RESEARCH ARTICLE

ECEL1 Mutation Causes Fetal Arthrogryposis Multiplex Congenita N. Dohrn,1* V.Q. Le,2 A. Petersen,3 P. Skovbo,4 I.S. Pedersen,2 A. Ernst,2 H. Krarup,2 and M.B. Petersen1,5 1

Department of Clinical Genetics, Aalborg University Hospital, Aalborg, Denmark

2

Section of Molecular Diagnostics, Clinical Biochemistry, Aalborg University Hospital, Aalborg, Denmark Department of Pathology, Aalborg University Hospital, Aalborg, Denmark

3 4

Department of Obstetrics and Gynecology, Aalborg University Hospital, Aalborg, Denmark

5

Department of Clinical Medicine, Aalborg University, Aalborg, Denmark

Manuscript Received: 15 May 2014; Manuscript Accepted: 28 January 2015

Arthrogryposis multiplex congenita (AMC) is a descriptor for the clinical finding of congenital fixation of multiple joints. We present a consanguineous healthy couple with two pregnancies described with AMC due to characteristic findings on ultrasonography of fixated knee extension and reduced fetal movement at the gestational age of 13 weeks þ 2 days and 12 weeks þ 4 days. Both pregnancies were terminated and postmortem examinations were performed. The postmortem examinations confirmed AMC and suggested a diagnosis of centronuclear myopathy (CNM) due to characteristic histological findings in muscle biopsies. Whole exome sequencing (WES) was performed on all four individuals and the outcome was filtered by application of multiple filtration parameters satisfying a recessive inheritance pattern. Only one gene, ECEL1, was predicted damaging and had previously been associated with neuromuscular disease or AMC. The variant found ECEL1 is a missense mutation in a highly conserved residue and was predicted pathogenic by prediction software. The finding expands the molecular basis of congenital contractures and the phenotypic spectrum of ECEL1 mutations. The histological pattern suggestive of CNM in the fetuses can expand the spectrum of genes causing CNM, as we propose that mutations in ECEL1 can cause CNM or a condition similar to this. Further investigation of this is needed and we advocate that future patients with similar clinical presentation or proven ECEL1 mutations are examined with muscle biopsy. Secondly, this study illustrates the great potential of the clinical application of WES in couples with recurrent abortions or stillborn neonates. Ó 2015 Wiley Periodicals, Inc.

Key words: fetus; arthrogryposis multiplex congenita; whole exome sequencing; ECEL1; central nuclei

INTRODUCTION Arthrogryposis Multiplex Congenita (AMC) Arthrogryposis multiplex congenita (AMC: OMIM 108110) is a descriptor for the clinical findings of a congenital joint fixation in at

Ó 2015 Wiley Periodicals, Inc.

How to Cite this Article: Dohrn N, Le VQ, Petersen A, Skovbo P, Pedersen IS, Ernst A, Krarup H, Petersen MB. 2015. ECEL1 mutation causes fetal arthrogryposis multiplex congenita. Am J Med Genet Part A 167A:731–743.

least two different body regions [Hall, 1997]. The term arthrogryposis is Greek and derives from the words “arthro” and “grypos” meaning “joint” and “hooked,” respectively. The occurrence of fixation of one or several joints is 1:200 of all live births, but the severity varies from isolated contractures (e.g., unilateral clubfoot) to implicating the entire body [Hall, 1997; Rink, 2011; Chen, 2012]. AMC is seen in 1:3,000 live births and with an equal gender ratio [Hall, 1997]. Distal arthrogryposis (DA) is a subgroup of AMC, originally delineated to be an autosomal dominant disorder with non-progressive, congenital contractures only affecting the distal joints (hands, feet, wrists, and ankles, elbows and knees) primarily as camptodactily and clubfoot and is not caused by a primary neurologic or muscle disease [Bamshad et al., 1996, 2009]. The complete spectrum of disorders causing AMC is heterogeneous and includes over 300 different entities. Regardless of the underlying disorder, an onset during fetal development with restriction of intrauterine movement and consequently, excess connective tissue deposition around the joints is always present [Hall, 1997]. The underlying etiologies consist of both genetic and Conflict of interest: none.  Correspondence to: Niclas Dohrn, Jagtvej 137, 4tv, DK-2200 Copenhagen N, Denmark. E-mail: [email protected] Article first published online in Wiley Online Library (wileyonlinelibrary.com): 23 February 2015 DOI 10.1002/ajmg.a.37018

731

732 environmental origins, but a typical subdivision into muscular diseases (congenital myopathies, muscular dystrophies, or mitochondrial abnormalities), neurological diseases (CNS or PNS), connective tissue disorders, fetal vascular compromise (secondary nerve and muscle damage), space limitations within the uterus (twin pregnancy, oligohydramnios, structural abnormalities of the uterus, fibroids, or other tumors), and maternal diseases (diabetes mellitus, multiple sclerosis, myasthenia gravis, infection, drugs, or trauma) is done [Hall, 1997; Polizzi et al., 2000; Kalampokas et al., 2012]. Prior to the great advances in ultrasonography (US) technology, AMC could not be recognized before birth. Today AMC is typically detected at routine US scans in second trimester [Navti et al., 2010; Kalampokas et al., 2012]. The characteristic findings on US consist of diminished fetal movement, abnormal fetal position, fixed flexion or extension deformities, dysmorphic features including micrognathia, altered amniotic fluid volume, limb deformities, a short umbilical cord, and growth retardation [Hyett et al., 1997; Lin et al., 2008; Navti et al., 2010; Dimitraki et al., 2011], but also in the late first and early second trimester findings of reduced fetal movement and increased nuchal translucency can be utilized to predict AMC [Hyett et al., 1997; Lin et al., 2008]. Today not only the diagnosis, but also family counseling concerning increased morbidity (physical handicaps and intellectual disability) and mortality is done prenatally [Witters et al., 2002], and parents can make an informed decision especially regarding termination of the pregnancy (TOP) [Navti et al., 2010]. Postmortem examination is recommended as it can confirm the prenatal findings and provide families with a specific diagnosis and recurrence risk in future pregnancies [Boyd et al., 2004].

AMERICAN JOURNAL OF MEDICAL GENETICS PART A hood or adolescence with hypotonia and weakness [Nicot et al., 2007; Claeys et al., 2010], while cases of CNMX have a more severe phenotype with neonatal presentation with marked weakness and hypotonia [Jungbluth et al., 2008; Romero, 2010; Romero and Bitoun, 2011] and occasionally fetal akinesia deformation sequence (FADS) with reduced fetal movements, AMC and respiratory insufficiency [Jungbluth et al., 2009; Romero, 2010; Romero and Bitoun, 2011; Lee et al., 2012]. Death usually occurs in infancy from respiratory failure and the families often have a history of miscarriages [Jungbluth et al., 2008]. Most cases follow the typical genotype–phenotype correlation, but a pronounced variation in clinical presentations exists [Bitoun et al., 2007; Jungbluth et al., 2009; Mejaddam et al., 2009; Bo¨hm et al., 2010; Hanisch et al., 2011].

