Abstract

Non-syndromic arthrogryposis multiplex congenita (AMC) is characterized by multiple congenital contractures resulting from reduced fetal mobility. Genetic mapping and whole exome sequencing (WES) were performed in 31 multiplex and/or consanguineous undiagnosed AMC families. Although this approach identified known AMC genes, we here report pathogenic mutations in two new genes. Homozygous frameshift mutations in CNTNAP1 were found in four unrelated families. Patients showed a marked reduction in motor nerve conduction velocity (<10 m/s) and transmission electron microscopy (TEM) of sciatic nerve in the index cases revealed severe abnormalities of both nodes of Ranvier width and myelinated axons. CNTNAP1 encodes CASPR, an essential component of node of Ranvier domains which underlies saltatory conduction of action potentials along the myelinated axons, an important process for neuronal function. A homozygous missense mutation in adenylate cyclase 6 gene (ADCY6) was found in another family characterized by a lack of myelin in the peripheral nervous system (PNS) as determined by TEM. Morpholino knockdown of the zebrafish orthologs led to severe and specific defects in peripheral myelin in spite of the presence of Schwann cells. ADCY6 encodes a protein that belongs to the adenylate cyclase family responsible for the synthesis of cAMP. Elevation of cAMP can mimic axonal contact in vitro and upregulates myelinating signals. Our data indicate an essential and so far unknown role of ADCY6 in PNS myelination likely through the cAMP pathway. Mutations of genes encoding proteins of Ranvier domains or involved in myelination of Schwann cells are responsible for novel and severe human axoglial diseases.

INTRODUCTION

Arthrogryposis multiplex congenita (AMC) is characterized by congenital contractures of at least two distinct joints of the body. The overall incidence is 1 in 3000 of live births (1,2). Some non-genetic factors may cause AMC, such as mechanical limitation of fetal movements or maternal autoimmune myasthenia. A number of genetic syndromes including AMC phenotype, collectively referred to as syndromic AMC, have been described under several conditions (3,4). Non-syndromic or isolated AMCs are the direct consequence of fetal akinesia/hypokinesia sequence which may lead, in addition to AMCs, to pterygia, lung hypoplasia, diaphragmatic defect or cleft palate. Isolated AMCs are genetically heterogeneous. Mutations of genes encoding components of the neuromuscular junctions (NMJ), including CHRNG (MIM 100730), CHRNA1 (MIM 100690), CHRND (MIM 100720), CHRNB1 (MIM 100710), DOK7 (MIM 610285), RAPSN (MIM 601592) or CHAT (MIM 118490), are responsible for lethal multiple pterygium syndromes, isolated AMC with neonatal myasthenia or Escobar syndrome. Fetal motor neuron diseases may also result in lethal congenital contractures caused by mutations in GLE1 (MIM 603371), PIP5K1C (MIM 606102) or ERBB3 (MIM 190151) genes. AMC also occasionally occurs in type I spinal muscular atrophy (SMA) caused by mutations of SMN1 (MIM 600354) or congenital SMA linked to TRPV4 mutations (MIM 605427). More recently, mutations of ECEL1 (MIM 605896) have been reported in distal AMC. Congenital myopathies associated with distal AMC may be caused by mutations in TPM2 (MIM 190990), MYH2 (MIM 160740), MYH3 (MIM 160720), MYH8 (MIM 160741), TNNI2 (MIM 191043), TNNT3 (MIM 600692) or MYBPC1 (MIM 160794) genes. Congenital myotonic dystrophy caused by abnormal triplet expansion of the DMPK (MIM 605377) gene, nemalin myopathy linked to ACTA1 (MIM 102610), TPM2 (MIM 190990) or NEB (MIM 161650), minicore myopathy linked to RYR1 mutations (MIM 180901) or more recently SYNE1 mutation (MIM 608441) have all been reported in non-syndromic AMC. Collectively, non-syndromic AMCs include a large spectrum of diseases of motor neurons, NMJs or skeletal muscle.

The difficulty in establishing a genetic diagnosis for AMC patients is likely due to the high genetic heterogeneity and/or to some not yet identified disease-causing genes. Moreover, there is a lack of suitable screening methods of all known AMC genes. In order to gain further insight into the underlying cause of these diseases, we took advantage of the added value of whole genome scanning using SNP microarrays alone or combined with whole exome sequencing (WES) to study a cohort of 31 multiplex and/or consanguineous families with undiagnosed non-syndromic AMC.

RESULTS

Phenotypic characterization of AMC patients

We enrolled 63 affected fetuses, children or adults belonging to 31 multiplex and/or consanguineous families with unexplained non-syndromic AMC (Supplementary Material, Fig. S1). The main clinical criteria were the gestational stage of AMC diagnosis based on the ultrasound examination (weeks’ gestation, wg, the earliest one in multiplex families), the topography of joint contractures, associated symptoms including reduced fetal mobility, pterygium, micrognathia and cleft palate, the amount of amniotic fluid, and cystic hygroma (Table 1). The stage of AMC diagnosis showed a marked heterogeneity of fetal onset starting from 11 wg and up to 32 wg. Reduced fetal mobility was associated with AMC in 24 out of 28 affected individuals (86%). No information was available in three families. Targeted gene tests when performed did not lead to a diagnosis (Supplementary Material, Table S1). The primary defects were established based on gene identification and function (see below) and were classified as skeletal muscle, NMJ or axoglial AMC. AMCs associated with pterygia, both distal and proximal contractures or cystic hygroma, are significantly more frequent in skeletal muscle or NMJ than axoglial classes (P < 0.02, Fisher's exact test, Fig. 1).

Table 1.

