Scholarly article on topic 'Clinical and molecular characteristics in three families with biallelic mutations in IGHMBP2'

Clinical and molecular characteristics in three families with biallelic mutations in IGHMBP2 Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
Academic journal
Neuromuscular Disorders
OECD Field of science
Keywords
{"Axonal polyneuropathy" / CMT2S / " IGHMBP2 " / SMARD1 / "Spinal muscular atrophy"}

Abstract of research paper on Clinical medicine, author of scientific article — Christeen Ramane J. Pedurupillay, Silja S. Amundsen, Tuva Barøy, Magnhild Rasmussen, Anne Blomhoff, et al.

Abstract Biallelic mutations in IGHMBP2 cause spinal muscular atrophy with respiratory distress type 1 (SMARD1) or Charcot–Marie–Tooth type 2S (CMT2S). We report three families variably affected by IGHMBP2 mutations. Patient 1, an 8-year-old boy with two homozygous variants: c.2T>C and c.861C>G, was wheelchair bound due to sensorimotor axonal neuropathy and chronic respiratory failure. Patient 2 and his younger sister, Patient 3, had compound heterozygous variants: c.983_987delAAGAA and c.1478C>T. However, clinical phenotypes differed markedly as the elder with sensorimotor axonal neuropathy had still unaffected respiratory function at 4.5 years, whereas the younger presented as infantile spinal muscular atrophy and died from relentless respiratory failure at 11 months. Patient 4, a 6-year-old girl homozygous for IGHMBP2 c.449+1G>T documented to result in two aberrant transcripts, was wheelchair dependent due to axonal polyneuropathy. The clinical presentation in Patients 1 and 3 were consistent with SMARD1, whereas Patients 2 and 4 were in agreement with CMT2S.

Academic research paper on topic "Clinical and molecular characteristics in three families with biallelic mutations in IGHMBP2"

Accepted Manuscript

Title: Clinical and molecular characteristics in three families with biallelic mutations in IGHMBP2

Author: Christeen Ramane J. Pedurupillay, Silja S. Amundsen, Tuva Baroy, Magnhild Rasmussen, Anne Blomhoff, Barbro Fossoy Stadheim, Kristin 0rstavik, Asbjorn Holmgren, Tahir Iqbal, Eirik Frengen, Doriana Misceo, Petter Stromme

PII: S0960-8966(15)30084-5

DOI: http://dx.doi.org/doi: 10.1016/j.nmd.2016.06.457

Reference: NMD 3210

To appear in:

Neuromuscular Disorders

Received date: Revised date: Accepted date:

23-10-2015 9-6-2016 20-6-2016

Please cite this article as: Christeen Ramane J. Pedurupillay, Silja S. Amundsen, Tuva Baroy, Magnhild Rasmussen, Anne Blomhoff, Barbro Fossoy Stadheim, Kristin 0rstavik, Asbjorn Holmgren, Tahir Iqbal, Eirik Frengen, Doriana Misceo, Petter Stromme, Clinical and molecular characteristics in three families with biallelic mutations in IGHMBP2, Neuromuscular Disorders (2016), http://dx.doi.org/doi: 10.1016/j.nmd.2016.06.457.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Case report

Neuromuscular Disorders

Clinical and molecular characteristics in three families with biallelic mutations in IGHMBP2

Christeen Ramane J. Pedurupillayf Silja S. Amundsen^, Tuva Bar0yf Magnhild Rasmussenb'c, Anne Blomhoff1, Barbro Foss0y Stadheima, Kristin 0rstavikd, Asbj0rn Holmgrena, Tahir Iqbale, Eirik Frengena,f, Doriana Misceoa,f and Petter Str0mmeb,f*

aDepartment of Medical Genetics, Oslo University Hospital and University of Oslo, Oslo, Norway; bWomen and Children's Division, Department of Clinical Neurosciences for Children, Oslo University Hospital, Oslo, Norway; cUnit for Congenital and Hereditary Neuromuscular Disorders, Department of Neurology, Oslo University Hospital, Oslo, Norway; dDepartment of Neurology, Oslo University Hospital, Oslo, Norway; eMolecular Biology laboratory, Department of Zoology, University of Gujrat, Gujrat, Pakistan;f Faculty of Medicine, University of Oslo, Oslo, Norway.

