Article Text

Original article
Agenesis of corpus callosum and optic nerve hypoplasia due to mutations in SLC25A1 encoding the mitochondrial citrate transporter
  1. Simon Edvardson1,
  2. Vito Porcelli2,
  3. Chaim Jalas3,
  4. Devorah Soiferman1,
  5. Yuval Kellner4,
  6. Avraham Shaag1,
  7. Stanley H Korman1,
  8. Ciro Leonardo Pierri2,
  9. Pasquale Scarcia2,
  10. Nitay D Fraenkel5,
  11. Reeval Segel6,
  12. Abraham Schechter7,
  13. Ayala Frumkin1,
  14. Ophry Pines4,
  15. Ann Saada1,
  16. Luigi Palmieri2,8,
  17. Orly Elpeleg1
  1. 1Monique and Jacques Roboh Department of Genetic Research, Hadassah, Hebrew University Medical Center, Jerusalem, Israel
  2. 2Department of Biosciences, Biotechnology and Biopharmaceutics, Laboratory of Biochemistry and Molecular Biology, University of Bari Aldo Moro, Bari, Italy
  3. 3Bonei Olam, Center for Rare Jewish Genetic Disorders, Brooklyn, New York, USA
  4. 4Department of Microbiology and Molecular Genetics, IMRIC, Faculty of Medicine, Hebrew University, Jerusalem, Israel
  5. 5Department of Respiratory Rehabilitation, Alyn Hospital, Jerusalem, Israel
  6. 6Institute of Medical Genetics, Shaare-Zedek Medical Center, Jerusalem, Israel
  7. 7Meuhedet Health Services, Jerusalem, Israel
  8. 8CNR Institute of Biomembranes and Bioenergetics, Bari, Italy
  1. Correspondence to Professor Orly Elpeleg, Monique and Jacques Roboh Department of Genetic Research, Hadassah, Hebrew University Medical Center, Ein karem, Jerusalem 91120, Israel; elpeleg{at}, Professor Luigi Palmieri, Department of Biosciences, Biotechnology and Biopharmaceutics, University of Bari Aldo Moro, Bari 70125, Italy; luigi.palmieri{at}


Background Agenesis of corpus callosum has been associated with several defects of the mitochondrial respiratory chain and the citric acid cycle. We now report the results of the biochemical and molecular studies of a patient with severe neurodevelopmental disease manifesting by agenesis of corpus callosum and optic nerve hypoplasia.

Methods and results A mitochondrial disease was suspected in this patient based on the prominent excretion of 2-hydroxyglutaric acid and Krebs cycle intermediates in urine and the finding of increased reactive oxygen species content and decreased mitochondrial membrane potential in her fibroblasts. Whole exome sequencing disclosed compound heterozygosity for two pathogenic variants in the SLC25A1 gene, encoding the mitochondrial citrate transporter. These variants, G130D and R282H, segregated in the family and were extremely rare in controls. The mutated residues were highly conserved throughout evolution and in silico modeling investigations indicated that the mutations would have a deleterious effect on protein function, affecting either substrate binding to the transporter or its translocation mechanism. These predictions were validated by the observation that a yeast strain harbouring the mutations at equivalent positions in the orthologous protein exhibited a growth defect under stress conditions and by the loss of activity of citrate transport by the mutated proteins reconstituted into liposomes.

Conclusions We report for the first time a patient with a mitochondrial citrate carrier deficiency. Our data support a role for citric acid cycle defects in agenesis of corpus callosum as already reported in patients with aconitase or fumarate hydratase deficiency.

  • Molecular Genetics
  • Neurology
View Full Text

Statistics from


Agenesis of corpus callosum (ACC) is a birth defect that occurs in over 50 different human congenital syndromes. The molecular basis of ACC, either isolated or syndromic, has been partly elucidated and involves a large number of genes encoding transcription factors, guidance molecules and their receptors, intracellular signalling molecules, growth factors, and patterning molecules reflecting the complexity of corpus callosum development.1 Indeed, the formation of this structure requires the presence of telencephalic hemispheres and their fusion in dorsal regions, where the callosal axons will grow and extend upon, eventually crossing, the midline. This fusion is governed by proteins expressed by the midline zipper glia. Pioneering callosal axons originating from the cingulate cortex create a pathway in which callosal axons derived from neocortical neurons will grow.2 ,3 Thus, callosal axons are required to grow over large distances to reach their final target and their path is under the influence of guidance factors which repel axons and of cues from midline glial structures. Not unexpectedly, this process is energy consuming and ACC, complete or partial, has been reported in several defects of the mitochondrial respiratory chain and the citric acid cycle including pyruvate dehydrogenase (E1α) deficiency, SLC25A19 (thiamine pyrophosphate transporter) defect, aconitase deficiency, fumarase deficiency, complex I assembly defect, and mitochondrial translation defects.4–10 We now report a defect in the mitochondrial citrate transporter as a new cause of ACC, suggesting that ACC is a common manifestation of impaired mitochondrial metabolism.

