Skip Navigation


Human Molecular Genetics Advance Access originally published online on July 21, 2008
Human Molecular Genetics 2008 17(20):3166-3179; doi:10.1093/hmg/ddn213
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/20/3166    most recent
ddn213v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Stum, M.
Right arrow Articles by Nicole, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stum, M.
Right arrow Articles by Nicole, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Evidence of a dosage effect and a physiological endplate acetylcholinesterase deficiency in the first mouse models mimicking Schwartz–Jampel syndrome neuromyotonia

Morgane Stum1,2, Emmanuelle Girard3,{dagger}, Marie Bangratz1,2,{dagger}, Véronique Bernard4,5, Marc Herbin6, Alban Vignaud2,7, Arnaud Ferry5,7, Claire-Sophie Davoine1,2, Andoni Echaniz-Laguna8,9,10, Frédérique René8,9, Christophe Marcel10, Jordi Molgó3, Bertrand Fontaine1,2,11, Eric Krejci4,5 and Sophie Nicole1,2,11,*

1 Inserm, U546, Paris, France 2 UPMC Univ Paris 06, UMR_S546, Paris, France 3 CNRS, Institut de Neurobiologie Alfred Fessard—FRC2118, Laboratoire de Neurobiologie Cellulaire et Moléculaire—UPR9040, Gif sur Yvette, France 4 Inserm, U686, Paris, France 5 Université Paris-Descartes, Paris, France 6 U.S.M. 0301-UMR 7179, MNHN/CNRS, Département Ecologie et Gestion de la Biodiversité, Muséum National d'Histoire Naturelle de Paris, Paris, France 7 Inserm, U787, Paris, France 8 Inserm, U692, Laboratoire de Signalisations Moléculaires et Neurodégénérescence, Strasbourg, France 9 Faculté de médecine, Université Louis Pasteur, UMRS692, Strasbourg, France 10 Département de Neurologie, Hôpital Civil de Strasbourg, Strasbourg, France 11 AP-HP, Groupe Hospitalier de la Pitié-Salpêtrière, Fédération des maladies du système nerveux et centre de référence ‘canalopathies musculaires’, Paris, France

* To whom correspondence should be addressed at: UMR S546 (INSERM-UPMC), Faculté de médecine Pierre et Marie Curie, site Pitié-Salpêtrière, 105 boulevard de l'Hôpital, 75634 Paris cedex 13, France. Tel: +33 140778158; Fax: +33 140778117; Email: nicole{at}chups.jussieu.fr

Received May 11, 2008; Accepted July 19, 2008


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Schwartz–Jampel syndrome (SJS) is a recessive neuromyotonia with chondrodysplasia. It results from hypomorphic mutations of the gene encoding perlecan, leading to a decrease in the levels of this heparan sulphate proteoglycan in basement membranes (BMs). It has been suggested that SJS neuromyotonia may result from endplate acetylcholinesterase (AChE) deficiency, but this hypothesis has never been investigated in vivo due to the lack of an animal model for neuromyotonia. We used homologous recombination to generate a knock-in mouse strain with one missense substitution, corresponding to a human familial SJS mutation (p.C1532Y), in the perlecan gene. We derived two lines, one with the p.C1532Y substitution alone and one with p.C1532Y and the selectable marker Neo, to down-regulate perlecan gene activity and to test for a dosage effect of perlecan in mammals. These two lines mimicked SJS neuromyotonia with spontaneous activity on electromyogramm (EMG). An inverse correlation between disease severity and perlecan secretion in the BMs was observed at the macroscopic and microscopic levels, consistent with a dosage effect. Endplate AChE levels were low in both lines, due to synaptic perlecan deficiency rather than major myofibre or neuromuscular junction disorganization. Studies of muscle contractile properties showed muscle fatigability at low frequencies of nerve stimulation and suggested that partial endplate AChE deficiency might contribute to SJS muscle stiffness by potentiating muscle force. However, physiological endplate AChE deficiency was not associated with spontaneous activity at rest on EMG in the diaphragm, suggesting that additional changes are required to generate such activity characteristic of SJS.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Schwartz–Jampel syndrome (SJS, MIM 255800 [OMIM] ) is a recessively inherited disorder characterized by permanent, generalized muscle stiffness, initially described as myotonia with chondrodysplasia. The disease appears early in life, progressing slowly until adulthood, and has no effect on lifespan (1). SJS results from hypomorphic mutations in the HSPG2 gene encoding perlecan (2). Perlecan is a complex (~800 kDa) heparan sulphate proteoglycan (HSPG) secreted into all basement membranes (BM) and involved in cell adhesion, growth factor signalling and BM maintenance (3). Thirty HSPG2 mutations have been described in SJS, including missense and truncating mutations, which are found along the entire length of HSPG2 (4,5). The hypomorphic effect of these mutations results in a decrease in the amount of apparently full-length perlecan in BMs, but not to the complete absence of this protein. Indeed, the complete absence of perlecan in humans leads to dissegmental dysplasia, Silverman-Handmaker type (DDSH, MIM 224410 [OMIM] )—a type of chondrodysplasia lethal at birth (6). It has been suggested that a gene dosage effect accounts for the difference between SJS and DDSH, as no genotype–phenotype relationship could be established.

The muscle stiffness in SJS is permanent and leads to a characteristic ‘mask-like’ face with blepharospasm and pursed lips. It is associated with the recording of complex repetitive discharges (CRDs) at rest on electromyography (EMG) (S. Nicole et al., manuscript in preparation). This spontaneous activity is probably pre-synaptic in origin as it is abolished by the blockade of neuromuscular transmission with curare, leading to the classification of SJS as a form of neuromyotonia (7,8). The persistence of this activity during sleep and anaesthesia suggests a peripheral origin. Perlecan is present in the muscle and nerve BMs. It is more abundant at the neuromuscular junction (NMJ), where it is crucial for anchoring the collagen-tailed form of acetylcholinesterase (AChE) to the synaptic BM (9,10). AChE limits the duration of the synaptic response by hydrolysing acetylcholine (ACh). It has therefore been suggested that SJS neuromyotonia may result from endplate AChE deficiency. Consistent with this hypothesis, we observed very low levels of endplate AChE in one SJS patient (S. Nicole et al., manuscript in preparation). However, endplate AChE deficiency has also been reported in congenital myasthenia due to mutations in COLQ, which encodes the collagen-tail of endplate AChE (11,12). This form of myasthenia is clearly different from SJS, as it is characterized by muscle fatigability without CRDs on EMG. Moreover, the physiological characteristics of endplate AChE deficiency—repetitive and decrementing compound muscle action potentials (CMAPs) in response to single and repetitive nerve stimulations, respectively—are not recorded in SJS (S. Nicole et al., manuscript in preparation). The physiological significance of the partial endplate AChE deficiency in SJS therefore remains unknown.

We aimed to develop a mouse model of SJS for studies of SJS neuromyotonia. We introduced the p.C1532Y substitution, corresponding to a familial SJS missense mutation, into the mouse Hspg2 gene for three reasons: (i) this mutation is homozygous in siblings with severe features of SJS (2); (ii) the C1532 residue is conserved across species and is required for a well-defined pattern of disulphide bonds in domain III of perlecan and (iii) it is associated with the intracellular retention of mutant perlecan, decreasing extracellular perlecan amount (4). We also aimed to investigate the possibility of a dosage effect of perlecan in mammals, by deriving two lines from the initial strain: one with and the other without the selectable marker Neo for the down-regulation of perlecan gene expression. An independently generated strain of p.C1532Y mice that did not reproduce SJS neuromyotonia was recently described (13). It was concluded in this previous study that the p.C1532Y substitution had no effect in mice because it did not cause chondrodysplasia when present in isolation. Our results contrast with these data since our two mutant lines developed a neuromuscular phenotype mimicking SJS at the macroscopic and microscopic levels. The mutant phenotype induced by p.C1532Y was worsened by Neo, consistent with a dosage effect. We further demonstrated the occurrence of endplate AChE deficiency in the absence of perlecan unrelated to major muscle or NMJ disorganization. Finally, we showed that endplate AChE deficiency may contribute to SJS neuromyotonia by potentiating muscle force, but probably not by inducing spontaneous activity at rest.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Generation of a series of alleles with decreasing levels of perlecan secretion
We introduced the c.4595G>A point substitution, encoding p.C1532Y, into exon 36 of the mouse Hspg2 gene by homologous recombination (Fig. 1A). We established two mouse lines to test for a gene dosage effect: Hspg2C1532Y-Neo containing the selection cassette Neo in intron 36, and Hspg2C1532Y derived from Hspg2C1532Y-Neo by mating with CMV-Cre transgenic mice. The heterozygous mice of both lines were normal and fertile, and homozygous mutant mice were obtained at the expected frequency.