CLINICAL REPORT A consanguineous couple originated from a small town in the Northern Sudan. The consanguinity is as illustrated in the pedigree (Fig. 1) with a mutual ancestor (I-1). The couple (IV-1 and IV-2) was healthy and denied any symptoms of systemic diseases, including myasthenia gravis or other immunologic disorders and they did not recall any hereditary diseases in close or distant relatives. The 24-year-old, first-time pregnant, woman (IV-2) attended our Prenatal Clinic in 2011 at the gestational age (GA) of 13 weeks þ 2 days (13 þ 2) for a routine nuchal scan with no reported events during the pregnancy. She was by conventional US (Voluson E8 Expert) shown to have an intrauterine fetus (V-1) with increased nuchal translucency and AMC. The description of AMC was based on findings of reduced fetal movements, fixated flexion in hips,

Centronuclear Myopathy (CNM) Centronuclear myopathy (CNM) is a rare group of congenital myopathies diagnosed by a characteristic histological appearance of myofibers with centralized nuclei, an absence of necrotic or regenerating fibers and clinical features of a congenital myopathy [Jungbluth et al., 2008; Biancalana et al., 2012]. This is in contrast to normal muscle fibers, where nuclei are at the periphery and organelles are in narrowly restricted locations. CNM is extremely rare with an estimated prevalence of about 1 in 400,000 in the United States [Amburgey et al., 2011]. CNM is a genetically and clinically heterogeneous disorder. The disorder can present sporadically or be inherited in an X-linked recessive (CNMX: OMIM 310400), autosomal dominant (CNM1: OMIM 160150), or autosomal recessive (CNM2: OMIM 255200) pattern. Currently, four genes are known to cause CNM. CNMX, also known as myotubular myopathy, is caused by mutations in the gene encoding myotubularin (MTM1: OMIM 300415), CNM1 is caused by mutations in the gene encoding dynamin 2 (DNM2: OMIM 602378), and finally CNM2 is caused by mutations in the genes encoding amphiphysin II (BIN1: OMIM 601248) or skeletal muscle ryanodine receptor (RYR1: OMIM 180901). However, a still large proportion (20–30%) of CNM patients are without any molecular genetic diagnosis [Dowling et al., 2008; Jungbluth et al., 2008; Romero, 2010; Biancalana et al., 2012]. The age of onset and severity of presentation differs between the different types. CNM1 and CNM2 will typically present in child-

FIG. 1. The family pedigree.

DOHRN ET AL.

733

FIG. 2. Ultrasonography of fetus V-2 showing AMC at GA 13 þ 4. Characteristic findings of extension in knees and flexion in upper limbs.

fixated extended knees, fixated flexed feet, crossed lower limbs and fixated flexion of the elbows. At a GA of 13 þ 4 a three-dimensional (3D) US anomaly scan was performed, showing a normal amount of amniotic fluid but unchanged reduced fetal movement and the

characteristically fixated extended knees bilaterally, confirming the findings of AMC in the initial scan (Fig. 2) (Table I). The couple was informed of the risks of mild to severe congenital physical handicaps associated with AMC and they applied for a TOP, which was permitted by an abortion committee. The TOP was performed at GA of 14 þ 2 with an intravaginal placement of 0.4 mg synthetic prostaglandin E1 (Cytotec1), and within 6 hr the fetus and placenta were delivered spontaneously. Subsequently, a vacuum evacuation of the uterus was performed. The postmortem examination of the fetus (V-1) including examination of the CNS and skeletal muscle was done. The examination revealed a female fetus, confirming the US findings of AMC with bilaterally fixated flexion in hips, bilaterally fixated extended knees, crossed lower extremities and webbing and fixated flexion contractures of the elbows. During the TOP the cranium and cerebrum were traumatized and only the cerebellum, brain stem, and spinal cord were available for examination. No abnormalities were discovered during the initial examination and no definite diagnosis was made (Table II). Six months later, the couple attended the Prenatal Clinic again for a routine nuchal scan of a new pregnancy at GA of 12 þ 4. The conventional US scan showed an intrauterine fetus (V-2) with increased nuchal translucency and AMC completely compatible with the first pregnancy. A 3D US scan at GA 12 þ 6 confirmed the former findings of AMC. The couple applied for TOP, which was

TABLE I. Ultrasound Findings FETUS 1 (V-1) Ultrasonograph Gestational age Heart activity Crown-rump length Nuchal translucency Amniotic fluid Fetal movement BPD Nose bone Cranium/brain Columna Heart Abdominal wall Gaster Kidneys/bladder Hands Feets Comments

GA at termination of pregnancy

Nuchal scan Conventional 13 þ 2 Yes 71.8 mm Increased—2.9 mm Reduced 26.1 mm Unclear Normal Normal No details Normal Visible Visible Visible Visible AMC—bilateral fixated extended knees, flexed hips, flexed feet and crossed lower limps. Fixated flexion of the elbows but normal hands

Malformation scan 3D 13 þ 4 Yes Normal Reduced

Confirmation of AMC and emphasis of bilateral fixated knee extension

14 þ 2

FETUS 2 (V-1) Nuchal scan Conventional 12 þ 4 Yes 61.6 mm Increased—3.3 mm Reduced 23.6 mm Unclear Uncertain Normal No details Normal Visible Visible Visible Visible AMC—bilateral fixated flexion in hips and extended knees. Bilateral fixated flexion in the elbows

Malformation scan 3D 12 þ 6 ND Normal Reduced

Confirmation of AMC and reports of possible abnormal head shape and increased intracraniel fluid 13 þ 4

734

AMERICAN JOURNAL OF MEDICAL GENETICS PART A

TABLE II. Postmortem Examination Gender Umbilical cord Weight Crown-heel Crown-rump Foot length Cranial circumference Cranium Face Nuchal thickness Upper extremities Lower extremities

