (perforated with bright lines or plaque-like), we find that nsSlcl2a6 ^A18/A18 post-synaptic terminals display a significantly more immature morphology than wildtype littermates. Quantification of endplate area, however, reveals similar sizes between mutant and control mice. Our results therefore show that pathological skeletal muscle and postsynaptic NMJ terminals are also found in mice where Slcl2a6 is specifically deleted in neurons, thus suggesting that the muscle and NMJ defects observed in ubiquitous Slcl2a6 ^~ ' ^~ mice stem from nerve-dependent events.

Patients with Andermann syndrome do not display neuromuscular transmission defects: We sought to evaluate our observations of pre- and post-synaptic NMJ defects in ACCPN patients by performing a commonly used functional test for the assessment of postsynaptic defects in NMJ transmission. As expected, healthy patients show no muscle fatigability. Myasthenia gravis patients were used as positive controls since a decremented electrical muscle response on repetitive motor nerve stimulation is observed in this postsynaptic NMJ disorder ²¹ . Although a severe muscle weakness was evident in ACCPN patients, there were no signs of muscle fatigability with repetitive electrical stimulation. These results thus support a pre-synaptic origin of KCC3 -induced neuropathy in both murine models of Andermann syndrome as well as in ACCPN patients.

Developmental expression of cation-chloride cotransporters in spinal cord is unaffected by KCC3 loss of function. Having uncovered importantly that defective presynaptic NMJ terminals occur early in Slcl2a6 ^~ ' ^~ mice, we next set out to determine how loss of KCC3 within motoneurons could lead to this aberrant synaptic maintenance. As the Andermann syndrome is a neurodevelopmental disease, we first assessed the pattern of KCC3 expression during spinal cord development. Using quantitative PCR (qPCR), we show that KCC3 transcript is expressed as early as embryonic day 12.5 (E12.5) and its expression significantly increases in adulthood (post-natal day 30 (P30)) when compared to E12.5 levels. KCC3 exists as two distinct isoforms designated as KCC3a and KCC3b, with the latter isoform being renal-specific ¹⁰ ' ²² ' ²³ . Moreover, an exon 2-deleted form of KCC3a is preferentially expressed in the nervous system ¹³ ' ^{24, 25} . We thus performed an RT-PCR analysis to compare expression of KCC3a isoforms in adult spinal cord and skeletal muscle. When using a primer set spanning exons 1 and 3 of KCC3a, the alternative splicing event was observed in both adult spinal cord and muscle, resulting in bands representing both the full- length (FL) and exon-deleted (Δ2) isoforms. However, the amount of exon 2-deleted transcript was clearly greater in spinal cord than in muscle. Using purified cultured motoneurons, we further confirmed the preferential expression of the exon 2-deleted isoform in neuronal cells. Indeed, using pan-primers designed to span exons 13-15 (KCC3ab), we observed a single band, supporting the neuronal specificity of the exon 2-deleted KCC3a isoform. There was no detectable transcript with primers specific for KCC3b, consistent with its non-neuronal expression. To determine whether KCC3 loss-of-function affects the chloride shift that occurs during development, we firstly assessed whether the expression of KCC2 and NKCC1 transcripts, the main cation-chloride cotransporters involved in [Cl ^" ]i maintenance, was modified in the spinal cord of Slcl2a6 ^~ ' ^~ mice. We compared transcript levels from E12.5, PI and P30 mice and normalized PI and P30 transcript levels to E12.5 values since KCC2 and NKCC1 transcript levels were not significantly different between groups at this time point (data not shown). Analysis of KCC2 transcript during development of Slcl2a6 ^+/+ spinal cord shows that the molecular signature of chloride shift, that is an increased expression of KCC2 ²⁶ , occurs between El 2.5 and PI . Further, we find that the expression of KCC2 transcript is not significantly different between Slcl2a6 ^+/+ and Slcl2a6 ^~ ' ^~ mice at any time point. More surprisingly, we did not observe the expected decrease in the expression of NKCC1 transcript with maturation of Slcl2a6 ^+I+ spinal cord, but rather a roughly two-fold increase at P30. Importantly, while NKCC1 levels are similar between Slcl2a6 ^+I+ and Slcl2a6 ^~ ' ^~ mice at E12.5 and PI, there is a mild increase of NKCC1 transcript in P30 mutant mice compared to aged- matched control mice. Seeing as NKCC1 plays an opposing role to that of KCC3 (inward chloride transport instead of outward), the increased expression of NKCC1 in KCC3-depleted mice may simply result from a non-specific compensatory mechanism or, alternatively, stem from other non-neuronal cell types within the spinal cord ²⁷ ' ²⁸ . Nevertheless, we have identified on one hand a developmental increased transcript expression of all three cation- chloride cotransporters (KCC3, KCC2, NKCC1) in the spinal cord and on the other hand, an abnormal expression of NKCC 1 in the spinal cord of adult Slcl2a6 ^~ ' ^~ mice.