Figure 1
View larger version (57K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Generation of a series of mutant alleles with decreasing levels of perlecan secretion. (A) The homologous recombination strategy used. The vector contained the c.4595G>A mutation in exon 36 and the PGK-neomycine cassette flanked by two loxP sites (black triangles) in intron 36, 86 bp downstream from exon 36. Two mouse lines were derived: Hspg2C1532Y-Neo and Hspg2C1532Y. The locations of primers used for RT–PCR analyses are indicated by arrows. (B) Perlecan immunostaining on transverse gastrocnemius sections (a–c, anti-domain V) and fibroblasts (d–i, anti-domain IV) demonstrated that the mutants (Hspg2C1532Y: b, e and h; Hspg2C1532Y-Neo: c, f and i) had smaller amounts of perlecan than the controls (a, d and g). Cell membrane permeabilization (g–i) showed the intracellular accumulation of perlecan within mutant cells. Bars = 50 µm. (C) Dot-blot quantification of perlecan in conditioned cell culture medium confirmed that perlecan was secreted in smaller amounts than fibronectin by Hspg2C1532Y (2, grey bar) and Hspg2C1532Y-Neo (3, white bar) cells than by controls (1, black bar; *P = 0.041; **P < 0.001). Repeated twice with comparable results. (D) Qualitative western blot analyses of proteoglycans purified from conditioned culture medium and digested with heparitinase (7 µg per lane) confirmed the residual secretion of apparently full-length perlecan in mutant fibroblasts (2 and 3) compared to a control (1; Bar = 200 kDa).

 
We first checked whether the p.C1532Y substitution in mice led to a pattern similar to that observed in humans. Immunostaining of transverse muscle sections with antibodies against domains I, IV and V of perlecan demonstrated smaller amounts of apparently full-length perlecan in the BMs of 2-month-old Hspg2C1532Y and Hspg2C1532Y-Neo homozygous mutant mice than in those of control mice (Fig. 1B). We investigated the origin of this decrease by analysing primary fibroblast cells obtained from skin biopsies. The extracellular network formed by perlecan was less extensive in Hspg2C1532Y than in control cells and was almost absent from Hspg2C1532Y-Neo cells. Furthermore, membrane permeabilization with 0.1% Triton X-100 before immunostaining showed the intracellular accumulation of perlecan in Hspg2C1532Y and Hspg2C1532Y-Neo cells (Fig. 1B). This accumulation did not result from a failure to form extracellular matrix, as the fibronectin network was normal (data not shown). Dot-blot quantification of perlecan present in conditioned culture media confirmed that perlecan secretion in Hspg2C1532Y samples was lower than that in controls by a factor of 1.8 (n = 3; t-test, P < 0.001; Fig. 1C). The perlecan secretion in Hspg2C1532Y-Neo samples was lower than that in Hspg2C1532Y samples by a factor of 1.4 (n = 3; t-test, P = 0.041). Consistent with the residual secretion of full-length perlecan, a weak band of the expected size was observed on western blot analyses of conditioned culture media (Fig. 1D).

We further analysed the effect of c.4595G>A and Neo on Hspg2 mRNAs by RT–PCR analysis. Quantitative real-time (QRT) PCR, using an assay directed against Hspg2 exons 11-12, showed no significant decrease in Hspg2 RNA levels of muscle samples from 2-month-old Hspg2C1532Y mutants compared with controls (n = 3; t-test, P = 0.1; Fig. 2A). RT–PCR with primers binding to exons 33 and 38 detected no abnormal splicing events and confirmed the production of a full-length Hspg2 mRNA containing the c.4595G>A substitution (data not shown). In contrast, Hspg2 mRNA levels in Hspg2C1532Y-Neo muscle samples were 44% lower than those in Hspg2C1532Y samples (n = 3; t-test, P = 0.003). The amount of Hspg2 mRNA with the c.4595G>A substitution generated from the Hspg2C1532Y-Neo allele was estimated to be 36% that generated from the Hspg2C1532Y allele, based on semi-quantitative analyses (n = 6; t-test, P = 0.003; Fig. 2B). Neo contains cryptic splice sites when positioned in the same orientation as the targeted gene (14). We therefore searched for splicing events between Neo and flanking Hspg2 exons in Hspg2C1532Y-Neo muscle samples. RT–PCR analyses detected three alternative splicing events resulting in the insertion of 820 and 907 bp of Neo between Hspg2 exons 36 and 37 (Fig. 2C). The hybrid mRNAs were predicted to encode mutant perlecan lacking the C-terminal domains IV and V (p.D1542PfsX62 and p.D1542LfsX7).


Figure 2
View larger version (34K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Hypomorphic effect of Neo on Hspg2 gene expression. (A) QRT–PCR demonstrated the presence of smaller amounts of Hspg2 mRNA in the skeletal muscles of Hspg2C1532Y-Neo mutants (white bar) than in those of Hspg2C1532Y mutants (black bar; *P = 0.003). Repeated twice with comparable results. (B) Semi-quantitative analyses of RT–PCR products digested with TseI. The wild-type RT–PCR product obtained using primers binding within exons 35 and 41 contained five TseI sites (WT, arrows). One disappeared with c.4595G>A (star). Agarose gel resolution of TseI-digested RT–PCR products obtained from skeletal muscles of Hspg2C1532Y and Hspg2C1532Y-Neo heterozygous mice distinguished the WT (318 bp) from the mutant (403 bp) cDNAs. The mutant versus WT ratio was estimated at 1 for the Hspg2C1532Y samples (black bar). This ratio was lower in Hspg2C1532Y-Neo samples (white bar; *P < 0.05). Repeated four times with comparable results. (C) Schematic representation of the splicing events (1–3) occurring between Hspg2 (exons 35, 36 and 37) and Neo in Hspg2C1532Y-Neo muscles. Agarose gel resolution and sequences of RT–PCR products obtained with primers binding within exon 35, exon 37 and Neo (arrows) identified at least two hybrid mRNAs resulting from splicing between two acceptor and two donor splice sites in Neo and the donor and acceptor splice sites of Hspg2 exons 36 and 37.

 
Thus, the p.C1532Y substitution in mice had an effect similar to that observed in humans, with the intracellular accumulation of apparently full-length p.C1532Y mutant protein, resulting in smaller amounts of perlecan in BMs. The hypomorphic effect of this mutation was aggravated by the presence of the Neo selection cassette in Hspg2C1532Y-Neo, resulting in the generation of a series of alleles associated with decreasing amounts of perlecan present in BMs.

Hspg2C1532Y and Hspg2C1532Y-Neo mutants developed a neuromuscular phenotype mimicking SJS
We examined homozygous mutants of both lines to determine whether they developed a phenotype mimicking SJS. The mutants grew more slowly than their wild-type littermates, reaching a mean adult weight of 22.45 ± 1.17 g for Hspg2C1532Y-Neo (n = 10) and 25.7 ± 0.77 g for Hspg2C1532Y (n = 11), whereas the controls weighed 29.07 ± 0.94 g (n = 12; t-test, P ≤ 0.02). X-ray analysis suggested the occurrence of chondrodysplasia with reduced body length (15% shorter than that of the controls; t-test, P < 0.01) and hip dysplasia in Hspg2C1532Y-Neo mutants (Supplementary Material, Fig. S1).

Interestingly, a progressive neuromuscular phenotype was observed from the age of 2 months, with delayed opening of the eyelids and stiffened flexion of hind limbs when the animal was suspended by the tail (Fig. 3A). This phenotype was observed in both lines but progressed more slowly in Hspg2C1532Y mice: 54% of Hspg2C1532Y-Neo mutants had a severe phenotype (constantly closed eyes and long flexions of hind limbs), 31% were mildly affected (partially closed eyes and mild flexions) and 15% did not appear to be affected at the age of 6 months. At the same age, 54% of Hspg2C1532Y mutants had developed a mild phenotype, but none was severely affected ({chi}2 test, P = 0.007). Consistent with their more severe phenotype, the lifespan of Hspg2C1532Y-Neo mice was shorter than that of Hspg2C1532Y mice (n = 10; log rank test, P < 0.001), with the lifespan of Hspg2C1532Y mice not significantly different from that of controls (Fig. 3B).


Figure 3
View larger version (69K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Neuromuscular phenotype of Hspg2C1532Y and Hspg2C1532Y-Neo mutants. (A) A representative, severely affected six-month-old Hspg2C1532Y-Neo mutant (c and d) with constantly closed eyes and stiffened flexion of hind limbs and clenched toes, compared with a control (a and b). (B) Survival plots showed lower levels of survival for Hspg2C1532Y-Neo (dashed curve) than for Hspg2C1532Y mice (full curve). (C) EMG of mice demonstrated spontaneous activities with a constant amplitude and abrupt end in the gastrocnemius of Hspg2C1532Y-Neo mutants (b), which were not observed in controls (a). No spontaneous activity was observed in the diaphragms of mutants (d) and controls (c) between respiratory bursts. (D) Haematoxylin and eosin (a–c) staining and ATPase reaction (d–f) on transverse soleus sections showed groups of fibres with centralized nuclei and predominance of type 1 fibres, with grouping of type 2 fibres in six-month-old Hspg2C1532Y-Neo mutants (c and f), whereas Hspg2C1532Y muscle sections (b and e) were similar to controls (a and d). Bar = 100 µm (a–c) and 50 µm (d–f). Bar graphs showed a higher percentage of myofibres with centralized nuclei (soleus) in Hspg2C1532Y-Neo (white bar) than in Hspg2C1532Y mutants (grey bar).

 
EMG was carried out in Hspg2C1532Y-Neo mutants, to confirm the occurrence of muscle hyperactivity. Spontaneous activity resembling CRDs was recorded at rest in all mutant muscles tested (gastrocnemius, biceps brachialis, rectus abdominus) other than the diaphragm (Fig. 3C). This activity was stable in amplitude (400 µV on average) and frequency, up to 250 Hz, and lasted up to 30 s. It was induced by movements of the recording needle in all limb muscles and was constant in head muscles (buccinator and levator palpebrae muscles), highlighting the relationship between spontaneous activity and the severity of muscle stiffness. Unlike rest EMG, CMAPs were normal. We observed no repetitive or decrementing responses following single or repetitive (3 Hz) supramaximal nerve stimulation (data not shown).

