Article Text

MEF2C haploinsufficiency caused by either microdeletion of the 5q14.3 region or mutation is responsible for severe mental retardation with stereotypic movements, epilepsy and/or cerebral malformations
  1. N Le Meur1,2,
  2. M Holder-Espinasse3,
  3. S Jaillard4,5,
  4. A Goldenberg1,
  5. S Joriot6,
  6. P Amati-Bonneau7,8,
  7. A Guichet7,
  8. M Barth7,
  9. A Charollais9,
  10. H Journel10,
  11. S Auvin11,
  12. C Boucher1,
  13. J-P Kerckaert12,
  14. V David5,13,
  15. S Manouvrier-Hanu3,
  16. P Saugier-Veber1,14,
  17. T Frébourg1,14,
  18. C Dubourg5,13,
  19. J Andrieux15,
  20. D Bonneau7,8
  1. 1Service de Génétique, CHU de Rouen, France
  2. 2Laboratoire de Cytogénétique, EFS-Normandie, Bois-Guillaume, France
  3. 3Service de Génétique Clinique, Hôpital Jeanne de Flandre, CHRU de Lille, France
  4. 4Laboratoire de Cytogénétique, CHU Pontchaillou, Rennes, France
  5. 5CNRS UMR 6061, Université de Rennes 1, IFR 140, Rennes, France
  6. 6Service de Neuropédiatrie, Hôpital Roger Salengro, CHRU de Lille, France
  7. 7Service de Génétique Médicale, CHU d'Angers, France
  8. 8Inserm U694, Université d'Angers, France
  9. 9Service de Médecine Néonatale, CHU de Rouen, France
  10. 10Service de Génétique Clinique, Centre Hospitalier Bretagne Atlantique, Vannes, France
  11. 11Service de Neurologie Pédiatrique, CHU Robert Debré - APHP, Paris, France
  12. 12Plateforme de Génomique Fonctionnelle, Université de Lille II, France
  13. 13Laboratoire de Génétique Moléculaire, CHU Pontchaillou, Rennes, France
  14. 14Inserm U614, IHU, Université de Rouen, France
  15. 15Laboratoire de Génétique Médicale, Hôpital Jeanne de Flandre, CHRU de Lille, France
  1. Correspondence to Dr N Le Meur, Laboratoire de Cytogénétique, EFS-Normandie, Bois-Guillaume, France; nathalie.lemeur{at}efs.sante.fr

Abstract

Background Over the last few years, array-comparative genomic hybridisation (CGH) has considerably improved our ability to detect cryptic unbalanced rearrangements in patients with syndromic mental retardation.

Method Molecular karyotyping of six patients with syndromic mental retardation was carried out using whole-genome oligonucleotide array-CGH.

Results 5q14.3 microdeletions ranging from 216 kb to 8.8 Mb were detected in five unrelated patients with the following phenotypic similarities: severe mental retardation with absent speech, hypotonia and stereotypic movements. Facial dysmorphic features, epilepsy and/or cerebral malformations were also present in most of these patients. The minimal common deleted region of these 5q14 microdeletions encompassed only MEF2C, the gene for a protein known to act in brain as a neurogenesis effector, which regulates excitatory synapse number. In a patient with a similar phenotype, an MEF2C nonsense mutation was subsequently identified.

Conclusion Taken together, these results strongly suggest that haploinsufficiency of MEF2C is responsible for severe mental retardation with stereotypic movements, seizures and/or cerebral malformations.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The clinical implementation of whole-genome array comparative genomic hybridisation (array-CGH) has revolutionised the diagnosis of patients with mental retardation, congenital anomalies or neuropsychiatric disorders. With high-density microarrays, chromosome imbalances are detected in up to 17% of cases of idiopathic developmental delay–mental retardation.1 On the basis of array-CGH data, the following new microdeletion and microduplication syndromes have been described: the 17q21.31 microdeletion involving the MAPT gene and resulting in syndromic mental retardation with typical facial features2; the 15q13.3 microdeletion associated with mental retardation and seizures3; and the 1q21.1 microdeletion, in the BP2–BP3 region, involving the PIAS3 gene in TAR syndrome.4 Array-CGH has also led to the identification of several genes involved in monogenic disorders, such as CHD7 in CHARGE and TCF4 in Pitt–Hopkins syndrome.5–7