Clinical characteristics of AMC families

Code Gene Tissue Target Onset (wg) AMC Hypo/akinesia Poly-Hydramnios Pterygium Micro-gnathia Cleft palate Hygroma 
A641 CNTNAP1 Axoglial 30 – – – – 
K182 CNTNAP1 Axoglial 32 – – – – 
K199 CNTNAP1 Axoglial 30 unk – – – unk 
B207 CNTNAP1 Axoglial 28 unk – – unk 
A649 ADCY6 Axoglial 32 – – – – – 
A635 ECEL1 Axoglial 16 P+D – – – – 
B192 ECEL1 Axoglial Uk – – – – – – 
B651 TRPV4 Axoglial 19 P+D – – – – – – 
A638 CHRNG NMJ 12 P+D – – 
A651 CHRNG NMJ 11 P+D – – 
A662 CHRNG NMJ 17 P+D – – – – – 
A650 RAPSN NMJ 23 P+D – – – 
A640 MYH3 Sk Mu 13 P+D – 
A646 NEB Sk Mu 14 P+D – – 
A631 NEB Sk Mu 13 P+D – – 
B415 RYR1 Sk Mu 20 P+D – 
A648 RYR1 Sk Mu 13 P+D – 
A636 RYR1 Sk Mu 15 P+D 
A656 RYR1 Sk Mu 20 P+D – 
K166 RYR1 Sk Mu 12 P+D – 
K168 SYNE1 Sk Mu 22 – – – – – – 
A663 TNNT3 Sk Mu 20 – – – – – – 
A642 TTN Sk Mu 26 P+D – – – 
A659 RBBP8 Other 24 P+D – 
A658 unk  22 – – – 
K180 unk  21 – – – 
K171 unk  14 P+D – 
K169 unk  35 P+D unk unk unk – unk 
K174 unk  22 P+D – – – –  
K165 unk  12 P+D – 
A657 unk  14 P+D – – 
Code Gene Tissue Target Onset (wg) AMC Hypo/akinesia Poly-Hydramnios Pterygium Micro-gnathia Cleft palate Hygroma 
A641 CNTNAP1 Axoglial 30 – – – – 
K182 CNTNAP1 Axoglial 32 – – – – 
K199 CNTNAP1 Axoglial 30 unk – – – unk 
B207 CNTNAP1 Axoglial 28 unk – – unk 
A649 ADCY6 Axoglial 32 – – – – – 
A635 ECEL1 Axoglial 16 P+D – – – – 
B192 ECEL1 Axoglial Uk – – – – – – 
B651 TRPV4 Axoglial 19 P+D – – – – – – 
A638 CHRNG NMJ 12 P+D – – 
A651 CHRNG NMJ 11 P+D – – 
A662 CHRNG NMJ 17 P+D – – – – – 
A650 RAPSN NMJ 23 P+D – – – 
A640 MYH3 Sk Mu 13 P+D – 
A646 NEB Sk Mu 14 P+D – – 
A631 NEB Sk Mu 13 P+D – – 
B415 RYR1 Sk Mu 20 P+D – 
A648 RYR1 Sk Mu 13 P+D – 
A636 RYR1 Sk Mu 15 P+D 
A656 RYR1 Sk Mu 20 P+D – 
K166 RYR1 Sk Mu 12 P+D – 
K168 SYNE1 Sk Mu 22 – – – – – – 
A663 TNNT3 Sk Mu 20 – – – – – – 
A642 TTN Sk Mu 26 P+D – – – 
A659 RBBP8 Other 24 P+D – 
A658 unk  22 – – – 
K180 unk  21 – – – 
K171 unk  14 P+D – 
K169 unk  35 P+D unk unk unk – unk 
K174 unk  22 P+D – – – –  
K165 unk  12 P+D – 
A657 unk  14 P+D – – 

wg, weeks of gestation; P+D, proximal and distal; D, distal; Y, yes; (–), No; unk, unknown; NMJ, neuromuscular junction; Sk Mu, skeletal muscle.

Figure 1.

Percentage of AMC patients with associated symptoms and classified upon the primary tissue targets (Sk Mu, skeletal muscle; NMJ, neuromuscular junction or axoglial). Asterisk indicates a significant difference (P < 0.02) between groups using Fisher's exact test. (P+D) AMC indicates that contractures involved both proximal and distal joints.

Figure 1.

Percentage of AMC patients with associated symptoms and classified upon the primary tissue targets (Sk Mu, skeletal muscle; NMJ, neuromuscular junction or axoglial). Asterisk indicates a significant difference (P < 0.02) between groups using Fisher's exact test. (P+D) AMC indicates that contractures involved both proximal and distal joints.

Identification of mutations in genes known to be responsible for AMC or neuromuscular disorders

The overall strategy for gene identification is depicted in Supplementary Material, Figure S2. Candidate loci were identified in each family by multipoint linkage analysis and homozygosity mapping by using SNP microarrays. When a single known AMC gene was identified in candidate disease loci, Sanger sequencing of exons and intron–exon junctions was performed. This was successfully applied in three families linked to mutations of RYR1 (family B415), TRPV4 (family B651, Table 2, Supplementary Material, Table S2 and Supplementary Material, Fig. S3) or ECEL1 (family B192, 5). When numerous or none of the AMC genes were identified in candidate loci, WES was performed using the DNA sample of one affected individual per family (n = 27). As a first filtering, homozygous or compound heterozygous variants mapping to the disease loci were selected in recessive and heterozygous variants in dominant forms, respectively (Table 2 and Supplementary Material, Table S2). Variants were then filtered against a list of genes known to be involved in AMC or neuromuscular disorders (NMDs). Mutations in NEB, RYR1, SYNE1, TNNT3, TTN or CHRNG genes were successfully identified in 10 out of 31 families (32%, Table 2 and Supplementary Material, Figs S3 and S4). Interestingly, homozygous stop gained mutation was found in exon 136 of SYNE-1 in family K168, which confirms that mutations of SYNE-1 are responsible for AMC since a single AMC family only was previously reported (6). In family A642, a non-sense mutation on one allele and a frameshift mutation on the other allele of TTN were found extending the clinical spectrum of TTN gene mutations. In a consanguineous family (A659), a homozygous splice mutation was found in RBBP8 (Table 2, Supplementary Material, Figs S3 and S4). Mutations of this gene have been reported in Seckel syndrome type 2, which may associate microcephaly, holoprosencephaly and arthrogryposis (7), suggesting that the affected fetus had a syndromic AMC that was not detected during pregnancy. In 16 families, allelic mutations were not identified using the above criteria. In four of them, a single heterozygous variant in AMC or NMD genes and mapping to candidate loci was identified. Non-covered exons visualized by the Integrative genomics viewer software (IGV, v1.5.64, 8) were then sequenced by the Sanger method. Allelic mutations were found in ECEL1, RAPSN, RYR1 and NEB (Table 2 and Supplementary Material, Fig. S3). When no mutations were identified using these filters, the hypothesis of dominant de novo mutation was tested by selecting variants within AMC or NMD genes without any linkage data filter. A single missense mutation of the MYH3 gene was identified in the affected fetuses of family A640, the father showing somatic mosaicism pattern (Table 2 and Supplementary Material, Fig. S3).

Table 2.