Developmental chloride shift in primary motoneuron cultures is unaffected by KCC3 loss of function: Since the spinal cord is a heterogeneous mixture of cells, we set out to functionally assess the impact of KCC3 loss of function on [Cl ^" ]i maintenance in primary cultures of purified motoneurons. We first investigated the effect of KCC3 depletion on KCC2 and NKCC 1 transcript expression during in vitro maturation of motoneurons isolated from E12.5 embryos between 1 and 7 days in vitro (DIV) . Our qPCR analysis reveals that NKCC1 transcripts are decreased while KCC2 transcripts are increased during normal maturation of Slcl2a6 ^+I+ motoneurons. In 7 DIV Slcl2a6 ^~ ' ^~ motoneurons, NKCC1 and KCC2 followed the same expression profiles than Slcl2a6 ^+I+ motoneurons, albeit with a significant reduction in the temporal increase of KCC2 transcript compared to control cells. Interestingly, expression changes observed in cation-chloride cotransporters of 7 DIV motoneurons are qualitatively similar to those of PI spinal cord.

The increase in KCC2 transcript together with the decrease in NKCC1 transcript observed between 1 and 7 DIV suggest that a chloride shift probably occurs during the maturation of cultured motoneurons. We therefore recorded the chloride reversal potential of GAB A current (EGABA A) with the gramicidin-perforated patch-clamp method to determine if loss of KCC3 would influence the motoneuron chloride. Consistent with the differential transcriptional expression of KCC2 and NKCC1 in motoneurons, EGABA-A showed a 22 mV hyperpolarization from -56 ± 3 mV (n = 12) at 1 DIV to -78 ± 3 mV (n = 10) at 7 DIV in Slcl2a6 ^+I+ motoneurons. In Slcl2a6 ^~ ' ^~ motoneurons, in vitro maturation still induced a 20 mV shift of EGABA-A from -62 ± 3 mV (n = 10) at 1 DIV to -82 ± 4 mV (n = 8) at 7 DIV. The corresponding [Cl ^" ]i, calculated with the Nernst equation, showed this concentration to be 17 ± 2 mM and 13 ± 2 mM at 1 DIV, decreasing to 7 ± 1 mM and 6 ± 1 mM, at 7 DIV, in Slcl2a6 ^+/+ and Slcl2a6 ^~ ' ^~ motoneurons, respectively (p < 0.001 between 1 and 7 DIV for both genotypes, t-test). Thus, loss of KCC3 does not alter the [Cl ^~ ]i of 1 and 7 DIV motoneurons, suggesting a potential chloride-independent function for KCC3 in motoneurons. Loss of KCC3 causes mislocalization and reduced activity of Na ⁺ /K ⁺ -ATPase al in soma of motoneurons: Since our results suggest that the defects observed at the presynaptic NMJ terminals of Slcl2a6 ^~ ' ^~ mice may not be related to changes in motoneuron [Cl ^" ]i, we hypothesized that KCC3 depletion may impact motoneuron function via perturbation of the al isoform of the Na ⁺ /K ⁺ -ATPase. Indeed, Na ⁺ /K ⁺ -ATPase al is specifically expressed in the large a-motoneurons of the spinal cord ventral horn area ²⁹ ' ³⁰ , interacts with the KCC3a isoform ¹⁶ and is key in the maintenance of cellular electrochemical gradient ¹¹ , and thus, in the regulation of ionic homeostasis. We first performed immunohistochemistry of lumbar spinal cord sections and, as previously reported ³⁰ , observed a specific expression of Na ⁺ /K ⁺ - ATPase al along the membrane of large Slcl2a6 ^+I+ motoneuron soma. Strikingly, the expression of Na ⁺ /K ⁺ -ATPase al in Slcl2a6 ^~ ' ^~ motoneurons displays a non-uniform pattern around the cell body perimeter and can also be aberrantly detected in the cytoplasm. Quantification of the standard deviation (SD) of Na ⁺ /K ⁺ - ATPase al staining intensity along the membrane perimeter shows a significantly greater uneven distribution in Slcl2a6 ^~ ' ^~ motoneurons than in Slcl2a6 ^+I+ cells. Seeing as Slcl2a6 ^~ ' ^~ mice display important pre-synaptic NMJ abnormalities, we evaluated the impact of KCC3 depletion on Na ⁺ /K ⁺ -ATPase al distribution at the nerve terminal. Initial immunohistochemistry on the TA muscle shows the expression of Na ⁺ /K ⁺ - ATPase al in nerve-like bundles, some of which branch out to innervate NMJ endplates. Since both motoneuron axons and Schwann cells are typically found together within these bundles, we performed co-labeling experiments to determine in which of these cell types Na ⁺ /K ⁺ -ATPase al was specifically expressed. Using NF to visualize the motoneuron axon and SI 00 to detect Schwann cells at the NMJ endplate, we find that the innervation pattern of the nerve terminal is quite different from that of Na ⁺ /K ⁺ - ATPase al and in fact, observe a co-localization with Schwann cells. Our results thus imply that Na ⁺ /K ⁺ -ATPase al is mostly expressed in the soma of motoneurons rather than at the axonal ending.