Internal organs CNS Skeletal muscle

Fetus 1 (V-1) Female Short: 4.5 cm. 30.47 g 12.5 cm 8.8 cm 15 mm ND because of traumatization Traumatized by the termination procedure Small flat nose Undeterminable due to maceration of the back skin Fixated flexion and webbing in elbows. No contractures in shoulder, hand or finger joints. No poly- or syndactyli Bilateral fixated flexion in hips, bilateral fixated extension in knees, crossed lower limbs, fixated flexion in right foot, left foot missing due to traumatic calcaneal tendon biopsy. No poly- or syndactyli Normal Cerebrum traumatized and not available for examination. Cerebellum, brain stem and spinal cord normal Similar changes as for fetus 2, but to a lesser extent

approved and performed at GA 13 þ 4 with the same procedure described for the first fetus. The postmortem examination showed a female fetus (V-2) and confirmed AMC with findings similar to the first pregnancy (V-1) (Fig. 3). CNS examination was normal but the muscle examination showed a greater variation in fiber diameter ranging from normal to double size and a greater amount of centrally placed nuclei than expected for the gestational age (Fig. 4a) (Table II). Glycogen accumulation was found in the center of the muscle fiber by light as well as electron microscopy (Fig. 4b). By electron microscopy the perinuclear zone was shown to be free of myofibrils and besides the accumulation of glycogen also accumulation of autolyzed mitochondria was observed (Fig. 5). The muscle histology from the first fetus (V-1) was revised and found to show the same alterations as the second fetus (V-2), although less prominent (Table II). For both fetuses, a diagnosis of CNM with autosomal recessive inheritance was proposed.

METHODS An informed consent was obtained from both the parents (IV-1 and IV-2). Calcaneal tendon biopsies were taken from both fetuses (V-1 and V-2) and cell cultures established. Blood samples were obtained and stabilized in EDTA. DNA was extracted from

Fetus 2 (V-2) Female Short: 7.5 cm 28.83 g 11.7 cm 8.5 cm 11 mm 8.5 cm Normal Anteverted nostrils, low set ears Undeterminable due to maceration of the back skin Fixated flexion and webbing in elbows and shoulders. No contractures in hand or finger joints. No poly- or syndactyli Bilateral fixated flexion in hips, bilateral fixated extension in knees and crossed lower limbs. Undeterminable positioning of feet due calcaneal tendon biopsy. No poly- or syndactyli Normal Normal Variation in fiber diameter greater than expected. Almost all fibers with central nuclei. Accumulation of glycogen and condensation of desmin around nucleus

blood using the Maxwell 16 Blood DNA purification kit (Promega1).

Cytogenetic Methods An examination for quantitative chromosomal abnormalities as copy-number variations (CNVs) by genome-wide array-based comparative genomic hybridization 180K oligo array comparative genomic hybridization (CGH) analysis was done on the DNA from both fetuses. Karyotype analysis was subsequently done on the cell culture from the second fetus in 10 metaphases from fibroblasts in Q-banding in a quality of 400–450 bands per metaphase.

Molecular Genetics Methods We performed bidirectional Sanger sequencing of coding regions of genomic DNA with flanking intronic sequences of the BIN1 gene based on blood samples and cell cultures from the parents and the two fetuses, respectively. A test for congenital myotonic dystrophy was done on DNA from the second fetus (V-2), which investigated for a repeat expansion of the trinucleotide sequence (CTG) in the 30 -untranslated region of the DMPK gene with PCR fragment analysis.

DOHRN ET AL.

735

FIG. 5. Electron microscopy of muscle biopsy from Fetus V-2. The central zone free of myofilaments and with accumulation of glycogen (arrow) and autolyzed mitochondria (arrowhead).

FIG. 3. Postmortem examination of fetus V-2 showing AMC. a: Side view with extension contractures in knees and flexion contractures in the hips. b: Top view showing crossed lower limbs and flexion contractures in elbows.

Whole exome sequencing (WES) was performed on the parents and fetuses on DNA extracted from the blood samples and cell cultures, respectively. In brief, 1 mg of DNA was fragmented through acoustic sonication on a Covaris1 S2 system and enriched using Roche Nimblegen SeqCap EZ Exome v3 prior to 75 bp

FIG. 4. Light microscopy of muscle biopsy from fetus 2. a: Cross-section showing majority of muscle fibers with centrally placed nuclei and a great variation in the muscle fiber diameter. b: Longitudinal section with central nuclei and accumulation of glycogen (arrow) around the nucleus.

736 forward sequencing on SOLiD 5500xl (Life Technologies, Carlsbad, CA) resulting in 78–153 million raw reads for each library. Individual DNA libraries were sequenced in separate lanes two times and the raw read libraries were combined to get these numbers. After a sequence alignment step, and before variant detection, duplicated reads were detected and marked with Picard’s MarkDuplicates tool (http://picard.sourceforge.net/). Variant detection and prioritization were done with various software tools such as GATK, GATK-Haplotype [DePristo et al., 2011], Freebayes [Garrison and Marth, 2012], Lifescope software version 2.5.1 (Life Technologies), and SVS 8 (Golden Helix, Bozeman, MT). Variant effect prediction of the variants was queried from dbNSFP database version 2.4 [Liu et al., 2011, 2013] which stores pre-calculated prediction data generated with the following prediction softwares; SIFT, PolyPhen2, MutationTaster, MutationAssessor, FATHMM, and RadialSVM. The inheritance pattern (heterozygosity in the parents and homozygosity in the fetuses) was used in the prioritization process. The coverage estimation for the exomes and individual genes of interest was done with chanjo software (https:// github.com/madcecilb/chanjo) on BAM files generated by Lifescope and Picard.

RESULTS No CNVs were identified by array-CGH in both fetuses and investigation for congenital myotonic dystrophy in fetus V-2 was negative. The karyotype of fetus V-2 was normal (46,XX). Sanger sequencing of BIN1 on all four individuals did not find any pathogenic variants. The WES analysis was concluded with at least 10 coverage in 91%, 92%, 87%, and 93% of the exome of the father (IV-1), mother (IV-2), fetus V-1, and fetus V-2, respectively (Online material only —WES coverage). This resulted in between 131,000 and 145,000 variants in each sample. A premise of recessive inheritance was proposed due to the consanguinity in the family, the healthy condition of the parents and female gender of the fetuses. Initially we examined the WES results for pathogenic variants in 42 genes, which have previously been associated with either AMC or CNM in the literature (Online material only—WES coverage) but none were found. These genes were sequenced adequately with a coverage ranging from 24 to 72, and a completeness ranging from 66% to 100% (coverage of at least 10). This suggested that the causative variant was not to be found amongst these genes and led us to expand the search to the whole exome. To prioritize from the 208,752 variants detected in the four exomes with four variant callers, we applied multiple filtration steps (Fig. 6). Five variants in five different genes were predicted damaging by at least four prediction softwares. These five genes were investigated further for an association with neuromuscular diseases (NMD) or AMC in the relevant literature. Only one gene, ECEL1 (OMIM 605896), fulfilled these criteria as it recently had been established to cause AMC and NMD [Dieterich et al., 2013; McMillin et al., 2013; Shaaban et al., 2014; Shaheen et al., 2014; Barnett et al., 2014]. The previous reports on ECEL1-associated AMC all described the same characteristic finding of fixated knee extensions and ECEL1 has, additionally, been proved to have an imperative function on the formation of the neuromuscular junc-