Identified mutations in genes associated with AMC

Code Inheritance Gene Transcript Ref.: nucleotide change MAF (dbSNP) MAF (EVS) Genotype Type Protein change Target 
A641 AR* CNTNAP1 NM_003632:exon18:c.2901_2902del 0.00016 homozyg. frameshift P967PfsX12 Axoglial 
K182 AR* CNTNAP1 NM_003632:exon19:c.3009_3010insT homozyg. frameshift F1003fs Axoglial 
K199 AR* CNTNAP1 NM_003632:exon19:c.3009_3010insT homozyg. frameshift F1003fs Axoglial 
B207 AR* CNTNAP1 NM_003632:intron18-exon19: c.2993-2_2994del homozyg. frameshift I999WfsX5 Axoglial 
A649 AR* ADCY6 NM_015270:exon20:c.C3346T homozyg. missense (d;1) R1116C Axoglial 
A635 AR ECEL1 NM_004826:exon4:c.925delA comp. het. frameshift K309fs Axoglial 
A635 AR ECEL1 NM_004826:exon2:c.C33G comp. het. stop gained Y11X Axoglial 
B192 AR* ECEL1 NM_004826:c.1685+1G>T homozyg. splice K552AfsX33 Axoglial 
B651 AD TRPV4 NM_021625:exon6:c.G947A heterozyg. missense (p) R316H Axoglial 
A638 AR* CHRNG NM_005199:exon2:c.117_118insC homozyg. frameshift P39fs NMJ 
A651 AR* CHRNG NM_005199:exon7:c.C715T homozyg. missense (p) R239C NMJ 
A662 AR* CHRNG NM_005199:exon7:c.C715T homozyg. missense (p) R239C NMJ 
A650 AR RAPSN NM_032645:exon2:c.C264A 0.0014 0.001 comp. het. missense (p) N88K NMJ 
A650 AR RAPSN NM_032645:EX1_2del comp. het. deletion Ex1_2del NMJ 
A640 AD MYH3 NM_002470:exon28:c.T3959C heterozyg. missense (d;0.99) L1320P Sk Mu 
A646 AR* NEB NM_001164508:exon41: c.4858_4866del homozyg. frameshift A1620fs Sk Mu 
A631 AR* NEB NM_001164508:c.9832-1G>A homozyg. splice Y3278MfsX22 Sk Mu 
B415 AR RYR1 NM_000540:EX70_71del comp. het. deletion EX70_71del Sk Mu 
B415 AR RYR1 NM_000540:exon8:c.G644A comp. het. missense (d;1) G215E Sk Mu 
A648 AR RYR1 NM_000540:exon46:c.G7373A comp. het. missense (p) R2458H Sk Mu 
A648 AR RYR1 NM_000540:c.14364+1G>A comp. het. splice W4768CfsX11 Sk Mu 
A636 AR RYR1 NM_000540:exon46:c.G7373A comp. het. missense (p) R2458H Sk Mu 
A636 AR RYR1 NM_000540:exon90:c.T12580C comp. het. missense (d;0.95) F4194L Sk Mu 
A656 AR* RYR1 NM_000540:exon26:c.G3449A homozyg. missense (d;1) C1150Y Sk Mu 
K166 AR* RYR1 NM_000540:exon57:c.C8758T homozyg. stop gained R2920X Sk Mu 
K168 AR* SYNE1 NM_182961:exon136:c.C24577T homozyg. stop gained R8193X Sk Mu 
A663 AD TNNT3 NM_006757:exon10:c.G188A heterozyg. missense (p) R63H Sk Mu 
A642 AR TTN NM_133378:exon195: c.37862_37863insA comp. het. frameshift Y12621_V12622 delinsX Sk Mu 
A642 AR TTN NM_133378:exon307:c.C96388T comp. het. stop gained R32130X Sk Mu 
A659 AR* RBBP8 NM_002894:c.2455-4T>G 0.00008 homozyg. splice Y819VfsX33 Other 
Code Inheritance Gene Transcript Ref.: nucleotide change MAF (dbSNP) MAF (EVS) Genotype Type Protein change Target 
A641 AR* CNTNAP1 NM_003632:exon18:c.2901_2902del 0.00016 homozyg. frameshift P967PfsX12 Axoglial 
K182 AR* CNTNAP1 NM_003632:exon19:c.3009_3010insT homozyg. frameshift F1003fs Axoglial 
K199 AR* CNTNAP1 NM_003632:exon19:c.3009_3010insT homozyg. frameshift F1003fs Axoglial 
B207 AR* CNTNAP1 NM_003632:intron18-exon19: c.2993-2_2994del homozyg. frameshift I999WfsX5 Axoglial 
A649 AR* ADCY6 NM_015270:exon20:c.C3346T homozyg. missense (d;1) R1116C Axoglial 
A635 AR ECEL1 NM_004826:exon4:c.925delA comp. het. frameshift K309fs Axoglial 
A635 AR ECEL1 NM_004826:exon2:c.C33G comp. het. stop gained Y11X Axoglial 
B192 AR* ECEL1 NM_004826:c.1685+1G>T homozyg. splice K552AfsX33 Axoglial 
B651 AD TRPV4 NM_021625:exon6:c.G947A heterozyg. missense (p) R316H Axoglial 
A638 AR* CHRNG NM_005199:exon2:c.117_118insC homozyg. frameshift P39fs NMJ 
A651 AR* CHRNG NM_005199:exon7:c.C715T homozyg. missense (p) R239C NMJ 
A662 AR* CHRNG NM_005199:exon7:c.C715T homozyg. missense (p) R239C NMJ 
A650 AR RAPSN NM_032645:exon2:c.C264A 0.0014 0.001 comp. het. missense (p) N88K NMJ 
A650 AR RAPSN NM_032645:EX1_2del comp. het. deletion Ex1_2del NMJ 
A640 AD MYH3 NM_002470:exon28:c.T3959C heterozyg. missense (d;0.99) L1320P Sk Mu 
A646 AR* NEB NM_001164508:exon41: c.4858_4866del homozyg. frameshift A1620fs Sk Mu 
A631 AR* NEB NM_001164508:c.9832-1G>A homozyg. splice Y3278MfsX22 Sk Mu 
B415 AR RYR1 NM_000540:EX70_71del comp. het. deletion EX70_71del Sk Mu 
B415 AR RYR1 NM_000540:exon8:c.G644A comp. het. missense (d;1) G215E Sk Mu 
A648 AR RYR1 NM_000540:exon46:c.G7373A comp. het. missense (p) R2458H Sk Mu 
A648 AR RYR1 NM_000540:c.14364+1G>A comp. het. splice W4768CfsX11 Sk Mu 
A636 AR RYR1 NM_000540:exon46:c.G7373A comp. het. missense (p) R2458H Sk Mu 
A636 AR RYR1 NM_000540:exon90:c.T12580C comp. het. missense (d;0.95) F4194L Sk Mu 
A656 AR* RYR1 NM_000540:exon26:c.G3449A homozyg. missense (d;1) C1150Y Sk Mu 
K166 AR* RYR1 NM_000540:exon57:c.C8758T homozyg. stop gained R2920X Sk Mu 
K168 AR* SYNE1 NM_182961:exon136:c.C24577T homozyg. stop gained R8193X Sk Mu 
A663 AD TNNT3 NM_006757:exon10:c.G188A heterozyg. missense (p) R63H Sk Mu 
A642 AR TTN NM_133378:exon195: c.37862_37863insA comp. het. frameshift Y12621_V12622 delinsX Sk Mu 
A642 AR TTN NM_133378:exon307:c.C96388T comp. het. stop gained R32130X Sk Mu 
A659 AR* RBBP8 NM_002894:c.2455-4T>G 0.00008 homozyg. splice Y819VfsX33 Other 