Overall, the homozygous mutants had a phenotype mimicking SJS, with chondrodysplasia, muscle stiffness (particularly pronounced in the facial area) and spontaneous activity at rest on EMG. The difference in phenotype between the two lines was consistent with a dosage effect, with an inverse correlation between severity and the amount of perlecan secreted into BMs.

The presence of smaller amounts of perlecan in extrasynaptic BMs does not lead to major damage to myofibres
Mild and non-specific histological changes in muscle are reported in SJS patients, with variability in the size of myofibres, centralized nuclei, vacuoles and a predominance of slow type 1 fibres with grouping of fast type 2 fibres (1). We searched for similar changes in mutant mice, and observed variability in fibre size, centralized nuclei and a predominance of type 1 fibres, with groups of type 2 fibres in the soleus of 6-month-old Hspg2C1532Y-Neo mutants (Fig. 3D). Despite the presence of centralized nuclei suggestive of myofibre regeneration, signs of acute muscle degeneration–regeneration processes, such as the infiltration of mononuclear cells and necrotic fibres, were not seen. No histological changes were seen in the other Hspg2C1532Y-Neo muscles studied. None of the Hspg2C1532Y muscles, including the soleus, appeared to be affected at any age.

Perlecan is important for the maintenance of BMs subject to mechanical stress and is required for the initiation and stabilization of sarcomere unit organization in Caenorhabditis elegans and zebrafish (15–17). Muscle degeneration–regeneration might therefore be expected in SJS, and we investigated this possibility in Hspg2C1532Y-Neo mice. We first ruled out the possibility of the changes observed being developmental defects, as they were not observed in 2-month-old mutants. Evans blue dye injection showed no signs of myofibre necrosis in the soleus of 6-month-old Hspg2C1532Y-Neo mutants (data not shown). We studied the expression of BM and dystrophin–glycoprotein complex (DGC) components, as the DGC is essential for the mechanical stability of the sarcolemma, linking the actin cytoskeleton to the BM (18). Immunostaining revealed no marked change other than the up-regulation of muscle agrin, another HSPG found in the extrasynaptic BM (Supplementary Material, Fig. S2A) (19). Finally, the organization of sarcomeres was investigated by immunostaining for {alpha}-actinin, the actin-binding protein present at the Z-line, on dilacerated myofibres. This organization appeared to have been preserved, with a typical striated pattern observed (data not shown). These findings are not consistent with major muscle degeneration events in SJS, and indicate that such events were not involved in the pathogenesis of SJS neuromyotonia.

Partial endplate AChE deficiency without major NMJ disorganization in Hspg2C1532Y and Hspg2C1532Y-Neo mutants
Endplate AChE deficiency is thought to be responsible for SJS neuromyotonia because perlecan is essential for the anchoring of the collagen-tail form of AChE to the synaptic BM (9,10). Consistent with this function, endplate AChE was lower in adult mutant mice of both lines than in controls, as shown by fasciculin-2 staining and the Koelle reaction in all muscles tested [soleus, gastrocnemius, tibialis anterior (TA), extensor digitorum longus (EDL), sternomastoid and diaphragm; Fig. 4A and Supplementary Material, Fig. S3]. These lower concentrations were associated with smaller amounts of perlecan in the synaptic BM in Hspg2C1532Y and Hspg2C1532Y-Neo mice (Fig. 4A).


Figure 4
View larger version (48K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Endplate AChE deficiency in Hspg2C1532Y and Hspg2C1532Y-Neo mutants. (A) Staining of post-synaptic nAChRs ({alpha}-BTX; a, d and g) and AChE (fasciculin-2; b, e and h) on dilacerated myofibres (EDL), and perlecan on unrelated longitudinal biceps sections (merged images of perlecan immunostaining in green and nAChR staining in red; c, f and i) showed the presence of smaller amounts of perlecan (anti-domain IV) in the NMJs with endplate AChE deficiency in 2-month-old Hspg2C1532Y (d–f) and Hspg2C1532Y-Neo (g–i) mutants than in controls (a–c). Bar = 10 µm. (B) Sucrose gradient analyses of whole muscle extracts (TA) confirmed the loss of collagen-tailed AChE forms (left peaks) in mutants compared with controls (in pink), this loss being more pronounced in Hspg2C1532Y-Neo (in yellow) than in Hspg2C1532Y (in blue) mutants.

 
AChE is present in three forms in muscle: collagen-tailed AChE anchored to the synaptic BM, prima-anchored AChE bound to the membrane and soluble (globular) AChE (20). The collagen-tailed form is the major AChE form at the NMJ. It consists of three ColQ units, which attach tetramers of AChE to the BM by binding to perlecan, itself attached to {alpha}-dystroglycan (21). Sedimentation velocity analyses confirmed the specific loss of collagen-tailed AChE in Hspg2C1532Y and Hspg2C1532Y-Neo mutants, with levels reduced to 52 (t-test; P = 0.001) and 18% (t-test; P < 0.001) of controls, respectively (Fig. 4B). The decrease in collagen-tailed AChE was more pronounced in Hspg2C1532Y-Neo compared with Hspg2C1532Y (t-test, P = 0.001). Other forms of AChE were not statistically modified compared with controls. This decrease probably reflected the loss of endplate AChE, because collagen-tailed AChE was detected only in muscle extracts containing NMJs in studies of wild-type mice (E. Krejci and V. Bernard, unpublished data). The possibility of the observed decrease at morphological level being due to redistribution of the collagen-tailed AChE along the myofibres was also excluded.

Denervation–reinnnervation processes occurred in SJS patients, potentially accounting for endplate AChE deficiency because synaptic AChE transcript levels are down-regulated by denervation (S. Nicole et al., manuscript in preparation; 22). We investigated this possibility in Hspg2C1532Y-Neo mutants. QRT–PCR detected no decrease in Ache and Colq gene expression in 2-month-old mutants (data not shown). Fluorescent double-staining of nerve terminals and AChE further excluded the possibility of endplate AChE deficiency being due to denervation, as weaker AChE staining was observed in well-innervated NMJs (Fig. 5A). Double-staining of nerve terminals and post-synaptic AChRs demonstrated the partial denervation of 12% of NMJs (n = 153) in 8-month-old mutants, with thin terminal nerves suggestive of nerve sprouting (Fig. 5B). This proportion was far lower than the 100% of NMJs showing AChE deficiency. Immunostaining of terminal Schwann cells (tSCs), the third cellular component of the NMJ, showed extension of cellular processes in 19% of mutant NMJs (n = 16; Fig. 5C), demonstrating the occurrence of reinnervation, as tSCs cells develop processes to guide nerve sprouting (23).


Figure 5
View larger version (65K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Denervation-reinnervation processes in Hspg2C1532Y-Neo mutants. (A) Double-staining of AChE (fasciculin-2, a and d) and nerve terminals (neurofilament and SV2; b and e) showed the amounts of endplate AChE to be smaller in well innervated NMJs in 2-month-old mutants (d–f) than in controls (a–c). (c and f): merged image. (B) Double-staining of nAChRs ({alpha}-BTX; a and d) and nerve terminals (neurofilament and synaptophysin; b and e) showed denervated patches of nAChRs (arrowheads in f) in 8-month-old mutants (d–f) not observed in controls (a–c). (c and f): merged image. (C) Double-staining of nAChRs (a and d) with Schwann cells (S100; b and e) showed tSC processes beyond residual clusters of nAChRs in mutant NMJs (d–f, arrows) not observed on the complete superimposition of staining seen in controls (a–c). (c and f): merged image. Bars = 10 µm (A–C).

 
In addition to denervation–reinnervation processes, we observed NMJ remodelling in 96% of mutant NMJs, associated with a streaky pattern, and weaker pretzel-like organization of nAChRs than for controls at the age of 8 months (n = 79; 1.56 ± 0.16 branching points versus 5.15 ± 0.28 in controls, P = 0.045; Supplementary Material, Fig. S4A). Mutant NMJs also covered only 58% the area of control NMJs (391 ± 42 versus 676 ± 64 µm2, respectively; t-test, P = 0.02). The diameter of myofibres was smaller in mutants (38.3 ± 2.3 versus 51.2 ± 3.4 µm in controls; t-test, P = 0.034), maintaining the proportionality between NMJ area and fibre diameter, probably to preserve the efficiency of neuromuscular transmission (Supplementary Material, Fig. S4B) (24). We investigated the distribution and level of major NMJ components by immunostaining, to determine whether any major molecular disorganization occurred. No changes other than utrophin down-regulation were observed (Supplementary Material, Fig. S2B). MuSK and dystroglycan, which attach AChE–ColQ and AChE–ColQ–perlecan complexes, respectively, to the synaptic sarcolemma, were unaffected (10,25). NMJ ultrastructure also appeared to be conserved (n = 31 from five mutants and n = 11 for controls; Supplementary Material, Fig. S4C). In particular, events seen in ColQ-deficient mice, such as invasion of the synaptic cleft by tSCs and post-synaptic necrosis, were not observed (26).

Overall, our data show that the endplate AChE deficiency resulted from a decrease in the amount of perlecan at the NMJ rather than any major NMJ disorganization.