AMERICAN JOURNAL OF MEDICAL GENETICS PART A

FIG. 6. The process of variant prioritization. The process started with a union of 208,752 variants identified from four subjects with four variant callers (Lifescope, GATK, GATK-haplotype, and Freebayes). A total of 1,328 variants followed a recessive inheritance pattern of homozygosity in the fetuses and heterozygosity in the parents. Three hundred ninety-one variants were found in exonic regions and none of them had stop-gain or frame-shift effect. One hundred ninety-six contained a nonsynonymous coding variant. Eighty-eight variants were predicted damaging or likely damaging by at least one prediction software. Five variants in five different genes were predicted damaging or likely damaging by at least four prediction softwares. Finally, only E CE L1 had previously been associated with neuromuscular diseases (NMD) or arthrogryposis multiplex congenita (AMC).

tions (NMJ) during the developmental stages of the nervous system in mice [Valdenaire et al., 2000; Nagata et al., 2006]. The ECEL1 variant, c.2023G>A, follows the suggested recessive inheritance pattern with herterozygosity in both of the parents and homozygosity in both of the fetuses. It is a novel missense mutation located in exon 15, resulting in p.Ala675Thr (Fig. 7) and six out of six prediction software classified the p.Ala675Thr variant as pathogenic and multi-protein sequence alignment around the variant showed that Ala675 is a phylogenetically highly conserved residue throughout multiple species (Fig. 8).

DISCUSSION Our main finding is a homozygous missense mutation in the ECEL1 gene in a family with two fetuses detected with AMC on US. To our knowledge, only 32 patients with confirmed ECEL1 mutations have been reported prior to this study (Table III). All of which showed a variety of DA diverging from the originally described criteria for DA by an autosomal recessive inheritance [Dieterich et al., 2013; McMillin et al., 2013; Shaaban et al., 2014; Shaheen et al., 2014; Barnett et al., 2014]. McMillin et al. [2013] designated the phenotype as distal arthrogryposis, type 5D (DA5D: OMIM 615065) due to the similarities with DA5 (OMIM 108145), but without the ocular abnormalities typically seen in DA5 patients. Even though not all previous studies use the terminology of DA5D, there is some consensus of phenotypic features of congenital extension contractures of the knees, camptodactyly of the hands and toes, wrist contractures, clubfoot, short neck, ptosis, round-shaped face, and a bulbous nose. However, some interfamilial variation in phenotype

DOHRN ET AL.

FIG. 7. Previously reported E CE L1 mutations. Exon 11 contains a LNAYY motif involved in substrate orientation; exon 13 contains the HExxH zinc binding motif and exon 15 contains the GExxxD zinc coordinating motif [Valdenaire et al., 1999]. LNAYY: motif involved in the orientation of the substrate peptide bond to be cleaved. HExxH: zinc binding motif. GExxxD: zinc coordinating motif.

FIG. 8. Multi-protein sequence alignment around the variant comparing human ECEL1 with ECEL1 orthologues showed a phylogenetically highly conserved residue throughout different species. The marked row depicts the conserved alanine residue at position 675 of human ECEL1.

737

Shaaban et al. [2014]

Dieterich et al. [2013]

Author McMillin et al. [2013]

DA: extension contractures of knees; camptodactyly, scoliosis, ophthalmoplegia, and astigmatism

Phenotype DA5D: extension contractures of the knee; camptodactyly of the hands, including adducted thumbs and wrists; mild camptodactyly of the toes; clubfoot and a calcaneovalgus deformity; distinctive facial features, including unilateral ptosis, a roundshaped face, arched eyebrows, a bulbous upturned nose, and micrognathia. No ophtalmoplegia DA: extension contractures the knees; congenital hip dislocation flexion contractures of fingers III–V; talus, talus valgus or varus; short neck; diminished muscle mass and central tongue atrophy; no pterygia Malian

Malian

Malian Malian Belgium Belgium Martinique

Turkey Morocco

Morocco

Lebanon Turkish

Turkish

F1P1

F1P2

F1P3

F1P4

F2P1

F2P2 F3P1

F4P1 F5P1

F5P2

F6P1 F1P1

F1P2

F4P1 F5P1

F3P2

F3P1

F2P2

F1P2 F1P3 F2P1

Origin Indian and East European Indian Indian European American European American European American European American Indian European American

Patient F1P1

AR AR

AR

A c.1819G>A

13

I. 4 13

I. 10

8 4 I 10

c.1470G>A c.874delG c.1685þ1G>T

c.1685þ1G>T

10 5

10



p.Gly607Ser

p.Try490 p.Val292Cysfs 51 p.Lys552AlafsX33 and p.Asp559AlafsX33 p.Lys552AlafsX33 and p.Asp559AlafsX33 p.Asp559AlafsX33 p.Gly607Ser

p.Ser550 p.Arg333



p.Ser550

ND

p.Cys760Arg

18 ND

p.Cys760Arg

p.Cys760Arg

Protein p.Tyr239 p.Asn115_Ala118 del p.Tyr239 p.Tyr239 p.Tyr290Cys p.Asp266-Glyfs 15 p.Tyr290Cys p.Asp266-Glyfs 15 p.Arg418Ser p.Arg418Ser p.Arg395Gln p.Gly197Asp p.Arg418Cys

18

18

2 2 4 3 4 3 7 I6 7 I6 6 2 7

Exon 2 2

c.1649C>G c.997C>T

c.1649C>G

ND

c.2278C>T

c.2278C>T

c.2278C>T

c.716dupA c.716dupA c.869A>G c.797_801delins GCT c.869A>G c.797_801delins GCT c.1252C>A c.1184þ3A>T c.1252C>A c.1184þ3A>T c.1184G>A c.590G>A c.1252C>T