Finally, we proceeded to examine whether mislocalization of Na ⁺ /K ⁺ - ATPase al affected its normal function by recording the electrical activity of large Slcl2a6 ^+/+ and Slcl2a6 ^~ ' ^~ 7-8 DIV primary motoneurons, an in vitro time point at which we have previously shown motoneuron cultures reach electrical maturation ³¹ . We find no significant differences in whole cell capacitance (54.3 ± 3.3 pF, n = 14 for Slcl2a6 ^+/+ motoneurons and 58.0 ± 3.7 pF, n = 19 for Slcl2a6 ^~ ' ^~ motoneurons, t-test) and in input resistance (218 ± 16 ΜΩ and 188 ± 17 ΜΩ for Slcl2a6 ^+/+ and Slcl2a6 ^~ ' ^~ motoneurons, respectively, t-test). Resting membrane potential (RMP) and peak amplitude of the action potential were also not significantly different. In addition, motoneurons from both genotypes were able to fire repetitive action potentials under application of a 500 ms depolarizing current. The frequencies versus injected current relationships were also similar. A salient characteristic of motoneurons is the appearance of a hyperpolarization following a train of action potentials, known as the post-tetanic hyperpolarization (PTH). This PTH was significantly reduced in ΞΙοΠαό ^' ^ motoneurons. Furthermore, while the amplitude of the hyperpolarization of each action potential remains constant in Slcl2a6 ^+/+ motoneurons during repetitive activity, it is significantly depolarized in Slcl2a6 ^+/+ motoneurons. Altogether, our results show that the endogenous sustained electrical activity of motoneurons activates a hyperpolarizing current that is inhibited in KCC3-depleted motoneurons. This result prompted us to analyze the effects of ouabain, a potent inhibitor of Na ⁺ /K ⁺ -ATPase ³² , on motoneuron electrical activity. Under our experimental conditions, a concentration of 300 μΜ of ouabain to inhibit the pump did not significantly modify the RMP (-64 ± 2 mV in control condition and -59 ± 6 mV 2 minutes after ouabain application, n = 5). During repetitive activity, ouabain induced a 4 mV shift of the plateau potential towards more depolarized values that subsequently remain constant. Depolarization of the RMP was only observed following intense repetitive activity. The PTH decreased from -1.19 ± 0.46 mV to - 0.45 ± 0.61 mV in the presence of ouabain. The depolarizing effects of ouabain during repetitive activity therefore strengthen our findings that perturbed localization of Na ⁺ /K ⁺ - ATPase al in Slcl2a6 ^~ ' ^~ motoneurons is responsible for a decreased activity of the Na ⁺ /K ⁺ - pump, leading to a progressive plateau potential depolarization. This aberrant distribution of Na ⁺ /K ⁺ -ATPase al, together with a decreased activity, may ultimately perturb the ionic homeostasis of the motoneuron, thus leading to the observed synaptic stripping events and histopathophysiology of Slcl2a6 ^~ ' ^~ mice.

