doi:10.1093/brain/awu272
Brain 2014: 137; 3160–3170
| 3160
BRAIN A JOURNAL OF NEUROLOGY
Adult-onset autosomal dominant centronuclear myopathy due to BIN1 mutations
1 IGBMC (Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire), 67404 Illkirch, France 2 Inserm, U964, 67404 Illkirch, France 3 CNRS, UMR7104, 67404 Illkirch, France 4 Universite´ de Strasbourg, 67404 Illkirch, France 5 Colle`ge de France, Chaire de Ge´ne´tique Humaine, 67404 Illkirch, France 6 Faculte´ de Me´decine, Laboratoire de Diagnostic Ge´ne´tique, Nouvel Hoˆpital Civil, 67000 Strasbourg, France 7 Universite´ Paris 6 UM76, Inserm UMR 974, CNRS UMR 7215, Institut de Myologie, Groupe Hospitalier La Pitie´-Salpeˆtrie`re, 75013 Paris, France 8 Centre de re´fe´rence de pathologie neuromusculaire Paris-Est, Groupe Hospitalier La Pitie´-Salpeˆtrie`re, 75013 Paris, France 9 Department of Neurological, Neurosurgical, and Behavioural Sciences, University of Siena, 53100 Siena, Italy 10 Department of Human Genetics, Julius-Maximilian University, 97074 Wu¨rzburg, Germany 11 Neurology, SHG Klinikum, 66663 Merzig, Germany 12 Institute for Neurological Research, FLENI, C1428AQK Buenos Aires, Argentina 13 Hospital Italiano de Buenos Aires, C1181ACH Buenos Aires, Argentina 14 Laboratoire de Neuropathologie, Groupe Hospitalier La Pitie´-Salpeˆtrie`re, 75013 Paris, France 15 Institut de Neuropatologia, IDIBELL-Hospital Universitari de Bellvitge, 08901 Hospitalet de Llobregat, Barcelona, Spain 16 Inserm, U929, 63000 Clermont-Ferrand, France 17 Universite´ Clermont 1, 63000 Clermont-Ferrand, France 18 CHU Clermont-Ferrand, 63000 Clermont-Ferrand, France 19 De´partement de Biochimie, Biochimie et Ge´ne´tique Mole´culaire, Toxicologie et Pharmacologie, CHU Grenoble, 38700 La Tronche, France 20 Institute of Neuropathology and JARA Brain Translational Medicine, RWTH Aachen University, 52062 Aachen, Germany *These authors contributed equally to this work. Correspondence to: Jocelyn Laporte, IGBMC, 1 Rue Laurent Fries, 67404 Illkirch, France E-mail: [email protected]
Centronuclear myopathies are congenital muscle disorders characterized by type I myofibre predominance and an increased number of muscle fibres with nuclear centralization. The severe neonatal X-linked form is due to mutations in MTM1, autosomal recessive centronuclear myopathy with neonatal or childhood onset results from mutations in BIN1 (amphiphysin 2), and dominant cases were previously associated to mutations in DNM2 (dynamin 2). Our aim was to determine the genetic basis and physiopathology of patients with mild dominant centronuclear myopathy without mutations in DNM2. We hence established and characterized a homogeneous cohort of nine patients from five families with a progressive adult-onset centronuclear myopathy without facial weakness, including three sporadic cases and two families with dominant disease inheritance. All patients had similar histological and ultrastructural features involving type I fibre predominance and hypotrophy, as well as
Received March 20, 2014. Revised July 27, 2014. Accepted July 29, 2014. Advance Access publication September 26, 2014 ß The Author (2014). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Johann Bo¨hm,1,2,3,4,5,* Vale´rie Biancalana,1,2,3,4,5,6,* Edoardo Malfatti,7,8,9 Nicolas Dondaine,6 Catherine Koch,1,2,3,4,5 Nasim Vasli,1,2,3,4,5 Wolfram Kress,10 Matthias Strittmatter,11 Ana Lia Taratuto,12 Hernan Gonorazky,13 Pascal Laforeˆt,8 Thierry Maisonobe,14 Montse Olive´,15 Laura Gonzalez-Mera,15 Michel Fardeau,7,8 Nathalie Carrie`re,16,17,18 Pierre Clavelou,16,17,18 Bruno Eymard,8 Marc Bitoun,7 John Rendu,19 Julien Faure´,19 Joachim Weis,20 Jean-Louis Mandel,1,2,3,4,5,6 Norma B. Romero7,8 and Jocelyn Laporte1,2,3,4,5
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3161
prominent nuclear centralization and clustering. We identified heterozygous BIN1 mutations in all patients and the molecular diagnosis was complemented by functional analyses. Two mutations in the N-terminal amphipathic helix strongly decreased the membrane-deforming properties of amphiphysin 2 and three stop-loss mutations resulted in a stable protein containing 52 supernumerary amino acids. Immunolabelling experiments revealed abnormal central accumulation of dynamin 2, caveolin-3, and the autophagic marker p62, and general membrane alterations of the triad, the sarcolemma, and the basal lamina as potential pathological mechanisms. In conclusion, we identified BIN1 as the second gene for dominant centronuclear myopathy. Our data provide the evidence that specific BIN1 mutations can cause either recessive or dominant centronuclear myopathy and that both disorders involve different pathomechanisms.
Keywords: centronuclear myopathy; BIN1; amphiphysin 2; DNM2; T-tubule Abbreviations: CNM = centronuclear myopathy
Introduction
Materials and methods Patients Patients originated from Germany (Patient AIZ35), Argentina (Patient AIZ39), Spain (Patients ATE18, ALX95 and API51) and France (Patients IL25, AIZ64, ATI45 and AMX48). Sample collection was performed with written informed consent from the patients according to the Declaration of Helsinki.
Histology and electron microscopy Transverse muscle cryostat sections (8 mm) were stained with haematoxylin-eosin and NADH-TR and then assessed for centralized nuclei, fibre morphology, fibre type distribution, cores, protein accumulations and cellular infiltrations. For electron microscopy, muscle sections were prepared by routine methods.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Centronuclear myopathies (CNM) are a group of congenital myopathies characterized by the presence of an abnormally high number of muscle fibres with central nuclei (Romero, 2010). Facial weakness is a clinical hallmark of CNM and often accompanied by bilateral ptosis and ophthalmoparesis. Centronuclear myopathies have been described as X-linked, autosomal dominant and autosomal recessive forms. However, the genetic basis of the disease is not known for all cases. Severe neonatal X-linked CNM (XLCNM, MIM# 310400) arises from mutations in MTM1, involving a strong decrease of the myotubularin protein level (Laporte et al., 1996, 2001). Morphological analysis of patients with X-linked CNM revealed centralized nuclei, fibre atrophy, and predominance of type 1 fibres (Romero, 2010), and in some cases fibres with basophilic rings (‘necklace fibres’) are seen (Bevilacqua et al., 2009). Autosomal dominant CNM (ADCNM, MIM# 160150) is frequently caused by mutations in DNM2 and was reported with severe neonatal and clinically milder adult-onset (Bitoun et al., 2005, 2007; Bohm et al., 2012). Histological findings on biopsies from patients with autosomal dominant CNM include fibre size heterogeneity, type I fibre hypotrophy, centralized nuclei and in particular radial arrangement of sarcoplasmic strands (Jeannet et al., 2004; Romero, 2010; Toussaint et al., 2011). CT and MRI assessment demonstrated a predominant involvement of distal muscles (Fischer et al., 2006; Schessl et al., 2007). Autosomal recessive CNM (ARCNM, MIM#255200) results from homozygous BIN1 mutations and occurs as a severe neonatal or childhood onset disorder predominantly involving the proximal musculature (Nicot et al., 2007; Bohm et al., 2010; Claeys et al., 2010). The muscle biopsies display atrophy, type I fibre predominance, prominent nuclear centralization, and often nuclear clustering (Romero, 2010). Amphiphysin 2, encoded by BIN1, is a key regulator of membrane curvature and trafficking. The muscle specific isoform contains an N-terminal amphipathic helix-inducing membrane curvature, a BAR (Bin/Amphiphysin/ Rvs) domain able to sense and maintain membrane curvature, a polybasic phosphoinositide (PI) binding motif, and a SH3 domain interacting with dynamin 2 (Sakamuro et al., 1996; Grabs et al., 1997; Lee et al., 2002; Itoh and De Camilli, 2006). In skeletal
muscle, amphiphysin 2 is localized at deep sarcolemmal invaginations, the T-tubules, implicated in excitation-contraction coupling (Butler et al., 1997). When exogenously expressed in cultured cells, the muscle-specific isoform induces membrane tubulation (Lee et al., 2002; Nicot et al., 2007), suggesting that amphiphysin 2 is implicated in T-tubule biogenesis. Accordingly, drosophila amph null mutants were shown to display an abnormal T-tubule structure (Razzaq et al., 2001). To date, only five autosomal recessive CNM families with homozygous BIN1 mutations have been described. In cultured cells, the BAR domain mutations were shown to repress the amphiphysin 2 membrane tubulating properties, and the SH3 truncating mutations impaired binding and recruitment of dynamin 2 (Nicot et al., 2007). In this study, we characterize the mild end of the centronuclear myopathy spectrum. Our homogeneous cohort includes patients with progressive adult-onset myopathy without facial weakness. Histological analyses revealed major nuclear centralization and clustering in muscle fibres. Although inherited as a dominant disease, DNM2 mutations were not found. Instead, we identified heterozygous BIN1 mutations in all patients. Functional analysis demonstrated a strong impact of the mutations on the biochemical properties of amphiphysin 2. We conclude that specific BIN1 mutations cause either dominant or recessive CNM and that both disorders involve different pathomechanisms.
3162
| Brain 2014: 137; 3160–3170
J. Bo¨hm et al.
CT/MRI
stained with Hoechst/DAPI (Sigma-Aldrich) and cells were mounted and viewed by epifluorescence.
CT studies included standard scans at hip, thigh and lower leg level. The scans were examined for muscle atrophy and abnormal signal densities. MRI of the lower extremities was performed using a 1.5 Tesla magnet, and muscle atrophy was assessed on T1-weighted images.
Results
Molecular genetics
RNA studies RNA was extracted from muscle biopsies from Patients AIZ35 and AIZ39 using Tri reagent (Molecular Research Center Inc), reverse transcribed using the SuperScriptÕ III kit (Invitrogen), amplified using BIN1-specific primers and Sanger-sequenced.
Protein studies Western blot and immunofluorescence were performed using routine protocols. The following antibodies were used for the study: R2444 (homemade rabbit anti-BIN1 SH3 domain) (Nicot et al., 2007), R2405 (homemade rabbit anti-BIN1 PI domain) (Nicot et al., 2007), R3062 (homemade rabbit anti-pan BIN1) (Bohm et al., 2013), mouse anti-B10 tag (homemade) (Royer et al., 2013), mouse anti-GAPDH (Merck Millipore), mouse anti b-tubulin (BAbCO), R2680 (homemade rabbit anti dynamin 2) (Koutsopoulos et al., 2013), mouse anti-DHPR (Abcam), mouse anti-ryanodine receptor 1 (Affinity BioReagents), rabbit anti-caveolin-3 (Gene Tex), mouse Anti-p62 Lck ligand (BD Biosciences). Nuclei were stained with Hoechst (Sigma-Aldrich). For immunohistofluorescence, sections were mounted with slowfade antifade reagent (Invitrogen) and viewed using a laser scanning confocal microscope (TCS SP2; Leica Microsystems).