Endplate AChE deficiency leads to the potentiation of muscle force, with a prolonged decay time of endplate potentials
The association of endplate AChE deficiency with congenital myasthenia called into question the physiological significance of endplate AChE deficiency in SJS. We addressed this question by studying 8-month-old Hspg2C1532Y-Neo mutants. We first studied the isometric contractile properties of the TA in situ in response to nerve stimulation, as AChE-null mice display potentiation with tetanic fade of muscle force (27). The specific maximal tetanic force (sP0, maximal tetanic force related to muscle mass) of mutants was not lower than those of controls, excluding muscle weakness (Table 1). Absolute (Pt) and specific (sPt) twitch forces were found to be potentiated in mutants, resulting in a higher twitch/tetanus ratio (Pt(%P0)), time to peak tension (CT) and half-relaxation time (1/2 RT). Potentiation was not limited to twitch forces, as specific submaximal tetanic force at 25 Hz (sP25) was 1.2 times higher in mutants than in controls. Moreover, analysis of the tetanic tension during a 500 ms nerve stimulation train showed a tetanic fade at low (12.5 and 25 Hz) but not at high (50 and 100 Hz) frequencies (Fig. 6A). Despite these modifications, the relaxation velocity, calculated from the tetanic force at 100 Hz, was unaffected. We measured muscle force evoked by direct muscle stimulation after blockade of post-synaptic AChRs with gallamine triethiodide (a synthetic non-depolarizing drug) to exclude a post-synaptic origin of the force changes. Twitch/tetanus ratio decreased from 44.8 ± 5.7% in response to nerve stimulation to 24.2 ± 1.6% in response to direct muscle stimulation (n = 3; t-test, P = 0.048,) in mutants, whereas it did not significantly change in controls (30.7 ± 8.4% versus 19.7 ± 1.9%; t-test, P = 0.251). Tetanic fade at 12.5 and 25 Hz were no more recorded in Hspg2C1532Y-Neo mutants when muscle stimulation was applied (Fig. 6A). These results agreed with endplate AChE loss as being at the origin of the muscle force changes.


Figure 6
View larger version (20K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Physiological effect of endplate AChE deficiency in 10-month-old Hspg2C1532Y-Neo mutants. (A) Representative force recordings for the TA muscle in situ of one control (a) and two mutants (b–d) in response to nerve (a–c) and direct muscle (d) stimulation. Note the potentiation of single twitch and tetanic fade at low nerve stimulation frequencies (12.5 and 25 Hz) in the mutants (b). Force recordings of the same mutant muscle showed that tetanic fade was evoked by nerve (c) but not by muscle (d) stimulation after inhibition of neuromuscular transmission. (B) Representative electrophysiological traces of MEPPs (a) and single EPPs (b) recorded at 30°C in a control and in a mutant hemidiaphragm muscle bathed in standard physiological solution containing µ-conotoxin GIIIB to block muscle contraction. The decay time of the EPP in the mutant was longer than that in the control, indicating AChE inhibition. Paired EPPs evoked at 30 ms intervals in control (c) and in mutant NMJ (d) showed prolongation of the decay phase of the second EPP with respect to the first EPP in the mutant, indicating AChE inhibition. The time to peak and the half-decay time are indicated in d by dotted lines. The resting membrane potential during measurements was –72 mV.

 


View this table:
[in this window]
[in a new window]

  		Table 1. Contractile properties of the TA muscle in situ evoked by nerve stimulation

 
We also studied isolated hemidiaphragm muscles ex vivo, as the diaphragm displayed no spontaneous activity on EMG, despite being deficient for endplate AChE. Supramaximal phrenic nerve stimulation evoked potentiated twitch and tetanic responses, depending on the stimulation frequency, as for TA. The twitch/tetanus ratio was up to 30% higher in mutant muscles than in controls at low frequencies (10–60 Hz), and was similar to that in controls at higher frequencies (up to 100 Hz, Supplementary Material, Fig. S5). The potentiation was not due to a difference in muscle weight, as it was similar in mutants and controls (144.9 ± 25.5 and 142.5 ± 6.4 g, respectively, t-test, P = 0.91). However, tetanic fade was not observed in any case, regardless of the frequency used for nerve stimulation, in contrast to what was observed for the TA of Hspg2C1532Y-Neo mutants and diaphragm of AChE-null mice (28). The comparison of the maximum force-time integral of twitch responses evoked by nerve stimulation to responses evoked by direct muscle stimulation (after suppressing neuromuscular transmission with 10 µM tubocurarine) revealed that twitch force-time integrals resulting from direct muscle stimulation were significantly reduced when compared with nerve stimulation in Hspg2C1532Y-Neo mutants (389.0 ± 64.4 versus 539.6 ± 47.7 mNms; n = 3; t-test, P = 0.049). Under the same conditions, no significant difference was observed in controls (P = 0.386).

We further evaluated neuromuscular transmission using intracellular recordings on isolated hemidiaphragm muscles in the presence of physiological Ca2+ concentrations (2 mM). Miniature endplate potentials (MEPP) and endplate potentials (EPP) were found to be prolonged in mutants when compared with controls, as expected for lower levels of endplate AChE activity (Fig. 6B) (29). Spontaneous MEPPs decayed 1.3 times more slowly in mutant NMJs than in controls at similar resting membrane potentials (range –60 to –70 mV; Table 2). Full-sized EPPs evoked by nerve stimulation at low frequency (0.1 Hz) also decayed more slowly in mutant NMJs than in control NMJs. Consistent with the partial, as opposed to complete AChE deficiency, mutant NMJs remained sensitive to AChE inhibition with 3 µM neostigmine. This treatment prolonged the half-decay time of MEPPs from 2.3 ± 0.08 to 3.4 ± 0.2 ms (n = 30). The prolonged decay time of EPPs and MEPPs in mutant NMJs probably reflects endplate AChE deficiency, even though EPPs and MEPPs give no direct measurement of the underlying change in conductance, because of the passive electrical properties of myofibres. MEPP frequency was significantly decreased, unlike the quantal content of EPPs (Table 2). These changes suggest that there are mechanisms compensating for endplate AChE deficiency, preserving neuromuscular transmission.


View this table:
[in this window]
[in a new window]

  		Table 2. Synaptic transmission parameters of the hemidiaphragm in vitro

 
Our data demonstrate a physiological effect of endplate AChE decrease with prolonged EPPs, which is probably responsible for the potentiation of isometric muscle tension evoked by nerve stimulation at low frequencies in TA and diaphragm. These changes may result from the slower disappearance of ACh within the synaptic cleft, which may have induced multiple myofibre action potentials due to repetitive binding to post-synaptic nAChRs.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
SJS consists of neuromyotonia with CRDs on EMG and chondrodysplasia due to a partial deficiency of perlecan within the BM. We describe here mouse models with low levels of perlecan in the BM, resulting not only in chondrodysplasia, as previously reported (13), but also in muscle stiffness with spontaneous activity on EMG. Using these models, we demonstrated for the first time that partial endplate AChE deficiency occurs as a result of synaptic perlecan deficiency in SJS, whereas myofibre and NMJ organizations are preserved. Moreover, we showed that the AChE deficiency exerts a physiological effect on neuromuscular transmission, which is most likely at the origin of the muscle force potentiation and fatigability evoked by low frequencies of nerve stimulation. However, our data suggest that endplate AChE deficiency is not sufficient to induce spontaneous activity on EMG.

We are confident that our mutants constitute accurate models of SJS for four reasons. First, the phenotype resulted from the intracellular retention of mutant perlecan in the Hspg2C1532Y line, leading to the secretion of smaller amounts of perlecan into the BMs, as in human cells homozygous for the orthologous mutation (4). Second, muscle stiffness was most marked for the head of the mutant mice, mimicking the ‘mask-like’ face of SJS patients. Third, spontaneous muscle activity, similar to the CRDs recorded for SJS patients, was recorded on EMG. Fourth, the muscle changes observed, with partial endplate AChE deficiency and NMJ remodelling with denervation–reinnervation and no major ultrastructural abnormalities, mimicked those reported to be associated with truncating HSPG2 mutations in humans (S. Nicole et al., manuscript in preparation). One difference was the progression of the disease, as it worsened with ageing in mice. The reason for this difference remains unclear. However, the patients homozygous for the p.C1532Y mutation were 15 and 18 years old at their most recent check-up, and the progression of the disease caused by this mutation in adults therefore remains unclear. Neuromyotonia would also be expected to lead to prolonged muscle relaxation, as reported in two patients with SJS (8). Our measurements of isometric tension in the TA in situ detected no prolongation of muscle relaxation in mutants, despite the presence of spontaneous activity on EMG. This suggests ‘latent’ neuromyotonia in our mutant mice, i.e. muscle hyperactivity that does not lead to disabling muscle stiffness (30), and is reminiscent of SJS cases with mask-like expression being the major or only sign of SJS neuromyotonia (4).

It was recently suggested that the p.C1532Y substitution acts as a splicing mutation in humans, as it does not induce severe chondrodysplasia in mice (13). Our results contrast with these findings, as our homozygous Hspg2C1532Y mutants developed a phenotype mimicking SJS at the macroscopic and microscopic levels. The p.C1532Y substitution was not associated with abnormal splicing events in mice (this study), or in lymphoblasts and fibroblasts derived from patients homozygous for p.C1532Y (S. Nicole, unpublished data). However, we observed intracellular accumulation of the mutant p.C1532Y protein, with lower than normal extracellular perlecan levels, as reported in humans (4). Domain III consists of modules with a uniform disulphide bridge pattern based on eight cysteine residues, giving rise to four loops (31). The p.C1532 residue is involved in this pattern, and its substitution probably results in the generation of a misfolded protein, which is retained within cells by quality control pathways (2). Consistent with this hypothesis, p.C1532Y prevents the secretion of recombinant mouse domain III by mouse cells (13). The different effects of p.C1532Y in our lines and in those previously published may result from the different ages of the animals studied. Rodgers’ group studied mice between birth and the age of 2 months (13). We studied adult mice and showed that they developed a neuromuscular phenotype with ageing. The different genetic backgrounds used (mixed 129Sv-R1 versus mixed 129Sv-BL/6J for our line) may also have contributed to the divergence of the results obtained, as genetic background may modify the phenotype resulting from a single mutation (32).