Together, our results highlight novel pathological hallmarks and molecular effectors within the motor circuitry of Andermann syndrome mice. Of utmost importance is the identification of reduced Na ⁺ /K ⁺ pump activity in motoneurons together with an aberrant localization of Na ⁺ /K ⁺ -ATPase l in the cell body of motoneurons in an Andermann syndrome mouse model, prior to any signs of motoneuron and axonal degeneration. These findings have major clinical implications as they bring forward novel symptomatic and molecular therapeutic targets for the treatment and clinical management of ACCPN patients. EXAMPLE 2: In vivo and in vitro treatment with carbamazepine partially restores motoneuron membrane expression of Na ⁺ /K ⁺ -ATPase al, firing plateau potential and innervation of NMJ: Based on the above results, we hypothesized that restoring motoneuron electrical activity should prevent the loss of innervation observed at the NMJ of Slcl2a6 ^~ ' ^~ mice. Carbamazepine (CBZ), a known pharmacological agent used in the treatment of epilepsy, bipolar disorders and neuropathic pain, is also reported to increase the expression and the activity of Na ⁺ /K ⁺ -ATPase l and a3 iso forms at doses ineffective on the action potential ³³ . This prompted us to investigate the effects of a chronic treatment of CBZ on Na ⁺ /K ⁺ -ATPase al expression, electrical activity and NMJ of Slcl2a6 ^~ ' ^~ mice.

Methods

For in vivo experiments, CBZ or vehicle were administered to post-natal P10 Slcl2a6 ^{~ /_} mice by intraperitoneal injection of 0.025 mg/g/day for 20 days as described in ³³ . Four- week old mice were then euthanized and tissues processed for neuromuscular junction and motoneuron Na ⁺ /K ⁺ -ATPase al analysis. A stock solution was prepared at 2.5 mg/ml using 12.5 mg CBZ diluted with 5 ml 2-Hydroxypropyl-P-cyclodextrin and stored at 4°C. For in vitro experiments, a stock solution of CBZ at 10 mM in DMSO was prepared and motoneuron cultures were incubated for 2 days with 10 μΜ CBZ or DMSO (1 : 1000) before electrophysiological recordings.

Results

Following a chronic treatment with CBZ beginning ten days after birth, the membrane expression of Na ⁺ /K ⁺ -ATPase al was restored in Slcl2a6 ^~ ' ^~ motoneurons (Fig. 1A, B). To ensure whether membrane Na ⁺ /K ⁺ -ATPase al expression was sufficient to correct the electrical activity, we determined the effects of CBZ on electrical activity. For this series of in vitro experiments, motoneurons were incubated in the presence of CBZ for two days before recordings at 9 DIV. As the aim of CBZ treatment was to correct the decrease in AHP amplitude during repetitive activity without affecting action potential amplitude, we first determined the maximal CBZ concentration for which no effects were observed on basal spontaneous synaptic-driven motoneuron electrical activity. At 100 μΜ, acute application of CBZ induced a decrease in spontaneous frequency, while at 10 μΜ CBZ, no effects on frequency and amplitude of wild-type motoneuron spontaneous activity were observed. Under whole-cell patch clamp, we confirmed that two days incubation in the presence of 10 μΜ CBZ did not modify action potential amplitude (Fig. 1C). Interestingly, under CBZ treatment, AHP amplitude in Sic 12a ^ ' motoneurons was maintained during repetitive activity (Fig. ID). Altogether these results demonstrate that CBZ is able to correct motoneuron Na ⁺ /K ⁺ - ATPase l expression and to maintain AHP amplitude during firing activity without significant effects on action potential amplitude.