AR, autosomal recessive; AD, autosomal dominant; *, consanguinity; transcript ref., transcript reference; homozyg., homozygous; comp. het., compound heterozygous; heterozyg., heterozygous; Type, (p) = known pathogenic mutation, (d) = damaging mutation as determined by polyphen-2 prediction with score; NMJ, neuromuscular junction; Sk Mu, skeletal muscle. Mutations were confirmed by Sanger sequencing or real-time PCR (Supplementary Material, Fig. S3) and splice mutations by RNA analysis (Supplementary Material, Fig. S4).

Identification of mutations in CNTNAP1 encoding CASPR, an essential component of node of Ranvier domains

In two consanguineous families (K182, A641), homozygous frameshift mutations were found in CNTNAP1 encoding CASPR (Fig. 2 and Table 2). In family A641, the three fetuses born from consanguineous parents carry a homozygous 1 bp deletion in CNTNAP1 exon 18 (c.2901_2902del), leading to frameshift and premature stop codon (P967PfsX12). In family K182, affected patients carry a homozygous 1 bp insertion in CNTNAP1 exon 19 (c.3009_3010insT) leading to frameshift (F1003fs). Two additional families with a similar phenotype (see below) were found to carry deleterious mutations in the same gene using either the combination of both linkage analysis with WES (family K199) or homozygosity mapping only (B207). In the consanguineous family K199, affected patients carry the same homozygous 1 bp insertion in CNTNAP1 exon 19 (c.3009_3010insT) as the unrelated family K182. In family B207, a homozygous frameshift mutation in exon 19 of CNTNAP1 was found in the patient (intron18–exon19:c.2993-2_2994del, I999WfsX5). In families K182 and A641, both parents were heterozygous for the mutation as expected. In families K199 and B207, DNA samples from parents were not available. These mutations were absent in 95 ethnically matched controls. The c.2901_2902del mutation was found at a very low minor allele frequency (MAF: 0.00016) in the current Exome Variant Server database (EVS, ESP6500SI-V2). The other mutations were found in neither EVS nor dbSNPv138.

Figure 2.

Sanger sequencing of mutations identified in CNTNAP1 and ADCY6 in AMC families. Arrows indicate mutant nucleotide positions. Open symbols: unaffected; filled symbols: affected. The nucleotide and amino acid changes are indicated as well as the genotype (homozygous) in affected individuals.

Figure 2.

Sanger sequencing of mutations identified in CNTNAP1 and ADCY6 in AMC families. Arrows indicate mutant nucleotide positions. Open symbols: unaffected; filled symbols: affected. The nucleotide and amino acid changes are indicated as well as the genotype (homozygous) in affected individuals.

In the four families, the fetal phenotype was quite similar (Table 3) and characterized by a late onset during pregnancy (from 28 wg), polyhydramnios and distal joint contractures, including talipes equinovarus and both proximal and distal interphalangeal joint contractures of the hands (Fig. 3). Proximal joints were not involved. At birth, the patients displayed severe hypotonia, facial diplegia and a lack of swallowing, autonomous respiratory function and deep tendon reflexes (Fig. 3). Motor nerve conduction velocity was markedly reduced (<10 m/s). In all patients, death occurred within the first 2 months of life. Since CASPR is known to play a key role in the delineation of the axonal domains of myelinated axons in mice (9), transmission electron microscopy (TEM) of sciatic nerve in the A641 and K182 index cases was performed and this revealed two major abnormalities (Fig. 4). First, examination of longitudinal sections revealed a marked widening of the nodes of Ranvier in patient A641 (3.98 ± 0.56; n = 5) when compared with control (1.12 ± 0.98; n = 4, P < 0.05; Fig. 4). Second, in transverse sections, myelinated axons of K182 patient nerve (n = 50) displayed a significant reduced surface associated with thinner myelin sheath when compared with an age-matched control case (n = 69, Fig. 4, P = 0.04 and P < 0.0003, respectively, Kruskal–Wallis test). Similar abnormalities of myelinated axons were found in K199 and B207 patient's nerve (data not shown, available on request).

Table 3.

Clinical characteristics of patients carrying CNTNAP1 or ADCY6 mutations

 A641 K182 K199 B207 A649 
Pregnancy 
 Age of discovery (wg) 30 31 30 28 32 
 Distal AMC 
 Polyhydramnios – 
Birth 
 Hypotonia 
 Respiratory distress 
 Facial diplegia Unk 
 Swallowing defect 
 Areflexia 
 Distal AMC 
 Motor NCV (m/s.) 10 7.8 No response 
 Age of death (postnatal days) 10 33 40 10 80 
 A641 K182 K199 B207 A649 
Pregnancy 
 Age of discovery (wg) 30 31 30 28 32 
 Distal AMC 
 Polyhydramnios – 
Birth 
 Hypotonia 
 Respiratory distress 
 Facial diplegia Unk 
 Swallowing defect 
 Areflexia 
 Distal AMC 
 Motor NCV (m/s.) 10 7.8 No response 
 Age of death (postnatal days) 10 33 40 10 80 

wg, week of gestation; Y, yes; (–), no; unk, unknown.

Figure 3.

Clinical features of arthrogryposis associated with mutations of CNTNAP1. Note the distal involvement of joint retractions and facial diplegia.

Figure 3.

Clinical features of arthrogryposis associated with mutations of CNTNAP1. Note the distal involvement of joint retractions and facial diplegia.

Figure 4.