We found that the PGK-Neo cassette worsened, but did not induce, the abnormal phenotype, by decreasing the production of mutant RNA. The lower levels of Hspg2 RNA generated from the Hspg2C1532Y-Neo allele were probably due to the presence of the transcription termination signal of PGK-Neo in addition to the splicing events we observed between Neo and Hspg2. The observed Hspg2-Neo hybrid mRNAs were predicted to encode truncated forms of perlecan. They were probably subject to nonsense-mediated decay, because of the premature stop codons they contain. This may account for the smaller amounts of Hspg2 mRNA detected in Hspg2C1532Y-Neo samples with a probe binding 5' end to the mutation. The lower levels of secretion of the mutant perlecan in Hspg2C1532Y-Neo mice were associated with a phenotype more severe than that of Hspg2C1532Y mice, suggesting dosage effect. A dosage effect associated with perlecan mutations has already been reported in Drosophila melanogaster and C. elegans, with phenotypes ranging from larval lethality to adult viability, depending on residual perlecan levels (1). Dosage effects may account for the difference between our lines and the lines of Rodgers' group. They may also account for the difference in severity for each mutant mouse. Indeed, our mutant lines were not established in a homogeneous genetic background, and the level of residual mutant perlecan secretion may vary between mutant mice. Establishing the same p.C1532Y mutation into different genetic backgrounds, then comparing the levels of perlecan and the severity of bone and muscle phenotype in mutants would resolve this issue. Eventually, it would allow to search for genetic factors modifying the expression of the disease. SJS clinical signs are also of variable severity in humans, ranging from almost asymptomatic neuromyotonia without bone changes to severe forms requiring the use of a wheelchair (33,34). The demonstration of a dosage effect in mammals is important, as it may open new possibilities for treatment in humans.

Perlecan is involved in BM maintenance in mammals, and is crucial for sarcomere organization in C. elegans and zebrafish (15–17). However, no muscle dystrophy with major muscle degeneration processes is observed in either humans or mice with SJS, consistent with the lack of muscle weakness detected on muscle force analyses. Perlecan is probably present in sufficient quantities to assume these muscle functions. Alternatively, the loss of perlecan may be compensated by agrin, probably produced in larger amounts due to denervation–reinnervation (19). The extrasynaptic isoform of agrin, which is not involved in NMJ formation, interacts with transmembranous receptors and BM components, and can rescue muscle dystrophy resulting from laminin {alpha}2 mutations (35). Crossing our perlecan-mutant line with mice deficient for the extrasynaptic form of agrin would determine whether this HSPG compensates for the loss of perlecan. Perlecan deficiency results in the redistribution of myofibre types, as observed in both mutant mice and SJS patients (N. Romero and M. Fardeau, unpublished observations). Our results demonstrated that this abnormal pattern was not developmental, as it appeared with age in mice. Instead, it probably resulted from denervation–reinnervation processes, as the redistribution of fibre types occurs after the partial crushing of nerves (36). The soleus was the only muscle studied to display a mosaic pattern. Thus, the grouping of type 2 myofibres may result from the sprouting of adjacent intact motor axons, whereas the predominance of type 1 myofibres may reflect the higher sprouting efficiency of motor nerve terminals innervating slow myofibres (37).

This study confirms that partial endplate AChE deficiency is a consistent feature of SJS. The accumulation of AChE at the NMJ results from a complex process involving the production of larger amounts of AChE and ColQ transcripts by subsynaptic nuclei and multiple interactions within the synaptic BM and sarcolemma (38). Our results demonstrate that the endplate AChE deficiency in SJS is related to the decrease in synaptic perlecan levels rather than myofibre degeneration, NMJ remodelling or denervation. It remains unclear where the binding interaction between perlecan and the AChE–ColQ complex occurs. Preliminary studies have suggested that the AChE–ColQ–perlecan complex assembles within the myofibre, at least during development (38). The huge decrease in endplate AChE levels may therefore result from a failure to secrete this complex, followed by intracellular degradation when perlecan is itself poorly secreted. Alternatively, the AChE–ColQ complex may be secreted and then diffused or degraded when not retained within the synaptic BM. Studies on perlecan-null cells are required to address this question.

In addition to highlighting the role of perlecan in endplate AChE anchoring, our results show a correlation between NMJ remodelling, denervation–reinnervation and decreases in perlecan levels. These changes may result from or reflect an adaptation to endplate AChE deficiency, as they are also observed in ColQ-null mice (26). However, they may also indicate a role for perlecan in NMJ maturation. No changes to the NMJ other than AChE deficiency were reported in perlecan-null embryos at E18, excluding an essential role for perlecan in NMJ formation (9). However, NMJ maturation occurs after birth, and could not be studied in perlecan-null mice due to embryonic lethality. NMJ maturation includes transformation of the rounded plaque into an elaborated pretzel-like shape of pre- and post-synaptic elements, with the formation of synaptic folds (39). The DGC plays an important role in these steps. NMJ remodelling with utrophin and AChE deficiencies are observed in mice deficient in dystroglycans (40). A major streaky pattern of nAChRs with a patchy distribution and undefined borders of gutters is observed in mice lacking {alpha}-dystrobrevin and {alpha}-syntrophin (41,42). These observations suggest that perlecan may be involved in NMJ maturation, as it is a major binding partner of {alpha}-dystroglycan (43).

AChE deficiency at the NMJ has already been linked to congenital myasthenia with loss-of-function mutations in COLQ (11,12). The clinical and physiological divergence between this form of myasthenia and SJS raised questions about the physiological effect of the endplate AChE deficiency in SJS. We demonstrated that the endplate AChE deficiency is of physiological significance, prolonging miniature and nerve-evoked EPPs and most likely potentiating muscle force. However, these changes were not associated with repetitive or decrementing responses on EMG. Nevertheless, we observed tetanic fade at 12.5 and 25 Hz in the TA of Hspg2C1532Y-Neo mutants. These frequencies are within the range of physiological frequencies (14–70 Hz in mice), suggesting that muscle fatigability may occur in SJS. The disappearance of tetanic fade at higher frequencies and its absence from the diaphragm probably result from pre- and post-synaptic modulations to compensate for ACh accumulation in the synaptic cleft. The preserved ultrastructural organization of NMJs may also account for the well sustained nerve-evoked muscle contraction at high frequencies in SJS (24,26,28).

Finally, could SJS neuromyotonia result from ACh accumulation in the synaptic cleft induced by endplate AChE deficiency? A distinction should be made between muscle stiffness and spontaneous activity on EMG at rest when addressing this question. Our data demonstrate potentiation of muscle force and increased twitch half-relaxation time, which most likely result from the endplate AChE deficiency documented by the prolongation of EPPs. Endplate AChE deficiency may contribute to SJS muscle stiffness through this potentiation. However, the endplate AChE deficiency does not appear to be responsible for the spontaneous activity observed on EMG since the diaphragm was devoid of spontaneous activity despite the demonstration of physiological AChE deficiency. Preliminary observations have indicated that spontaneous activity in Hspg2C1532Y-Neo mutants originates from the distal motor nerve (A. Echaniz-Laguna and S. Nicole, unpublished data). Perlecan is present in axonal BMs. Axonal changes may therefore result from decreases in perlecan levels, accounting for spontaneous activity on EMG in isolation or together with endplate AChE deficiency. One plausible hypothesis is a dysfunction of pre-synaptic potassium channels, a known cause of neuromyotonia (44). Our mouse models will be valuable tools for investigating these hypotheses, for gaining insight into SJS neuromyotonia, and for testing new therapeutic approaches.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Mouse lines
The mutant mouse lines were generated from two independent targeted ES cell clones established at the ICS-MCI (Mouse Clinical Institute, Illkirch, France) by homologous recombination. Briefly, two fragments encompassing introns 33 to 37 of the Hspg2 gene were amplified by long-range PCR from 129Sv genomic DNA for introduction of the c.4595G>A (p.C1532Y) point mutation into exon 36, and were inserted into an ICS-MCI proprietary vector. The PGK-neomycin resistance cassette (Neo), containing a polyadenylation signal, was flanked by two loxP sites and inserted into intron 36 in the same orientation as Hspg2, 86 bp downstream from exon 36. Targeted 129Sv ES clones were identified by PCR, with confirmation by Southern blotting. Male chimeras resulting from the injection of the targeted ES clones into C57BL/6J blastocysts were mated with C57BL/6 females. Mice were genotyped by PCR amplification, using primers binding within Hspg2 exon 36, intron 36 and Neo (sequences available on request). All animal procedures were performed in accordance with institutional guidelines (agreements 75-952 and 3959). Wild-type littermates were used as controls. At least three mice from each group were analysed in all experiments.