Email addresses:

CRJP: p.j.c.ramane@medisin.uio.no SSA: s.s.amundsen@medisin.uio.no TB: tuva.baroy@medisin.uio.no MR: marasmus@ous-hf.no AB: uxnnbl@ous-hf.no BFS: barsta@ous-hf.no

K0: krorstav@ous-hf.no

AH: asbjorn.holmgren@medisin.uio.no

TI: taq333@gmail.com

EF: eirik.frengen@medisin.uio.no

DM: doriana.misceo@medisin.uio.no

PS: petter.stromme@medisin.uio.no

^These authors contributed equally to the work

Corresponding author Prof. Petter Stromme MD PhD

Email: petter.stromme@medisin.uio.no; telephone: +47 23015234, Fax: +47 23015790 University of Oslo, P.O. Box 1036, Blindern, N-0315 Oslo, Norway

Running title: IGHMBP2 mutations in three families

Keywords: axonal polyneuropathy; CMT2S; IGHMBP2; SMARD1; spinal muscular atrophy

Highlights

• Novel IGHMBP2 variant in boy with SMARD1

• Two siblings with IGHMBP2 defect show discordant phenotypes

• IGHMBP2 c.449+1G>T result in two aberrant transcripts and axonal neuropathy

Abstract

Biallelic mutations in IGHMBP2 cause spinal muscular atrophy with respiratory distress type 1 (SMARD1) or Charcot Marie Tooth type 2S (CMT2S). We report three families variably affected by IGHMBP2 mutations. Patient 1, an 8-year-old boy with two homozygous variants: c.2T>C and c.861C>G, was wheelchair bound due to sensorimotor axonal neuropathy and chronic respiratory failure. Patient 2 and his younger sister Patient 3 had compound heterozygous variants: c.983_987delAAGAA and c.1478C>T. However, clinical phenotypes differed markedly as the elder with sensorimotor axonal neuropathy had still unaffected respiratory function at 4.5 years, whereas the younger presented as infantile spinal muscular atrophy and died from relentless respiratory failure at 11 months. Patient 4, a 6-year-old girl homozygous for IGHMBP2 c.449+1G>T documented to result in two aberrant transcripts, was wheelchair dependent due to axonal polyneuropathy. The clinical presentation in Patients 1 and 3 were consistent with SMARD1, whereas Patients 2 and 4 were in agreement with CMT2S.

1. Introduction

Immunoglobulin mu-binding protein 2 (IGHMBP2) is a ubiquitously expressed DNA and RNA helicase [1-3]. Biallelic mutations in this gene result in degeneration of a-motor neurons in the brain stem and the anterior horns of the spinal cord, causing spinal muscular atrophy with respiratory distress type 1 (SMARD1; OMIM 604320) or Charcot Marie Tooth type 2S (CMT2S; OMIM 616155). IGHMBP2 is implicated in transcription [4, 5], pre-mRNA processing [2, 6], translation [5, 7, 8], and immunoglobulin class switching [2]. Mutated IGHMBP2 is suggested to result in a deficiency of maturation of mRNAs in neurons, thereby leading to neuronal degeneration [1].

SMARD1 is characterized by infantile onset severe axonal polyneuropathy and diaphragmatic paralysis [4, 9-11]. Typically SMARD1 patients die within the first year of life due to respiratory failure [4, 10, 11]. However, the disease severity in SMARD1 may be variable, and juvenile onset of respiratory distress [12-15] with survival up to 20 years has been reported [14, 16]. At the mild end of the IGHMBP2 mutation disease spectrum, patients present with CMT2S, a sensorimotor axonal polyneuropathy, usually with less pronounced neurophysiological changes compared to SMARD1, relatively spared respiratory function, and longer survival [1, 15, 17].

Here we describe four individuals from three families with biallelic mutations in IGHMBP2. The common denominator for referral of these patients was gross motor delay. We identified four mutations in IGHMBP2. The clinical presentation in the patients varied from SMARD1 to CMT2S, even within one family. We also show for the first time that the IGHMBP2 c.449+1G>T variant results in two aberrant transcripts, both containing a premature termination codon (PTC).

