Article Text


Original article
Phenotypic spectrum associated with CASK loss-of-function mutations
  1. Ute Moog1,
  2. Kerstin Kutsche2,
  3. Fanny Kortüm2,
  4. Bettina Chilian2,
  5. Tatjana Bierhals2,
  6. Neophytos Apeshiotis3,
  7. Stefanie Balg4,
  8. Nicolas Chassaing5,
  9. Christine Coubes6,
  10. Soma Das7,
  11. Hartmut Engels8,
  12. Hilde Van Esch9,
  13. Ute Grasshoff10,
  14. Marisol Heise2,
  15. Bertrand Isidor11,
  16. Joanna Jarvis12,
  17. Udo Koehler4,
  18. Thomas Martin13,
  19. Barbara Oehl-Jaschkowitz13,
  20. Els Ortibus14,
  21. Daniela T Pilz15,
  22. Prab Prabhakar16,
  23. Gudrun Rappold1,
  24. Isabella Rau2,
  25. Günther Rettenberger17,
  26. Gregor Schlüter18,
  27. Richard H Scott19,
  28. Moonef Shoukier20,
  29. Eva Wohlleber8,
  30. Birgit Zirn21,
  31. William B Dobyns22,
  32. Gökhan Uyanik2
  1. 1Institute of Human Genetics, Heidelberg University, Heidelberg, Germany
  2. 2Institute of Human Genetics, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
  3. 3Privatpraxis für Humangenetik, Braunschweig, Germany
  4. 4Medizinisch Genetisches Zentrum, MGZ München, München, Germany
  5. 5Medical Genetics Department, Toulouse University Hospital and UPS III EA4555, Toulouse, France
  6. 6Genetic Services, A. de Villeneuve Hospital, Montpellier, France
  7. 7Department of Human Genetics, University of Chicago, Chicago, Illinois, USA
  8. 8Institute of Human Genetics, University of Bonn, Bonn, Germany
  9. 9Centre for Human Genetics, University Hospitals Leuven, Leuven, Belgium
  10. 10Institute of Human Genetics, Department of Medical Genetics, Tübingen University, Tübingen, Germany
  11. 11Clinical Genetics Unit, University Hospital Nantes, Nantes, France
  12. 12Clinical Genetics Unit, Birmingham Women's Hospital, Birmingham, UK
  13. 13Gemeinschaftspraxis für Humangenetik, Homburg, Germany
  14. 14Department of Pediatric Neurology, University Hospitals Leuven, Leuven, Belgium
  15. 15Institute of Medical Genetics, University Hospital of Wales, Cardiff, UK
  16. 16Department of Neurology, Great Ormond Street Hospital for Children, London, UK
  17. 17Genetikum, Neu-Ulm, Germany
  18. 18Pränatalmedizin und Genetik Nürnberg/Bayreuth (MVZ), Nürnberg, Germany
  19. 19Department of Clinical Genetics, Great Ormond Street Hospital, London and Institute of Child Health, London, UK
  20. 20Institute of Human Genetics, University of Goettingen, Goettingen, Germany
  21. 21Department of Neuropediatrics, University of Goettingen, Goettingen, Germany
  22. 22Departments of Pediatrics and Neurology, and Center for Integrative Brain Research - Seattle Children's Research Institute, Seattle, Washington, USA
  1. Correspondence to Gökhan Uyanik, Institute of Human Genetics, University Medical Center Hamburg-Eppendorf, Martinistraße 52, 20246 Hamburg, Germany; g.uyanik{at}


Background Heterozygous mutations in the CASK gene in Xp11.4 have been shown to be associated with a distinct brain malformation phenotype in females, including disproportionate pontine and cerebellar hypoplasia.

Methods The study characterised the CASK alteration in 20 new female patients by molecular karyotyping, fluorescence in situ hybridisation, sequencing, reverse transcriptase (RT) and/or quantitative real-time PCR. Clinical and brain imaging data of a total of 25 patients were reviewed.

Results 11 submicroscopic copy number alterations, including nine deletions of ∼11 kb to 4.5 Mb and two duplications, all covering (part of) CASK, four splice, four nonsense, and one 1 bp deletion are reported. These heterozygous CASK mutations most likely lead to a null allele. Brain imaging consistently showed diffuse brainstem and cerebellar hypoplasia with a dilated fourth ventricle, but of remarkably varying degrees. Analysis of 20 patients in this study, and five previously reported patients, revealed a core clinical phenotype comprising severe developmental delay/intellectual disability, severe postnatal microcephaly, often associated with growth retardation, (axial) hypotonia with or without hypertonia of extremities, optic nerve hypoplasia, and/or other eye abnormalities. A recognisable facial phenotype emerged, including prominent and broad nasal bridge and tip, small or short nose, long philtrum, small chin, and/or large ears.