Reverse transcription-polymerase chain reaction
Total RNA was extracted from mouse quadriceps or TA using Trizol reagent (InvitrogenTM) and was reverse-transcribed with random hexamers, according to the manufacturer's protocol (ThermoscriptTM RT–PCR System, InvitrogenTM). The presence of the c.4595G>A mutation in Hspg2 was confirmed by sequencing cDNAs after PCR amplification with primers binding within Hspg2 exons 33 and 38 or 35 and 41. Semi-quantitative PCR and TseI digestion were done using muscle samples of 2-month-old heterozygous mice. Quantization of digested products was performed using the free Image J software. Splicing events between Neo and Hspg2 were studied using primers binding within exons 35, 37 and Neo, respectively (all primer sequences are available on request). The RT–PCR products were sequenced using the BigDye v3.1 Terminator kit (Applied Biosystems) and an automated sequencer (ABI3100, Applied Biosystems).

QRT–PCR analyses were performed using a TaqMan Real-Time PCR detection system (ABI7000, Applied Biosystems). The products of the reverse transcription reaction (100 µg), QPCR Mastermix (Eurogentec) and the prevalidated TaqMan® Gene expression assays (Applied Biosystems) Hs99999901_s1 (universal 18S RNA, used for the normalization of the results), Mm00464544_g1 (Hspg2), Mm 00477275_m1 (Ache), and Mm01211248_m1 (Colq) were used according to the manufacturers’ protocols. Colq and Ache probes were considered to detect all mRNA isoforms as they bind to sequences encoding the triple helical domain of ColQ and the esterase domain of AChE, respectively. All samples were measured in triplicate and the results were expressed as 2–{Delta}{Delta}Ct to determine the amount of Hspg2 mRNA relative to controls (User Bulletin #2 ABI PRISM 7700, Applied Biosystems, December 1997).

Cell culture
Primary fibroblast cells were established from mouse skin biopsy specimens, as previously described (4).

Biochemistry
For dot-blotting, 5.104 fibroblasts were incubated in serum-free medium for 2 days. Serial dilutions of conditioned medium were applied to nitrocellulose membrane (Amersham) using a dot-blot apparatus (BioRad Laboratories). The membrane was blocked with non-fat milk in phosphate-buffered saline (pH 7.4; PBS) and incubated with monoclonal antibodies to perlecan domain IV, or polyclonal antibodies to fibronectin (Sigma-Aldrich) as control of equal deposition of conditioned medium. After intensive washings, membranes were incubated with secondary horseradish peroxidase-conjugated antibodies (Jackson Immunoresearch laboratories). Chemiluminescence (PerbioScience) was detected by autoradiography. Dot quantification was performed using ImageJ software, and perlecan protein levels were normalized to fibronectin.

Western blotting for perlecan was carried out as previously described (45). Briefly, proteoglycans were purified by ion-exchange chromatography from conditioned culture media. Proteins (7 µg) were digested overnight at 37°C with 4 mU of heparitinase (Seikagaku Corporation) in 100 mM sodium acetate buffer, pH 7.0, 10 mM calcium acetate, 0.1% bovine serum albumin (BSA) with protease inhibitors (10 mM EDTA, 10 mM ethyl maleimide, 1.25 mM phenylmethanesulfonyl fluoride, 1 µM pepstatin A). Digestion products were separated by SDS–PAGE and blotted onto a membrane, which was then processed as for dot-blotting.

Sucrose gradient sedimentation was performed on TA ground in liquid nitrogen. The powder was homogenized in 25 mM Tris–HCl, pH 7.4, 10 mM EDTA, 800 mM NaCl, and 1% CHAPS supplemented with protease inhibitors. Sedimentation analyses of AChE forms were carried out with 5–20% (wt/vol) sucrose gradients containing 50 mM Tris–HCl, pH 7, 10 mM MgCl2, 800 mM NaCl and 1% Brij-97. We collected 48 fractions from each gradient, which were assayed for AChE activity in the presence of 50 mM tetraisopropylpyro-phosphoramide (Iso-OMPA, Sigma-Aldrich). Optical density at 414 nm was determined several times for each fraction. The area of the peaks was used to estimate the quantity of collagen-tailed form of AChE in muscle extracts.

Muscle specimens
Fresh skeletal muscles were snap-frozen in liquid nitrogen-cooled isopentane (Sigma-Aldrich) at –160°C and stored at –80°C for immunostaining on transverse (8 µm) or longitudinal (20 µm) cryosections. For whole-mount preparations, mice were transcardially perfused with a mixture of 2% paraformaldehyde (PFA, Sigma-Aldrich) and 0.2% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4), and dissected muscles were post-fixed by overnight incubation in 2% PFA. EDL of transcardially perfused mice was post-fixed by overnight incubation in 2% PFA for electronic microscopy. The NMJs were identified by detecting nAChRs with {alpha}-bungarotoxin ({alpha}-BTX) coupled to biotin (Molecular Probes), using a pre-embedding immunogold method. Muscles were treated with 1% osmium, dehydrated and embedded in resin. Ultrathin sections were cut, stained with lead citrate and examined in a Philips CM120 EM.

Histo- and immunochemistry
Transverse sections of the gastrocnemius, soleus, TA and EDL were stained with haematoxylin–eosin and labelled for myofibrillar ATPase by standard methods. Mice were intraperitonally injected with Evans blue dye (Sigma-Aldrich, 10 µg per 10 g body weight) and killed 9 h later to study muscle necrosis. Then gastrocnemius and soleus were dissected and processed for transverse cryosectioning. For immunohistochemistry, muscle sections were fixed (acetone, methanol or PFA, depending on the primary antibody), blocked with 3% goat serum (Jackson ImmunoResearch Laboratories), incubated with antibodies in 3% BSA in PBS, counterstained with Hoechst 33258 (Sigma-Aldrich) and mounted in Vectashield mounting medium (Vector Laboratories). EDL-teased muscle fibres were used for whole-mount fluorescence staining of NMJs in 3% PBS–BSA.

Staining was photographed under conventional (Leica DMRA, Leica Microsystems) or confocal (Leica SP2) fluorescence microscopes. Maximal z-projections were analysed with Image J software. The area covered by nAChR clusters was evaluated by {alpha}-BTX staining. NMJ shape was evaluated by measuring the number of branching points (46).

Antibodies
The following primary antibodies were used: rat anti-perlecan domain IV (MAB1948, Chemicon International), rabbit anti-perlecan domains I and V (kindly provided by the late Professor R. Timpl), rabbit anti-agrin (gift from Professor Dr MA Rüegg), rat anti-laminin-2 (4H8-2, Sigma-Aldrich), mouse anti-dystrophin (MANDRA1, Sigma-Aldrich), mouse anti-{gamma}-sarcoglycan (35DAG/21B5, Novocastra Laboratories), mouse anti {alpha}-dystroglycan (VIA4-1, Novocastra Laboratories), mouse anti-β-dystroglycan (NCL-b-DG, Novocastra Laboratories), rabbit anti-laminin {alpha}5 (gift from Dr J.H. Miner), mouse anti-syntrophin (1351, Affinity Bioreagents), mouse anti-SV2 (Developmental Studies Hybridoma Bank), mouse anti-2H3 (Developmental Studies Hybridoma Bank), mouse anti-rapsyn (1234, Affinity Bioreagents), rabbit anti-MuSK (Affinity Bioreagents), mouse anti-utrophin (clone DRP3/20C4, Novocastra Laboratories), mouse anti-neurofilament 200 (NE14, Sigma-Aldrich), mouse anti-synaptophysin (SVP-38, Sigma-Aldrich), mouse anti {alpha}-actinin (A7811, Sigma-Aldrich) and rabbit anti-S100 (Z0311, DakoCytomation). The fluorescein isothiocyanate (FITC)-, Texas Red (TR)- and Cy3-conjugated secondary antibodies were obtained from Jackson ImmunoResearch Laboratories. Biotinylated fasciculin-2 and Alexa 488-conjugated streptavidin (Molecular Probes, Invitrogen Detection Technologies) were used to label AChE. NMJs were counterstained with tetramethylrhodamine (TRITC) or FITC-conjugated {alpha}-BTX.

Electromyography
All recordings were made with a standard EMG apparatus (Dantec, Les Ulis, France) in accordance with the guidelines of the American Association of Electrodiagnostic Medicine. Six and 10-month-old mice (n = 12 for mutants and n = 10 for controls) were anaesthetized with 1 mg/kg ketamine chlorhydrate and 0.5 mg/kg xylazine. A monopolar needle electrode (diameter, 0.3 mm; 9013R0312; Medtronic) was inserted into the tail of the mouse to ground the system. Recordings were made with a concentric needle electrode (diameter, 0.3 mm; 9013S0011; Medtronic). Muscles were monitored on both sides (right and left) for at least 2 min. Ten supramaximal nerve stimulations at slow rate (3 Hz) were applied to the sciatic nerve and the motor unit action potential (MUAP) was recorded to search for decrementing responses to repetitive stimulation. A sixth response 10% weaker than the first response was considered pathological. A more drastic protocol was also applied in which MUAPs were recorded in response to five supramaximal nerve stimulations at low frequency (3 Hz) after a series of three sets of 100 supramaximal stimulations at high frequency (30 Hz).