2. Case report

2.1 Clinical description

Family A

Patient 1, a boy, was the second child of healthy first-degree cousins of Pakistani origin (Fig.

1a). When assessed at a routine health check-up at 6 months he was noted to be hypotonic

with protrusion of the right side of his thorax. Motor development was significantly delayed

and at 2.5 years he could only walk a few steps using an orthosis. He had distal muscular

atrophy of the upper and lower extremities, hypermobile joints and with absent reflexes. At 5

years he had profound distal weakness and poor head control. He was no longer able to sit or

walk, and had become wheelchair dependent. He had dysphagia and a weak cough and suffered from chronic respiratory distress that required ventilatory support during sleep. From 5.5 years he was completely dependent on ventilatory support and also required a cough assist machine. At 6 years his extremities were paralyzed, except preservation of some movement with his fingers. Feeding was difficult, necessitating a gastric feeding tube. Neurophysiological examinations at 2 years showed lack of responses from both motor nerves (median, tibial and peroneal in his right extremities) and from the median sensory nerve. At the age of 5 years he still lacked both motor responses from the median, ulnar, tibial and peroneal nerves and sensory responses from the median, ulnar, sural and superficial peroneal nerve. However, there were responses from the sensory radial nerves (amplitude 3.0-5.2 |aV and nerve conduction velocities (NCV) 35-36 m/s). Needle-electromyography (EMG) of the anterior tibial and lateral vastus muscles at age 2 years and the lateral vastus and deltoid muscles at age 5 years showed spontaneous activity with fibrillation potentials in at least one muscle each time and at age 2 also some neurogenic motor unit potentials. However, due to muscle atrophy it was generally difficult to judge activity in the muscles. A biopsy of the sural nerve at 6 years showed lack of myelinated fibers. The results of these examinations were compatible with a pronounced sensorimotor axonal neuropathy. Ultrasound examination of his chest revealed reduced diaphragm motility. Echocardiography and cognitive function were normal.

Family B

Patients 2 and 3 were siblings, born to non-consanguineous healthy parents of Norwegian origin (Fig. 1b).

Patient 2, a boy, appeared normal at birth, but soon developed contractures of the Achilles

tendons, progressing to pes cavus with forefoot adduction between 4 and 8 months. Although

he learned to walk before 15 months, peripheral neurogenic problems persisted making him dependent on orthosis for walking at 20 months. At this stage, atrophy of the calf muscles and weakness for dorsiflexion of the ankles were noted. The upper extremities had also become mildly affected with contractures of the fingers. The deep tendon reflexes were absent. Pectus excavatum was noted at 4.5 years, but he had not shown signs of respiratory dysfunction. Neurophysiological examination performed at 18 month showed pathologically low motor amplitudes from all nerves tested (right median 2.2mV, right ulnar 0.3 mV and the tibial nerves bilaterally 0.2-07mV, no response from peroneal nerves). It was only possible to measure the NCV in the median nerve and this was slightly delayed (34.6 m/s; reference values for NCV used in this study are based on Miller et al. [18]). Sensory nerve conduction velocities ranged between 38.5 m/s in the ulnar nerve (normal range between 13 - 24 months is 41.3-63.5 m/s) and 56.3 m/s in the median nerve, 38.7 m/s in the sural nerve and 57.3 m/s in the peroneal superficial nerve. The sensory amplitudes were also pathologically low, ranging from 3.6-3.9^V in the median and ulnar nerves, respectively, and 2.9-3.6 |aV in the sural and peroneal superficial nerves. EMG of the anterior tibial and lateral vastus muscles showed spontaneous activity. These findings suggested sensorimotor axonal polyneuropathy with motor nerve fibers more affected than sensory nerve fibers. This was confirmed by similar findings in a repeated neurophysiological examination less than 3 months later when EMG also showed chronic neurogenic findings in the anterior tibial muscles bilaterally. Ultrasound examination of the diaphragm at 3 years showed normal motility and overnight monitoring of pO2 and pCO2 was also normal. His cognitive development was normal.

Patient 3, a girl, was born at 36 weeks of gestation after an uncomplicated pregnancy, with birth weight and length at the 3rd centiles, and head circumference 1 cm below the 3rd centile.

She was admitted to the neonatal intensive care unit with feeding difficulties and hypotonia.