Constructs The BIN1-GFP construct (muscle-specific isoform 8) was a kind gift from Pietro de Camilli (Howard Hughes Medical Institute) and the BIN1 coding sequence was cloned into the pSG5-B10 vector using the Gateway system (Invitrogen). The mutations (c.61_63delAAG and c.70C 4 T) were introduced by site-directed mutagenesis using the Pfu DNA polymerase (Stratagene).
Membrane tubulation assay COS-1 cells were seeded on coverslips and transfected at 50–60% confluency with the BIN1 wild-type or mutant constructs. The next day, cells were fixed in 4% paraformaldehyde in PBS and stained using a home made anti-B10 antibody by routine methods. Nuclei were
The patients characterized in this study belong to five unrelated families and constitute a homogeneous cohort of mild and adultonset muscle disorders diagnosed as CNM. Patients AIZ35 (Family 1), AIZ39 (Family 2) and IL25 (Family 4) were sporadic cases, while Families 3 and 5 had an ancestral history of a mild muscle phenotype segregating as a dominant disease over at least three generations. All patients presented with a mildly progressive muscle weakness predominantly affecting the proximal muscles of the lower limbs (Table 1). Age of onset was between 22 (Patient ATI45, Family 5) and 50 years (Patient API51, Family 3) and at the last medical examination, most patients had walking difficulties and two patients were wheelchair-bound (Patient ATE18 from Family 3 and Patient ATI45 from Family 5). Centronuclear myopathy generally involves respiratory insufficiency, opththalmoplegia/paresis, ptosis and facial weakness. However, breathing was substantially normal in our patients, mild ptosis was only observed in Patients AIZ39 (Family 2) and API51 (Family 3), vertical gaze palsy in Patients AIZ64 and ATI45 from Family 5, and facial weakness was absent in all patients. Cognitive impairment or cardiac involvement was not noted for any of the patients. EMG revealed a clear myopathic pattern for all analysed patients and CPK levels were elevated in two patients (Patient AIZ39 from Family 2 and Patients AMX48 from Family 5). MRI/CT scans were performed for all families except Family 1 (Patient AIZ35) and essentially revealed a predominant involvement of the distal muscles and a selective involvement of the proximal muscles of the lower limbs in most patients (Fig. 1 and Supplementary material). In summary, the nine patients had comparable clinical findings of a mild adultonset muscle disorder predominantly affecting the lower limbs.
Muscle histology and ultrastructural findings For all patients, haematoxylin-eosin staining revealed prominent nuclear centralization and clusters of myonuclei, as well as significant fibre size variability (Fig. 1A). Oxidative staining revealed dense accumulations in the centre of the fibres, type I fibre predominance (65–92%) and hypotrophy and moderate disorganization of the myofibrillar network. We also infrequently found fibres with radial arrangements of sarcoplasmic strands (‘spoke of wheels’), moderate endomysial fibrosis, vacuoles and deep invaginations of the sarcolemma pointing to the centre of the fibre. Electron microscopy was performed on muscle biopsies from all patients. We observed clustered central nuclei surrounded by sarcoplasm containing accumulated glycogen granules, mitochondria and other organelles (Fig. 1B). We furthermore noted enlarged vacuoles adjacent to centralized nuclei, moderate Z-
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Genomic DNA was prepared from peripheral blood by routine procedures and sequenced for all coding exons and intron/exon boundaries of MTM1, DNM2 and BIN1 as described elsewhere (Laporte et al., 1996; Bitoun et al., 2005; Nicot et al., 2007). RYR1 was excluded by targeted deep sequencing for Families 1 and 4, and by cDNA sequencing for Families 3 and 5. Samples from Family 2 were not available for further investigations. Segregation analysis was performed by PCR and Sanger sequencing of the BIN1 exons harbouring the mutations. The mutations were numbered according to GenBank NM_139343.2 and NP_647593.1. Nucleotide position reflects cDNA numbering with + 1 corresponding to the A of the ATG translation initiation codon.
Clinical reports
Proximo-distal Axial, proximo-distal
Axial, proximo-distal
n.d.
n.d.
AP151 IL25
AIZ64
AMX48
AT145
Family 4
Family 5
n.d. = not determined.
Proximal Proximal Axial, proximo-distal
AIZ39 ATE18 ALX95
Family 2 Family 3
Muscle involvement
Patient
Proximal
M
ATI45
AIZ35
M
AMX48
France
France
Argentina Spain
Germany
Origin
Family 1
F
AIZ64
Family 5
F
F
API51
IL25
F
ALX95
Family 4
M F
AIZ39 ATE18
Family 2 Family 3
F
AIZ35
Family 1
Gender
Normal
Normal
Slightly abnormal
Normal Normal
Normal Dyspnea Normal
Normal
Breathing
c.1780delT
c.1779delA
c.70C 4 T c.1778delC
No
No
No
No No
No No No
No
Vertical gaze palsy
n.d.
Vertical gaze palsy
Mild ptosis No
Mild ptosis No No
No
Ocular involvement
p.X594AspfsX53
p.X594AspfsX53
p.Arg24Cys p.Pro593HisfsX54
p.Lys21del
Predicted protein impact
Facial weakness
c.61_63delAAG
Mutation
n.d.
No
No
No No
No No No
No
Cardiac involvement
Dominant
Sporadic
Sporadic Dominant
Sporadic
Disease segregation
n.d.
Myopathic
Myopathic
Myopathic Myopathic
Myopathic n.d. Myopathic
Myopathic
EMG
22
32
43
30
50
31
35 ?
48
Onset (age)
n.d.
Elevated (4)
Normal
Normal Normal
Central nuclei, nuclear clustering, fibre size variability, enlarged endomysium, radial sarcoplasmic strands Central nuclei, nuclear clustering n.d. Central nuclei, nuclear clustering, fibre size variability n.d. Central nuclei, nuclear clustering, fibre size variability, enlarged endomysium, radial sarcoplasmic strands Central nuclei, nuclear clustering, fibre size variability, type I fibre atrophy, radial sarcoplasmic strands, sarcolemmal invaginations Central nuclei, nuclear clustering, fibre size variability, type I fibre atrophy, enlarged vacuoles, sarcolemmal invaginations n.d.
Calf hypertrophy
CPK elevated, myalgia, photophobia, weakness of the hands, deglutition problems Contractures
Bilateral pes cavus
Bilateral scapula alata Chronic lymphocytic thyroiditis
Contractures
Histopathology hallmarks
Associated findings
Walking difficulties, requires cane or wheelchair Wheelchair-bound
Cane from age 54, wheelchair for long distances Requires wheelchair
Waddling gate
Waddling gate
Progressive, lower limbs
Slightly waddling gate, slight paresis of the left leg Cane from age 45 Wheelchair-bound
Progressive, lower limbs Progressive, predominantly lower limbs Progressive, first lower limbs, then upper limbs Progressive, predominantly lower limbs Progressive, predominantly lower limbs Progressive, predominantly lower limbs Progressive, predominantly lower limbs Progressive, lower limbs
Muscle weakness
Motor function
Elevated (10 ) n.d. Normal
Normal
CPK
22
41
Deceased age 65 61
65
46
56 88
53
Age at last medical examination
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Patient
Table 1 Main clinical, histological and molecular findings in nine patients with adult-onset dominant centronuclear myopathy
BIN1 mutations in dominant CNM Brain 2014: 137; 3160–3170
| 3163
3164
| Brain 2014: 137; 3160–3170
J. Bo¨hm et al.
Haematoxylin-eosin (HE) staining of muscle sections demonstrated prominent nuclear centralization and clustering, as well as fibre size variability and endomysial fibrosis. Oxidative NADH-TR staining revealed central accumulations in type I fibres and some radial arrangements of sarcoplasmic strands. (B) Ultrastructural analysis confirmed nuclear centralization and clustering, and revealed structural triad abnormalities (arrow), mild Z-band streaming, accumulations of glycogen granules and mitochondria, and enlarged vacuoles adjacent to centralized nuclei. (C) MRI/CT scan of Patients AIZ39 and IL25 demonstrated involvement of all distal muscles and selective proximal muscles, particularly of adductor longus (AL) and sartorius (S).
band streaming, and enlargement of membrane structures. Additional unspecific findings included autophagosomes and lipofuscin granules. Cores were not noted. In conclusion, histology and electron microscopy were suggestive of centronuclear myopathy.
Molecular genetics and functional impact of the BIN1 mutations Mutations in the CNM genes DNM2 and MTM1 were excluded in all patients of our cohort of mild and late-onset centronuclear myopathy, and RYR1 was excluded for Families 1, 3, 4 and 5.
We sequenced all coding exons and adjacent splice-relevant intronic regions of BIN1 and identified heterozygous mutations in all five families (Fig. 2A). RNA analysis was performed for two sporadic families and did not identify a second BIN1 mutation, indicating a dominant disorder. Both alleles were equably amplified and sequenced, suggesting that the heterozygous mutations did not affect mRNA stability. A muscle biopsy for RNA analysis was not available for the third sporadic case (Family 4). For the dominant Families 3 and 5, sequencing of the affected family members Patients ATE18, ALX95, API51 (Family 3), and Patients AIZ64, ATI45, AMX48 (Family 5) demonstrated disease segregation with the heterozygous BIN1 mutations. None of the identified
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 1 Histopathology, ultrastructural analysis and MRI/CT scan of patients with mild adult-onset centronuclear myopathy. (A)
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3165
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 2 Heterozygous BIN1 mutations in five families with adult-onset dominant centronuclear myopathy. (A) Patients AIZ35 and AIZ39 are sporadic cases with heterozygous missense mutations in the amphiphysin 2 amphipathic helix. Patients from Families 3, 4 and 5 harbour heterozygous single nucleotide deletions at the 3’ end of the coding sequence, affecting the regular stop codon. Healthy members with available DNA samples and without BIN1 mutation are marked as wild-type (WT). (B) The affected residues K21 and R24 (dominant CNM) and K35 (recessive CNM) are highly conserved. (C) The single nucleotide deletions at the 3’ end are predicted to induce a shift of the stop codon, resulting in a protein with 52 supernumerary amino acids. (D) Western blot analysis of muscle protein extracts from Patients IL25 (Family 4) and AIZ64 (Family 5) revealed a normal and a higher molecular weight protein compared to the healthy control. (E) Schematic representation of the amphiphysin 2 protein domains and position of the autosomal recessive CNM (black) and autosomal dominant CNM (red) mutations.
BIN1 mutations are listed in databases such as 1000 Genomes, Exome Variant Server or dbSNP, nor in our in-house database containing exome data from 276 individuals and Sanger data from 100 individuals with matched ethnicity. Taken together, these findings are clearly in favour of a dominant disease inheritance and genetically evidence the pathogenicity of the identified heterozygous BIN1 mutations.