Patient and methods

Patient history

The subject of the present study is an 18-month-old female, the third child to non-consanguineous Ashkenazi-Jewish parents. The pregnancy, not monitored by prenatal ultrasound, ended at term by an uneventful delivery with the birth of a 3.2 kg baby girl with a 1 min Apgar score of 9. At the age of 10 days she was admitted because of poor sucking and apathy. On admission her weight was 2.8 kg and there was marked hypotonia but no syndromatic stigmata. Because of erratic breathing and prolonged apnoeas mechanical ventilation was initiated. The respiratory events were not considered to be epileptic in nature as several electroencephalograms (EEGs) were normal, they were not accompanied by other paroxysmal manifestations, and other forms of seizures were not noted at the time. In view of the persistence of her symptoms, feeding gastrostomy and tracheostomy were inserted and the patient was discharged. On follow-up a near total lack of psychomotor development was noted; generalised epilepsy became evident at 5 months of age, mainly manifesting as tonic-clonic seizures which responded favourably to the administration of clonazepam. At 18 months the patient could tolerate 4–6 h of unassisted breathing. There were no spontaneous voluntary movements and eye-contact was minimal. Ophthalmologic examination revealed hypoplastic optic nerves. Weight and length have been increasing on the third centile. Head circumference, initially at the third centile, has been plateauing since 10 months of age and at 18 months was 42.2 cm (−3 SD), consistent with acquired microcephaly. There have been no metabolic decompensation episodes and no indications of other organ involvement.

Basic laboratory investigation was normal including blood count, liver and renal functions, plasma lactate, creatine kinase, and amino acid profile. Urinary organic acid analysis revealed a large peak of 2-hydroxyglutaric acid and Krebs cycle intermediates. Endocrine studies were within the normal range, indicating normal hypophyseal function. Echocardiogram, performed in the neonatal period, disclosed a small ventricular septal defect and a patent foramen ovale. Brain MRI at 3 weeks of age demonstrated a complete ACC but was otherwise normal for age. Brainstem auditory evoked potential testing revealed complete absence of waves beyond the first wave which occurred at a normal latency and thresholds. Regrettably, muscle biopsy was uninformative because of poor tissue quality and cultured skin fibroblasts were the only tissue available for studies.

Studies on the patient's cultured skin fibroblasts

Cell viability, oxygen free radicals (reactive oxygen species (ROS)) production, ATP content and mitochondrial membrane potential (ΔΨ) were assessed essentially as previously described.11 Briefly, cells were seeded on 96 microtitre wells in glucose containing medium overnight. The following day the medium was changed to glucose-free, galactose containing medium (GAL). After 72 h, growth was assessed by methylene blue (MB) stain. ROS production was measured by 2′,7′-dichlorodihydrofluorescein diacetate and ATP content by luciferin-luciferase relative to MB. ΔΨ was estimated by tetramethylrhodamine ethyl ester relative to mitotracker green. Luminescence, fluorescence and absorbance measurements were performed with a Synergy HT microplate reader (Bio-Tek instruments, Vinoosky, Vermont, USA).

Cytogenetic analysis was performed with Affymetrix CytoScan HD array according to the manufacturer instructions.

Whole exome analysis

Exonic sequences were enriched in the DNA sample of the patient using SureSelect Human All Exon 50 Mb Kit (Agilent Technologies, Santa Clara, California, USA). Sequences were determined by HiSeq2000 (Illumina, San Diego, California, USA) as 100 bp paired-end runs. Data analysis included read alignment and variant calling was performed by DNAnexus software (Palo Alto, California, USA) using the default parameters with the human genome assembly hg19 (GRCh37) as reference, as previously described.12 Parental consent was given for all the studies where cells or DNA of the patient and other family members were used. The study was performed with the approval of the ethical committee of Hadassah Medical Center.