Isometric tension analyses
The contractile properties of the TA in situ and of isolated phrenic nerve-hemidiaphragm preparations ex vivo were studied as previously described (27,28,47). Fatigue resistance was measured as the time taken for submaximal tetanic force to decrease by 20% during prolonged stimulation at 50 Hz. For TA, muscle tension was recorded in the same muscle in response to nerve stimulation, then in response to direct muscle stimulation after blockade of neuromuscular transmission with intramuscular injection of 30 µl of gallamine triethiodide (20 mg/ml), a synthetic non-depolarizing drug. PBS was injected as control in the controlateral muscle. To determine the contractile force for hemidiaphragms stimulated directly, a bath of 4 ml with an electrode array placed on each side of the muscle was used. Under this condition, neuromuscular transmission was blocked with 10 µM tubocurarine to avoid any pre-synaptic stimulation influence on the muscle tension recordings. The force-time integrals were calculated from twitch force signals using the software kindly provided by Dr John Dempster (Department of Physiology and Pharmacology, University of Strathclyde, Scotland).

Ex vivo electrophysiological studies
Electrophysiological recordings were carried out with conventional techniques (48). Isolated hemidiaphragm muscles were mounted in thermoregulated and oxygenated standard physiological solution [154 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 5 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) and 11 mM glucose, pH 7.4]. Membrane and synaptic potentials were recorded from endplate regions at 31 ± 1°C with intracellular microelectrodes (filled with 3 M KCl solution, 8–12 M{Omega} resistance) and an Axoclamp-2A system (Axon Instruments). The distal part of the phrenic nerve was stimulated by a suction electrode with current pulses of 0.1 ms duration and a supramaximal voltage (typically 3–8 V) at 0.1 Hz unless otherwise specified. Electrical signals were collected after amplification and digitized at a sampling rate of 25 kHz with a computer equipped with an analogue-to-digital interface board (Digidata 1200, Axon Instruments), using software kindly provided by Dr John Dempster. Studies on nerve-evoked EPPs were performed in standard physiological solution containing 1.6 µM µ-conotoxin-GIIIB (Alomone Labs, Jerusalem, Israel), a selective muscle Na+ channel blocker, to inhibit the generation of muscle action potentials (49). The amplitudes of full-sized EPPs and MEPPs recorded on NMJs were normalized to a membrane potential of –75 mV. EPP amplitudes were corrected for non-linear summation (50). The quantal content of full-sized EPPs was calculated at each NMJ by the direct method, by dividing the mean normalized and corrected EPP amplitude by the mean normalized MEPP amplitude. Giant MEPPs were excluded from the amplitude analysis as they did not contribute to the evoked EPPs. We generally used 20 consecutive EPPs and 50–80 MEPPs in the calculations. Preparations were equilibrated by incubation for 60 min with 3 µM neostigmine methylsulphate (France Biochem) for studies of AChE inhibition.

Statistical analyses
Data are expressed as the mean ± SEM. The Student's t-test (two-tailed), {chi}2 test, Mann–Whitney rank sum test, log rank test and Kolmogorov–Smirnov two-sample test were used to determine the statistical significance of differences between control and mutant values. Differences were considered significant when P < 0.05.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
FUNDING
 REFERENCES
 
Supplementary Material is available at HMG Online.


   FUNDING
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
REFERENCES
 
This work was funded by Institut National de la Santé et de la Recherche Médicale, Groupement d'Intêret scientifique-Institut des maladies rares, Association Française contre les Myopathies and Agence Nationale de la Recherche-programme Maladies Rares. S.N. holds a Contrat d'interface from Assistance Publique-Hôpitaux de Paris.


   ACKNOWLEDGEMENTS
 
We thank the Targeted Mutagenesis and Transgenesis Department of the Institut Clinique de la Souris (Strasbourg, France) for knock-in mouse generation, Eric Pellé for technical assistance with X rays (CNRS UMR 7179), Roy Combe for X-ray analysis, Dr N. Tabti (Inserm UMR S546) and Professor J-C. Willer (AP-HP) for preliminary EMG investigations, J-G. Cormier and H. Marques for immunohistochemistry, and Dr B. Nait-Oumesmar for critical reading of the manuscript. We also thank Drs P. Guicheney and A. Rouche (Inserm U582, Institut de Myologie, Paris, France) and the P3S facility for use of the electron microscope and automated sequencer. We also thank Drs J.H. Miner (Washington University School of Medicine, St Louis, USA) and M.A. Ruegg (University of Basel, Basel, Switzerland) for providing antibodies directed against laminin {alpha}5 and agrin, respectively. The SV2 and 2H3 antibodies developed by K.M. Buckley, and T.M Jessell and J. Dodd, respectively, were obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA 52242.

Conflict of Interest statement. None declared.


   FOOTNOTES
 
{dagger} The authors wish it to be known that, in their opinion, the second and third authors should be regarded as joint second authors. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 

  1. Nicole S., Stum M., Fontaine B. Perlecan: Schwartz–Jampel syndrome (SJS, MIM 255800) and dyssegmental dysplasia, Silverman-Handmaker type (DDSH, MIM 224410). In: Inborn Errors of Development. The Molecular Basis of Clinical Disorders of Morphogenesis—Epstein C., Erickson R., Wynshaw-Boris A., eds. (2008) 2nd edn. Oxford University Press. 1435–1442.

  2. Nicole S., Davoine C.-S., Topaloglu H., Cattolico L., Barral D., Beighton P., Ben Hamida C., Hammouda H., Cruaud C., White P., et al. Perlecan, the major proteoglycan of basement membranes, is altered in patients with Schwartz–Jampel syndrome (chondrodystrophic myotonia). Nature Genet. (2000) 26:480–483.[CrossRef][Web of Science][Medline]

  3. Iozzo R.V. Basement membrane proteoglycans: from cellar to ceiling. Nat. Rev. Mol. Cell. Biol. (2005) 6:646–656.[CrossRef][Web of Science][Medline]

  4. Stum M., Davoine C.S., Vicart S., Guillot-Noel L., Topaloglu H., Carod-Artal F.J., Kayserili H., Hentati F., Merlini L., Urtizberea J.A., et al. Spectrum of HSPG2 (Perlecan) mutations in patients with Schwartz–Jampel syndrome. Hum. Mutat. (2006) 27:1082–1091.[CrossRef][Web of Science][Medline]

  5. Arikawa-Hirasawa E., Le A.H., Nishino I., Nonaka I., Ho N.C., Francomano C.A., Govindraj P., Hassell J.R., Devaney J.M., Spranger J., et al. Structural and functional mutations of the perlecan gene cause Schwartz–Jampel syndrome, with myotonic myopathy and chondrodysplasia. Am. J. Hum. Genet. (2002) 70:1368–1375.[CrossRef][Web of Science][Medline]

  6. Arikawa-Hirasawa E., Wilcox W.R., Le A., Silverman N., Govindraj P., Hassell J., Yamada Y. Dyssegmental dysplasia, Silverman-Handmaker type, is caused by functional null mutations of the perlecan gene. Nature Genet. (2001) 27:431–434.[CrossRef][Web of Science][Medline]

  7. Cadilhac J., Baldet P., Greze J., Duday H. E.M.G. studies of two family cases of the Schwartz and Jampel syndrome (osteo-chondro-muscular dystrophy with myotonia). Electromyogr. Clin. Neurophysiol. (1975) 15:5–12.[Medline]

  8. Taylor R.G., Layzer R.B., Davis H.S., Fowler W.M. Jr. Continuous muscle fiber activity in the Schwartz–Jampel syndrome. Electroencephalogr. Clin. Neurophysiol. (1972) 33:497–509.[CrossRef][Web of Science][Medline]

  9. Arikawa-Hirasawa E., Rossi S.G., Rotundo R.L., Yamada Y. Absence of acetylcholinesterase at the neuromuscular junctions of perlecan-null mice. Nat. Neurosci. (2002) 5:119–123.[CrossRef][Web of Science][Medline]

  10. Peng H.B., Xie H., Rossi S.G., Rotundo R.L. Acetylcholinesterase clustering at the neuromuscular junction involves perlecan and dystroglycan. J. Cell. Biol. (1999) 145:911–921.[Abstract/Free Full Text]

  11. Donger C., Krejci E., Serradell A.P., Eymard B., Bon S., Nicole S., Chateau D., Gary F., Fardeau M., Massoulie J., et al. Mutation in the human acetylcholinesterase-associated collagen gene, COLQ, is responsible for congenital myasthenic syndrome with end-plate acetylcholinesterase deficiency (Type Ic). Am. J. Hum. Genet. (1998) 63:967–975.[CrossRef][Web of Science][Medline]

  12. Ohno K., Brengman J., Tsujino A., Engel A.G. Human endplate acetylcholinesterase deficiency caused by mutations in the collagen-like tail subunit (ColQ) of the asymmetric enzyme. Proc. Natl. Acad. Sci. USA (1998) 95:9654–9659.[Abstract/Free Full Text]

  13. Rodgers K.D., Sasaki T., Aszodi A., Jacenko O. Reduced perlecan in mice results in chondrodysplasia resembling Schwartz–Jampel syndrome. Hum. Mol. Genet. (2007) 16:515–528.[Abstract/Free Full Text]

  14. Nagy A., Moens C., Ivanyi E., Pawling J., Gertsenstein M., Hadjantonakis A.K., Pirity M., Rossant J. Dissecting the role of N-myc in development using a single targeting vector to generate a series of alleles. Curr. Biol. (1998) 8:661–664.[CrossRef][Web of Science][Medline]

  15. Rogalski T.M., Gilchrist E.J., Mullen G.P., Moerman D.G. Mutations in the unc-52 gene responsible for body wall muscle defects in adult Caenorhabditis elegans are located in alternatively spliced exons. Genetics (1995) 139:159–169.[Abstract]