For Families 3, 4 and 5 we identified heterozygous single nucleotide deletions in the last codon or in the regular stop codon of BIN1. The mutations c.1778delC (p.Pro593HisfsX54; Family 3) and c.1780delT (p.X594AspfsX53; Family 5) segregated as an autosomal dominant disease and the c.1779delA (p.X594AspfsX53; Family 4) deletion was found in a singleton case. All three single nucleotide deletions are predicted to shift the reading frame of the regular stop
3166
| Brain 2014: 137; 3160–3170
As amphiphysin 2 was proposed to be implicated in T-tubule biogenesis (Razzaq et al., 2001; Lee et al., 2002), we examined the localization of the T-tubule marker DHPR and the junctional sarcoplasmic calcium channel RYR1. The localization of both proteins was strongly altered, showing focal accumulations in the muscle fibres. Accordingly, we observed structural triad abnormalities on the patient biopsies (Fig. 1). To explore if the abnormal triad structure reflects general membrane defects, we analysed caveolin-3, a key regulator of membrane trafficking in muscle (Parton and del Pozo, 2013) and noted a striking mislocalization within the cytoplasm as well as prominent perinuclear accumulations, potentially corresponding to the enlarged perinuclear vacuoles observed by electron microscopy. By labelling of laminin, we furthermore confirmed the presence of characteristic deep invaginations of the sarcolemma and the basal lamina often pointing towards central nuclei (Fig. 4B). These defects were not seen on muscle sections from CNM patients with MTM1 or DNM2 mutations (Supplementary Fig. 2), and might serve as distinctive markers for BIN1-related CNM. In order to assess whether the overall structural aberrations within the muscle fibres involve autophagy, we monitored the autophagic marker p62 on muscle sections from Patient IL25 (Family 4) and we found significant positive signals around central nuclei (Fig. 5). Taken together, these data demonstrate that BIN1-dominant CNM is associated with general membrane alterations of the triad, the sarcolemma and the basal lamina, and potentially the autophagic pathway.
Discussion In this study we identified heterozygous BIN1 mutations as a genetic cause of adult-onset autosomal dominant CNM. We validated our data at the clinical, histological, genetic and functional levels. BIN1 is the second gene implicated in autosomal dominant CNM after DNM2. The consistent clinical, histological and ultrastructural features of the nine patients with autosomal dominant CNM with dominant BIN1 mutations described in this study clearly differ from BIN1related autosomal recessive CNM or DNM2-related autosomal dominant CNM. Our patients presented with an exclusively adult-onset muscle weakness affecting selected proximal muscles and all distal muscles and of the lower limbs. Predominant proximal muscle weakness was reported for BIN1-related autosomal recessive CNM, whereas DNM2-autosomal dominant CNM patients displayed involvement of the distal muscles (Fischer et al., 2006; Bohm et al., 2010). All reported BIN1-related autosomal recessive CNM cases had neonatal/childhood disease onset and often marked respiratory insufficiency and facial weakness, which are not seen in our clinically milder patients. Similarly, ocular involvement is classically seen in the other CNM forms, and was not or only barely present in our cohort. Histological and ultrastructural analyses of the BIN1-related autosomal dominant CNM patients revealed nuclear centralization and clustering, consistent with BIN1-related autosomal recessive CNM and contrasting the muscle morphology described for dominant CNM cases with DNM2 mutations. The ratio of fibres with centralized
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
codon and to generate an extended amphiphysin 2 protein containing 52 supernumerary C-terminal amino acids (Fig. 2C). The presence of a protein of higher molecular weight was confirmed by western blot of protein extracts from muscle biopsies (Families 4 and 5; Fig. 2D) or lymphoblasts (Family 3; Supplementary Fig. 1). We conclude that the stop-loss mutations do not fundamentally affect mRNA or protein stability. For Family 1 we identified a heterozygous c.61_63delAAG (p.Lys21del) in-frame deletion of three nucleotides, and for Family 2 we found a heterozygous c.70C 4 T (p.Arg24Cys) missense mutation. In both cases, the mutation was not found in the healthy mothers, and the fathers’ DNAs were not available. Both mutations affect essential and conserved residues in the N-terminal amphipathic helix (Fig. 2B), which inserts into the membrane bilayer and plays a crucial role in curvature induction (Peter et al., 2004; Gallop et al., 2006; Low et al., 2008; Jao et al., 2010). The mutations are predicted to impact on the composition and the orientation of the hydrophobic face of the amphipathic helix (Fig. 3A). The mutations do not seem to influence protein stability, as western blot revealed a comparable band size and intensity for cells expressing GFP-BIN1 wild-type, GFP-BIN1 K21del, or GFPBIN1 R24C (Supplementary Fig. 1). In cultured cells, exogenously expressed wild-type GFP-BIN1 promotes membrane tubulation (Fig. 3B) (Lee et al., 2002; Nicot et al., 2007). In contrast, GFP-BIN1 constructs harbouring the p.Lys21del (K21del) or the p.Arg24Cys (R24C) mutations did not induce tubulation. The p.Lys35Asn (K35N) and the p.Asp151Asn (D151N) mutations, both found in patients with severe autosomal recessive CNM, showed no or only partial membrane tubulation (Nicot et al., 2007). Membrane tubulation is mediated through dimerization of the BAR domains of amphiphysin 2 (Peter et al., 2004; Gallop et al., 2006). To assess a potential different effect of the dominant compared to the recessive BIN1 mutations, we co-transfected wild-type GFP-BIN1 and mutant B10-BIN1 constructs (Fig. 3C). The recessive K35N and D151N mutants were efficiently recruited to the membrane tubules induced by the wild-type construct, demonstrating that these loss-of-function mutations do not interfere with dimerization. In contrast, the K21del and R24C constructs barely localized at the tubules induced by the wild-type protein, suggesting that they affect recruitment of mutated/wild-type heterodimers to the membranes and/or amphiphysin 2 dimerization. Importantly, and in contrast to the recessive BIN1 K35N and D151N mutations, the impact of the dominant K21del and R24C mutations cannot be rescued by the co-expressed wild-type protein. These results clearly indicate a different pathomechanism in BIN1-related autosomal dominant CNM versus autosomal recessive CNM. To further characterize the physiopathology of BIN1-related autosomal dominant CNM, we performed immunolabelling on muscle biopsies from our patients with dominant BIN1 mutations (Fig. 4). We noted scattered amphiphysin 2 accumulations, most often surrounding centralized nuclei. Similarly, dynamin 2 accumulated around centralized nuclei. Dynamins are known to interact with amphiphysins (David et al., 1996; Kojima et al., 2004). Our data therefore suggest that the dynamin 2 accumulations in our patients are a direct consequence of the BIN1 mutations, and provide a molecular link between both dominant CNM forms.
J. Bo¨hm et al.
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3167
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 3 Functional impact of the amphipathic helix mutations. (A) The identified BIN1 mutations are predicted to impact on the amphipathic helix properties. (B) Overexpression of GFP-BIN1-WT in COS-1 cells induces membrane tubulation. The amphipathic helix mutations strongly impair membrane tubulation. (C) Co-expression of wild-type GFP-BIN1 and mutant B10-BIN1 constructs show that the dominant K21del and R24C mutations, but not the recessive K35N and D151N mutations, impair recruitment to the membrane tubules induced by the wild-type protein. Nuclei are stained with Hoechst and are shown in blue.
nuclei does not seem to correlate with disease severity, as our clinically mild patients presented central nuclei in the vast majority of the fibres. A similar observation was made for X-linked CNM (Pierson et al., 2007). Comparison of our data with the literature suggests that this cohort represents the mild end of the centronuclear myopathy spectrum.
The dominant BIN1 mutations reported in this study differ from the homozygous BIN1 mutations causing recessive CNM (Nicot et al., 2007; Bohm et al., 2010; Claeys et al., 2010). In seven of our patients we found single nucleotide deletions inducing the expression of a stable amphiphysin 2 protein with 52 supernumerary amino acids. This extension follows the SH3 domain and
3168
| Brain 2014: 137; 3160–3170
J. Bo¨hm et al.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 4 Abnormal localization of proteins and structural defects of the muscle fibre. (A) Immunolocalization on transversal sections from patients revealed an abnormal pattern of the CNM proteins amphiphysin 2 and dynamin 2, the T-tubule marker DHPR, the junctional sarcoplasmic calcium channel RYR1, and the sarcolemmal marker caveolin-3 (all indicated by arrows). (B) Immunolabelling of laminin unravelled deep invaginations of the sarcolemma and basal lamina pointing towards the centre of fibres (arrows). The electron microscopy picture shows a close-up of an invagination.
potentially impacts on the domain folding or the interaction with other proteins. Amphiphysin recruits dynamin to the plasma membrane (Takei et al., 1999; Lee et al., 2002; Praefcke and McMahon 2004) and accordingly, we observed a striking mislocalization of both proteins around central nuclei on muscle sections from our patients. We also found two missense mutations
in the N-BAR domain impairing the membrane tubulation properties of amphiphysin 2. In contrast to the recessive mutations in the N-BAR domain, the dominant mutations additionally affect the recruitment to membrane tubules induced by wild-type amphiphysin 2. These observations suggest that the BIN1 autosomal dominant CNM mutations interfere with membrane binding and/or
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3169
Figure 5 p62 accumulations around centralized nuclei. p62 accumulations were found around centralized nuclei in muscle sections from Patient IL25 (Family 3, arrow). Muscle sections from a patient with X-linked myopathy with excessive autophagy (MEAX) and an unaffected individual were used as controls.
X-linked CNM (Al-Qusairi et al., 2013) and might contribute to the muscle pathology in BIN1-related CNM. In conclusion, BIN1 is the second gene implicated in dominant centronuclear myopathy. This is of high importance for genetic diagnosis and counselling as BIN1 mutations should not only be considered for recessive families, but also for dominant families and sporadic adult cases, especially if the biopsy displays prominent nuclear clustering in patients with mild adult-onset muscle disorders. BIN1 mutations can cause autosomal dominant or autosomal recessive CNM and both conditions occur with similar frequency. The late onset, the mild clinical course, the histopathologial features and the dominant disease inheritance of BIN1-related autosomal dominant CNM clearly differ from the reported CNM forms. Our study broadens the phenotypic spectrum of muscle disorders caused by BIN1 mutations and provides novel insights into the pathogenesis of centronuclear myopathies. The disorganization of the skeletal muscle triad and the basal lamina, and the cytoplasmic accumulation of caveolin-3 and p62 suggest that a general disorganization of membrane trafficking is the main pathological cause of this disorder.
Acknowledgements We thank the patients for their cooperation and interest in this study, Pietro de Camilli for the GFP-BIN1 construct, and Linda Mane`re, Osorio Abath Neto, Adeline Normand, Peggy Therier and Lena Kristina Beilschmidt for technical assistance.
Funding This work was supported by the INSERM, CNRS, University of Strasbourg, Colle`ge de France, Agence Nationale de la Recherche (ANR-11-BSV1-026), Association Franc¸aise contre les Myopathies (AFM 15352), Muscular Dystrophy Association (MDA186985) and Myotubular Trust.