Cardoso et al8 recently reported on three cases of 5q14.3–q15 deletions accompanied by severe mental retardation, hypotonia, seizures, minor dysmorphic features and periventricular heterotopia, and an additional case has been reported in the Decipher database (https://decipher.sanger.ac.uk/application/). Here, we report the detection of five distinct 5q14 deletions in five unrelated patients with severe mental retardation, absent speech and stereotypic movements and one duplication of the 5q14 region in a patient with mental retardation. These deletions, which are different from those reported by Cardoso et al,8 allowed us to define a minimal critical region encompassing only the MEF2C gene and led us to sequence this gene in patients with similar phenotypes. We provide evidence that this severe mental retardation results from MEF2C haploinsufficiency.

Patients and methods

Patients

All seven patients included in this study have been examined by a clinical geneticist in the context of aetiological investigations in children with developmental delay. Array-CGH analysis was performed because mental retardation was associated with at least two of the following criteria: dysmorphic facial features, family history of mental retardation, growth anomaly or congenital malformation. For the seven patients, high-resolution chromosome analysis and extensive biochemical metabolic screening (lactic acid, pyruvic acid, ammonia, plasma amino acids, urinary organic acids, purine metabolism, blood and urinary creatine and guanidinoacetic acid) were normal. For each patient, blood samples for genetic analysis were collected after written informed consent had been obtained from the parents. Table 1 summarises the clinical presentations of the patients.

Table 1

Clinical findings in the six patients with MEF2C deletions or mutation

Case 1

BV is the third child of healthy unrelated parents. She was born at 36 weeks' gestation with normal growth parameters. At 3 days of age, she experienced a single episode of cyanosis with eye revulsion. Frequent crying, sleep disturbance, hypotonia and poor visual contact were noted at 3 months. From the age of 4 months, myoclonic jerks of the upper limbs were noted and were followed, several weeks later, by brief episodes of eye revulsion concomitant with the jerks. Epilepsy was diagnosed at the age of 7 months and characterised by the association of several bilateral isolated spasms and frequent synchronous myoclonus with abnormal and slow background EEG pattern. Clinical examination at 4 years 9 months showed normal growth and head circumference. Developmental delay was severe. She sat unaided and was able to crawl and manipulate toys. Eye contact was present although transient. Speech was absent. She had stereotypic repetitive movements, rocking her head and rubbing her chin with her hands. Subtelomeric rearrangements were excluded by quantitative multiplex PCR of short fluorescent fragments (QMPSF). Analysis of the CDKL5 gene revealed no alteration.

Case 2

CA is the second child of healthy unrelated parents. At birth, at 40 weeks' gestation, weight and length were normal, but head circumference was along the −3 standard deviation (SD). Since the first day of life, she developed tonicoclonic seizures. EEG revealed frequent bursts with no basic rhythm and a very unstructured pattern. At 9 months of age, severe hypotonia and poor eye contact were noted. Awakening stages were short. Despite treatment, the seizures occurred every day.

Case 3

ED is the second child of healthy unrelated parents. Delivery was provoked at 41 weeks' gestation because of abnormal fetal cardiac rhythm. At birth, weight and length were on +3 SD, but head circumference was small (−2.5 SD). At the age of 2 months, severe hypotonia and absent eye contact were noted. Cortical blindness was subsequently diagnosed. EEG showed a slow basic rhythm with infraclinical temporoparietal paroxystic discharges. At 18 months of age, weight gain was insufficient (−2 SD) leading to placement of a gastrostomy tube. He had severe hypotonia, transient eye contact and sleep disturbance.

Case 4

WD is the second child of healthy unrelated parents. Delivery occurred at 38 weeks' gestation and birth growth parameters were within the normal range. Failure to thrive and severe hypotonia were observed and led to several neurological investigations at the age of 4 months. Eye contact was difficult to obtain during the first year of life. He never experienced seizures. EEG was normal. He sat unaided at age 18 months and crawled at age 2 years. When referred to the genetic clinic at 3 years of age, he was able to stand, cruise along the furniture and manipulate toys. Speech was absent. Eye contact was transient. He presented with repetitive hand flapping and clapping movements. A diagnosis of Angelman syndrome was considered to be unlikely because of a normal SNRPN methylation pattern.