She exhibited dysmorphic features that included widely spaced eyes, elongated palpebral fissures, long eyelashes, high-arched and interrupted eyebrows, fetal pads, and pectus excavatum, suspected to be due to Kabuki syndrome (OMIM 147920). She was hospitalized at 1 month due to feeding difficulties and at 5 months due to urosepsis. Motor development was severely delayed and at 8 months she could not crawl, whereas the deep tendon reflexes were difficult to elicit. When hospitalized due to a pulmonary infection at age 8 months, she showed paradoxical movements of the diaphragm, indicating muscle weakness. Upon the acute respiratory distress caused by the pulmonary infection, she developed a chronic respiratory failure with increasing need for ventilator support. Neurophysiological examinations were suggestive of motor axonal polyneuropathy with low motor amplitudes both for the median (1.1mV) and ulnar nerve (2.2 mV) in the right upper extremity and for the tibial nerves bilaterally in the lower extremities (0.4-2.2 mV). The NCV were within normal range (38.5-59 m/s). No motor responses were detected from the peroneal nerves bilaterally. Only two sensory nerves were examined, the right median and left peroneal superficial nerves and they were within normal range for her age (amplitudes 15 and 4.7 respectively and NCV 48 and 40 m/s). EMG-recordings in the anterior tibial muscles bilaterally and from the left vastus and right deltoid muscles revealed no spontaneous activity. She died at 11 months due to chronic respiratory failure. Cerebral MRI was normal.

Family C

Patient 4, a girl, was the first child of healthy parents of Kurdish origin, reported to be unrelated (Fig. 1c).

She was referred from the public health station for children at 13 months due to delayed

motor development, as she was hypotonic and could sit only with support. She walked with

support at 22 months. On examination at 2 years, distal muscle atrophy, foot deformities with

valgus positioning of the right foot, varus positioning of the left foot and hyperextension of the knees were observed. At 6 years, the deep tendon reflexes were noted to be absent and she required orthosis in order to stand and walk. She was easily fatigued and mostly wheelchair dependent. At the same age she was evaluated by cough peak expiratory flow (PEF), spirometry and nocturnal transcutaneous registration of carbon dioxide and oxygen saturation, showing a weak cough, but no signs of hypoventilation (movements of the diaphragm were not evaluated). She started to use a cough assist machine. Slow recovery from respiratory tract infections became evident. She underwent surgery for bilateral clubfeet. The ability to swallow liquids and solids was impaired to such an extent she needed the insertion of a percutaneous gastric tube. Neurophysiological examination at 28 months showed low motor amplitudes in the right median (2.5 mV) and bilaterally in the tibial nerves (0.6 and 2.0 mV respectively) and with only a very small response (0.1 mV) in one of the two peroneal nerves tested. The conduction velocities were delayed, 38.7 m/s in the median nerve and from 25.128.7 m/s in the lower extremities. There was no sensory response from the sural nerve, but a slightly lowered response from the median nerve (5.9^V) with a conduction velocity of 40 m/s. EMG from both anterior tibial muscles showed spontaneous activity and during voluntary activity there were a few neurogenic motor unit potentials. The findings were compatible with sensorimotor axonal polyneuropathy. Echocardiography and cognitive development were normal.

2.2 Molecular analyses Family A

Whole exome sequencing (WES) was performed in Patient 1 and his parents. Data analysis assuming autosomal recessive mode of inheritance identified two homozygous variant in

IGHMBP2 (NM_002180.2): c.2T>C; p.Met1? (chr11:hg19.g:68,671,422 bp) and c.861C>G;