Supplementary material Supplementary material is available at Brain online.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
dimerization, whereas the membrane tubulation defects caused by the BIN1 autosomal recessive CNM mutations can be rescued by co-transfection of wild-type BIN1. Indeed, heterozygous carriers of BIN1 autosomal recessive CNM mutations are healthy, suggesting different pathomechanisms for BIN1-related autosomal recessive CNM and autosomal dominant CNM. We hypothesize that both disorders arise from loss-of-function with a partial loss-of-function for the recessive BIN1 mutations, and a full lossof-function for the dominant BIN1 mutations. The pathological mechanisms can be inferred from our functional in cellulo data and from the histological and immunolabelling experiments on patient muscles. First, we noted concurrent mislocalizations of amphiphysin 2 and dynamin 2, both implicated in dominant CNM. Second, the disorganization of triad markers and the abnormal triad structure seen by electron microscopy is in accordance with a role of amphiphysin 2 at the T-tubules (Razzaq et al., 2001; Kojima et al., 2004), and suggests that impaired excitation-contraction coupling and subsequent calcium homeostasis defects are the most likely cause of the muscle weakness. Indeed, triad defects were seen on biopsies from patients with other CNM forms (Dowling et al., 2009; Romero, 2010; Toussaint et al., 2011) and abnormal intracellular calcium release was detected in isolated murine muscle fibres following BIN1 shRNA-mediated knock-down (Tjondrokoesoemo et al., 2011). Third, we noted a general disorganization of membrane structures and proteins regulating membrane trafficking. Most strikingly, the patients of our BIN1-related autosomal dominant CNM cohort displayed deep invaginations of the basal lamina and sarcolemma, as well as cytoplasmic accumulations of caveolin-3. Deep invaginations pointing towards centralized nuclei and abnormal localization of caveolin-3 were previously observed in several BIN1-related autosomal recessive CNM cases (Toussaint et al., 2011; Bohm et al., 2013). The considerable perinuclear accumulation of caveolin-3, mutated in limb-girdle muscular dystrophy 1C and hypertrophic cardiomyopathy, indicates a potential role of amphiphysin 2 in the transport of caveolin-3. Caveolin-3 is a key regulator of membrane trafficking in skeletal muscle and has been implicated in T-tubule maturation (Parton et al., 1997; Galbiati et al., 2001; Al-Qusairi and Laporte, 2011). In addition, we found significant accumulation of p62/SQSTM1 around central nuclei in a patient with BIN1-related autosomal dominant CNM. Altered autophagy has previously been observed in the clinically similar X-linked myopathy with excessive autophagy (MEAX, MIM# 310440; Sugie et al., 2005), and in MTM1-related
3170
| Brain 2014: 137; 3160–3170
References
Jeannet PY, Bassez G, Eymard B, Laforet P, Urtizberea JA, Rouche A, et al. Clinical and histologic findings in autosomal centronuclear myopathy. Neurology 2004; 62: 1484–90. Kojima C, Hashimoto A, Yabuta I, Hirose M, Hashimoto S, Kanaho Y, et al. Regulation of Bin1 SH3 domain binding by phosphoinositides. EMBO J 2004; 23: 4413–22. Koutsopoulos OS, Kretz C, Weller CM, Roux A, Mojzisova H, Bohm J, et al. Dynamin 2 homozygous mutation in humans with a lethal congenital syndrome. Eur J Hum Genet 2013; 21: 637–42. Laporte J, Hu LJ, Kretz C, Mandel JL, Kioschis P, Coy JF, et al. A gene mutated in X-linked myotubular myopathy defines a new putative tyrosine phosphatase family conserved in yeast. Nat Genet 1996; 13: 175–82. Laporte J, Kress W, Mandel JL. Diagnosis of X-linked myotubular myopathy by detection of myotubularin. Ann Neurol 2001; 50: 42–6. Lee E, Marcucci M, Daniell L, Pypaert M, Weisz OA, Ochoa GC, et al. Amphiphysin 2 (Bin1) and T-tubule biogenesis in muscle. Science 2002; 297: 1193–6. Low C, Weininger U, Lee H, Schweimer K, Neundorf I, BeckSickinger AG, et al. Structure and dynamics of helix-0 of the N-BAR domain in lipid micelles and bilayers. Biophys J 2008; 95: 4315–23. Nicot AS, Toussaint A, Tosch V, Kretz C, Wallgren-Pettersson C, Iwarsson E, et al. Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat Genet 2007; 39: 1134–9. Parton RG, del Pozo MA. Caveolae as plasma membrane sensors, protectors and organizers. Nat Rev Mol Cell Biol 2013; 14: 98–112. Parton RG, Way M, Zorzi N, Stang E. Caveolin-3 associates with developing T-tubules during muscle differentiation. J Cell Biol 1997; 136: 137–54. Peter BJ, Kent HM, Mills IG, Vallis Y, Butler PJ, Evans PR, et al. BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 2004; 303: 495–9. Pierson CR, Agrawal PB, Blasko J, Beggs AH. Myofiber size correlates with MTM1 mutation type and outcome in X-linked myotubular myopathy. Neuromuscul Disord 2007; 17: 562–8. Praefcke GJ, McMahon HT. The dynamin superfamily: universal membrane tubulation and fission molecules? Nat Rev Mol Cell Biol 2004; 5: 133–47. Razzaq A, Robinson IM, McMahon HT, Skepper JN, Su Y, Zelhof AC, et al. Amphiphysin is necessary for organization of the excitation-contraction coupling machinery of muscles, but not for synaptic vesicle endocytosis in Drosophila. Genes Dev 2001; 15: 2967–79. Romero NB. Centronuclear myopathies: a widening concept. Neuromuscul Disord 2010; 20: 223–8. Royer B, Hnia K, Gavriilidis C, Tronchere H, Tosch V, Laporte J. The myotubularin-amphiphysin 2 complex in membrane tubulation and centronuclear myopathies. EMBO Rep 2013; 14: 907–15. Sakamuro D, Elliott KJ, Wechsler-Reya R, Prendergast GC. BIN1 is a novel MYC-interacting protein with features of a tumour suppressor. Nat Genet 1996; 14: 69–77. Schessl J, Medne L, Hu Y, Zou Y, Brown MJ, Huse JT, et al. MRI in DNM2-related centronuclear myopathy: evidence for highly selective muscle involvement. Neuromuscul Disord 2007; 17: 28–32. Sugie K, Noguchi S, Kozuka Y, Arikawa-Hirasawa E, Tanaka M, Yan C, et al. Autophagic vacuoles with sarcolemmal features delineate Danon disease and related myopathies. J Neuropathol Exp Neurol 2005; 64: 513–22. Takei K, Slepnev VI, Haucke V, De Camilli P. Functional partnership between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat Cell Biol 1999; 1: 33–9. Tjondrokoesoemo A, Park KH, Ferrante C, Komazaki S, Lesniak S, Brotto M, et al. Disrupted membrane structure and intracellular Ca(2)( + ) signaling in adult skeletal muscle with acute knockdown of Bin1. PLoS One 2011; 6: e25740. Toussaint A, Cowling BS, Hnia K, Mohr M, Oldfors A, Schwab Y, et al. Defects in amphiphysin 2 (BIN1) and triads in several forms of centronuclear myopathies. Acta Neuropathol 2011; 121: 253–66.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Al-Qusairi L, Laporte J. T-tubule biogenesis and triad formation in skeletal muscle and implication in human diseases. Skelet Muscle 2011; 1: 26. Al-Qusairi L, Prokic I, Amoasii L, Kretz C, Messaddeq N, Mandel JL, et al. Lack of myotubularin (MTM1) leads to muscle hypotrophy through unbalanced regulation of the autophagy and ubiquitin-proteasome pathways. FASEB J 2013; 27: 3384–94. Bevilacqua JA, Bitoun M, Biancalana V, Oldfors A, Stoltenburg G, Claeys KG, et al. “Necklace” fibers, a new histological marker of late-onset MTM1-related centronuclear myopathy. Acta Neuropathol 2009; 117: 283–91. Bitoun M, Bevilacqua JA, Prudhon B, Maugenre S, Taratuto AL, Monges S, et al. Dynamin 2 mutations cause sporadic centronuclear myopathy with neonatal onset. Ann Neurol 2007; 62: 666–70. Bitoun M, Maugenre S, Jeannet PY, Lacene E, Ferrer X, Laforet P, et al. Mutations in dynamin 2 cause dominant centronuclear myopathy. Nat Genet 2005; 37: 1207–9. Bohm J, Biancalana V, Dechene ET, Bitoun M, Pierson CR, Schaefer E, et al. Mutation spectrum in the large GTPase dynamin 2, and genotype-phenotype correlation in autosomal dominant centronuclear myopathy. Hum Mutat 2012; 33: 949–59. Bohm J, Vasli N, Maurer M, Cowling B, Shelton GD, Kress W, et al. Altered splicing of the BIN1 muscle-specific exon in humans and dogs with highly progressive centronuclear myopathy. PLoS Genet 2013; 9: e1003430. Bohm J, Yis U, Ortac R, Cakmakci H, Kurul SH, Dirik E, et al. Case report of intrafamilial variability in autosomal recessive centronuclear myopathy associated to a novel BIN1 stop mutation. Orphanet J Rare Dis 2010; 5: 35. Butler MH, David C, Ochoa GC, Freyberg Z, Daniell L, Grabs D, et al. Amphiphysin II (SH3P9; BIN1), a member of the amphiphysin/Rvs family, is concentrated in the cortical cytomatrix of axon initial segments and nodes of Ranvier in brain and around T tubules in skeletal muscle. J Cell Biol 1997; 137: 1355–67. Claeys KG, Maisonobe T, Bohm J, Laporte J, Hezode M, Romero NB, et al. Phenotype of a patient with recessive centronuclear myopathy and a novel BIN1 mutation. Neurology 2010; 74: 519–21. David C, McPherson PS, Mundigl O, de Camilli P. A role of amphiphysin in synaptic vesicle endocytosis suggested by its binding to dynamin in nerve terminals. Proc Natl Acad Sci USA 1996; 93: 331–5. Dowling JJ, Vreede AP, Low SE, Gibbs EM, Kuwada JY, Bonnemann CG, et al. Loss of myotubularin function results in T-tubule disorganization in zebrafish and human myotubular myopathy. PLoS Genet 2009; 5: e1000372. Fischer D, Herasse M, Bitoun M, Barragan-Campos HM, Chiras J, Laforet P, et al. Characterization of the muscle involvement in dynamin 2-related centronuclear myopathy. Brain 2006; 129: 1463–9. Galbiati F, Engelman JA, Volonte D, Zhang XL, Minetti C, Li M, et al. Caveolin-3 null mice show a loss of caveolae, changes in the microdomain distribution of the dystrophin-glycoprotein complex, and t-tubule abnormalities. J Biol Chem 2001; 276: 21425–33. Gallop JL, Jao CC, Kent HM, Butler PJ, Evans PR, Langen R, et al. Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J 2006; 25: 2898–910. Grabs D, Slepnev VI, Songyang Z, David C, Lynch M, Cantley LC, et al. The SH3 domain of amphiphysin binds the proline-rich domain of dynamin at a single site that defines a new SH3 binding consensus sequence. J Biol Chem 1997; 272: 13419–25. Itoh T, De Camilli P. BAR, F-BAR (EFC) and ENTH/ANTH domains in the regulation of membrane-cytosol interfaces and membrane curvature. Biochim Biophys Acta 2006; 1761: 897–912. Jao CC, Hegde BG, Gallop JL, Hegde PB, McMahon HT, Haworth IS, et al. Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J Biol Chem 2010; 285: 20164–70.