  16. Zoeller J.J., McQuillan A., Whitelock J., Ho S.Y., Iozzo R.V. A central function for perlecan in skeletal muscle and cardiovascular development. J. Cell. Biol. (2008) 181:381–394.[Abstract/Free Full Text]

  17. Costell M., Gustafsson E., Aszodi A., Morgelin M., Bloch W., Hunziker E., Addicks K., Timpl R., Fassler R. Perlecan maintains the integrity of cartilage and some basement membranes. J. Cell. Biol. (1999) 147:1109–1122.[Abstract/Free Full Text]

  18. Lapidos K.A., Kakkar R., McNally E.M. The dystrophin glycoprotein complex: signaling strength and integrity for the sarcolemma. Circ. Res. (2004) 94:1023–1031.[Abstract/Free Full Text]

  19. Eusebio A., Oliveri F., Barzaghi P., Ruegg M.A. Expression of mouse agrin in normal, denervated and dystrophic muscle. Neuromuscul. Disord. (2003) 13:408–415.[CrossRef][Web of Science][Medline]

  20. Massoulie J. The origin of the molecular diversity and functional anchoring of cholinesterases. Neurosignals (2002) 11:130–143.[CrossRef][Web of Science][Medline]

  21. Deprez P., Inestrosa N.C., Krejci E. Two different heparin-binding domains in the triple-helical domain of ColQ, the collagen tail subunit of synaptic acetylcholinesterase. J. Biol. Chem. (2003) 278:23233–23242.[Abstract/Free Full Text]

  22. Sketelj J., Crne-Finderle N., Brzin M. Influence of denervation on the molecular forms of junctional and extrajunctional acetylcholinesterase in fast and slow muscles of the rat. Neurochem. Int. (1992) 21:415–421.[CrossRef][Web of Science][Medline]

  23. O'Malley J.P., Waran M.T., Balice-Gordon R.J. In vivo observations of terminal Schwann cells at normal, denervated, and reinnervated mouse neuromuscular junctions. J. Neurobiol. (1999) 38:270–286.[CrossRef][Web of Science][Medline]

  24. Wood S.J., Slater C.R. Safety factor at the neuromuscular junction. Prog. Neurobiol. (2001) 64:393–429.[CrossRef][Web of Science][Medline]

  25. Cartaud A., Strochlic L., Guerra M., Blanchard B., Lambergeon M., Krejci E., Cartaud J., Legay C. MuSK is required for anchoring acetylcholinesterase at the neuromuscular junction. J. Cell. Biol. (2004) 165:505–515.[Abstract/Free Full Text]

  26. Feng G., Krejci E., Molgo J., Cunningham J.M., Massoulie J., Sanes J.R. Genetic analysis of collagen Q: roles in acetylcholinesterase and butyrylcholinesterase assembly and in synaptic structure and function. J. Cell. Biol. (1999) 144:1349–1360.[Abstract/Free Full Text]

  27. Mouisel E., Blondet B., Escourrou P., Chatonnet A., Molgo J., Ferry A. Outcome of acetylcholinesterase deficiency for neuromuscular functioning. Neurosci. Res. (2006) 55:389–396.[CrossRef][Web of Science][Medline]

  28. Girard E., Barbier J., Chatonnet A., Krejci E., Molgo J. Synaptic remodeling at the skeletal neuromuscular junction of acetylcholinesterase knockout mice and its physiological relevance. Chem. Biol. Interact. (2005) 157–158:87–96.

  29. Katz B., Miledi R. The binding of acetylcholine and its removal from the synaptic cleft. J. Physiol. (1973) 231:549–574.[Abstract/Free Full Text]

  30. Sun C., Tranebjaerg L., Torbergsen T., Holmgren G., Van Ghelue M. Spectrum of CLCN1 mutations in patients with myotonia congenita in Northern Scandinavia. Eur. J. Hum. Genet. (2001) 9:903–909.[CrossRef][Web of Science][Medline]

  31. Schulze B., Mann K., Battistutta R., Wiedemann H., Timpl R. Structural properties of recombinant domain III-3 of perlecan containing a globular domain inserted into an epidermal-growth-factor- like motif. Eur. J. Biochem. (1995) 231:551–556.[Web of Science][Medline]

  32. Threadgill D.W., Dlugosz A.A., Hansen L.A., Tennenbaum T., Lichti U., Yee D., LaMantia C., Mourton T., Herrup K., Harris R.C., et al. Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype. Science (1995) 269:230–234.[Abstract/Free Full Text]

  33. Figuera L.E., Jimenez-Gil F.J., Garcia-Cruz M.O., Cantu J.M. Schwartz–Jampel syndrome: an atypical form? Am. J. Med. Genet. (1993) 47:526–528.[CrossRef][Web of Science][Medline]

  34. Spranger J., Hall B.D., Hane B., Srivastava A., Stevenson R.E. Spectrum of Schwartz–Jampel syndrome includes micromelic chondrodysplasia, kyphomelic dysplasia, and Burton disease. Am. J. Med. Genet. (2000) 94:287–295.[CrossRef][Web of Science][Medline]

  35. Moll J., Barzaghi P., Lin S., Bezakova G., Lochmuller H., Engvall E., Muller U., Ruegg M.A. An agrin minigene rescues dystrophic symptoms in a mouse model for congenital muscular dystrophy. Nature (2001) 413:302–307.[CrossRef][Web of Science][Medline]

  36. Karpati G., Engel W.K. ‘Type grouping’ in skeletal muscles after experimental reinnervation. Neurology (1968) 18:447–455.[Free Full Text]

  37. De Winter F., Vo T., Stam F.J., Wisman L.A., Bar P.R., Niclou S.P., van Muiswinkel F.L., Verhaagen J. The expression of the chemorepellent Semaphorin 3A is selectively induced in terminal Schwann cells of a subset of neuromuscular synapses that display limited anatomical plasticity and enhanced vulnerability in motor neuron disease. Mol. Cell. Neurosci. (2006) 32:102–117.[CrossRef][Web of Science][Medline]

  38. Rotundo R.L., Rossi S.G., Kimbell L.M., Ruiz C., Marrero E. Targeting acetylcholinesterase to the neuromuscular synapse. Chem. Biol. Interact. (2005) 157–158:15–21.

  39. Sanes J.R., Lichtman J.W. Development of the neuromuscular junction. Ann. Rev. Neurosci. (1999) 22:389–442.[CrossRef][Web of Science][Medline]

  40. Cote P.D., Moukhles H., Lindenbaum M., Carbonetto S. Chimaeric mice deficient in dystroglycans develop muscular dystrophy and have disrupted myoneural synapses. Nature Genet. (1999) 23:338–342.[CrossRef][Web of Science][Medline]

  41. Adams M.E., Kramarcy N., Krall S.P., Rossi S.G., Rotundo R.L., Sealock R., Froehner S.C. Absence of alpha-syntrophin leads to structurally aberrant neuromuscular synapses deficient in utrophin. J. Cell. Biol. (2000) 150:1385–1398.[Abstract/Free Full Text]

  42. Grady R.M., Zhou H., Cunningham J.M., Henry M.D., Campbell K.P., Sanes J.R. Maturation and maintenance of the neuromuscular synapse: genetic evidence for roles of the dystrophin-glycoprotein complex. Neuron (2000) 25:279–293.[CrossRef][Web of Science][Medline]

  43. Peng H.B., Ali A.A., Daggett D.F., Rauvala H., Hassell J.R., Smalheiser N.R. The relationship between perlecan and dystroglycan and its implication in the formation of the neuromuscular junction. Cell. Adhes. Commun. (1998) 5:475–489.[Web of Science][Medline]

  44. Maddison P. Neuromyotonia. Clin. Neurophysiol. (2006) 117:2118–2127.[CrossRef][Web of Science][Medline]

  45. Brown C.T., Nugent M.A., Lau F.W., Trinkaus-Randall V. Characterization of proteoglycans synthesized by cultured corneal fibroblasts in response to transforming growth factor beta and fetal calf serum. J. Biol. Chem. (1999) 274:7111–7119.[Abstract/Free Full Text]

  46. Tomas i Ferre J., Mayayo E., Fenoll i Brunet R. Morphometric study of the neuromuscular synapses in the adult rat with special reference to the remodelling concept. Biol. Cell. (1987) 60:133–144.[Web of Science][Medline]

  47. Vignaud A., Cebrian J., Martelly I., Caruelle J.P., Ferry A. Effect of anti-inflammatory and antioxidant drugs on the long-term repair of severely injured mouse skeletal muscle. Exp. Physiol. (2005) 90:487–495.[Abstract/Free Full Text]

  48. Minic J., Chatonnet A., Krejci E., Molgo J. Butyrylcholinesterase and acetylcholinesterase activity and quantal transmitter release at normal and acetylcholinesterase knockout mouse neuromuscular junctions. Br. J. Pharmacol. (2003) 138:177–187.[CrossRef][Web of Science][Medline]

  49. Cruz L., Gray W., Olivera B., Zeikus R., Kerr L., Yoshikami D., Moczydlowski E. Conus geographus toxins that discriminate between neuronal and muscle sodium channels. J. Biol. Chem. (1985) 260:9280–9288.[Abstract/Free Full Text]

  50. McLachlan E.M., Martin A.R. Non-linear summation of end-plate potentials in the frog and mouse. J. Physiol. (1981) 311:307–324.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/20/3166    most recent
ddn213v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Stum, M.
Right arrow Articles by Nicole, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stum, M.
Right arrow Articles by Nicole, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?