p.Ser287Arg (chr11:hg19.g:68,682,440 bp) in the patient, while both parents were found to be heterozygous for both variants (Fig. 1a) (Complete list of gene variants is provided in Table S1). Additional experimental WES details are in Supplementary file 1. Sanger sequencing verified the variants (all primer sequences used in this study are given in Table S2). The c.2T>C variant was previously identified in two patients clinically presenting with CMT1 by targeted gene panel sequencing [19]. IGHMBP2 c.2T>C alters the start codon and is predicted to abolish protein production. The variant was not reported in the ExAC database (http://exac.broadinstitute.org), including more than 60,000 exomes. The closest alternative in-frame initiation codon was found at p.338, which is likely to be too distant to be utilized [20], and if used it would result in protein truncation in the crucial helicase domain [4, 5, 8]. Also, no alternative Kozak consensus sequence, 5'GCCGCCRMCATGGCG'3 (R=A or G and M=A or C), suggestive of an alternative start of translation [19, 20], was found. In addition, another IGHMBP2 variant, c.861C>G, was homozygous in Patient 1 and heterozygous in the parents. This missense variant in exon 6 changes a highly conserved amino acid in the helicase domain (Fig. 1d). ExAC database reported that IGHMBP2 c.861C>G was found in heterozygosity in 55 out of 8255 individuals from South Asia (allele frequency 0.0033). Neither of the c.2T>C and c.861C>G variants were identified by Sanger sequencing of 166 healthy Pakistani individuals. In order to investigate if the c.2T>C variant would affect the transcript level, the IGHMBP2 mRNA level was measured in Patient 1, his parents and six controls. The analysis showed similar levels in all individuals tested (Fig. 2).

Family B

Sanger sequencing of IGHMBP2 was performed in Patient 3 based on the clinical

presentation. A frameshift deletion in IGHMBP2 exon 7, c.983_987delAAGAA;

p.Lys328Thrfs46* (chr11:hg19.g:68,685,274-68,685,278), and a missense in IGHMBP2 exon

10, c.1478C>T; p.Thr493Ile (chr11: hg19.g:68,701,322) were identified. Sanger sequencing showed that Patient 2 also had both variants, and that the c.983_987delAAGAA and c.1478C>T were of maternal and paternal origin, respectively (Fig. 1b). Both variants were previously reported in SMARD1 patients [5, 10, 12, 21]. Patient 3 also had Kabuki syndrome (OMIM 147920), due to a de novo c.14092_14093insTA mutation in myeloid/lymphoid or mixed-lineage leukemia protein 2 (MLL2) (identified by Sanger sequencing). In order to investigate if the IGHMBP2 c.983_987delAAGAA or c.1478C>T variants result in nonsensemediated mRNA decay (NMD), the IGHMBP2 mRNA levels were measured by qRT-PCR in Patient 2, his parents and six controls, showing a higher level in the patient and similar levels in parents and controls (Fig. 2).

Family C

In Patient 4 a homozygous IGHMBP2 c.449+1G>T (chr11:hg19.g:68,675,806) splice donor

site variant in intron 3 was identified by WES after analyzing a set of candidate genes (Table

S3) based on clinical presentation. Sanger sequencing confirmed that Patient 4 was

homozygous and the parents were heterozygous for the variant (Fig.1c). The c.449+1G>T

variant was previously reported in patients with SMARD1 and CMT2S [15, 22]. Exons 3-5 in

the IGHMBP2 transcript were amplified by reverse-transcriptase PCR (RT-PCR) in Patient 4,

her parents and six controls in order to identify possible aberrant transcripts resulting from the

c.449+1G>T variant. The gel image showed a band representing the normal transcript in the

controls (Fig. 3 band 1), which was absent in Patient 4. Instead the patient had two bands

representing abnormal transcripts that were not present in the controls (Fig. 3 band 2 and 3).

All three bands were found also in the heterozygous parents (Fig. 3). Sanger sequencing of

band 2 in the patient showed a 100 bp intron 3 partial retention (chr11:hg19.g:68,675,806-

68,675,906 bp), and Sanger sequencing of band 3 showed full 196 bp intron 3 retention

(chr11:hg19.g:68,675,806-68,676,001 bp). Both aberrant transcripts have a PTC at c.449+2-c.449+4 (p.Lys150Asnfs*0), and are expected to result in NMD or a truncated protein. The likely disease causing IGHMBP2 variants reported in the study are present in the Clinical Variant database (http://www.ncbi.nlm.nih.gov/clinvar/) with the following accession numbers: c.2T>C: SCV000255975; c.983_987delAAGAA: SCV000255976; c.1478C>T: SCV000255977; c.983_987delAAGAA: SCV000255978; c.1478C>T: SCV000255979; c.449+1G>T: SCV000255980.