TEM analysis of nerve in AMC fetuses carrying deleterious mutations of CNTNAP1 (A and C) and ADCY6 (E). (A) Longitudinal section of sciatic nerve in AMC fetus A641 and control (B) of the same gestational age (32 wg). Note the marked widening of the node of Ranvier in AMC fetus (white arrow). (C) Transverse ultra-thin section showing the absence of large myelinated axons associated with thinner myelin sheaths in the AMC fetus K182 when compared with control (D) of the same gestational age (34 wg). (E) Lack of myelin surrounding axon (white arrow), while Schwann cell morphology appeared to be normal (black arrow) in AMC fetus A649 when compared with control of the same age (F, 40 wg). Scale bars: A and B; E and F: 3.8 µm; C and D: 20 µm.

Figure 4.

TEM analysis of nerve in AMC fetuses carrying deleterious mutations of CNTNAP1 (A and C) and ADCY6 (E). (A) Longitudinal section of sciatic nerve in AMC fetus A641 and control (B) of the same gestational age (32 wg). Note the marked widening of the node of Ranvier in AMC fetus (white arrow). (C) Transverse ultra-thin section showing the absence of large myelinated axons associated with thinner myelin sheaths in the AMC fetus K182 when compared with control (D) of the same gestational age (34 wg). (E) Lack of myelin surrounding axon (white arrow), while Schwann cell morphology appeared to be normal (black arrow) in AMC fetus A649 when compared with control of the same age (F, 40 wg). Scale bars: A and B; E and F: 3.8 µm; C and D: 20 µm.

Identification of mutations in ADCY6 reveals an essential and so far unknown role of ADCY6 in myelination of Schwann cells

In the consanguineous family A649, a homozygous missense mutation predicted to be damaging using Polyphen-2 (10) with a high score (1.00) was found in ADCY6 (c.C3346T, p. R1116C, Fig. 2 and Table 2). This mutation was found in neither EVS nor dbSNPv138. The phenotype of patients was similar to that carrying CNTNAP1 mutations (Table 3). Patient nerve immunohistochemistry using an S100 protein antibody revealed Schwann cells but all nerve fascicles were negative for MBP antibodies (Supplementary Material, Fig. S5). TEM of nerve revealed no myelinated axons (Fig. 4). Some redundant basal lamina of Schwann cells were also observed, suggesting inability of Schwann cells to properly myelinate axons. Antisense morpholino oligonucleotides and their corresponding ‘mismatch’ controls were designed to specifically knockdown the ADCY6 orthologs in zebrafish (adcy6a and adcy6b, Supplementary Material, Table S3). Knockdown of the ADCY6 orthologs in zebrafish led to a very similar phenotype as observed in patients (Fig. 5) with a loss of mbp expression in the peripheral nervous system (PNS, showed here is the Posterior Lateral Line nerve, n = 36/38) while the central nervous system mbp expression was comparable with controls. Furthermore, we also found no defects in Schwann cell migration and axonal growth in the morphants by using transgenic line foxd3::GFP (11) and acetylated tubulin antibody staining, respectively (Fig. 5).

Figure 5.

Adcy6a and b genes are essential for PNS myelination. (A, B, E, F, I and J) mbp mRNA expression by in situ hybridization in WT (A and B), embryos injected with the two control five mismatched morpholinos (MOs) (0.3 pmol 5 mis adcy6a MO + 0.3 pmol 5 mis adcy6b MO per embryo, E and F), and embryos injected with adcy6a and adcy6b MOs (0.3 pmol adcy6a MO + 0.3 pmol adcy6b MO per embryo, I and J), respectively. For convenience, we used adcy6 MO instead of adcy6a and b MOs. Note that mbp expression is severely reduced specifically in the PNS in embryos injected with both adcy6a and b MOs (compare I with A and E), while its expression is normal and comparable with controls in the CNS (compare J with B and F). Arrows in (A) and (E) indicate mbp expression along the posterior lateral line nerve (PLLn). Arrows in (B, F and J) indicate the expression of mbp in the hindbrain and spinal cord, while the arrowheads indicate its expression in the PLL ganglion (PLLg). Note the absence of mbp expression in the PLLg in J. (C, D, G, H, K and L) Schwann cells and PLLn labeling using the transgenic line foxd3::GFP and acetylated tubulin (ac Tub) antibody, respectively. Note the presence of Schwann cells and PLLn in the adcy6 morphants as observed in controls (compare K with C and G; L with D and H). Arrows in (C, G and K) indicate Schwann cells along the PLLn and in (D, H and L) indicate the PLLn. Scale bar: 200 μm in A, B, E, F, I and J; 100 μm in C, D, G, H, K and L.

Figure 5.

Adcy6a and b genes are essential for PNS myelination. (A, B, E, F, I and J) mbp mRNA expression by in situ hybridization in WT (A and B), embryos injected with the two control five mismatched morpholinos (MOs) (0.3 pmol 5 mis adcy6a MO + 0.3 pmol 5 mis adcy6b MO per embryo, E and F), and embryos injected with adcy6a and adcy6b MOs (0.3 pmol adcy6a MO + 0.3 pmol adcy6b MO per embryo, I and J), respectively. For convenience, we used adcy6 MO instead of adcy6a and b MOs. Note that mbp expression is severely reduced specifically in the PNS in embryos injected with both adcy6a and b MOs (compare I with A and E), while its expression is normal and comparable with controls in the CNS (compare J with B and F). Arrows in (A) and (E) indicate mbp expression along the posterior lateral line nerve (PLLn). Arrows in (B, F and J) indicate the expression of mbp in the hindbrain and spinal cord, while the arrowheads indicate its expression in the PLL ganglion (PLLg). Note the absence of mbp expression in the PLLg in J. (C, D, G, H, K and L) Schwann cells and PLLn labeling using the transgenic line foxd3::GFP and acetylated tubulin (ac Tub) antibody, respectively. Note the presence of Schwann cells and PLLn in the adcy6 morphants as observed in controls (compare K with C and G; L with D and H). Arrows in (C, G and K) indicate Schwann cells along the PLLn and in (D, H and L) indicate the PLLn. Scale bar: 200 μm in A, B, E, F, I and J; 100 μm in C, D, G, H, K and L.