Comparative modeling and docking investigations

Multiple sequence alignment of citrate carrier (CIC) orthologs from various species and other mitochondrial carriers was obtained using ClustalW.13 The sequence of the bovine ADP/ATP carrier (BtAAC1)14 was introduced in the alignment in order to use the secondary structure information of BtAAC1 to weigh gap insertions in the alignment and locate them preferentially in correspondence of loops.15 Modeller16 was used to calculate three dimensional structural models of human CIC using the structure of the bovine BtAAC1 (protein data bank accession code: 1okc) as template for the comparative modeling. The structural properties of the comparative models of wild-type (WT) human CIC and of the G130D and R282H variants were evaluated using the biochemical/computational tools of the WHAT IF Web server.17 Q-site Finder18 was used to predict the potential binding sites of the structures with the best energy function. For docking analysis, citrate ligand was docked into the predicted binding sites using Autodock

Study of the pathogenicity of the mutations in yeast

Saccharomyces cerevisiae strains BY4742 (Mat α; his3Δ1; leu2Δ0; lys2Δ0; ura3Δ0) and Δctp1 (Mat α; his3Δ1; leu2Δ0; lys2Δ0; ura3Δ0YBR291c::kanMX4) were obtained from Euroscarf. CTP1, the yeast orthologue of SLC25A1, was cloned into plasmid YEp51 under the control of the GAL promoter. A double mutant G117D/R276H corresponding to the human mutations G130D/R282H was created using the QuickChange site-directed mutagenesis kit II (Stratagene). Strains harbouring the appropriate plasmids were grown in synthetic depleted medium containing 0.67% (w/v) yeast nitrogen base, 2% galactose (w/v), supplemented with the appropriate amino acids, and 2% agar was added to plates. Where indicated, hydroxyurea (HU) was added to a final concentration of 200 mM.

Construction of the expression plasmids and bacterial expression

The coding sequence of the yeast CIC (CTP1, NM_001178639.1) was amplified from S. cerevisiae genomic DNA via PCR and the mutations were introduced by site-directed mutagenesis (for details see online supplementary material). Ctp1p and its mutated forms (G117D and R276H) were overexpressed as inclusion bodies in the cytosol of Escherichia coli strain BL21 (DE3). Inclusion bodies were purified on a sucrose density gradient and washed at 4°C, first with Tris-EDTA buffer (10 mM Tris–HCl, 1 mM EDTA, pH 7.0), then twice with a buffer containing Triton X-114 (3%, w/v), 1 mM EDTA, 20 mM Na2SO4 and 10 mM PIPES pH 7.0, and finally with the Tris-EDTA buffer pH 7.0. The proteins were solubilised in 1.7% sarkosyl (w/vol). Eventual small residues were removed by centrifugation (20800×g for 10 min at 4°C).

Reconstitution of the recombinant proteins into liposomes and transport assays

The recombinant proteins in sarkosyl were reconstituted by cyclic removal of detergent as described.20 The reconstitution mixture consisted of protein solution (25 µl, about 10 µg), 10% Triton X-114 (70 µl), 10 mM citrate (35 µl), cardiolipin (0.7 mg), 10 mM PIPES/NaOH pH 7.0 and water (final volume 700 µl). The mixture was recycled 13-fold through an Amberlite column pre-equilibrated with 10 mM PIPES-NaOH (pH 7.0) and 50 mM NaCl. All operations were performed at 4°C except the passages through Amberlite, which were carried out at room temperature. Approximately 20% of WT or mutated CIC proteins were reconstituted. External citrate was removed from proteoliposomes on Sephadex G-75 columns pre-equilibrated with 50 mM NaCl and 10 mM PIPES at pH 7.0 (buffer A). Transport at 25°C was started by adding 0.1 mM (14C) citrate (from PerkinElmer) to proteoliposomes, and terminated by addition of 20 mM pyridoxal 5′-phosphate and 20 mM of bathophenanthroline. In controls, inhibitors were added with the labelled substrate. The external substrate was removed by Sephadex G-75 columns pre-equilibrated with buffer A, and the entrapped radioactivity was counted. The experimental values were corrected by subtracting control values and the initial transport rate was calculated from the time course of substrate uptake by proteoliposomes.20


In view of the severe neurodevelopmental disorder and the organic acid profile, mitochondrial respiratory chain defect was suspected in this patient. The study of mitochondrial functions in cultured skin fibroblasts disclosed increased ROS production and partially decreased mitochondrial membrane potential in the patient cells compared to the control cells. ATP content was comparable between the patient and the control cells (figure 1A–C). These findings indicated a functional respiratory chain, increased oxidative stress with possibly decreased ΔΨ.