3. Discussion

We identified four likely disease causing mutations in IGHMBP2 in four patients from three independent families (c.2T>C, c.449+1G>T, c.1478C>T and c.983_987delAAGAA). Patient 1 (consanguineous family; Fig. 1a) had neurophysiological findings indicating a severe sensorimotor axonal neuropathy with onset of respiratory distress at 5 years, consistent with SMARD1. His neurophysiological findings could also be compatible with spinal muscular atrophy with an additional component of sensory neuropathy as previously described in SMARD1 [4, 10]. At 8 years, he was wheelchair bound and in need of continuous respiratory support. He was found by WES to have a homozygous c.2T>C mutation in IGHMBP2, previously reported in two patients clinically presenting with CMT1 [19]. The lack of alterative in-frame initiation codons in vicinity of the original initiation codon indicates that the c.2T>C variant likely abolishes protein translation, thus it was considered to be the cause of the disease in this patient. In addition, he had a second homozygous variant c.861C>G, which has been reported with an allele frequency of 0.0033 in the South Asian population in the ExAC database. Since SMARD1 is recognized as an extremely rare disorder [16], the relatively high allele frequency of the c.861C>G variant indicates it is a non-pathogenic rare

genetic variant. In any case, the c.861C>G is not expected to have an impact on IGHMBP2 in Patient 1, because the c.2T>C variant likely abolishes protein production. The siblings, Patient 2 and 3 (Fig. 1b), were compound heterozygous for two IGHMBP2 mutations: c.1478C>T and c.983_987delAAGAA. There was considerable clinical variability as Patient 3 had SMARD1 and died at 11 months due to respiratory failure, whereas her brother, Patient 2, had a more typical axonal CMT2S phenotype without signs of respiratory distress at 4.5 years. In two families with IGHMBP2 mutations a large degree of intrafamilial variability was described [15, 23]. Intrafamilial variability may result from modifier variants in other genes or from variation in the residual level of IGHMBP1. In patient 3 Kabuki syndrome may have predisposed to the pulmonary infection [24], which caused acute respiratory distress at age 8 months, after which the patient developed chronic respiratory distress.

The c.1478C>T was previously described in five patients from four families with compound heterozygous mutations [5, 12, 23, 25]. The onset of respiratory distress ranged from 6 months to 10 years in four of these [5, 23, 25], and the fifth patient had normal respiratory function at 12 years [12]. A single patient with the c.983_987delAAGAA mutation in a homozygous state had SMARD1 and infantile onset of respiratory distress [10]. In our study, expression of the IGHMBP2 mRNA indicated that the c.2T>C, c.983_987delAAGAA and c.1478C>T variants did not result in NMD (Fig. 2). This is in accordance with previous analysis showing that IGHMBP2 mutations do not lead to reduced mRNA level [12, 26]. In the absence of NMD, the c.983_987delAAGAA is predicted to result in protein truncation, the c.2T>C is predicted to abolish protein translation and the c.1478C>T is predicted to directly affect the protein function.

Patient 4 (Fig. 1c), had predominantly axonal motor polyneuropathy and was mostly

wheelchair dependent. At 6 years of age, only a weak cough indicated she might have mild

respiratory muscle dysfunction. She was homozygous for IGHMBP2 c.449+1G>T, previously reported in three patients: one with typical SMARD1 and onset of respiratory distress at 4 months [22] and two siblings with axonal sensory motor neuropathy, one with late-onset respiratory distress at 15 years, and one still without respiratory problems at 22 years [15]. IGHMBP2 c.449+1G>T is a rare variant, since it is not reported in the ExAC database (search October 21th 2015), but it was found in two consanguineous families, one from Lebanon [15] and one from Kurdistan [22] (similar ethnicity as family C). This variant may therefore have a founder in the Middle East. In this report we show that IGHMBP2 c.449+1G>T variant results in a full intron 3 retention, in addition to a previously described partial intron 3 retention [15] (Fig. 3), both predicted to result in NMD or a truncated protein with lack of function. Two of our patients presented with SMARD1 (Patient 1 and 3), whereas the clinical phenotypes in the other two patients were in agreement with CMT2S (Patient 2 and 4), but late onset of respiratory distress cannot be excluded. However, prognostication based on the genotype is difficult, as neurophysiological findings overlap [1, 15, 22], and respiratory distress may present at a substantially later age, or even not occur at all [14, 15].