DISCUSSION

To our knowledge, this is the first study of a large cohort of undiagnosed AMC individuals investigated by combining genetic mapping with WES. This strategy markedly improves the genetic diagnosis of AMC, since pathogenic mutations were detected in known AMC or NMD genes in 19 families (61%). Importantly, this approach allowed identifying mutations in CNTNAP1 and ADCY6 in five families revealing novel axoglial human diseases. The fetal phenotype was quite similar and mainly characterized by a late onset during pregnancy, polyhydramnios (in four out of five families), distal joint contractures and very severe motor paralysis at birth leading to death within the first 3 months of life. A causative role of mutations in the CNTNAP1 gene encoding CASPR was established by the identification of distinct homozygous frameshift mutations in four unrelated families. In addition, dramatic reduction of motor nerve conduction velocities associated with severe abnormalities of the nodes of Ranvier and myelinated axons was observed. The node of Ranvier, the flanking paranodal junctions and the juxtaparanodes underlie saltatory conduction of action potentials along the myelinated axons, an essential process for neuronal function. The paranodal junctions consist of a complex containing the axonal proteins Caspr (12) and contactin (13) and the glial isoform of neurofascin (14). Mice that lack Caspr exhibit severe motor paresis and most of the mutant mice die at weaning (P21, 9). In these mice, normal paranodal junctions fail to form leading to disruption of the paranodal loops. Our data show similar defects of myelinated axons, indicating a critical role for CASPR in the delineation of specific axonal domains and the axon–glia interactions required for normal saltatory conduction in both mouse and human.

A homozygous missense mutation in ADCY6 has been found in a consanguineous family. Ultrastructural morphology of patient nerve sample revealed the presence of Schwann cells but the lack of myelin in the PNS. Although mutation of this gene has been found in a single family, knock down of the orthologous genes in zebrafish provided evidence for major myelin defects in the PNS, data similar to those found in patients. These results strongly support an essential and so far unknown role of ADCY6 in this process. ADCY6 encodes a protein that belongs to adenylate cyclase family responsible for the synthesis of cAMP (15). During development, promyelinating Schwann cells associate with one segment of an axon and differentiate into myelinating Schwann cells to form the myelin sheath. Elevation of cAMP can mimic axonal contact in vitro and is thought to upregulate myelinating signals (16–18). The G protein-coupled receptor Gpr126 has been recently shown to be required in Schwann cells for myelination (19). Elevation of cAMP in gpr126 zebrafish mutants could restore myelination, suggesting that Gpr126 drives the differentiation of promyelinating Schwann cells by elevating cAMP levels (19). Our data suggest that ADCY6 is involved in the same Gpr126-cAMP pathway for myelination of Schwann cells.

Hence, our study extends our knowledge of the pathogenic mechanisms in AMC and indicates that Caspr and Adcy6 are essential for correct differentiation of peripheral myelinated axons in human. These genes may be regarded as strong candidates in undiagnosed peripheral neuropathies.

In spite of some pitfalls, most likely due to the lack of coverage of mutations by WES, parallel next-generation sequencing (NGS) targeting previously known AMC genes, and the identification of two novel genes in this study should now lead to a molecular diagnosis in >75% of AMC cases, since our study concentrated on undiagnosed AMC only, without previous integration of muscle or nerve morphological studies. Such targeted NGS approach should avoid the identification of clinically relevant mutations unrelated to AMC. This will be particularly useful for diagnosis purpose and appropriate management.

MATERIALS AND METHODS

Patients

All cases were evaluated by obstetricians, fetal pathologists, clinical geneticists, neonatologists or neuropediatricians. The main clinical features of the affected individuals are summarized in Table 1. The gestational age of AMC diagnosis was based on ultrasound examination. Any other defective organ was regarded as exclusion criteria. The parents of all patients provided written informed consents for genetic analysis of their children or fetuses and themselves in accordance with the ethical standards of our institutional review boards.

Methods

Genome-wide linkage analysis

Genomic DNA was isolated from blood or frozen tissue with the use of a QiaAmp DNA midi or mini kit, respectively (Qiagen). Whole genome SNP scanning was carried out according to the Affymetrix 250 K GeneChip Mapping Assay manual. Multipoint linkage analysis and homozygosity mapping of SNP data were performed using the Alohomora (20) and Merlin software (21) with the following parameters: autosomal recessive or dominant inheritance, 100% penetrance and disease gene frequency in the population of 1:1000.

WES

WES was performed using the DNA sample of one affected individual per family. The Illumina TruSeq DNA Sample Prep kit v1 and the NimbleGen SeqCap EZ Human Exome Library v2.0 (targeting 44 Mb, from A631 to A648 DNA samples) or v3.0 (targeting 64 Mb, from A649 to K199 DNA samples) were used for library preparation and exome enrichment, respectively, as previously described (22). Sequencing was performed on a Genome Analyzer IIx instrument using paired-end 75 bp reads and following the Illumina's protocol. The median coverage of the WES was 33–138 (average 77).

Reads were aligned to the human reference genome sequence (UCSC hg19, NCBI build 37.3) via the BWA program (23). Variants were selected using the SAMtools (24), then annotated using Annovar software (25). Reads with a coverage of at least 2× were filtered against dbSNPv131 database. Variants with a MAF <0.0015 and located in coding regions, intron–exon junctions or short coding insertions or deletions were selected. Then variants mapping to the candidate regions as determined by linkage analysis were selected. MAF was updated using dbSNPv138 database and EVS (ESP6500SI-V2). Integrative genomics viewer (IGV, v1.5.64, 8) was used as a visualization tool of WES variants. In silico prediction of the functional effect at the amino acid level was calculated using Polyphen-2 software (version 2.2.2, 10).

Real-time PCR amplification of genomic DNA

Real-time PCR amplification was conducted using genomic DNA on a 7300 real-time PCR system (Applied Biosystems) using a Power SYBR Green PCR master mix (Applied Biosystems). Albumin gene was used as an internal control. Genomic deletion was defined when the ratio of tested DNA to control DNA was <0.5. Real-time PCR amplification of each sample was performed in duplicate using primers within selected exons (Supplementary Material, Table S3).

Reverse transcription-PCR amplification

Total RNAs were extracted by a TRI Reagent LS method (Sigma). One microgram RNA was used to synthesize Complementary DNA (cDNA) by using random primers following the manufacturer's manual (SuperScript III reverse Transcriptase, Invitrogen) in a final volume of 20 µl. PCR amplification was carried out using 1.5 mm MgCl2, 0.6 U DNA polymerase (Invitrogen), 0.2 µm each primer and 1 µl cDNA. After an initial cycle of denaturation at 94°C for 5 min., 30 cycles were performed consisting in denaturation at 94°C for 30 s., annealing at 60°C for 1 min. and extension at 72°C for 1 min., followed by a final extension for 7 min. at 72°C, in a ABI9700 Thermal Cycler. RT-PCR products were separated by agarose gel electrophoresis and labelled with ethidium bromide. To determine the effect of mutations closed to intron–exon junctions, PCR amplification analysis from single-strand cDNA was performed using primers flanking exons (Supplementary Material, Table S3). As internal control for PCR amplification, β-Actin cDNA was coamplified (Supplementary Material, Table S3). Sanger sequencing was performed from the RT-PCR products.