Case 5

LD is the second child of healthy parents who were first cousins. She was born at 40 weeks' gestation, with growth parameters below −2SD. Hypotonia and developmental delay were observed during the first months of life. At 3 years of age, she experienced tonicoclonic febrile seizures, which were well controlled by valproate. At 7 years, she was unable to walk and had not acquired any language skills. She made repetitive hand washing and hand-to-mouth movements as well as frequent bouts of hyperventilation. Subtelomeric rearrangements were excluded by fluorescence in situ hybridisation (FISH). Screening of the MECP2 gene revealed no deleterious mutation.

Case 6

TM is the third child of healthy unrelated parents. Growth parameters at birth at term were in the normal range. He was first referred at the age of 2 years 6 months for slight global developmental delay. He was able to walk unaided since the age of 2 years. Cerebral MRI and EEG results were normal. Clinical examination at 6 years of age revealed microcephaly (−3 SD) with normal height and weight. He had mental retardation; IQ was evaluated as 50–60 (WISC IV). Speech was severely delayed but understandable; he was not able to produce short sentences. Eye contact, behaviour and social skills were normal. Special education was required. Subtelomeric rearrangements and the most common microdeletion syndromes were excluded using the multiplex ligation-dependent probe amplification method.

Case 7

DG is the first girl born to healthy non-consanguineous parents with an unremarkable family history. She was born at term after an uneventful pregnancy. At birth, growth parameters were normal. The neonatal period was normal apart from difficulties in breast feeding. She appeared to develop normally until the age of 5 months. From the age of 5 months, she started to regress and to lose previously acquired skills. She was unable to use her hands purposefully and failed to acquire vocalisation with intonation. She walked unaided at age 3 years. She had behavioural disorders, including decreased eye contact, lack of emotional reciprocity, lack of interest in her surroundings and hand and hand–mouth stereotypic movements. She also had severe feeding difficulties which started at age 5 months and were still present at 7 years of age. Generalised tonicoclonic seizures, well controlled on sodium valproate, started at age 9 months. At 7 years of age, she was severely mentally impaired and presented with poor eye contact and no speech. Neurological examination showed an unstable, wide-based gait, without any objective cerebellar signs. Sequencing of the MECP2 and CDKL5 genes revealed no mutation.

Standard and molecular karyotyping

Karyotyping of RHG-banded chromosomes from lymphocytes at 550-band resolution was performed according to standard procedures. High-molecular-mass genomic DNA was extracted from patient's peripheral blood lymphocytes using the QIAamp DNA Blood Midi kit (Qiagen, Valencia, California, USA) according to the manufacturer's instructions. DNA concentration was determined with a NanoDrop ND-1000 spectrophotometer and software (NanoDrop Technologies, Berlin, Germany). Gene copy number was determined by array-CGH following standard and manufacturer's recommendations (Agilent Technologies, Santa Clara, California, USA) using 44 000 oligo probes approximately spaced at 35–40 kb intervals across the genome (Human Genome CGH microarray 44B kit; Agilent) or 244 000 oligo probes approximately spaced at 10 kb intervals across the genome (Human Genome CGH microarray 244B kit; Agilent). Commercial (Promega, Madison, Wisconsin, USA) or non-commercial female and male genomic DNAs were used as references. Hybridisation results were extracted with Feature extraction software and analysed with the DNA-analytics software by applying a Z-score or ADM 2 segmentation algorithm to identify chromosome aberrations. Copy-number gains and losses were determined from thresholds of 0.3 and −0.3, respectively. Aberrant signals obtained with three or more neighbouring oligonucleotides were considered indicative of genomic aberrations and further evaluated by FISH, multiplex PCR/liquid chromatography (MP-LC), quantitative PCR (qPCR) and/or QMPSF, unless they matched with a published DNA copy-number variant, as listed in the Database of Genomic Variants (http://projects.tcag.ca/variation/?source=hg18, version June 2008).