Next, we verified whether CBZ could restore pre- and post-synaptic neuromuscular abnormalities. The percentage of denervation was significantly decreased following chronic CBZ treatment and amounted 40 % in CBZ-treated Slcl2a6 ^~ ' ^~ mice compared with 51 % in vehicle-treated Slcl2a6 ^~ ' ^~ (Fig. 2A). However, the percentage of post-synaptic disorganized NMJ was not modified by CBZ at this time point (Fig. 2B). DISCUSSION:

Our results strongly support that early in development motoneuron electrical activity is a key factor in the formation and maintenance of the neuromuscular synapses ³⁴ and influences the progression of neurodegenerative diseases in which synaptic degeneration is an early sign ³⁵ . The use of CBZ at a concentration that does not inhibit voltage-gated Na ⁺ channels reveals specific effects on the activity and expression of Na ⁺ /K ⁺ -ATPase in motoneurons. In addition to Andermann syndrome, such specificity makes CBZ a therapeutically valuable compound in the treatment of Na ⁺ /K ⁺ pump related diseases.

REFERENCES:

Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. 1. Larbrisseau A, Vanasse M, Brochu P, Jasmin G. The Andermann syndrome: agenesis of the corpus callosum associated with mental retardation and progressive sensorimotor neuronopathy. Can J Neurol Sci. 1984;11 :257-261 2. Mathieu J, Bedard F, Prevost C, Langevin P. [Motor and sensory neuropathies with or without agenesis of the corpus callosum: a radiological study of 64 cases]. Can J Neurol Sci. 1990;17: 103-108

3. De Braekeleer M, Dallaire A, Mathieu J. Genetic epidemiology of sensorimotor polyneuropathy with or without agenesis of the corpus callosum in northeastern Quebec. Hum

Genet. 1993;91 :223-227

4. Howard HC, Mount DB, Rochefort D et al. The K-Cl cotransporter KCC3 is mutant in a severe peripheral neuropathy associated with agenesis of the corpus callosum. Nat Genet. 2002;32:384-392.

5. Boettger T, Rust MB, Maier H et al. Loss of K-Cl co-transporter KCC3 causes deafness, neurodegeneration and reduced seizure threshold. Embo J. 2003;22:5422- 5434

6. Shekarabi M, Moldrich RX, Rasheed S et al. Loss of neuronal potassium/chloride cotransporter 3 (KCC3) is responsible for the degenerative phenotype in a conditional mouse model of hereditary motor and sensory neuropathy associated with agenesis of the corpus callosum. J Neurosci. 2012;32:3865-3876

7. Sun YT, Lin TS, Tzeng SF et al. Deficiency of electroneutral K+-C1- cotransporter 3 causes a disruption in impulse propagation along peripheral nerves. Glia. 2010;58: 1544-1552

8. Byun N, Delpire E. Axonal and periaxonal swelling precede peripheral neurodegeneration in KCC3 knockout mice. Neurobiol Dis. 2007;28:39-51

9. Dupre N, Howard HC, Mathieu J et al. Hereditary motor and sensory neuropathy with agenesis of the corpus callosum. Ann Neurol. 2003;54:9-18

10. Mount DB, Mercado A, Song L et al. Cloning and characterization of KCC3 and KCC4, new members of the cation-chloride cotransporter gene family. J Biol Chem. 1999;274: 16355-16362

11. Gagnon KB, Di Fulvio M. A molecular analysis of the Na(+)-independent cation chloride cotransporters. Cell Physiol Biochem. 2013;32: 14-31

12. Blaesse P, Airaksinen MS, Rivera C, Kaila K. Cation-chloride cotransporters and neuronal function. Neuron. 2009;61 :820-838

13. Lucas O, Hilaire C, Delpire E, Scamps F. KCC3-dependent chloride extrusion in adult sensory neurons. Mol Cell Neurosci. 2012;50:211-220

14. Allain AE, Le Corronc H, Delpy A et al. Maturation of the GABAergic transmission in normal and pathologic motoneurons. Neural Plast. 2011;2011 :905624 15. Ikeda K, Onimaru H, Yamada J et al. Malfunction of respiratory-related neuronal activity in Na+, K+- ATPase alpha2 subunit-deficient mice is attributable to abnormal CI- homeostasis in brainstem neurons. J Neurosci. 2004;24: 10693-10701