Mutation c.716dupA c.344_355del

Missense

Splice site Missense

Splice site

Nonsense Frameshift, PT Splice site

Nonsense Nonsense

Nonsense

ND

Missense

Missense

Missense

Missense Missense Missense

Missense

Frameshift, PT Frameshift, PT Missense Frameshift Missense Frameshift Missense

Consequence Frameshift, PT Deletion

738 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

Phenotype DA5: extension contractures of the knees; limited extension contractures of the fingers and toes with flexion deformity; scoliosis, camptodactyly of the hands; adducted thumbs, mild adducted wrist; limited hip extension; calcaneovalgus; ptosis and strabismus DA5D: extension contractures of the hips, dislocatable hips, extension contractures of knees, adducted thumbs, stiff wrists and camptodactyly, clubfoot, bilateral ptosis, delayed motor milestones, multiple pterygia Saudi Saudi

F3P2

F3P3

Caucasian Australian and Filipino

Upper Egypt Saudi

F2P2 F3P1

F1P2

Saudi Arabia Upper Egypt

F1P4 F2P1

Caucasian Australian and Filipino

Saudi Arabia Saudi Arabia

F1P2 F1P3

F1P1

Origin Saudi Arabia

Patient F1P1

AR AR

A c.1797-1G>A

c.1210C>T

c.1210C>T

c.1057dupC c.1210C>T

c.1221_1223dup c.1057dupC

c.1221_1223dup c.1221_1223dup

Mutation c.1221_1223dup

9 I. 12

9 I. 12

7

7

5 7

7 5

7 7

Exon 7

Missense

Missense Splice site

p.Gly511Ser -

Missense

Missense

Truncation Missense

In frame dub Truncation

In frame dub In frame dub

Consequence In frame dub

p.Gly511Ser Splice site -

p.Arg404Cys

p.Arg404Cys

p.Arg404Cys

-

-

Protein -

F, family; P, patient; ND, no data available; AR, autosomal recessive; RFM, reduced fetal movement; NFM, normal fetal movement; BAT, born at term; PT, premature truncation of protein; BW, birth weight; IUGR, intrauterine growth retardation; C-section, caesarian section; GA, gestational age; I, intron; AMC, arthrogryposis multiplexa congenita.

Barnett et al. [2014]

Author Shaheen et al. [2014]

Age at onset Birth

DOHRN ET AL. 739

740 has been described, such as micrognathia [McMillin et al., 2013], tongue atrophy and hip dislocation [Dieterich et al., 2013], ophthalmoplegia and astigmatism [Shaaban et al., 2014], and a short stature [Shaheen et al., 2014]. The ECEL1 variant found in this study has not been reported in any of the current publicly available databases. The strength of this observation is limited by the fact that these databases do not include Sudanese chromosomes. However, based on the high value of predicted pathogenicity in conjunction with the previous reports with similar phenotypes resulting from different variants in ECEL1, the causality of the reported variant is highly probable. Further studies with findings of additional families with identical clinical characteristics and identical variant in ECEL1 or alternatively, findings of identical clinical characteristics and a different variant in ECEL1 are needed to finally determine the variant in ECEL1 as causal.

Severe Phenotype Seemingly, the AMC in Fetuses V-1 and V-2 were more comprehensive than typically described in DA patients as more proximally joints were affected. However, as both pregnancies were terminated at an early GA it is impossible to determine if the fetuses by term or in infancy would have had the same characteristic features as the previous ECEL1 patients (e.g., round shaped face, unilateral ptosis, bulbous upturned nose, etc.). We speculate that the phenotype of the fetuses in our family is more severe than the previously reported ECEL1 patients based on the early GA at the discovery of AMC and increased nuchal translucency. Studies have shown that fetuses diagnosed with AMC at an early GA have a higher mortality rate, particularly if immobility is evident [Pakkasja¨rvi et al., 2006] and that the GA at onset of immobility can predict the severity of future contractures [Rink, 2011]. We only have sparse knowledge on the prenatal findings from the previous ECEL1 patients but only one was reported with prenatal AMC [Barnett et al., 2014], none were reported with increased nuchal translucency, only 5/32 were reported with reduced fetal movements (i.e., a lesser extent of prenatal involvement), and 27/32 were reported with congenital DA (Table III). The pathogenesis of congenital DA always includes reduced fetal movement, and consequently, we speculate that a greater proportion of the patients were affected prenatally with reduced fetal movement than actually reported. Four patients were described with fixed knee extension on US in the third trimester, but without further reduction in fetal movement, any mentioning of AMC, and at a significantly older GA [Dieterich et al., 2013] than the Fetuses V-1 and V-2 (Table III).

The ECEL1 Protein The ECEL1 gene encodes for endothelin-converting enzyme-like 1 (ECEL1) and consists of 18 exons [Valdenaire et al., 2000] (Fig. 7). ECEL1 was identified as a type II integral transmembrane zinc metalloprotease, and a member of the neprilysin family (M13) [Valdenaire et al., 1999; Kiryu-seo et al., 2000]. Sequence analysis predicted that the protein consists of 775 amino acids with a short N-terminal cytoplasmic domain of 59 residues and a large extracellular domain that contains a characteristic zinc-binding motif of HEXXH. The motif contains two ZINC-binding His residues, and

AMERICAN JOURNAL OF MEDICAL GENETICS PART A a Glu that has a catalytic role. There is also a more C-terminal Glu residue that works as the third binding-site of the zinc atom [Valdenaire et al., 1999]. ECEL1 is considered to have enzymatic activity such as processing and degradation of neuropeptides and peptide hormones, which are crucial mediators of interneuronal signaling, however, the exact substrate of ECEL1 is unknown. ECEL1 plays an important role in nerve regeneration [Kiryu-seo et al., 2000] and embryonic neural and subsequently NMJ development during fetal life in mice and human [Schweizer et al., 1999; Valdenaire et al., 1999, 2000; Kiryu-seo et al., 2000; Nagata et al., 2006, 2010]. Mice deficient of the homologous gene, Ecel1, die of respiratory failure immediately after birth [Schweizer et al., 1999; Nagata et al., 2010]. Interestingly, these mice do not exhibit any macroscopic abnormalities [Schweizer et al., 1999; Nagata et al., 2010], but histochemical analysis demonstrates a significant decrease of peripheral nerve terminal arborization resulting in deprived formation of NMJs and secondary muscle atrophy of the muscle, including the diaphragm [Nagata et al., 2010]. The expression of ECEL1 and Ecel1 is mainly limited to neurons within CNS and PNS in both humans and rodents [Kiryu-seo et al., 2000; Nagata et al., 2006, 2010; Dieterich et al., 2013], and to a diminutive extent in human skeletal muscle [Valdenaire et al., 1999] and human fetal skeletal muscle [Dieterich et al., 2013]. Dieterich et al. [2013] found that ECEL1 expression was present in both the CNS and skeletal muscle during embryogenesis in humans, and that it was remarkably increased in fetal muscle compared to human adult muscle. It was hypothesized that the difference between mice and human expression patterns could explain the highly diverse phenotypic presentation seen between the two species. The mechanism by which ECEL1 dysfunction affects axonal arborization in skeletal muscle and the exact localization of ECEL1 expression in humans at different stages of life remains unknown.