FISH analysis

Aberrations were validated by FISH experiments (see supplementary material online). FISH was performed with RP11-1006G2 (case 1), RP11-1147F22 (cases 2 and 3), CTD-2328P23, RP11-1147F22 and RP11-110I3 (case 5) and CTD-2328P23 (case 6).

Quantitative multiplex PCR of short fluorescent fragments

MEF2C exon 2 was PCR-amplified using the dye-labelled primers MEF2C-F (5′-CGTTAGATAGTGGGAACTGAGCTGTGCAAGT-3′) and MEF2C-R (5′-GATAGGGTTACGTTCATCCATAATCCTCGTAATC-3′). Exon 13 of HMBS located on chromosome 11 and exon 4 of MECP2 located on chromosome Xq28 were co-amplified as controls using the dye-labelled primers HMBS-F (5′-CGTTAGATAGACGGCTCAGATAGCATACAAG-3′) and HMBS-R (5′-GATAGGGTTAATGCCTACCAACTGTGGGTCA-3′) and MECP2-F (5′-CGTTAGATAGTTTCGCTCTAAAGTGGAGTTGAT-3′) and MECP2-R (5′-GATAGGGTTAGGGCTTCTTAGGTGGTTTCTG-3′), respectively. A detailed description of the QMPSF analyses is available in supplementary material online.

Multiplex PCR/liquid chromatography

MEF2C intron 2 was PCR-amplified using the following primers: MEF2C-MPLC-F (5′-TAATCCAGGAGCCACAGGTC-3′) and MEF2C-MPLC-R (5′-GAGAAAGAGCATTTAGGAGGG-3′). An additional fragment, corresponding to the HMBS gene located on chromosome 11, was co-amplified as a control (primers are available on request). A detailed description of the MP-LC analyses is available in supplementary material online.

qPCR

qPCR of MEF2C was performed using SYBR green (see supplementary material online). The RPPH1 gene was used for normalisation. MEF2C exon 1 was amplified with the primer MEF2C-E1-F (5′-TGTCAGGGTTTGGACAACAA-3′) and MEF2C-E1-R (5′-TTGGCAATTGAAACTCACCA-3′) and MEF2C exon 2 was amplified with MEF2C-E2-F (5′-CACATTTGAAGGTGCCAAAG-3′) and MEF2C-E2-R (5′-CAATCATTTGCCTTCCGTTC-3′).

Sequencing analysis of the MEF2C gene

The 10 coding exons of the MEF2C gene, including the exon–intron junctions, were PCR-amplified (primers are available on request) from 100 ng of genomic DNA in 50 µl containing 1.5 mM MgCl2, 75 mM Tris/HCl (pH 9 at 25°C), 20 mM (NH4)2SO4, 0.01% Tween 20, 50 pmol of each primer, 200 µM each dNTP and 2 units of Hot GoldStar (Eurogentec, Seraing, Belgium). PCR conditions include one cycle for 4 min at 94°C followed by 30 cycles at 94°C for 30 s, 58°C for 30 s, 72°C for 1 min, and one last cycle at 72°C for 5 min. The purified PCR products were then sequenced using a Ceq2000/8000 DNA sequencer (CEQ DTCS-Quick Start Kit; Beckman Coulter, Fullerton, California, USA).

Results

In the course of systematic molecular karyotyping of patients with syndromic mental retardation, we identified five 5q14 deletions and one 5q14 duplication. All patients with a 5q14 deletion (cases 1–5; table 1) presented with early and severe developmental delay and hypotonia. Speech was absent in all cases. None of the children was able to walk unaided. Stereotypic movements were present in three of the patients and absent in the two youngest children. Different types of epilepsy were observed in three of the patients, from well-controlled generalised seizures to early refractory tonicoclonic or myoclonic epilepsy. Miscellaneous dysmorphic facial features were present in all cases (figure 1), but some common features were noticed, eg, high and wide forehead, pronounced eyebrows, anteverted nostrils, short and prominent philtrum, down-turned corners of the mouth and small chin. In three cases, palpebral fissures were up-slanted.