J. Bo¨hm et al.
Brain 2014: 137; 3160–3170
| 3160
BRAIN A JOURNAL OF NEUROLOGY
Adult-onset autosomal dominant centronuclear myopathy due to BIN1 mutations
1 IGBMC (Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire), 67404 Illkirch, France 2 Inserm, U964, 67404 Illkirch, France 3 CNRS, UMR7104, 67404 Illkirch, France 4 Universite´ de Strasbourg, 67404 Illkirch, France 5 Colle`ge de France, Chaire de Ge´ne´tique Humaine, 67404 Illkirch, France 6 Faculte´ de Me´decine, Laboratoire de Diagnostic Ge´ne´tique, Nouvel Hoˆpital Civil, 67000 Strasbourg, France 7 Universite´ Paris 6 UM76, Inserm UMR 974, CNRS UMR 7215, Institut de Myologie, Groupe Hospitalier La Pitie´-Salpeˆtrie`re, 75013 Paris, France 8 Centre de re´fe´rence de pathologie neuromusculaire Paris-Est, Groupe Hospitalier La Pitie´-Salpeˆtrie`re, 75013 Paris, France 9 Department of Neurological, Neurosurgical, and Behavioural Sciences, University of Siena, 53100 Siena, Italy 10 Department of Human Genetics, Julius-Maximilian University, 97074 Wu¨rzburg, Germany 11 Neurology, SHG Klinikum, 66663 Merzig, Germany 12 Institute for Neurological Research, FLENI, C1428AQK Buenos Aires, Argentina 13 Hospital Italiano de Buenos Aires, C1181ACH Buenos Aires, Argentina 14 Laboratoire de Neuropathologie, Groupe Hospitalier La Pitie´-Salpeˆtrie`re, 75013 Paris, France 15 Institut de Neuropatologia, IDIBELL-Hospital Universitari de Bellvitge, 08901 Hospitalet de Llobregat, Barcelona, Spain 16 Inserm, U929, 63000 Clermont-Ferrand, France 17 Universite´ Clermont 1, 63000 Clermont-Ferrand, France 18 CHU Clermont-Ferrand, 63000 Clermont-Ferrand, France 19 De´partement de Biochimie, Biochimie et Ge´ne´tique Mole´culaire, Toxicologie et Pharmacologie, CHU Grenoble, 38700 La Tronche, France 20 Institute of Neuropathology and JARA Brain Translational Medicine, RWTH Aachen University, 52062 Aachen, Germany *These authors contributed equally to this work. Correspondence to: Jocelyn Laporte, IGBMC, 1 Rue Laurent Fries, 67404 Illkirch, France E-mail: [email protected]
Centronuclear myopathies are congenital muscle disorders characterized by type I myofibre predominance and an increased number of muscle fibres with nuclear centralization. The severe neonatal X-linked form is due to mutations in MTM1, autosomal recessive centronuclear myopathy with neonatal or childhood onset results from mutations in BIN1 (amphiphysin 2), and dominant cases were previously associated to mutations in DNM2 (dynamin 2). Our aim was to determine the genetic basis and physiopathology of patients with mild dominant centronuclear myopathy without mutations in DNM2. We hence established and characterized a homogeneous cohort of nine patients from five families with a progressive adult-onset centronuclear myopathy without facial weakness, including three sporadic cases and two families with dominant disease inheritance. All patients had similar histological and ultrastructural features involving type I fibre predominance and hypotrophy, as well as
Received March 20, 2014. Revised July 27, 2014. Accepted July 29, 2014. Advance Access publication September 26, 2014 ß The Author (2014). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Johann Bo¨hm,1,2,3,4,5,* Vale´rie Biancalana,1,2,3,4,5,6,* Edoardo Malfatti,7,8,9 Nicolas Dondaine,6 Catherine Koch,1,2,3,4,5 Nasim Vasli,1,2,3,4,5 Wolfram Kress,10 Matthias Strittmatter,11 Ana Lia Taratuto,12 Hernan Gonorazky,13 Pascal Laforeˆt,8 Thierry Maisonobe,14 Montse Olive´,15 Laura Gonzalez-Mera,15 Michel Fardeau,7,8 Nathalie Carrie`re,16,17,18 Pierre Clavelou,16,17,18 Bruno Eymard,8 Marc Bitoun,7 John Rendu,19 Julien Faure´,19 Joachim Weis,20 Jean-Louis Mandel,1,2,3,4,5,6 Norma B. Romero7,8 and Jocelyn Laporte1,2,3,4,5
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3161
prominent nuclear centralization and clustering. We identified heterozygous BIN1 mutations in all patients and the molecular diagnosis was complemented by functional analyses. Two mutations in the N-terminal amphipathic helix strongly decreased the membrane-deforming properties of amphiphysin 2 and three stop-loss mutations resulted in a stable protein containing 52 supernumerary amino acids. Immunolabelling experiments revealed abnormal central accumulation of dynamin 2, caveolin-3, and the autophagic marker p62, and general membrane alterations of the triad, the sarcolemma, and the basal lamina as potential pathological mechanisms. In conclusion, we identified BIN1 as the second gene for dominant centronuclear myopathy. Our data provide the evidence that specific BIN1 mutations can cause either recessive or dominant centronuclear myopathy and that both disorders involve different pathomechanisms.
Keywords: centronuclear myopathy; BIN1; amphiphysin 2; DNM2; T-tubule Abbreviations: CNM = centronuclear myopathy
Introduction
Materials and methods Patients Patients originated from Germany (Patient AIZ35), Argentina (Patient AIZ39), Spain (Patients ATE18, ALX95 and API51) and France (Patients IL25, AIZ64, ATI45 and AMX48). Sample collection was performed with written informed consent from the patients according to the Declaration of Helsinki.
Histology and electron microscopy Transverse muscle cryostat sections (8 mm) were stained with haematoxylin-eosin and NADH-TR and then assessed for centralized nuclei, fibre morphology, fibre type distribution, cores, protein accumulations and cellular infiltrations. For electron microscopy, muscle sections were prepared by routine methods.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Centronuclear myopathies (CNM) are a group of congenital myopathies characterized by the presence of an abnormally high number of muscle fibres with central nuclei (Romero, 2010). Facial weakness is a clinical hallmark of CNM and often accompanied by bilateral ptosis and ophthalmoparesis. Centronuclear myopathies have been described as X-linked, autosomal dominant and autosomal recessive forms. However, the genetic basis of the disease is not known for all cases. Severe neonatal X-linked CNM (XLCNM, MIM# 310400) arises from mutations in MTM1, involving a strong decrease of the myotubularin protein level (Laporte et al., 1996, 2001). Morphological analysis of patients with X-linked CNM revealed centralized nuclei, fibre atrophy, and predominance of type 1 fibres (Romero, 2010), and in some cases fibres with basophilic rings (‘necklace fibres’) are seen (Bevilacqua et al., 2009). Autosomal dominant CNM (ADCNM, MIM# 160150) is frequently caused by mutations in DNM2 and was reported with severe neonatal and clinically milder adult-onset (Bitoun et al., 2005, 2007; Bohm et al., 2012). Histological findings on biopsies from patients with autosomal dominant CNM include fibre size heterogeneity, type I fibre hypotrophy, centralized nuclei and in particular radial arrangement of sarcoplasmic strands (Jeannet et al., 2004; Romero, 2010; Toussaint et al., 2011). CT and MRI assessment demonstrated a predominant involvement of distal muscles (Fischer et al., 2006; Schessl et al., 2007). Autosomal recessive CNM (ARCNM, MIM#255200) results from homozygous BIN1 mutations and occurs as a severe neonatal or childhood onset disorder predominantly involving the proximal musculature (Nicot et al., 2007; Bohm et al., 2010; Claeys et al., 2010). The muscle biopsies display atrophy, type I fibre predominance, prominent nuclear centralization, and often nuclear clustering (Romero, 2010). Amphiphysin 2, encoded by BIN1, is a key regulator of membrane curvature and trafficking. The muscle specific isoform contains an N-terminal amphipathic helix-inducing membrane curvature, a BAR (Bin/Amphiphysin/ Rvs) domain able to sense and maintain membrane curvature, a polybasic phosphoinositide (PI) binding motif, and a SH3 domain interacting with dynamin 2 (Sakamuro et al., 1996; Grabs et al., 1997; Lee et al., 2002; Itoh and De Camilli, 2006). In skeletal
muscle, amphiphysin 2 is localized at deep sarcolemmal invaginations, the T-tubules, implicated in excitation-contraction coupling (Butler et al., 1997). When exogenously expressed in cultured cells, the muscle-specific isoform induces membrane tubulation (Lee et al., 2002; Nicot et al., 2007), suggesting that amphiphysin 2 is implicated in T-tubule biogenesis. Accordingly, drosophila amph null mutants were shown to display an abnormal T-tubule structure (Razzaq et al., 2001). To date, only five autosomal recessive CNM families with homozygous BIN1 mutations have been described. In cultured cells, the BAR domain mutations were shown to repress the amphiphysin 2 membrane tubulating properties, and the SH3 truncating mutations impaired binding and recruitment of dynamin 2 (Nicot et al., 2007). In this study, we characterize the mild end of the centronuclear myopathy spectrum. Our homogeneous cohort includes patients with progressive adult-onset myopathy without facial weakness. Histological analyses revealed major nuclear centralization and clustering in muscle fibres. Although inherited as a dominant disease, DNM2 mutations were not found. Instead, we identified heterozygous BIN1 mutations in all patients. Functional analysis demonstrated a strong impact of the mutations on the biochemical properties of amphiphysin 2. We conclude that specific BIN1 mutations cause either dominant or recessive CNM and that both disorders involve different pathomechanisms.
3162
| Brain 2014: 137; 3160–3170
J. Bo¨hm et al.
CT/MRI
stained with Hoechst/DAPI (Sigma-Aldrich) and cells were mounted and viewed by epifluorescence.
CT studies included standard scans at hip, thigh and lower leg level. The scans were examined for muscle atrophy and abnormal signal densities. MRI of the lower extremities was performed using a 1.5 Tesla magnet, and muscle atrophy was assessed on T1-weighted images.
Results
Molecular genetics
RNA studies RNA was extracted from muscle biopsies from Patients AIZ35 and AIZ39 using Tri reagent (Molecular Research Center Inc), reverse transcribed using the SuperScriptÕ III kit (Invitrogen), amplified using BIN1-specific primers and Sanger-sequenced.
Protein studies Western blot and immunofluorescence were performed using routine protocols. The following antibodies were used for the study: R2444 (homemade rabbit anti-BIN1 SH3 domain) (Nicot et al., 2007), R2405 (homemade rabbit anti-BIN1 PI domain) (Nicot et al., 2007), R3062 (homemade rabbit anti-pan BIN1) (Bohm et al., 2013), mouse anti-B10 tag (homemade) (Royer et al., 2013), mouse anti-GAPDH (Merck Millipore), mouse anti b-tubulin (BAbCO), R2680 (homemade rabbit anti dynamin 2) (Koutsopoulos et al., 2013), mouse anti-DHPR (Abcam), mouse anti-ryanodine receptor 1 (Affinity BioReagents), rabbit anti-caveolin-3 (Gene Tex), mouse Anti-p62 Lck ligand (BD Biosciences). Nuclei were stained with Hoechst (Sigma-Aldrich). For immunohistofluorescence, sections were mounted with slowfade antifade reagent (Invitrogen) and viewed using a laser scanning confocal microscope (TCS SP2; Leica Microsystems).