Sanger sequencing

PCR primer pairs were designed from genomic DNA to amplify and sequence each variant (Supplementary Material, Table S3). PCR amplification was carried out using 1.5 mm MgCl2, 0.6 U DNA polymerase, 0.25 µm each primer and 50 ng DNA. After an initial cycle of denaturation at 94°C for 5 min, 30 cycles were performed consisting of denaturation at 95°C for 30 s., annealing at 60°C (±3°C) for 1 min and extension at 72°C for 1 min. and final extension 7 min. at 72°C, on an ABI9700 Thermal Cycler. PCR products were then purified on P100 columns (Bio-Gel P-2 Gel fine, Biorad) sequenced using the forward or reverse primers and the Big Dye Terminator V3.1 Cycle sequencing kit (Applied Biosystems). The sequencing reaction products were purified on G50 columns (Sephadex G-50 Superfine, GE Healthcare), and then migrated on an automated fluorescent DNA sequencer (ABI Prism 3100 Genetic analyzer, Applied Biosystems). The obtained DNA sequences were compared with published sequences (BLAST, NCBI). Variants fulfilling prioritization criteria were validated by Sanger sequencing. Sanger sequencing was also performed to establish the genotype of each family member, and to analyze the segregation of the variants within each family. A cohort of 95 ethnically matched controls was analyzed for the selected variants when they were not known to be pathogenic so far.

Morphology of the neuromuscular system

Muscle and nerve biopsy samples were processed as previously described (26). Nerve immunohistochemistry was carried out using a phosphorylated neurofilament monoclonal antibody (1:200), myelin basic protein (MBP, 1:200) and S100 protein polyclonal antibodies (1:2500, Dakopatts). Muscle and nerve ultrastructural studies were carried out according to the standardized protocols. Briefly, tissue samples were fixed in a 2% glutaraldehyde fixative solution, post-fixed with osmium tetroxide and embedded in resin epoxy. Semi-thin sections were stained with toluidine blue. Ultra-thin sections were contrasted with uranyl acetate and lead citrate, and examined under a PHILIPS CM10 transmission electron microscope.

Knockdown of ADCY6 in zebrafish

Embryos were cared for according to the standard protocols (27). Foxd3::GFP transgenic line was used here (11). Antisense morpholino oligonucleotides (adcy6 MOs) were purchased from Gene Tools LLC and designed against the two corresponding orthologous zebrafish genes: adcy6a and adcy6b. The adcy6a MO was designed to target the 5′ UTR of adcy6a mRNA (XM_001922714, Supplementary Material, Table S3). The corresponding control ‘mismatch’ morpholino (Supplementary Material, Table S3) had five nucleotides altered along its sequence. The same applies to adcy6b MO targeting adcy6b mRNA (XM_002666490, Supplementary Material, Table S3). 0.3 mm of each of the two morpholinos, respectively, designed against the two genes were mixed (either a mix of the 5′ UTR MOs or the controls ones; final concentration: 0.6 mm). Morpholino (1 nl) was injected into one to four cell-stage embryos as previously described (28).

In situ hybridization

Embryos were fixed in 4% paraformaldehyde and stained as whole mounts following standard in situ protocols and using mbp probe.

Immunohistochemistry

For immunostaining, embryos were fixed in 4% paraformaldehyde and stained as whole mounts. An anti-acetylated tubulin antibody (Sigma) was used at a 1:1000 dilution. Primary antibodies were detected with appropriate secondary antibodies conjugated to Alexa 568 (Molecular probes) at a 1:200 dilution.

Confocal image analysis

Image acquisition was performed using a Zeiss confocal microscope and Zeiss LSM imaging software. Image analysis was performed offline using ImageJ and Adobe photoshop CS6.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

FUNDING

This work was supported by a grant from the French Ministry of Health (PHRC 2010, AOM10181), the Association Française contre les Myopathies (AFM, DAJ1891), Inserm and the Alliance Arthrogrypose to J.M. and sponsored by Assistance Publique- Hôpitaux de Paris. L.F. is a recipient of the University Paris 11 scholarship. K.D. was a recipient of the Fédération pour la Recherche Médicale scholarship and was supported by a grant from the PHRC. The authors thank the Biomedical Institute of Bicêtre for providing them with Sanger sequencing facilities, the Clinical Research Unit of Bicêtre Hospital, the Direction of Clinical Research and Development (AP-HP) and the NHLBI GO Exome Sequencing Project and its ongoing studies that produced and provided exome variant calls for comparison.

ACKNOWLEDGEMENTS

We thank all families for participating in this study.

Conflict of Interest statement. None declared.