Figure 1

Frontal and profile views of the five patients carrying a 5q14 microdeletion or MEF2C mutation. Note the high and wide forehead, thick eyebrows, short nose with anteverted nostrils, short and prominent philtrum, down-turned corners of the mouth and small chin. (A) Case 1, note “question mark ears” (constriction at the junction between the lower and the middle thirds of the pinna); (B) case 2; (C) case 3; (D) case 4; (E) case 5; (F) case 7 harbouring the MEF2C mutation.

All patients with the 5q14 deletion displayed MRI abnormalities (table 1), including either corpus callosum agenesis (2/5) or increased corpus callosum thickness and shortness (1/5), abnormal gyration (1/5), frontoparietal atrophy and enlarged pericerebral spaces (1/5), enlarged lateral ventricle (2/5) or enlarged fourth ventricle (2/5). As shown in figures 2 and 3, the 5q14 deletions measured approximately 2.68 Mb (case 1), 3.5 Mb (case 2), 8.8 Mb (case 3), 1.57 Mb (case 4) and 216 kb (case 5). In case 6, we detected a 5q14 duplication which was estimated to be 4.6 Mb in size. These 5q14 rearrangements were all confirmed by a second independent method, ie, QMPSF (case 1), MP-LC (cases 5 and 6), qPCR (case 4) and/or FISH analysis (cases 1, 2, 3, 5 and 6), and segregation analyses revealed that they all had occurred de novo (figure 4). The smallest rearrangement detected in case 5, presenting a Rett-like phenotype with developmental delay, poor eye contact, epilepsy, stereotypic hand–mouth movements, episodes of hyperventilation and apnoea, corresponded to a 216 kb deletion that encompassed a single gene, MEF2C, therefore restricting the minimal common deleted region to this gene.

Figure 2

Array-comparative genomic hybridisation profiles on chromosome 5 using Agilent 44 K or 244 K microarray (DNA analytics software display) showing the five microdeletions and the microduplication. In case 1, the 244 K microarray shows that the proximal and distal breakpoint of the deletion are respectively located between 86 939 816 and 87 005 072 and between 89 690 632 and 89 709 694. Using 44 K microarrays, the proximal breakpoints of the four other deletions and of the duplication were located between 87 770 283 and 88 051 970 in case 2, between 86 142 271 and 86 412 812 in case 3, between 88 185 407 and 88 268 343 in case 4, between 87 770 283 and 88 051 970 in case 5 and between 85 951 601 and 86 142 512 in case 6. Note that, in patient 4, the proximal breakpoint has been localised by qPCR, between 88 221 326 and 88 235 476. The distal breakpoints of these rearrangements are respectively located between 91 578 247 and 91 730 827 in case 2, between 95 315 261 and 95 494 937 in case 3, between 89 843 194 and 89 966 438 in case 4, between 88 268 402 and 88 629 033 in case 5 and between 90 712 814 and 90 731 163 in case 6. The Human Gene Assembly used to define the extensions of the deletion is Hg18.

Figure 3

Schematic representation of the 5q14 genomic region. The MEF2C gene is indicated by a black box. BAC clones appear above the genomic representation and black bars. The minimal extents of deletions and the duplication are, respectively, shown by grey boxes and a black box below the chromosome scheme. The maximal extents of the region implicated are shown by dotted lines. Note that the smallest copy-number variation detected is a deletion of a single gene, MEF2C, located from approximately 88 051 970 to 88 268 402 in Hg18 (in patient 5).

Figure 4

Confirmation of the 5q14 deletion by quantitative multiplex PCR of short fluorescent fragments (QMPSF) and fluorescence in situ hybridisation (FISH) in case 1. (A) Electropherogram in case 1 (in red) was superimposed on that of a normal female (in blue) by adjusting to the same level the peaks obtained for the HMBS and MECP2 control amplicons. Case 1 is a female. The y axis displays fluorescence in arbitrary units, and the x axis indicates the size in bp. Heterozygous 5q14 deletions are easily detected by a 50% reduction in the MEF2C peak in the patient compared with a normal control. (B) Results of two-colour FISH analysis in case 1 with RP11-1006G2 labelled with Spectrum Green in combination with a cri-du-chat syndrome probe used as control (5p15.2, Spectrum Green, 5q31 EGR1 Spectrum Orange; Abbott, Abbott Park, Illinois, USA). Note that RP11-1006G2 is lacking on one chromosome 5 in the patient and is present on both homologous chromosomes in the parents (arrows), demonstrating the de novo origin of the deletion. (C) Sequencing DNA chromatogram in case 7. The arrow indicates the c.683C>G mutation within exon 7 of MEF2C.