Figure 1

Measurement of reactive oxygen species (ROS), ATP, and mitochondrial membrane potential (ΔΨ) in control and patient fibroblasts. Fibroblasts from controls (n=3) and the patient were grown in microtitre wells in GAL for 72 h. (A) ROS was measured by dichlorodihydrofluorescein diacetate and expressed as relative fluorescent units divided by the amount of cells measured by methylene blue (MB). (B) ATP measured by luciferin-luciferase is expressed as relative luminescence units divided by the amount of cells measured by MB. (C) ΔΨ is presented as the ratio of tetramethylrhodamine ethyl ester red to mitotracker green fluorescence. Values are presented as mean of at least triplicates±SEM *p<0.05, Student's t test.

In order to identify the disease causing gene, we first searched for a chromosomal copy number variant and for homozygous genomic regions using a dense DNA marker array. Normal female karyotype was seen and no chromosomal rearrangements or homozygous genomic regions >2.5 Mb were detected.

We then looked for single nucleotide variants and indels by whole exome analysis. This resulted in 104 235 672 reads of which 88.06% were mapped confidently. Following alignment to a reference sequence, we removed variants with a minor allele frequency >0.02 which were present in dbSNP137, in the in-house variant database or in the Exome Variant Server, NHLBI Exome Sequencing Project (ESP), Seattle, Washington, USA (, accessed June 2012). We also removed variants having a reading depth < ×7, synonymous and deep intronic variants. Eight genes carrying homozygous and one gene carrying two compound heterozygous variants remained (see online supplementary table S1) but the conservation of all the homozygous variants was low, none of the genes was targeted to the mitochondria, and none predicted pathogenic by Mutation Taster software.21 The remaining two SLC25A1 (NM_005984) heterozygous variants, chr. 22:19163734 C->T resulting in Arg282His (R282H) and chr.22:19165292 C->T resulting in Gly130Asp (G130D), were further studied. These mutations segregated in the family (see online supplementary figure S1) and determination of their carrier rate among 1032 anonymous Ashkenazi Jews using allele specific discrimination real-time PCR revealed that none carried the G130D mutation whereas three were heterozygous for the R282H mutation.

SLC25A1 consists of eight exons which encode the mitochondrial citrate transporter, a 311 amino acid protein. In order to study the effect of the mutations on the protein function, we first performed in silico comparative modeling and docking investigations. Gly130 is located in transmembrane helix 3 at the PG-level 1 (figure 2B), a mitochondrial carrier region involved in conformational changes occurring during the translocation mechanism,22 and is also conserved in several other mitochondrial carriers (figure 2A). Arg282 is located in transmembrane helix 5, three residues below the contact point III of the putative common substrate binding site of the mitochondrial carrier family (figure 2B).22 Its position together with the observation that Arg282 is not conserved in other mitochondrial carriers suggests a role in substrate specificity.23 ,24 The Q-site Finder tool6 was used to predict the binding sites in the WT and in the mutated CIC proteins. It was found that the predicted binding sites contained charged residues of the even helices at the middle point of the carrier cavity (figure 2B), as previously proposed.24 A gridbox involving residues of the predicted binding region was built in order to investigate the binding of citrate at this region in the WT and in the mutated proteins. We found that the substrate binds the WT CIC by interacting with residues belonging to the common binding site including R282, R285 and K97 (figure 3A). In the G130D protein (figure 3B), the binding site is apparently unaffected. However, the mutation leads to the replacement of the small neutral amino acid glycine with the bulkier negatively charged aspartic acid which protrudes towards transmembrane helix 4, introducing an important perturbation of local secondary structures. In the mutated R282H protein, the substrate cannot reach the proposed binding site on the bottom of the carrier cavity and is trapped at the level of K97 (see figure 3C) and now interacts with R198 (not shown).

Figure 2

(A) Sequence alignment of mitochondrial citrate carrier (CIC) from different organisms (upper) and other members of the mitochondrial carrier family. Accession numbers for each species are given (B) Structural homology model of human CIC and docking of citrate. Prolines (magenta spheres) and glycines (yellow spheres) of the PG levels (or the corresponding residues) are reported as reference residues that surround the binding region (Palmieri and Pierri22). The citrate ligand is shown in green sticks. This figure is only reproduced in colour in the online version.