Impact of Mutation on the Protein Our suggestion of a more severely affected family led us to hypothesize that the mutation could have a more destructive impact on protein structure and functioning compared to the previously reported ECEL1 mutations. As illustrated in Figure 7, the c.2023G>A mutation is found in exon 15, which encodes the zinc-coordinating motif (GExxxD), and, hence, involved in the catalytic function. The resulting amino acid change, p.Ala675Thr, is located within a highly conserved stretch of 12 residues, ENI ADMGGLKLA, where “ A” marks the mutated Alanine (Fig. 8). The stretch is the last motif of the signature region of the Neprilysin metalloprotease (M13) family consisting of ECE1, ECEL1, etc. ECE1 is a close homologue to ECEL1 and the threedimensional structure of ECE1 has previously been studied [Schulz et al., 2009]. The signature can be found on InterPro database with ID IPR000718 and on PRINTS-S [Attwood et al., 2003] database version 20 with ID PR00786. By using PROSITE [Sigrist et al., 2013] we identified the amino acid sequence “NITD”—where “T” is introduced as a result of the substitution of Alanine to Threonine. This sequence is a potential site for N-glycolysation, and, consequently, susceptible for post-translational modification. All these characteristics strongly suggest that the mutation may

DOHRN ET AL. interfere with proper functioning of ECEL1 protease activity and folding of the protein. Comprehensive three-dimensional protein structure analysis is, however, required in order to investigate the impact of the mutation on ECEL1 protein structure more accurately. Several ECEL1 mutations with nonsense, frame-shift or splice site mutations resulting in a truncated protein deprived of the catalytic site or nonsense-mediated mRNA decay leading to a lack of ECEL1 protein, have been reported in the literature. Interestingly, none of which had any clear signs of increased clinical severity compared to the cases without protein truncation (Table III) [Dieterich et al., 2013; McMillin et al., 2013; Shaheen et al., 2014]. Until this report, phenotypic variability amongst human patients has been sparse and limited to variability related to different subtypes of DA and without establishment of a clear genotype–phenotype correlation. Coexisting mutations in other genes than ECEL1 could, hypothetically, be implicated in the increased severity of the phenotype. The WES analysis found four other genes incl. ITPR3 [Schabhu¨ttl et al., 2014], which was predicted damaging by prediction software (Online material only—Variants predicted highly damaging). Although none of these genes are currently established to cause a NMD or AMC, they might supplement ECEL1 in the pathogenesis.

Centronuclear Myopathy Our second main finding was the atypical histopathological pattern of CNM in both fetuses that indicated a primary muscle disorder. None of the prior ECEL1 patients have been diagnosed with CNM or any other myopathy; however only in few of the patients (6/32) a muscle biopsy with histological examination was performed. Moreover, only two of these were taken from affected distal muscles. Fibrous tissue was found in biopsies from the affected muscles [Shaaban et al., 2014] and type 1-predominance, fiber size variation and altered lipid storage were found in the biopsies from unaffected muscle with neither major structural abnormalities of the sarcolemma or signs of denervation (Table III) [Dieterich et al., 2013]. Noticeably, none of the biopsies were normal but neither had central nuclei nor other specific abnormalities. The literature has no specific previous reports of fetuses diagnosed with CNM, but is limited to reports of neonates diagnosed with CNM and prenatal US findings of AMC, and consequently, it is uncertain at what point the pathogenesis initiates. During the fetal myogenesis, centralized nuclei are a normal finding in myotubes (i.e., a progenitor to myofibers). CNM was, in fact, originally named myotubular myopathy by Spiro et al. [1966] due to the histological resemblance of fetal myotubes with physiological central nuclei and a muscle biopsy from a 12-year-old boy with CNM. It was postulated that the finding of myotube-like fibers indicated an arrest in development of the muscle at a cellular level before the 20th week, but recent evidence suggests that CNM evolves after myogenesis is completed [Dowling et al., 2008]. It is important to emphasize that the current knowledge on human fetal physiology and gene expression at different developmental stages is very sparse and that more work is needed on the subject. Even though centralized nuclei are considered as normal findings in fetuses of this GA, the proportion

741 of centralized nuclei was significantly increased in both fetuses compared to normal fetuses of the same GA. Despite differences in the CNM phenotype, all known CNM genes result in generally the same histopathological pattern, due to abnormalities in proteins implicated in the membrane trafficking machinery or membrane remodeling (MTM1, DNM2, and BIN1) or the excitation–contraction coupling (RYR1) intrinsically in the myofiber [Dowling et al., 2008; Jungbluth et al., 2009; Romero and Bitoun, 2011]. ECEL1 is predominantly expressed in neural tissue [Valdenaire et al., 2000], thus, having little implication in the intrinsic function of muscular tissue. This implies that the fetuses presented here have a neurological rather than muscular disorder, which was initially implicated by the findings of muscle histopathology. It is plausible that a failure of NMJ formation due to the diminished arborization of the neurons, as seen in mice studies, can result in reduced maturation of the muscles (e.g., a lack of trophic factor from the motor nerve) and thus mimicking the histopatholgical appearance of CNM. Innervation may play a role in the maturation and orientation of the sarcoplasmic reticulum-transverse tubule system [Ambler et al., 1984]. On the contrary, the findings by Dieterich et al. [2013] with increased expression of ECEL1 in muscular tissue during embryogenesis compared to adult muscle could imply that ECEL1 has an important function during the fetal stage, and thereby a possibility of a primary muscular disorder. These hypotheses are, however, challenged by the six previous ECEL1 patients with muscle biopsy without reports of specific histopathology or sign of immaturities in muscles. Based on our findings, we advocate that future patients with similar clinical presentation or proven ECEL1 mutations are examined with muscle biopsies of affected muscle. Ultimately, our study illustrates the excessive potential of clinical application of WES in couples with recurrent abortions or stillborn neonates in accordance with current literature [Filges et al., 2014]. Prenatal genetic diagnosis may be offered to the family regarding the ECEL1 mutation as a consequence of our findings.