In cases 1, 2, 3 and 5, the deletion removed the MEF2C gene entirely. In case 4, qPCR showed a partial deletion of MEF2C removing the first exon, the breakpoint being located within intron 1. These observations led us to perform sequencing analysis of the MEF2C gene in five additional patients affected with severe mental retardation, in whom the diagnosis of the Rett phenotype had been suggested and for whom the array-CGH was normal. As shown in figure 4C, in one patient (case 7), we identified a point mutation within exon 7 (NM_002397.2:c.683C>G) predicted to result in a premature stop codon (p.Ser228X). Sequencing analysis performed in the parents revealed that the mutation had occurred de novo.

Discussion

We here report on five 5q14 submicroscopic deletions, one 5q14 duplication and one nonsense MEF2C mutation in seven unrelated children, all presenting with mental retardation.

The clinical significance of the 5q14.3 duplication remains uncertain and, despite its de novo occurrence, it may represent a benign variant. Such non-pathogenic duplications have been described, contrasting with the deleterious deletion mirror event.8 Alternatively, the 5q14.3 duplication may be responsible for a mild phenotype distinct from that resulting from the deletions. Indeed, 7q11.23, 16p13.11 and 17p11.2 duplications result in a relatively mild phenotype compared with deletions.9–11 Study of additional patients will be required to determine the pathogenic effect of the 5q14.3 duplication.

In the five cases with a 5q14.3 microdeletion, the developmental delay was severe, associated with hypotonia, poor eye contact and stereotypic hand movements in three of them. Moreover, in cases 4 and 5, classical Rett syndrome had initially been suspected because of repetitive clapping hand–mouth movements. However, the two latter patients did not display any period of normal development or progressive loss of motor or communication skills. Epilepsy was present in three cases. In one case, refractory myoclonic epilepsy with atypical spasms beginning before 6 months of age led us to consider initially the hypothesis of an early-onset seizure variant of Rett syndrome due to CDKL5 mutation. However, the presence of dysmorphic facial features in all five cases was strongly suggestive of a chromosomal aberration.

Only a few interstitial 5q14 deletions detected by standard chromosomal analysis had previously been reported12–18 and shown to be associated with severe to moderate mental retardation, growth retardation, deep hypotonia, facial dysmorphism and malformations. The facial features included prominent forehead, epicanthal folds, brachycephaly, hypertelorism, flat nasal bridge, anteverted nostrils, abnormal ears and short neck. Malformations were miscellaneous and included renal abnormalities, cleft palate, club feet, heart defect and dislocated hips. Among these cases, no clinical phenotype could actually be delineated, and epilepsy had not previously been reported. The boundaries of the deletions in the published cases cannot really be compared because of the low resolution level of chromosome analyses.

The three 5q14.3–q15 deletion cases recently reported by Cardoso et al8 and an additional case reported in the Decipher database partially overlap with the 5q14.3 deletions that we report here. The critical regions, defined by the latter cases, are different. The clinical features common to the four previously described patients and the five described here include mental retardation, seizures and dysmorphic facial features. However, in contrast with the patients described by Cardoso et al,8 none of ours had periventricular heterotopia. Still, in our patients, a developmental defect of the neuronal migration with ectopic neurons cannot be formally excluded as a cause of the severe and early epilepsy associated with mental retardation.