4. Conclusion

Our study confirms that biallelic IGHMBP2 mutations either may cause SMARD1 or

CMT2S. In fact, phenotypic diversity may occur within the same family, as identical

compound heterozygous mutations in siblings segregated as SMARD1 and CMT2S,

exhibiting the severe and moderate phenotypes of the disease spectrum. We did not detect

NMD of IGHMBP2 mRNA resulting from the c.2T>C, c.983_987delAAGAA or c.1478C>T

variants, indicating these variants affect the protein level or function directly. The IGHMBP2

c.449+1G>T variant was shown to result in two aberrant transcripts, which may be degraded by NMD or alternatively produce a truncated protein.

Acknowledgement

We are grateful to the families who participated in this study. The Norwegian South-East Regional Health Authority supported this study (grant # 2011071). The Norwegian Sequencing Centre at Oslo University Hospital, Oslo, Norway provided the whole exome sequencing service in Patient 1. We thank the Diagnostic Section at the Department of Medical Genetics, Oslo University Hospital, Oslo, Norway for Sanger sequencing of MLL2 in Patient 3. We thank Section of Medical Genetics, Department of Laboratory Medicine, Telemark Hospital, Skien, Norway, for whole exome sequencing in Patient 4.

References

[1] Cottenie E, Kochanski A, Jordanova A, et al. Truncating and missense mutations in IGHMBP2 cause Charcot-Marie Tooth disease type 2. Am J Hum Genet 2014;95:590-601.

[2] Fukita Y, Mizuta TR, Shirozu M, Ozawa K, Shimizu A, Honjo T. The human S mu bp-2, a DNA-binding protein specific to the single-stranded guanine-rich sequence related to the immunoglobulin mu chain switch region. J Biol Chem 1993;268:17463-17470.

[3] Grohmann K, Rossoll W, Kobsar I, et al. Characterization of Ighmbp2 in motor neurons and implications for the pathomechanism in a mouse model of human spinal muscular atrophy with respiratory distress type 1 (SMARD1). Hum Mol Genet 2004;13:2031-2042.

[4] Grohmann K, Schuelke M, Diers A, et al. Mutations in the gene encoding immunoglobulin mu-binding protein 2 cause spinal muscular atrophy with respiratory distress type 1. Nat Genet 2001;29:75-77.

[5] Lim SC, Bowler MW, Lai TF, Song HW. The Ighmbp2 helicase structure reveals the molecular basis for disease-causing mutations in DMSA1. Nucleic Acids Res 2012;40:11009-11022.

[6] Molnar GM, Crozat A, Kraeft SK, Dou QP, Chen LB, Pardee AB. Association of the mammalian helicase MAH with the pre-mRNA splicing complex. Proc Natl Acad Sci U S A 1997;94:7831-7836.

[7] de Planell-Saguer M, Schroeder DG, Rodicio MC, Cox GA, Mourelatos Z. Biochemical and genetic evidence for a role of IGHMBP2 in the translational machinery. Hum Mol Genet 2009;18:2115-2126.

[8] Guenther UP, Varon R, Schlicke M, et al. Clinical and mutational profile in spinal muscular atrophy with respiratory distress (SMARD): defining novel phenotypes through hierarchical cluster analysis. Hum Mutat 2007;28:808-815.

[9] Bertini E, Gadisseux JL, Palmieri G, et al. Distal infantile spinal muscular atrophy associated with paralysis of the diaphragm: a variant of infantile spinal muscular atrophy. Am J Med Genet 1989;33:328-335.

[10] Grohmann K, Varon R, Stolz P, et al. Infantile spinal muscular atrophy with respiratory distress type 1 (SMARD1). Ann Neurol 2003;54:719-724.

[11] Kaindl AM, Guenther UP, Rudnik-Schoneborn S, et al. Spinal muscular atrophy with respiratory distress type 1 (SMARD1). J Child Neurol 2008;23:199-204.

[12] Guenther UP, Handoko L, Varon R, et al. Clinical variability in distal spinal muscular atrophy type 1 (DSMA1): determination of steady-state IGHMBP2 protein levels in five patients with infantile and juvenile disease. J Mol Med (Berl) 2009;87:31-41.