CONCLUSION With WES we found a homozygous missense mutation in exon 15 of the ECEL1 gene in two fetuses described with AMC. Based on previous reports on similar phenotypes and a high value of predicted pathogenicity we conclude that the ECEL1 gene is highly probable to be the causative gene in the presented family. However, further studies with findings of additional families with the identical clinical characteristics and possible pathogenic variants in ECEL1 are needed to finally determine the mutation in ECEL1 as causal. We propose the possibility of an expansion of the molecular basis and the phenotypic spectrum of the ECEL1 associated congenital contracture syndromes. Our supplementary finding of a histopathological pattern suggestive of CNM in the fetuses can possibly expand the spectrum of genes causing CNM as we propose that mutations in ECEL1 can be causative of CNM or a similar condition. Further investigation is, however, needed and we advocate that future patients with similar clinical presentation or proven ECEL1 mutations are examined with muscle biopsy of affected muscle.

742

ACKNOWLEDGMENTS The authors would like to thank the family described in the publication for their participation and Christian Gilissen for his excellent advice in navigating the data analysis.

REFERENCES Ambler MW, Neave C, Singer DB. 1984. X-linked recessive myotubular myopathy: II. Muscle morphology and human myogenesis. Hum Pathol 15:1107–1120. Amburgey K, McNamara N, Bennett LR, McCormick ME, Acsadi G, Dowling JJ. 2011. Prevalence of congenital myopathies in a representative pediatric united states population. Ann Neurol 70:662–665. Attwood TK, Bradley P, Flower DR, Goulton A, Maudling N, Mitchell AL, Moulton G, Nordle A, Paine K, Taulor P, Uddin A, Zygoury C. 2003. PRINTS and its automatic supplement, prePRINTS. Nucleic Acids Res 31:400–402. Bamshad M, Jorde LB, Carey JC. 1996. A revised and extended classification of the distal arthrogryposes. Am J Med Genet 65:277–281. Bamshad M, Van Heest AE, Pleasure D. 2009. Arthrogryposis: A review and update. J Bone Joint Surg Am 91:40–46. Barnett CP, Todd EJ, Ong R, Davis MR, Atkinson V, Allcock R, Laing N, Ravenscroft G. 2014. Distal arthrogryposis type 5D with novel clinical features and compound heterozygous mutations in ECEL1. Am J Med Genet Part A 164:1846–1849. Biancalana V, Beggs AH, Das S, Jungbluth H, Kress W, Nishino I, North K, Romero NB, Laporte J. 2012. Clinical utility gene card for: Centronuclear and myotubular myopathies. Eur J Hum Genet 20, doi: 10.1038/ ejhg.2012.91 Bitoun M, Bevilacqua Ja, Prudhon B, Maugenre S, Taratuto AL, Monges S, Lubieniecki F, Cances C, Uro-Coste E, Mayer M, Fardeau M, Romero NB, Guicheney P. 2007. Dynamin 2 mutations cause sporadic centronuclear myopathy with neonatal onset. Ann Neurol 62:666–670. Bo¨hm J, Yis¸ U, Ortac¸ R, Cakmakc¸ı H, Kurul SH, Dirik E, Laporte J. 2010. Case report of intrafamilial variability in autosomal recessive centronuclear myopathy associated to a novel BIN1 stop mutation. Orphanet J Rare Dis 5:35. Boyd Pa, Tondi F, Hicks NR, Chamberlain PF. 2004. Autopsy after termination of pregnancy for fetal anomaly: Retrospective cohort study. BMJ 328:137. Chen C-P. 2012. Prenatal diagnosis and genetic analysis of fetal akinesia deformation sequence and multiple pterygium syndrome associated with neuromuscular junction disorders: A review. Taiwan J Obstet Gynecol 51:12–117. Claeys K, Maisonobe T, Bo¨hm J, Laporte J, Hezode M, Romero N, Brochier G, Bitoun M, Carlier RY, Stojkovic T. 2010. Phenotype of a patient with recessive centronuclear myopathy and a novel BIN1 mutation. Neurology 74:519–523. DePristo Ma, Banks E, Poplin R, Garimella KV, Maguire JR, Hartl C, Philippakis Aa, del Angel G, Rivas Ma, Hanna M, McKenna A, Fennell TJ, Kernytsky AM, Sivachenko AY, Cibulskis K, Gabriel SB, Altshuler D, Daly MJ. 2011. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 43:491–498. Dieterich K, Quijano-Roy S, Monnier N, Zhou J, Faure´ J, Smirnow DA, Carlier R, Laroche C, Marcorelles P, Mercier S, Me´garbane´ A, Odent S, Romero N, Sternberg D, Marty I, Estournet B, Jouk P-S, Melki J, Lunardi J. 2013. The neuronal endopeptidase ECEL1 is associated with a distinct form of recessive distal arthrogryposis. Hum Mol Genet 22:1483–1492.