In our patients, the 5q14 deletion sizes ranged from 216 kb to 8.8 Mb. The absence of low copy repeat sequences flanking the breakpoints, as well as the absence of recurrent breakpoints, suggested that these rearrangements did not result from non-allelic homologous recombination or involve another mechanism such as non-homologous end joining of DNA breaks. The detection of a 216 kb deletion, removing only the MEF2C gene, is the first evidence that haploinsufficiency of this gene contributes to the 5q14 microdeletion syndrome. MEF2C deletion is not a common copy number variant (http://projects.tcag.ca/variation/?source=hg18), and the de novo origin of the five deletions reported in this study is in agreement with their causal role in the patient phenotypes. Although we cannot formally exclude that haploinsufficiency of other 5q14 genes contribute to the phenotype of patients harbouring larger deletions, the fact that the five patients reported in this study with different deletion sizes showed striking phenotypic similarities (table 1) strongly suggests that the phenotype is mainly due to the MEF2C deletion. Finally, the identification of an inactivation point mutation within the MEF2C gene in case 7 constitutes a key argument that haploinsufficiency of this gene results in mental retardation. The causal role of the MEF2C alteration in the phenotype observed is in agreement with the biological function of the MEF2C protein and the murine models of MEF2C inactivation. Indeed, myocyte enhancer factor 2 (MEF2) transcription factors act in the brain as neurogenesis effectors which regulate excitatory synapse number,19 dendrite morphogenesis and differentiation of postsynaptic structures.20 The role of MEF2 proteins in synaptic plasticity is consistent with a role in learning and memory. MEF2C is the predominant isoform in the developing cerebrocortex and is highly expressed in frontal cortex, entorhinal cortex, cerebellum, dentate gyrus and amygdale.21 22 Conditional Mef2c-null mice were generated, as knockout of the Mef2c gene is embryonic lethal.23 24 Late embryonic deletion of Mef2c in the forebrain causes hippocampus-dependent learning and memory impairment associated with a dramatic increase in the number of excitatory synapses.24 The role of Mef2c in synaptic plasticity in mice, limiting the excessive increase in the number of excitatory synapses, is consistent with its possible involvement in seizures in humans. In mutant mice with earlier embryonic deletions, abnormal aggregation and compaction of neurons migrating into the lower layers of the neocortex during development were observed.23 As a consequence, the cortical plate in postnatal/adult neocortex in these conditional Mef2c-null mice displays disorganisation, and the neurons exhibit immature electrophysiological properties characteristic of an immature neuronal network. These murine phenotypes provide convincing arguments for MEF2C haploinsufficiency being the cause of mental retardation in all our cases and epilepsy in four of them. Furthermore, the role of MEF2C in neuronal migration is of particular interest with regard to the periventricular heterotopia described by Cardoso et al.8 Although MEF2C was deleted in only one of their three cases, a position effect on MEF2C of the two other deletions cannot be excluded. This hypothesis is emphasised by the report, in a patient sharing a strikingly similar phenotype, of a de novo balanced translocation between chromosomes 5 and 8, the breakpoint being located near the MEF2C gene.25 Moreover, Mef2c-null mice display behavioural phenotypes with abnormal anxiety, decreased cognitive function, and marked paw wringing/clasping stereotypy, resulting in a Rett-like phenotype as observed in mutant Mecp2 mouse models.23 24 26 In addition, according to the Transfac Matrix Database, the Mecp2 gene contains multiple putative MEF2-binding sites. Another gene, DIA1 (deleted in autism-1 or C3orf58), containing putative MEF2-binding sites has recently been implicated in autism and epilepsy.27 Indeed, a homozygous deletion of DIA1 was found in a patient with striking resemblance to patients with an MEF2C deletion, ie, early seizures, poor eye contact, absent speech and stereotypic movements. Additional findings are needed to assess the role of MEF2C, DIA1 and MECP2 in a common biological pathway essential for normal neurological development.

In conclusion, our results indicate that MEF2C haploinsufficiency caused by either 5q14.3 microdeletion or mutation is responsible for a severe mental retardation with stereotypic movements, epilepsy and/or cerebral malformations.

Acknowledgments

We are grateful to the patients and their families who participated in this study.

References

Supplementary materials

Footnotes

  • Supplementary methods are published online at http://jmg.bmj.com/content/vol47/issue1

  • Competing interests None.

  • Ethics approval Obtained.

  • Patient consent Parental consent obtained.

  • Provenance and Peer review Not commissioned; externally peer reviewed.