Constructs The BIN1-GFP construct (muscle-specific isoform 8) was a kind gift from Pietro de Camilli (Howard Hughes Medical Institute) and the BIN1 coding sequence was cloned into the pSG5-B10 vector using the Gateway system (Invitrogen). The mutations (c.61_63delAAG and c.70C 4 T) were introduced by site-directed mutagenesis using the Pfu DNA polymerase (Stratagene).
Membrane tubulation assay COS-1 cells were seeded on coverslips and transfected at 50–60% confluency with the BIN1 wild-type or mutant constructs. The next day, cells were fixed in 4% paraformaldehyde in PBS and stained using a home made anti-B10 antibody by routine methods. Nuclei were
The patients characterized in this study belong to five unrelated families and constitute a homogeneous cohort of mild and adultonset muscle disorders diagnosed as CNM. Patients AIZ35 (Family 1), AIZ39 (Family 2) and IL25 (Family 4) were sporadic cases, while Families 3 and 5 had an ancestral history of a mild muscle phenotype segregating as a dominant disease over at least three generations. All patients presented with a mildly progressive muscle weakness predominantly affecting the proximal muscles of the lower limbs (Table 1). Age of onset was between 22 (Patient ATI45, Family 5) and 50 years (Patient API51, Family 3) and at the last medical examination, most patients had walking difficulties and two patients were wheelchair-bound (Patient ATE18 from Family 3 and Patient ATI45 from Family 5). Centronuclear myopathy generally involves respiratory insufficiency, opththalmoplegia/paresis, ptosis and facial weakness. However, breathing was substantially normal in our patients, mild ptosis was only observed in Patients AIZ39 (Family 2) and API51 (Family 3), vertical gaze palsy in Patients AIZ64 and ATI45 from Family 5, and facial weakness was absent in all patients. Cognitive impairment or cardiac involvement was not noted for any of the patients. EMG revealed a clear myopathic pattern for all analysed patients and CPK levels were elevated in two patients (Patient AIZ39 from Family 2 and Patients AMX48 from Family 5). MRI/CT scans were performed for all families except Family 1 (Patient AIZ35) and essentially revealed a predominant involvement of the distal muscles and a selective involvement of the proximal muscles of the lower limbs in most patients (Fig. 1 and Supplementary material). In summary, the nine patients had comparable clinical findings of a mild adultonset muscle disorder predominantly affecting the lower limbs.
Muscle histology and ultrastructural findings For all patients, haematoxylin-eosin staining revealed prominent nuclear centralization and clusters of myonuclei, as well as significant fibre size variability (Fig. 1A). Oxidative staining revealed dense accumulations in the centre of the fibres, type I fibre predominance (65–92%) and hypotrophy and moderate disorganization of the myofibrillar network. We also infrequently found fibres with radial arrangements of sarcoplasmic strands (‘spoke of wheels’), moderate endomysial fibrosis, vacuoles and deep invaginations of the sarcolemma pointing to the centre of the fibre. Electron microscopy was performed on muscle biopsies from all patients. We observed clustered central nuclei surrounded by sarcoplasm containing accumulated glycogen granules, mitochondria and other organelles (Fig. 1B). We furthermore noted enlarged vacuoles adjacent to centralized nuclei, moderate Z-
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Genomic DNA was prepared from peripheral blood by routine procedures and sequenced for all coding exons and intron/exon boundaries of MTM1, DNM2 and BIN1 as described elsewhere (Laporte et al., 1996; Bitoun et al., 2005; Nicot et al., 2007). RYR1 was excluded by targeted deep sequencing for Families 1 and 4, and by cDNA sequencing for Families 3 and 5. Samples from Family 2 were not available for further investigations. Segregation analysis was performed by PCR and Sanger sequencing of the BIN1 exons harbouring the mutations. The mutations were numbered according to GenBank NM_139343.2 and NP_647593.1. Nucleotide position reflects cDNA numbering with + 1 corresponding to the A of the ATG translation initiation codon.
Clinical reports
Proximo-distal Axial, proximo-distal
Axial, proximo-distal
n.d.
n.d.
AP151 IL25
AIZ64
AMX48
AT145
Family 4
Family 5
n.d. = not determined.
Proximal Proximal Axial, proximo-distal
AIZ39 ATE18 ALX95
Family 2 Family 3
Muscle involvement
Patient
Proximal
M
ATI45
AIZ35
M
AMX48
France
France
Argentina Spain
Germany
Origin
Family 1
F
AIZ64
Family 5
F
F
API51
IL25
F
ALX95
Family 4
M F
AIZ39 ATE18
Family 2 Family 3
F
AIZ35
Family 1
Gender
Normal
Normal
Slightly abnormal
Normal Normal
Normal Dyspnea Normal
Normal
Breathing
c.1780delT
c.1779delA
c.70C 4 T c.1778delC
No
No
No
No No
No No No
No
Vertical gaze palsy
n.d.
Vertical gaze palsy
Mild ptosis No
Mild ptosis No No
No
Ocular involvement
p.X594AspfsX53
p.X594AspfsX53
p.Arg24Cys p.Pro593HisfsX54
p.Lys21del
Predicted protein impact
Facial weakness
c.61_63delAAG
Mutation
n.d.
No
No
No No
No No No
No
Cardiac involvement
Dominant
Sporadic
Sporadic Dominant
Sporadic
Disease segregation
n.d.
Myopathic
Myopathic
Myopathic Myopathic
Myopathic n.d. Myopathic
Myopathic
EMG
22
32
43
30
50
31
35 ?
48
Onset (age)
n.d.
Elevated (4)
Normal
Normal Normal
Central nuclei, nuclear clustering, fibre size variability, enlarged endomysium, radial sarcoplasmic strands Central nuclei, nuclear clustering n.d. Central nuclei, nuclear clustering, fibre size variability n.d. Central nuclei, nuclear clustering, fibre size variability, enlarged endomysium, radial sarcoplasmic strands Central nuclei, nuclear clustering, fibre size variability, type I fibre atrophy, radial sarcoplasmic strands, sarcolemmal invaginations Central nuclei, nuclear clustering, fibre size variability, type I fibre atrophy, enlarged vacuoles, sarcolemmal invaginations n.d.
Calf hypertrophy
CPK elevated, myalgia, photophobia, weakness of the hands, deglutition problems Contractures
Bilateral pes cavus
Bilateral scapula alata Chronic lymphocytic thyroiditis
Contractures
Histopathology hallmarks
Associated findings
Walking difficulties, requires cane or wheelchair Wheelchair-bound
Cane from age 54, wheelchair for long distances Requires wheelchair
Waddling gate
Waddling gate
Progressive, lower limbs
Slightly waddling gate, slight paresis of the left leg Cane from age 45 Wheelchair-bound
Progressive, lower limbs Progressive, predominantly lower limbs Progressive, first lower limbs, then upper limbs Progressive, predominantly lower limbs Progressive, predominantly lower limbs Progressive, predominantly lower limbs Progressive, predominantly lower limbs Progressive, lower limbs
Muscle weakness
Motor function
Elevated (10 ) n.d. Normal
Normal
CPK
22
41
Deceased age 65 61
65
46
56 88
53
Age at last medical examination
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Patient
Table 1 Main clinical, histological and molecular findings in nine patients with adult-onset dominant centronuclear myopathy
BIN1 mutations in dominant CNM Brain 2014: 137; 3160–3170
| 3163
3164
| Brain 2014: 137; 3160–3170
J. Bo¨hm et al.
Haematoxylin-eosin (HE) staining of muscle sections demonstrated prominent nuclear centralization and clustering, as well as fibre size variability and endomysial fibrosis. Oxidative NADH-TR staining revealed central accumulations in type I fibres and some radial arrangements of sarcoplasmic strands. (B) Ultrastructural analysis confirmed nuclear centralization and clustering, and revealed structural triad abnormalities (arrow), mild Z-band streaming, accumulations of glycogen granules and mitochondria, and enlarged vacuoles adjacent to centralized nuclei. (C) MRI/CT scan of Patients AIZ39 and IL25 demonstrated involvement of all distal muscles and selective proximal muscles, particularly of adductor longus (AL) and sartorius (S).
band streaming, and enlargement of membrane structures. Additional unspecific findings included autophagosomes and lipofuscin granules. Cores were not noted. In conclusion, histology and electron microscopy were suggestive of centronuclear myopathy.
Molecular genetics and functional impact of the BIN1 mutations Mutations in the CNM genes DNM2 and MTM1 were excluded in all patients of our cohort of mild and late-onset centronuclear myopathy, and RYR1 was excluded for Families 1, 3, 4 and 5.
We sequenced all coding exons and adjacent splice-relevant intronic regions of BIN1 and identified heterozygous mutations in all five families (Fig. 2A). RNA analysis was performed for two sporadic families and did not identify a second BIN1 mutation, indicating a dominant disorder. Both alleles were equably amplified and sequenced, suggesting that the heterozygous mutations did not affect mRNA stability. A muscle biopsy for RNA analysis was not available for the third sporadic case (Family 4). For the dominant Families 3 and 5, sequencing of the affected family members Patients ATE18, ALX95, API51 (Family 3), and Patients AIZ64, ATI45, AMX48 (Family 5) demonstrated disease segregation with the heterozygous BIN1 mutations. None of the identified
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 1 Histopathology, ultrastructural analysis and MRI/CT scan of patients with mild adult-onset centronuclear myopathy. (A)
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3165
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 2 Heterozygous BIN1 mutations in five families with adult-onset dominant centronuclear myopathy. (A) Patients AIZ35 and AIZ39 are sporadic cases with heterozygous missense mutations in the amphiphysin 2 amphipathic helix. Patients from Families 3, 4 and 5 harbour heterozygous single nucleotide deletions at the 3’ end of the coding sequence, affecting the regular stop codon. Healthy members with available DNA samples and without BIN1 mutation are marked as wild-type (WT). (B) The affected residues K21 and R24 (dominant CNM) and K35 (recessive CNM) are highly conserved. (C) The single nucleotide deletions at the 3’ end are predicted to induce a shift of the stop codon, resulting in a protein with 52 supernumerary amino acids. (D) Western blot analysis of muscle protein extracts from Patients IL25 (Family 4) and AIZ64 (Family 5) revealed a normal and a higher molecular weight protein compared to the healthy control. (E) Schematic representation of the amphiphysin 2 protein domains and position of the autosomal recessive CNM (black) and autosomal dominant CNM (red) mutations.
BIN1 mutations are listed in databases such as 1000 Genomes, Exome Variant Server or dbSNP, nor in our in-house database containing exome data from 276 individuals and Sanger data from 100 individuals with matched ethnicity. Taken together, these findings are clearly in favour of a dominant disease inheritance and genetically evidence the pathogenicity of the identified heterozygous BIN1 mutations.