REFERENCES

1
Hall
J.G.
Genetic aspects of arthrogryposis
Clin. Orthop.
 , 
1985
, vol. 
194
 (pg. 
44
-
53
)
2
Fahy
M.J.
Hall
J.G.
A retrospective study of pregnancy complications among 828 cases of arthrogryposis
Genet. Couns.
 , 
1990
, vol. 
1
 (pg. 
3
-
11
)
3
Bamshad
M.
Van Heest
A.E.
Pleasure
D.
Arthrogryposis: a review and update
J. Bone Joint Surg. Am.
 , 
2009
, vol. 
91
 (pg. 
40
-
46
)
4
Navti
O.B.
Kinning
E.
Vasudevan
P.
Barrow
M.
Porter
H.
Howarth
E.
Konje
J.
Khare
M.
Review of perinatal management of arthrogryposis at a large UK teaching hospital serving a multiethnic population
Prenat. Diagn.
 , 
2010
, vol. 
30
 (pg. 
49
-
56
)
5
Dieterich
K.
Quijano-Roy
S.
Monnier
N.
Zhou
J.
Fauré
J.
Smirnow
D.A.
Carlier
R.
Laroche
C.
Marcorelles
P.
Mercier
S.
, et al.  . 
The neuronal endopeptidase ECEL1 is associated with a distinct form of recessive distal arthrogryposis
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
1483
-
1492
)
6
Attali
R.
Warwar
N.
Israel
A.
Gurt
I.
McNally
E.
Puckelwartz
M.
Glick
B.
Nevo
Y.
Ben-Neriah
Z.
Melki
J.
Mutation of SYNE-1, encoding an essential component of the nuclear lamina, is responsible for autosomal recessive arthrogryposis
Hum. Mol. Genet.
 , 
2009
, vol. 
18
 (pg. 
3462
-
3469
)
7
Sarici
D.
Akin
M.A.
Kara
A.
Doganay
S.
Kurtoglu
S.
Seckel syndrome accompanied by semilobar holoprosencephaly and arthrogryposis
Pediatr. Neurol.
 , 
2012
, vol. 
46
 (pg. 
189
-
191
)
8
Robinson
J.T.
Thorvaldsdóttir
H.
Winckler
W.
Guttman
M.
Lander
E.S.
Getz
G.
Mesirov
J.P.
Integrative Genomics Viewer
Nat. Biotechnol.
 , 
2011
, vol. 
29
 (pg. 
24
-
26
)
9
Bhat
M.A.
Rios
J.C.
Lu
Y.
Garcia-Fresco
G.P.
Ching
W.
St Martin
M.
Li
J.
Einheber
S.
Chesler
M.
Rosenbluth
J.
, et al.  . 
Axon–glia interactions and the domain organization of myelinated axons requires neurexin IV/Caspr/Paranodin
Neuron
 , 
2001
, vol. 
30
 (pg. 
369
-
383
)
10
Adzhubei
I.A.
Schmidt
S.
Peshkin
L.
Ramensky
V.E.
Gerasimova
A.
Bork
P.
Kondrashov
A.S.
Sunyaev
S.R.
A method and server for predicting damaging missense mutations
Nat. Methods
 , 
2010
, vol. 
7
 (pg. 
248
-
249
)
11
Gilmour
D.T.
Maischein
H.M.
Nüsslein-Volhard
C.
Migration and function of a glial subtype in the vertebrate peripheral nervous system
Neuron
 , 
2002
, vol. 
34
 (pg. 
577
-
588
)
12
Peles
E.
Nativ
M.
Lustig
M.
Grumet
M.
Schilling
J.
Martinez
R.
Plowman
G.D.
Schlessinger
J.
Identification of a novel contactin-associated transmembrane receptor with multiple domains implicated in protein-protein interactions
EMBO J.
 , 
1997
, vol. 
16
 (pg. 
978
-
988
)
13
Rios
J.C.
Melendez-Vasquez
C.V.
Einheber
S.
Lustig
M.
Grumet
M.
Hemperly
J.
Peles
E.
Salzer
J.L.
Contactin-associated protein (Caspr) and contactin form a complex that is targeted to the paranodal junctions during myelination
J. Neurosci.
 , 
2000
, vol. 
20
 (pg. 
8354
-
8364
)
14
Tait
S.
Gunn-Moore
F.
Collinson
J.M.
Huang
J.
Lubetzki
C.
Pedraza
L.
Sherman
D.L.
Colman
D.R.
Brophy
P.J.
An oligodendrocyte cell adhesion molecule at the site of assembly of the paranodal axo–glial junction
J. Cell Biol.
 , 
2000
, vol. 
150
 (pg. 
657
-
666
)
15
Edelhoff
S.
Villacres
E.C.
Storm
D.R.
Disteche
C.M.
Mapping of adenylyl cyclase genes type I, II, III, IV, V, and VI in mouse
Mamm. Genome
 , 
1995
, vol. 
6
 (pg. 
111
-
113
)
16
Morgan
L.
Jessen
K.R.
Mirsky
R.
The effects of cAMP on differentiation of cultured Schwann cells: progression from an early phenotype (04+) to a myelin phenotype (P0+, GFAP-, N-CAM-, NGF-receptor-) depends on growth inhibition
J. Cell Biol.
 , 
1991
, vol. 
112
 (pg. 
457
-
467
)
17
Monuki
E.S.
Weinmaster
G.
Kuhn
R.
Lemke
G.
SCIP: a glial POU domain gene regulated by cyclic AMP
Neuron
 , 
1989
, vol. 
3
 (pg. 
783
-
793
)
18
Scherer
S.S.
Wang
D.Y.
Kuhn
R.
Lemke
G.
Wrabetz
L.
Kamholz
J.
Axons regulate Schwann cell expression of the POU transcription factor SCIP
J. Neurosci.
 , 
1994
, vol. 
14
 (pg. 
1930
-
1942
)
19
Monk
K.R.
Naylor
S.G.
Glenn
T.D.
Mercurio
S.
Perlin
J.R.
Dominguez
C.
Moens
C.B.
Talbot
W.S.
A G protein-coupled receptor is essential for Schwann cells to initiate myelination
Science
 , 
2009
, vol. 
325
 (pg. 
1402
-
1405
)
20
Ruschendorf
F.
Nurnberg
P.
ALOHOMORA: a tool for linkage analysis using 10 K SNP array data
Bioinformatics
 , 
2005
, vol. 
21
 (pg. 
2123
-
2125
)
21
Abecasis
G.R.
Cherny
S.S.
Cookson
W.O.
Cardon
L.R.
Merlin-rapid analysis of dense genetic maps using sparse gene flow trees
Nat. Genet.
 , 
2002
, vol. 
30
 (pg. 
97
-
101
)
22
Zhou
J.
Tawk
M.
Tiziano
F.D.
Veillet
J.
Bayes
M.
Nolent
F.
Garcia
V.
Servidei
S.
Bertini
E.
Castro-Giner
F.
, et al.  . 
Spinal muscular atrophy associated with progressive myoclonic epilepsy is caused by mutations in ASAH1
Am. J. Hum. Genet.
 , 
2012
, vol. 
91
 (pg. 
5
-
14
)
23
Li
H.
Durbin
R.
Fast and accurate short read alignment with Burrows–Wheeler transform
Bioinformatics
 , 
2009
, vol. 
25
 (pg. 
1754
-
1760
(a)
24
Li
H.
Handsaker
B.
Wysoker
A.
Fennell
T.
Ruan
J.
Homer
N.
Marth
G.
Abecasis
G.
Durbin
R.
1000 Genome Project Data Processing Subgroup
The Sequence alignment/map format and SAMtools
Bioinformatics
 , 
2009
, vol. 
25
 (pg. 
2078
-
2079
(b)
25
Wang
K.
Li
M.
Hakonarson
H.
ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data
Nucleic Acids Res.
 , 
2010
, vol. 
38
 pg. 
e164
 
26
Dubowitz
V.
Sewry
C.A.
Fitzsimons
R.
Muscle biopsy: a practical approach
 , 
1985
2nd edn
London. Philadelphia
Baillière Tindall
27
Westerfield
M.
The zebrafish book. A guide for the laboratory use of zebrafish (Danio rerio)
1995
Eugene, OR
University of Oregon Press
28
Nasevicius
A.
Ekker
S.C.
Effective targeted gene ‘knockdown’ in zebrafish
Nat. Genet.
 , 
2000
, vol. 
26
 (pg. 
216
-
220
)

Author notes

Equal contribution.
Present address: UMR-986, Inserm and University Paris 11, 94276 Le Kremlin Bicêtre, France.

Supplementary data