AMERICAN JOURNAL OF MEDICAL GENETICS PART A Dimitraki M, Tsikouras P, Bouchlariotou S, Dafopoulos A, Konstantou E, Liberis V. 2011. Prenatal assessment of arthrogryposis. A review of the literature. J Matern Fetal Neonatal Med 24:32–36. Dowling JJ, Gibbs EM, Feldman EL. 2008. Membrane traffic and muscle: Lessons from human disease. Traffic 9:1035–1043. Filges I, Nosova E, Bruder E, Tercanli S, Townsend K, Gibson WT, Ro¨thlisberger B, Heinimann K, Hall JG, Gregory-Evans CY, Wasserman WW, Miny P, Friedman JM. 2014. Exome sequencing identifies mutations in KIF14 as a novel cause of an autosomal recessive lethal fetal ciliopathy phenotype. Clin Genet 86:220–228. Garrison E, Marth G. 2012. Haplotype-based variant detection from shortread sequencing. arXiv Prepr. arXiv1207.3907:1–9. Hall JG. 1997. Arthrogryposis Mulitplex Congenita: Etiology, genetics, classification, diagnostic approach and general aspects. J. Pediatr. Orthopeadics Part B 6:159–166. Hanisch F, Mu¨ller T, Dietz A, Bitoun M, Kress W, Weis J, Stoltenburg G, Zierz S. 2011. Phenotype variability and histopathological findings in centronuclear myopathy due to DNM2 mutations. J Neurol 258:1085– 1090. Hyett J, Noble P, Sebire NJ, Snijders R, Nicolaides KH. 1997. Lethal congenital arthrogryposis presents with increased nuchal translucency at 10-14 weeks of gestation. Ultrasound Obstet Gynecol 9:310–313. Jungbluth H, Wallgren-Pettersson C, Laporte J. 2008. Centronuclear (myotubular) myopathy. Orphanet J Rare Dis 3:26. Jungbluth H, Wallgren-Pettersson C, Laporte JF. 2009. 164th ENMC International workshop: 6th workshop on centronuclear (myotubular) myopathies, 16–18 January 2009, Naarden, The Netherlands. Neuromuscul Disord 19:721–729. Kalampokas E, Kalampokas T, Sofoudis C, Deligeoroglou E, Botsis D. 2012. Diagnosing arthrogryposis multiplex congenita: A review. ISRN Obstet Gynecol 2012:264918. doi: 10.5402/2012/264918. Epub 2012 Sep 23. Kiryu-seo S, Sasaki M, Yokohama H, Nakagomi S, Hirayama T, Aoki S. 2000. Damage-induced neuronal endopeptidase (DINE) is a unique metallopeptidase expressed in response to neuronal damage and activates superoxide scavengers. Proc Natl Acad Sci 97:4345–4350. Lee I-C, Su P-H, Chen J-Y, Hu J-M, Lu J-J, Ng Y-Y. 2012. Congenital myotubular myopathy with a novel MTM1 gene mutation in a premature infant presenting with ventilator dependency and intrahepatic cholestasis. J Child Neurol 27:99–104. Lin I, Chueh H, Chang S, Cheng P. 2008. The application of threedimensional ultrasonography in the prenatal diagnosis of arthrogryposis. Taiwan J Obs Gynecol 47:75–78. Liu X, Jian X, Boerwinkle E. 2011. dbNSFP: a lightweight database of human nonsynonymous SNPs and their functional predictions. Hum. Mutat. 32:894–9. Liu X, Jian X, Boerwinkle E. 2013. dbNSFP v2.0: a database of human nonsynonymous SNVs and their functional predictions and annotations. Hum. Mutat. 34:E2393–402. McMillin MJ, Below JE, Shively KM, Beck AE, Gildersleeve HI, Pinner J, Gogola GR, Hecht JT, Grange DK, Harris DJ, Earl DL, Jagadeesh S, Mehta SG, Robertson SP, Swanson JM, Faustman EM, Mefford HC, Shendure J, Nickerson Da, Bamshad MJ. 2013. Mutations in ECEL1 cause distal arthrogryposis type 5D. Am J Hum Genet 92:150–156. Mejaddam AY, Nennesmo I, Sejersen T. 2009. Severe phenotype of a patient with autosomal recessive centronuclear myopathy due to a BIN1 mutation. Acta Myol 28:91–93. Nagata K, Kiryu-Seo S, Kiyama H. 2006. Localization and ontogeny of damage-induced neuronal endopeptidase mRNA-expressing neurons in the rat nervous system. Neuroscience 141:299–310.

DOHRN ET AL.

743

Nagata K, Kiryu-Seo S, Maeda M, Yoshida K, Morita T, Kiyama H. 2010. Damage-induced neuronal endopeptidase is critical for presynaptic formation of neuromuscular junctions. J Neurosci 30:6954–6962.

Schulz H, Dale GE, Karimi-Nejad Y, Oefner C. 2009. Structure of human endothelin-converting enzyme I complexed with phosphoramidon. J Mol Biol 385:178–187.

Navti O, Kinning E, Vasudevan P. 2010. Review of perinatal management of arthrogryposis at a large UK teaching hospital serving a multiethnic population. Prenat Diagn 30:49–56.

Schweizer A, Valdenaire O, Lang Y, Schmitt G, Lenz B, Bluethmann H, Ko A, Rohrer J. 1999. Neonatal lethality in mice deficient in XCE, a novel member of the endothelin-converting enzyme and neutral endopeptidase family. J Biol Chem 274:20450–20456.

Nicot A-S, Toussaint A, Tosch V, Kretz C, Wallgren-Pettersson C, Iwarsson E, Kingston H, Garnier J-M, Biancalana V, Oldfors A, Mandel J-L, Laporte J. 2007. Mutations in amphiphysin 2 (BIN1) disrupt interaction with dynamin 2 and cause autosomal recessive centronuclear myopathy. Nat Genet 39:1134–1139. Pakkasja¨rvi N, Ritvanen A, Herva R, Peltonen L, Kestila¨ M, Ignatius J. 2006. Lethal congenital contracture syndrome (LCCS) and other lethal arthrogryposes in Finland—An epidemiological study. Am J Med Genet Part A 140A:1834–1839. Polizzi A, Huson SM, Vincent A. 2000. Teratogen update: Maternal myasthenia gravis as a cause of congenital arthrogryposis. Teratology 62:332–341. Rink BD. 2011. Arthrogryposis: A review and approach to prenatal diagnosis. Obestetrical Gynecol Surv 66:369–377. Romero NB. 2010. Centronuclear myopathies: A widening concept. Neuromuscul Disord 20:223–228. Romero NB, Bitoun M. 2011. Centronuclear myopathies. Semin Pediatr Neurol 18:250–256. Schabhu¨ttl M, Wieland T, Senderek J, Baets J, Timmerman V, De Jonghe P, Reilly MM, Stieglbauer K, Laich E, Windhager R, Erwa W, Trajanoski S, Strom TM, Auer-Grumbach M. 2014. Whole-exome sequencing in patients with inherited neuropathies: Outcome and challenges. J Neurol 261:970–982.

Shaaban S, Duzcan F, Yildirim C, Chan W-M, Andrews C, Akarsu N, Engle E. 2014. Expanding the phenotypic spectrum of ECEL1-related congenital contracture syndromes. Clin Genet 85:562–567. Shaheen R, Al-Owain M, Khan A, Zaki M, Hossni H, Al-Tassan R, Eyaid W, Alkuraya F. 2014. Identification of three novel ECEL1 mutations in three families with distal arthrogryposis type 5D. Clin Genet 85:568–572. Sigrist CJa, de Castro E, Cerutti L, Cuche Ba, Hulo N, Bridge A, Bougueleret L, Xenarios I. 2013. New and continuing developments at PROSITE. Nucleic Acids Res 41:D344–D347. Spiro AJ, Shy M, Gonatas N. 1966. Myotubular myopathy. Arch Neurol 14:1–14. Valdenaire O, Richards JG, Faull RLM, Schweizer A. 1999. XCE, a new member of the endothelin-converting enzyme and neutral endopeptidase family, is preferentially expressed in the CNS. Mol Brain Res 64:211– 221. Valdenaire O, Rohrbacher E, Langeveld A, Schweizer A, Meijers C. 2000. Organization and chromosomal localization of the human ECEL1 (XCE) gene encoding a zinc metallopeptidase involved in the nervous control of respiration. Biochem J 346:611–616. Witters I, Moerman P, Fryns J-P. 2002. Fetal akinesia deformation sequence: A study of 30 consecutive in utero diagnoses. Am J Med Genet 113:23–28.