For Families 3, 4 and 5 we identified heterozygous single nucleotide deletions in the last codon or in the regular stop codon of BIN1. The mutations c.1778delC (p.Pro593HisfsX54; Family 3) and c.1780delT (p.X594AspfsX53; Family 5) segregated as an autosomal dominant disease and the c.1779delA (p.X594AspfsX53; Family 4) deletion was found in a singleton case. All three single nucleotide deletions are predicted to shift the reading frame of the regular stop
3166
| Brain 2014: 137; 3160–3170
As amphiphysin 2 was proposed to be implicated in T-tubule biogenesis (Razzaq et al., 2001; Lee et al., 2002), we examined the localization of the T-tubule marker DHPR and the junctional sarcoplasmic calcium channel RYR1. The localization of both proteins was strongly altered, showing focal accumulations in the muscle fibres. Accordingly, we observed structural triad abnormalities on the patient biopsies (Fig. 1). To explore if the abnormal triad structure reflects general membrane defects, we analysed caveolin-3, a key regulator of membrane trafficking in muscle (Parton and del Pozo, 2013) and noted a striking mislocalization within the cytoplasm as well as prominent perinuclear accumulations, potentially corresponding to the enlarged perinuclear vacuoles observed by electron microscopy. By labelling of laminin, we furthermore confirmed the presence of characteristic deep invaginations of the sarcolemma and the basal lamina often pointing towards central nuclei (Fig. 4B). These defects were not seen on muscle sections from CNM patients with MTM1 or DNM2 mutations (Supplementary Fig. 2), and might serve as distinctive markers for BIN1-related CNM. In order to assess whether the overall structural aberrations within the muscle fibres involve autophagy, we monitored the autophagic marker p62 on muscle sections from Patient IL25 (Family 4) and we found significant positive signals around central nuclei (Fig. 5). Taken together, these data demonstrate that BIN1-dominant CNM is associated with general membrane alterations of the triad, the sarcolemma and the basal lamina, and potentially the autophagic pathway.
Discussion In this study we identified heterozygous BIN1 mutations as a genetic cause of adult-onset autosomal dominant CNM. We validated our data at the clinical, histological, genetic and functional levels. BIN1 is the second gene implicated in autosomal dominant CNM after DNM2. The consistent clinical, histological and ultrastructural features of the nine patients with autosomal dominant CNM with dominant BIN1 mutations described in this study clearly differ from BIN1related autosomal recessive CNM or DNM2-related autosomal dominant CNM. Our patients presented with an exclusively adult-onset muscle weakness affecting selected proximal muscles and all distal muscles and of the lower limbs. Predominant proximal muscle weakness was reported for BIN1-related autosomal recessive CNM, whereas DNM2-autosomal dominant CNM patients displayed involvement of the distal muscles (Fischer et al., 2006; Bohm et al., 2010). All reported BIN1-related autosomal recessive CNM cases had neonatal/childhood disease onset and often marked respiratory insufficiency and facial weakness, which are not seen in our clinically milder patients. Similarly, ocular involvement is classically seen in the other CNM forms, and was not or only barely present in our cohort. Histological and ultrastructural analyses of the BIN1-related autosomal dominant CNM patients revealed nuclear centralization and clustering, consistent with BIN1-related autosomal recessive CNM and contrasting the muscle morphology described for dominant CNM cases with DNM2 mutations. The ratio of fibres with centralized
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
codon and to generate an extended amphiphysin 2 protein containing 52 supernumerary C-terminal amino acids (Fig. 2C). The presence of a protein of higher molecular weight was confirmed by western blot of protein extracts from muscle biopsies (Families 4 and 5; Fig. 2D) or lymphoblasts (Family 3; Supplementary Fig. 1). We conclude that the stop-loss mutations do not fundamentally affect mRNA or protein stability. For Family 1 we identified a heterozygous c.61_63delAAG (p.Lys21del) in-frame deletion of three nucleotides, and for Family 2 we found a heterozygous c.70C 4 T (p.Arg24Cys) missense mutation. In both cases, the mutation was not found in the healthy mothers, and the fathers’ DNAs were not available. Both mutations affect essential and conserved residues in the N-terminal amphipathic helix (Fig. 2B), which inserts into the membrane bilayer and plays a crucial role in curvature induction (Peter et al., 2004; Gallop et al., 2006; Low et al., 2008; Jao et al., 2010). The mutations are predicted to impact on the composition and the orientation of the hydrophobic face of the amphipathic helix (Fig. 3A). The mutations do not seem to influence protein stability, as western blot revealed a comparable band size and intensity for cells expressing GFP-BIN1 wild-type, GFP-BIN1 K21del, or GFPBIN1 R24C (Supplementary Fig. 1). In cultured cells, exogenously expressed wild-type GFP-BIN1 promotes membrane tubulation (Fig. 3B) (Lee et al., 2002; Nicot et al., 2007). In contrast, GFP-BIN1 constructs harbouring the p.Lys21del (K21del) or the p.Arg24Cys (R24C) mutations did not induce tubulation. The p.Lys35Asn (K35N) and the p.Asp151Asn (D151N) mutations, both found in patients with severe autosomal recessive CNM, showed no or only partial membrane tubulation (Nicot et al., 2007). Membrane tubulation is mediated through dimerization of the BAR domains of amphiphysin 2 (Peter et al., 2004; Gallop et al., 2006). To assess a potential different effect of the dominant compared to the recessive BIN1 mutations, we co-transfected wild-type GFP-BIN1 and mutant B10-BIN1 constructs (Fig. 3C). The recessive K35N and D151N mutants were efficiently recruited to the membrane tubules induced by the wild-type construct, demonstrating that these loss-of-function mutations do not interfere with dimerization. In contrast, the K21del and R24C constructs barely localized at the tubules induced by the wild-type protein, suggesting that they affect recruitment of mutated/wild-type heterodimers to the membranes and/or amphiphysin 2 dimerization. Importantly, and in contrast to the recessive BIN1 K35N and D151N mutations, the impact of the dominant K21del and R24C mutations cannot be rescued by the co-expressed wild-type protein. These results clearly indicate a different pathomechanism in BIN1-related autosomal dominant CNM versus autosomal recessive CNM. To further characterize the physiopathology of BIN1-related autosomal dominant CNM, we performed immunolabelling on muscle biopsies from our patients with dominant BIN1 mutations (Fig. 4). We noted scattered amphiphysin 2 accumulations, most often surrounding centralized nuclei. Similarly, dynamin 2 accumulated around centralized nuclei. Dynamins are known to interact with amphiphysins (David et al., 1996; Kojima et al., 2004). Our data therefore suggest that the dynamin 2 accumulations in our patients are a direct consequence of the BIN1 mutations, and provide a molecular link between both dominant CNM forms.
J. Bo¨hm et al.
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3167
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 3 Functional impact of the amphipathic helix mutations. (A) The identified BIN1 mutations are predicted to impact on the amphipathic helix properties. (B) Overexpression of GFP-BIN1-WT in COS-1 cells induces membrane tubulation. The amphipathic helix mutations strongly impair membrane tubulation. (C) Co-expression of wild-type GFP-BIN1 and mutant B10-BIN1 constructs show that the dominant K21del and R24C mutations, but not the recessive K35N and D151N mutations, impair recruitment to the membrane tubules induced by the wild-type protein. Nuclei are stained with Hoechst and are shown in blue.
nuclei does not seem to correlate with disease severity, as our clinically mild patients presented central nuclei in the vast majority of the fibres. A similar observation was made for X-linked CNM (Pierson et al., 2007). Comparison of our data with the literature suggests that this cohort represents the mild end of the centronuclear myopathy spectrum.
The dominant BIN1 mutations reported in this study differ from the homozygous BIN1 mutations causing recessive CNM (Nicot et al., 2007; Bohm et al., 2010; Claeys et al., 2010). In seven of our patients we found single nucleotide deletions inducing the expression of a stable amphiphysin 2 protein with 52 supernumerary amino acids. This extension follows the SH3 domain and
3168
| Brain 2014: 137; 3160–3170
J. Bo¨hm et al.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Figure 4 Abnormal localization of proteins and structural defects of the muscle fibre. (A) Immunolocalization on transversal sections from patients revealed an abnormal pattern of the CNM proteins amphiphysin 2 and dynamin 2, the T-tubule marker DHPR, the junctional sarcoplasmic calcium channel RYR1, and the sarcolemmal marker caveolin-3 (all indicated by arrows). (B) Immunolabelling of laminin unravelled deep invaginations of the sarcolemma and basal lamina pointing towards the centre of fibres (arrows). The electron microscopy picture shows a close-up of an invagination.
potentially impacts on the domain folding or the interaction with other proteins. Amphiphysin recruits dynamin to the plasma membrane (Takei et al., 1999; Lee et al., 2002; Praefcke and McMahon 2004) and accordingly, we observed a striking mislocalization of both proteins around central nuclei on muscle sections from our patients. We also found two missense mutations
in the N-BAR domain impairing the membrane tubulation properties of amphiphysin 2. In contrast to the recessive mutations in the N-BAR domain, the dominant mutations additionally affect the recruitment to membrane tubules induced by wild-type amphiphysin 2. These observations suggest that the BIN1 autosomal dominant CNM mutations interfere with membrane binding and/or
BIN1 mutations in dominant CNM
Brain 2014: 137; 3160–3170
| 3169
Figure 5 p62 accumulations around centralized nuclei. p62 accumulations were found around centralized nuclei in muscle sections from Patient IL25 (Family 3, arrow). Muscle sections from a patient with X-linked myopathy with excessive autophagy (MEAX) and an unaffected individual were used as controls.
X-linked CNM (Al-Qusairi et al., 2013) and might contribute to the muscle pathology in BIN1-related CNM. In conclusion, BIN1 is the second gene implicated in dominant centronuclear myopathy. This is of high importance for genetic diagnosis and counselling as BIN1 mutations should not only be considered for recessive families, but also for dominant families and sporadic adult cases, especially if the biopsy displays prominent nuclear clustering in patients with mild adult-onset muscle disorders. BIN1 mutations can cause autosomal dominant or autosomal recessive CNM and both conditions occur with similar frequency. The late onset, the mild clinical course, the histopathologial features and the dominant disease inheritance of BIN1-related autosomal dominant CNM clearly differ from the reported CNM forms. Our study broadens the phenotypic spectrum of muscle disorders caused by BIN1 mutations and provides novel insights into the pathogenesis of centronuclear myopathies. The disorganization of the skeletal muscle triad and the basal lamina, and the cytoplasmic accumulation of caveolin-3 and p62 suggest that a general disorganization of membrane trafficking is the main pathological cause of this disorder.
Acknowledgements We thank the patients for their cooperation and interest in this study, Pietro de Camilli for the GFP-BIN1 construct, and Linda Mane`re, Osorio Abath Neto, Adeline Normand, Peggy Therier and Lena Kristina Beilschmidt for technical assistance.
Funding This work was supported by the INSERM, CNRS, University of Strasbourg, Colle`ge de France, Agence Nationale de la Recherche (ANR-11-BSV1-026), Association Franc¸aise contre les Myopathies (AFM 15352), Muscular Dystrophy Association (MDA186985) and Myotubular Trust.