Figure 3

Citrate docking to the binding region of wild type (A), G130D (B), and R282H (C) citrate carrier proteins. The citrate ligand is shown in green sticks. Positively charged residues (K97, R282, and R285) involved in the binding of the citrate are reported in cyan sticks. Mutated residues are shown in red sticks. This figure is only reproduced in colour in the online version.

To test the functional relevance of the identified mutations we took advantage of the absolute interspecies conservation of both G130 and R282 and the high similarity between the human CIC and its fungal ortholog, pCtp1. S. cerevisae has already proved useful for studying the functional effect of pathogenic mutations in human mitochondrial transporters, introducing them at equivalent positions in the yeast orthologous protein.25 ,26

In preliminary experiments, we found a distinct difference in the growth of a yeast deletion strain devoid of the endogenous citrate transporter (Δctp1) versus the WT in galactose media (respiration required), in the presence of HU (cell cycle and DNA damage) at 37°C (heat stress) as shown in figure 4. Under these stress conditions, yeast cells expressing a double mutant G117D/R276H, corresponding to the human mutations G130D/R282H, exhibited a severe growth phenotype when compared to the WT gene (figure 4). These results corroborated the in silico prediction, indicating functional significance of the SLC25A1 mutations.

Figure 4

CTP1-G117D+R276H double mutant harbouring the mutations identified in human SLC25A1 exhibits a growth defect under stress conditions. Wild-type yeast and a mutant deleted for the CTP1 gene (Δctp1) were transformed with the indicated plasmids. Cultures were streaked onto galactose medium plates containing 200 mM hydroxyurea, which were incubated at 37°C for 5 days. This result represents three similar experiments.

Kinetic analysis of R276H and G117D mutants compared to WT Ctp1p was performed upon reconstitution of purified recombinant proteins into liposomes. As shown in figure 5, the substitution of Arg276, corresponding to Arg282 in human CIC, with histidine caused a complete loss of activity in accordance with the role of this residue in substrate binding. Similarly, the G117D mutant protein also exhibited a marked decrease in transport activity (figure 5) consistent with the predicted involvement of Gly117, corresponding to Gly123 in human CIC, in protein conformational changes occurring during substrate translocation.

Figure 5

Functional characterisation of wild-type (WT) and mutated Ctp1p forms. The uptake rate of (14C) citrate was measured by adding 0.1 mM of (14C) citrate to proteoliposomes reconstituted with purified WT Ctp1p or with the mutated G117D and R276H forms. The proteoliposomes were preloaded internally with 10 mM of citrate. The means and SDs from five independent experiments are shown.


Our patient suffered from a severe neurodevelopmental syndrome accompanied by the radiological findings of ACC and optic nerve hypoplasia and by a distinctive organic acid profile in urine with prominent 2-hydroxyglutaric acid and Krebs cycle intermediates. The finding of increased ROS content and partially decreased mitochondrial membrane potential further indicated that the primary defect in this patient resided in the mitochondrial respiratory chain. Whole exome sequencing of the patient's DNA disclosed a large number of potentially pathogenic variants and after a careful in silico filtering process we focused on the two variants in the SLC25A1 gene. The fact that the mutated gene encoded a mitochondrial transporter added confidence to these genomic findings. The deleterious effect of the mutations was first predicted in silico by comparative modeling and substrate docking studies and then demonstrated by the defective growth phenotype of a yeast strain harbouring the mutations and direct transport measurements in proteoliposomes. The citrate transporter catalyses the efflux of citrate/isocitrate from the mitochondrial matrix in an electroneutral counter-exchange for cytosolic malate. Thus, the CIC is a key component of the citrate–malate shuttle, as the exported citrate is cleaved in the cytosol to acetyl-coenzyme A and oxaloacetate by ATP-citrate lyase. The acetyl-coenzyme A is subsequently incorporated into fatty acid synthesis27 and the oxaloacetate is reduced to malate (providing cytosolic NAD+ for glycolysis) which is recycled back to mitochondria. Alternatively, malate can be converted to pyruvate by the malic enzyme producing cytosolic NADPH. Isocitrate can also generate NAPDH in the cytosol via isocitrate dehydrogenase which has been proposed to be a component of the isocitrate/oxoglutarate NADPH redox shuttle (see Palmieri27 and references therein), which involves the citrate transporter (exchanging isocitrate and malate) and the oxoglutarate/malate transporter (exchanging malate and oxoglutarate). Furthermore, by regulating glycolysis at the level of phosphofruktokinase 1, citrate also directs carbon flux to the pentose pathway which is a major producer of NADPH molecules. A synergistic role for NADP dependent isocitrate dehydrogenase and glucose-6-phosphate dehydrogenase in NADPH production and resistance to stress conditions most likely due to reduced glutathione generation is well documented in yeast28 and may at least in part explain the Δctp1 phenotype. Thus impaired mitochondrial citrate transport may lead to low cytosolic NADPH content in the patient cells which may underlie the abnormally high oxidative stress, even in the presence of low membrane potential.