[13] Messina MF, Messina S, Gaeta M, et al. Infantile spinal muscular atrophy with respiratory distress type I (SMARD 1): an atypical phenotype and review of the literature. Eur J Paediatr Neurol 2012;16:90-94.

[14] Rudnik-Schoneborn S, Stolz P, Varon R, et al. Long-term observations of patients with infantile spinal muscular atrophy with respiratory distress type 1 (SMARD1). Neuropediatrics 2004;35:174-182.

[15] Schottmann G, Jungbluth H, Schara U, et al. Recessive truncating IGHMBP2 mutations presenting as axonal sensorimotor neuropathy. Neurology 2015.

[16] Vanoli F, Rinchetti P, Porro F, Parente V, Corti S. Clinical and molecular features and therapeutic perspectives of spinal muscular atrophy with respiratory distress type 1. J Cell Mol Med 2015.

[17] Wilmshurst JM, Bye A, Rittey C, et al. Severe infantile axonal neuropathy with respiratory failure. Muscle & Nerve 2001;24:760-768.

[18] Miller RG, Kuntz NL. Nerve-Conduction Studies in Infants and Children. Journal of Child Neurology 1986;1:19-26.

[19] Antoniadi T, Buxton C, Dennis G, et al. Application of targeted multi-gene panel testing for the diagnosis of inherited peripheral neuropathy provides a high diagnostic yield with unexpected phenotype-genotype variability. BMC Med Genet 2015;16:84.

[20] Wolf A, Caliebe A, Thomas NST, et al. Single Base-Pair Substitutions at the Translation Initiation Sites of Human Genes as a Cause of Inherited Disease. Hum Mutat 2011;32:1137-1143.

Legend to figures and supplementary data

Supplementary file 1. Additional details on methods. Table S1

The complete list of variants identified in Patient 1, Family A from analysis of WES data according to an autosomal recessive inheritance model. Table S2

Primer sequences used in this study for Sanger sequencing and for qRT-PCR Table S3

List of candidate genes analyzed in the WES data in Patient 2 from family B.

Fig. 1

a) Pedigree of family A showing the segregation of two variants in IGHMBP2 c.2T>C and c.861C>G. Both IGHMBP2 variants were found in cis in heterozygosity in the parents and in homozygosity in Patient 1. b) Pedigree of family B showing the segregation of the IGHMBP2 c.1478C>T and c.983-987delAAGAA variants (indicated in black), parents were carriers. Patient 3 also had a de novo c.14092_14093insTA in MLL2 (indicated in grey). c) Pedigree of family C showing the segregation of the IGHMBP2 c.449+1G>T variant. d) Image from Alamut Visual Software (Rouen, France) showing the evolutionary conservation of the amino acid changes resulting from the two variants c.2T>C (p.1M?) and c.861C>G (p.Ser287Arg) in Patient 1 Family A (black stars).

Fig. 2

Relative IGHMBP2 mRNA levels measured by qRT-PCR in patients and parents in family A and B compared to the mean level in six controls, showing similar levels to controls in family A, and similar to increased levels in family B.

Family A: Patient 1 with the two homozygous variants c.2T>C and c.861C>G and his parents heterozygous for both variants. Family B: Patient 2 with compound heterozygous variants c.1478C>T and c.983-987delAAGAA, and his heterozygous parents for the c.1478C>T and c.983-987delAAGAA variants, respectively. The mean level in the six controls is shown with 95% confidence interval. Fig. 3

a) Gel image of the RT-PCR products for exon 3-5 of the IGHMBP2 transcript from the

patient and parents from family C and six controls, showing that the band representing the

normal transcript is absent in the patient with the homozygous splice variant c.449+1G>T

(band 1), and instead the presence of band 2 and 3 not seen in the controls. All three bands

were detected in the parents with the heterozygous c.449+1G>T variant. This shows that the splice site variant c.449+1G>T results in two aberrant transcripts, and by Sanger sequencing of cDNA from band 2 and 3, these were shown to have a full 196 bp intron 3 retention and a partial 100 bp intron 3 retention, respectively. b) Drawing of the content of bands 1, 2, and 3 based on Sanger sequencing results.