Supplementary material Supplementary material is available at Brain online.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
dimerization, whereas the membrane tubulation defects caused by the BIN1 autosomal recessive CNM mutations can be rescued by co-transfection of wild-type BIN1. Indeed, heterozygous carriers of BIN1 autosomal recessive CNM mutations are healthy, suggesting different pathomechanisms for BIN1-related autosomal recessive CNM and autosomal dominant CNM. We hypothesize that both disorders arise from loss-of-function with a partial loss-of-function for the recessive BIN1 mutations, and a full lossof-function for the dominant BIN1 mutations. The pathological mechanisms can be inferred from our functional in cellulo data and from the histological and immunolabelling experiments on patient muscles. First, we noted concurrent mislocalizations of amphiphysin 2 and dynamin 2, both implicated in dominant CNM. Second, the disorganization of triad markers and the abnormal triad structure seen by electron microscopy is in accordance with a role of amphiphysin 2 at the T-tubules (Razzaq et al., 2001; Kojima et al., 2004), and suggests that impaired excitation-contraction coupling and subsequent calcium homeostasis defects are the most likely cause of the muscle weakness. Indeed, triad defects were seen on biopsies from patients with other CNM forms (Dowling et al., 2009; Romero, 2010; Toussaint et al., 2011) and abnormal intracellular calcium release was detected in isolated murine muscle fibres following BIN1 shRNA-mediated knock-down (Tjondrokoesoemo et al., 2011). Third, we noted a general disorganization of membrane structures and proteins regulating membrane trafficking. Most strikingly, the patients of our BIN1-related autosomal dominant CNM cohort displayed deep invaginations of the basal lamina and sarcolemma, as well as cytoplasmic accumulations of caveolin-3. Deep invaginations pointing towards centralized nuclei and abnormal localization of caveolin-3 were previously observed in several BIN1-related autosomal recessive CNM cases (Toussaint et al., 2011; Bohm et al., 2013). The considerable perinuclear accumulation of caveolin-3, mutated in limb-girdle muscular dystrophy 1C and hypertrophic cardiomyopathy, indicates a potential role of amphiphysin 2 in the transport of caveolin-3. Caveolin-3 is a key regulator of membrane trafficking in skeletal muscle and has been implicated in T-tubule maturation (Parton et al., 1997; Galbiati et al., 2001; Al-Qusairi and Laporte, 2011). In addition, we found significant accumulation of p62/SQSTM1 around central nuclei in a patient with BIN1-related autosomal dominant CNM. Altered autophagy has previously been observed in the clinically similar X-linked myopathy with excessive autophagy (MEAX, MIM# 310440; Sugie et al., 2005), and in MTM1-related
3170
| Brain 2014: 137; 3160–3170
References
Jeannet PY, Bassez G, Eymard B, Laforet P, Urtizberea JA, Rouche A, et al. Clinical and histologic findings in autosomal centronuclear myopathy. Neurology 2004; 62: 1484–90. Kojima C, Hashimoto A, Yabuta I, Hirose M, Hashimoto S, Kanaho Y, et al. Regulation of Bin1 SH3 domain binding by phosphoinositides. EMBO J 2004; 23: 4413–22. Koutsopoulos OS, Kretz C, Weller CM, Roux A, Mojzisova H, Bohm J, et al. Dynamin 2 homozygous mutation in humans with a lethal congenital syndrome. Eur J Hum Genet 2013; 21: 637–42. Laporte J, Hu LJ, Kretz C, Mandel JL, Kioschis P, Coy JF, et al. A gene mutated in X-linked myotubular myopathy defines a new putative tyrosine phosphatase family conserved in yeast. Nat Genet 1996; 13: 175–82. Laporte J, Kress W, Mandel JL. Diagnosis of X-linked myotubular myopathy by detection of myotubularin. Ann Neurol 2001; 50: 42–6. Lee E, Marcucci M, Daniell L, Pypaert M, Weisz OA, Ochoa GC, et al. Amphiphysin 2 (Bin1) and T-tubule biogenesis in muscle. Science 2002; 297: 1193–6. Low C, Weininger U, Lee H, Schweimer K, Neundorf I, BeckSickinger AG, et al. Structure and dynamics of helix-0 of the N-BAR domain in lipid micelles and bilayers. Biophys J 2008; 95: 4315–23. Nicot AS, Toussaint A, Tosch V, Kretz C, Wallgren-Pettersson C, Iwarsson E, et al. Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat Genet 2007; 39: 1134–9. Parton RG, del Pozo MA. Caveolae as plasma membrane sensors, protectors and organizers. Nat Rev Mol Cell Biol 2013; 14: 98–112. Parton RG, Way M, Zorzi N, Stang E. Caveolin-3 associates with developing T-tubules during muscle differentiation. J Cell Biol 1997; 136: 137–54. Peter BJ, Kent HM, Mills IG, Vallis Y, Butler PJ, Evans PR, et al. BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science 2004; 303: 495–9. Pierson CR, Agrawal PB, Blasko J, Beggs AH. Myofiber size correlates with MTM1 mutation type and outcome in X-linked myotubular myopathy. Neuromuscul Disord 2007; 17: 562–8. Praefcke GJ, McMahon HT. The dynamin superfamily: universal membrane tubulation and fission molecules? Nat Rev Mol Cell Biol 2004; 5: 133–47. Razzaq A, Robinson IM, McMahon HT, Skepper JN, Su Y, Zelhof AC, et al. Amphiphysin is necessary for organization of the excitation-contraction coupling machinery of muscles, but not for synaptic vesicle endocytosis in Drosophila. Genes Dev 2001; 15: 2967–79. Romero NB. Centronuclear myopathies: a widening concept. Neuromuscul Disord 2010; 20: 223–8. Royer B, Hnia K, Gavriilidis C, Tronchere H, Tosch V, Laporte J. The myotubularin-amphiphysin 2 complex in membrane tubulation and centronuclear myopathies. EMBO Rep 2013; 14: 907–15. Sakamuro D, Elliott KJ, Wechsler-Reya R, Prendergast GC. BIN1 is a novel MYC-interacting protein with features of a tumour suppressor. Nat Genet 1996; 14: 69–77. Schessl J, Medne L, Hu Y, Zou Y, Brown MJ, Huse JT, et al. MRI in DNM2-related centronuclear myopathy: evidence for highly selective muscle involvement. Neuromuscul Disord 2007; 17: 28–32. Sugie K, Noguchi S, Kozuka Y, Arikawa-Hirasawa E, Tanaka M, Yan C, et al. Autophagic vacuoles with sarcolemmal features delineate Danon disease and related myopathies. J Neuropathol Exp Neurol 2005; 64: 513–22. Takei K, Slepnev VI, Haucke V, De Camilli P. Functional partnership between amphiphysin and dynamin in clathrin-mediated endocytosis. Nat Cell Biol 1999; 1: 33–9. Tjondrokoesoemo A, Park KH, Ferrante C, Komazaki S, Lesniak S, Brotto M, et al. Disrupted membrane structure and intracellular Ca(2)( + ) signaling in adult skeletal muscle with acute knockdown of Bin1. PLoS One 2011; 6: e25740. Toussaint A, Cowling BS, Hnia K, Mohr M, Oldfors A, Schwab Y, et al. Defects in amphiphysin 2 (BIN1) and triads in several forms of centronuclear myopathies. Acta Neuropathol 2011; 121: 253–66.
Downloaded from http://brain.oxfordjournals.org/ by guest on November 15, 2015
Al-Qusairi L, Laporte J. T-tubule biogenesis and triad formation in skeletal muscle and implication in human diseases. Skelet Muscle 2011; 1: 26. Al-Qusairi L, Prokic I, Amoasii L, Kretz C, Messaddeq N, Mandel JL, et al. Lack of myotubularin (MTM1) leads to muscle hypotrophy through unbalanced regulation of the autophagy and ubiquitin-proteasome pathways. FASEB J 2013; 27: 3384–94. Bevilacqua JA, Bitoun M, Biancalana V, Oldfors A, Stoltenburg G, Claeys KG, et al. “Necklace” fibers, a new histological marker of late-onset MTM1-related centronuclear myopathy. Acta Neuropathol 2009; 117: 283–91. Bitoun M, Bevilacqua JA, Prudhon B, Maugenre S, Taratuto AL, Monges S, et al. Dynamin 2 mutations cause sporadic centronuclear myopathy with neonatal onset. Ann Neurol 2007; 62: 666–70. Bitoun M, Maugenre S, Jeannet PY, Lacene E, Ferrer X, Laforet P, et al. Mutations in dynamin 2 cause dominant centronuclear myopathy. Nat Genet 2005; 37: 1207–9. Bohm J, Biancalana V, Dechene ET, Bitoun M, Pierson CR, Schaefer E, et al. Mutation spectrum in the large GTPase dynamin 2, and genotype-phenotype correlation in autosomal dominant centronuclear myopathy. Hum Mutat 2012; 33: 949–59. Bohm J, Vasli N, Maurer M, Cowling B, Shelton GD, Kress W, et al. Altered splicing of the BIN1 muscle-specific exon in humans and dogs with highly progressive centronuclear myopathy. PLoS Genet 2013; 9: e1003430. Bohm J, Yis U, Ortac R, Cakmakci H, Kurul SH, Dirik E, et al. Case report of intrafamilial variability in autosomal recessive centronuclear myopathy associated to a novel BIN1 stop mutation. Orphanet J Rare Dis 2010; 5: 35. Butler MH, David C, Ochoa GC, Freyberg Z, Daniell L, Grabs D, et al. Amphiphysin II (SH3P9; BIN1), a member of the amphiphysin/Rvs family, is concentrated in the cortical cytomatrix of axon initial segments and nodes of Ranvier in brain and around T tubules in skeletal muscle. J Cell Biol 1997; 137: 1355–67. Claeys KG, Maisonobe T, Bohm J, Laporte J, Hezode M, Romero NB, et al. Phenotype of a patient with recessive centronuclear myopathy and a novel BIN1 mutation. Neurology 2010; 74: 519–21. David C, McPherson PS, Mundigl O, de Camilli P. A role of amphiphysin in synaptic vesicle endocytosis suggested by its binding to dynamin in nerve terminals. Proc Natl Acad Sci USA 1996; 93: 331–5. Dowling JJ, Vreede AP, Low SE, Gibbs EM, Kuwada JY, Bonnemann CG, et al. Loss of myotubularin function results in T-tubule disorganization in zebrafish and human myotubular myopathy. PLoS Genet 2009; 5: e1000372. Fischer D, Herasse M, Bitoun M, Barragan-Campos HM, Chiras J, Laforet P, et al. Characterization of the muscle involvement in dynamin 2-related centronuclear myopathy. Brain 2006; 129: 1463–9. Galbiati F, Engelman JA, Volonte D, Zhang XL, Minetti C, Li M, et al. Caveolin-3 null mice show a loss of caveolae, changes in the microdomain distribution of the dystrophin-glycoprotein complex, and t-tubule abnormalities. J Biol Chem 2001; 276: 21425–33. Gallop JL, Jao CC, Kent HM, Butler PJ, Evans PR, Langen R, et al. Mechanism of endophilin N-BAR domain-mediated membrane curvature. EMBO J 2006; 25: 2898–910. Grabs D, Slepnev VI, Songyang Z, David C, Lynch M, Cantley LC, et al. The SH3 domain of amphiphysin binds the proline-rich domain of dynamin at a single site that defines a new SH3 binding consensus sequence. J Biol Chem 1997; 272: 13419–25. Itoh T, De Camilli P. BAR, F-BAR (EFC) and ENTH/ANTH domains in the regulation of membrane-cytosol interfaces and membrane curvature. Biochim Biophys Acta 2006; 1761: 897–912. Jao CC, Hegde BG, Gallop JL, Hegde PB, McMahon HT, Haworth IS, et al. Roles of amphipathic helices and the bin/amphiphysin/rvs (BAR) domain of endophilin in membrane curvature generation. J Biol Chem 2010; 285: 20164–70.
J. Bo¨hm et al.
![[Centronuclear myopathy with autosomal dominant inheritance(author's transl)].](/assets/img/hold.png)