The mitochondrial citrate transporter is believed to play an important role in fatty acid and sterol synthesis, gluconeogenesis and glycolysis.27 Recently, the mitochondrial CIC has also been proposed to play a role in the maintenance of chromosome integrity29 and in the regulation of autophagy.30 The identification of a patient with deficiency of the mitochondrial citrate transporter offers a unique insight into the importance of this protein in vivo. The lack of lactic acidaemia and hypoglycaemia in the early neonatal period, when the patient was failing to thrive, and her normal fat distribution, suggests that glycolysis, gluconeogenesis and fatty acid synthesis are largely unaffected by the defect. Rather, the severe aberration of early brain development, manifesting as ACC, optic nerve hypoplasia, epilepsy and profound psychomotor retardation, underscores the important consequences of impaired citrate/isocitrate export from mitochondria which could result in decreased citric acid cycle flux, low NADPH content, oxidative stress, and ROS toxicity. In this context, it should be noted that normal ATP content in fibroblasts can be observed even in the presence of severe defects in ATP production in other tissues of the patient, as observed in other mitochondrial transporters.31 The prominent excretion of 2-hydroxyglutaric acid is also supportive of depletion of cytosolic citrate/isocitrate in our patient. In fact isocitrate dehydrogenase has the ability to produce 2-hydroxyglutarate from 2-oxoglutarate.32 Although this activity is low compared to isocitrate oxidation, we suggest that in the absence of citrate/isocitrate in the cytosol, 2-hydoxyglutarate production becomes predominant and its presence could serve as a biomarker for this condition. Of note, the association of 2-hydroxyglutaric aciduria with the absence of corpus callosum has been described previously.33 Thus ACC, already reported in patients with aconitase or fumarate hydratase deficiency, seems to be part of the clinical presentation in defects involving the citric acid cycle. The preferential involvement of the brain could be attributed to the a priori low mitochondrial citrate transport activity in the brain compared, for example, to the liver.

In summary, a neurodevelopmental disorder manifesting as profound psychomotor retardation, hypotonia, epilepsy, postnatal microcephaly, sensorineural deafness, ACC, and hypoplastic optic nerves is associated with impaired activity of the mitochondrial citrate transporter. The abnormally increased oxidative stress and the increased excretion of 2-hydroxyglutaric acid could be secondary to impaired citrate/isocitrate export from mitochondria.


Parts of this work were supported by Rabbi Bochner of the Bonei Olam organisation and a research grant of the Manackerman Charitable Trust, UK. We thank Shamir Zenvirt and Rachel Dahan for excellent technical assistance. Reproductive Medicine Associates of New Jersey are acknowledged for genotyping services.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • SE and VP contributed equally.

  • Contributors SE, AS, OP, SK, LP and OE conceived and designed the experiments; VP, DS, YK, AS, PS and CLP performed the experiments; SE, AS, OP, SK, LP, OE, AS, SK, AF, VP, PS, CLP and CJ analysed the data; RS, NDF and AS contributed reagents/materials/analysis tools; SE, AS, OP, SK, LP, VP, CLP and OE wrote the paper; SE, AS, RS and NDF undertook patient management, collection of samples, and delineation of the phenotype.

  • Competing interests None.

  • Funding This work was supported by the Italian Human ProteomeNet (Grant no. RBRN07BMCT_009 MIUR) and the Center of Excellence in Comparative Genomics (CEGBA).

  • Ethics approval Hadassah ethical committee.

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

  • Data sharing statement The CEL file of the chromosomal microarray and the exome FASTQ file are available upon request.

  • Patient consent Parental consent obtained.

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.