16. Fujii T, Takahashi Y, Itomi Y et al. K+-C1- Cotransporter-3a Up-regulates Na+,K+- ATPase in Lipid Rafts of Gastric Luminal Parietal Cells. J Biol Chem. 2008;283:6869-6877

17. Kaplan JH. Biochemistry of Na,K- ATPase. Annu Rev Biochem. 2002;71 :511-535

18. Doi M, Iwasaki K. Na+/K+ ATPase regulates the expression and localization of acetylcholine receptors in a pump activity-independent manner. Mol Cell Neurosci. 2008;38:548-558

19. Murray LM, Talbot K, Gillingwater TH. Review: neuromuscular synaptic vulnerability in motor neurone disease: amyotrophic lateral sclerosis and spinal muscular atrophy. Neuropathol Appl Neurobiol. 2010;36: 133-156

20. Bowerman M, Murray LM, Beauvais A et al. A critical smn threshold in mice dictates onset of an intermediate spinal muscular atrophy phenotype associated with a distinct neuromuscular junction pathology. Neuromuscul Disord. 2012;22:263-276

21. Berrih-Aknin S, Frenkian-Cuvelier M, Eymard B. Diagnostic and clinical classification of autoimmune myasthenia gravis. J Autoimmun. 2014

22. Hiki K, DAndrea RJ, Furze J et al. Cloning, characterization, and chromosomal location of a novel human K+-C1- cotransporter. J Biol Chem. 1999;274: 10661-10667

23. Race JE, Makhlouf FN, Logue PJ et al. Molecular cloning and functional characterization of KCC3, a new K-Cl cotransporter. Am J Physiol. 1999;277:0210-1219

24. Mercado A, Vazquez N, Song L et al. NH2 -terminal heterogeneity in the KCC3 K+-C1- cotransporter. Am J Physiol Renal Physiol. 2005 ;289:F 1246- 1261

25. Le Rouzic P, Ivanov TR, Stanley PJ et al. KCC3 and KCC4 expression in rat adult forebrain. Brain Res. 2006;1110:39-45

26. Rivera C, Voipio J, Payne JA et al. The K+/C1- co-transporter KCC2 renders GABA during neuronal maturation. Nature. 1999;397:251-255

27. Hubner CA, Lorke DE, Hermans-Borgmeyer I. Expression of the Na-K-2Clcotransporter NKCC1 during mouse development. Mech Dev. 2001;102:267-269

28. Yan Y, Dempsey RJ, Sun D. Expression of Na(+)-K(+)-Cl(-) cotransporter in rat brain during development and its localization in mature astrocytes. Brain Res. 2001;911 :43-55 29. Mata M, Siegel GJ, Hieber V et al. Differential distribution of (Na,K)-ATPase alpha isoform mRNAs in the peripheral nervous system. Brain Res. 1991;546:47-54 30. Edwards IJ, Bruce G, Lawrenson C et al. Na+/K+ ATPase alphal and alpha3 isoforms are differentially expressed in alpha- and gamma-motoneurons. J Neurosci. 2013;33:9913-9919

31. Camu W, Tremblier B, Plassot C et al. Vitamin D confers protection to motoneurons and is a prognostic factor of amyotrophic lateral sclerosis. Neurobiol Aging. 2014;35: 1198-1205 32. Hall C, Ruoho A. Ouabain-binding-site photoaffinity probes that label both subunits of Na+,K+-ATPase. Proc Natl Acad Sci U S A. 1980;77:4529-4533

33. Li B, Hertz L, Peng L. Cell-specific mRNA alterations in Na+, K+- ATPase a and β isoforms and FXYD in mice treated chronically with carbamazepine, an anti-bipolar drug. Neurochem Res. 2013 Apr;38(4): 834-41.

34. Tintignac LA, Brenner HR, Ruegg MA. Mechanisms regulating neuromuscular junction development and function and causes of muscle wasting. Physiol Rev 2015; 95: 809-852. 35. Brown R, Hynes-Allen A, Swan AJ, Dissanayake KN, Gillingwater TH, Ribchester RR. Activity-Dependent Degeneration Of Axotomized Neuromuscular Synapses In Wlds Mice. Neuroscience 2015 290:300-320.