Conclusions These findings define the phenotypic spectrum associated with CASK loss-of-function mutations. The combination of developmental and brain imaging features together with mild facial dysmorphism is highly suggestive of this disorder and should prompt subsequent testing of the CASK gene.

  • CASK
  • microcephaly
  • intellectual disability
  • pontocerebellar hypoplasia
  • X linked mental retardation
  • other neurology
  • molecular genetics
  • chromosomal
  • clinical genetics
  • developmental, genetics
  • visual development
  • academic medicine
  • diagnostics
  • diagnostics tests
  • copy-number
  • epigenetics
  • epilepsy and seizures
  • genetic screening/counselling
  • neurology
  • neurosciences, neurology

Statistics from


In 2008, we identified heterozygous loss-of-function mutations in the X-linked CASK gene, including disruption of the gene by a paracentric X-chromosomal inversion, submicroscopic deletions, and sequence changes in four female patients with severe intellectual disability (ID), postnatal microcephaly, and pontocerebellar hypoplasia (MICPCH; MIM 300749). We also described a partly penetrant CASK splice mutation in a severely affected male patient who died at the age of 2 weeks, and hypothesised that hypomorphic CASK alleles could be compatible with life in males.1 Two reports on female patients with heterozygous de novo deletions of ∼3.2 and ∼4.0 Mb covering part of CASK provided further evidence for CASK as a gene for an X-linked form of microcephaly and severe ID manifesting mainly in females.2 3 Very recently, loss-of-function mutations of CASK have been reported in 10 females with MICPCH.4

CASK is a calcium/calmodulin dependent serine protein kinase and belongs to the membrane associated guanylate kinase (MAGUK) protein family.5 Although expressed in different tissues, CASK is widely distributed in different regions of the brain.6 The CASK protein interacts with about 18 different binding partners and plays a critical role in brain development and synaptic function. It is important in: (1) presynaptic organisation and regulation of neurotransmitter release; (2) maintaining the morphology of dendritic spines and regulating ion channels at the postsynaptic site; and (3) regulating the expression of genes involved in cortical development, such as RELN and GRIN2B, by entering the nucleus of neurons.7 8

Since the first description, CASK mutations have also been described in males with a milder phenotype comprising variable ID, nystagmus, and microcephaly.9–11 Preliminary data suggest that the severe brain malformation phenotype is caused by CASK null alleles, whereas the milder phenotype is due to hypomorphic missense mutations. However, data on both groups are limited so far due to the small number of patients.

Since 2008, we ascertained 20 novel female patients with severe ID, microcephaly, and pontine and cerebellar hypoplasia, and characterised all of them. Here we summarise the clinical, brain imaging, and molecular data of novel and previously published CASK mutation positive patients to define better the associated phenotypic spectrum.



We obtained clinical data, brain imaging studies and results of laboratory testing, as well as blood or DNA samples, from 20 patients with ID, microcephaly, and pontocerebellar hypoplasia, who were assessed by experienced clinical geneticists and/or neurologists. Two patients were recruited through the MRNET consortium ( High resolution molecular karyotyping was performed on a clinical basis for eight female patients using different array platforms (detailed information is available on request). All available brain imaging studies from these 20 patients and from five female patients with a CASK mutation published previously1 2 were reviewed by two of the authors (GU and WBD). Clinical data and facial phenotypes were reviewed by UM and GU. The clinical data and samples were obtained with informed consent, including consent to use the photographs in this report, under protocols approved by Institutional Review Boards at all participating institutions.

Fluorescence in situ hybridisation

Fluorescence in situ hybridisation (FISH) experiments were performed to confirm molecular karyotyping data. Metaphase spreads from peripheral blood lymphocytes were prepared by standard procedure. Bacterial artificial chromosomes (BACs) (RP11 human BAC library), P-1 derived artificial chromosomes (PACs) (RPCI human PAC libraries 1 and 5), and fosmid clones (WIBR-2 human fosmid library [G248P8]) were received from the BACPAC Resource Center, Children's Hospital Oakland, California, USA. The BAC clone CTD-3115N12 (CalTech human BAC library) was obtained from Invitrogen, Karlsruhe, Germany. BAC, PAC, and fosmid DNA was prepared using the NucleoBond Xtra Midi kit (Macherey-Nagel, Düren, Germany). DNA was labelled by nick translation using the CGH Nick Translation Kit and Spectrum Green-dUTP (Vysis, Downers Grove, Illinois, USA) according to the protocol provided. Chromosomes were counterstained using 4',6-diamidino-2-phenylindole (DAPI) (Serva Feinbiochemica, Heidelberg, Germany) and mounted in antifading solution (Vector Labs, Burlingame, California, USA). Slides were analysed with a Leica Axioscope fluorescence microscope. Images were merged using a cooled CCD camera (Pieper, Schwerte, Germany) and CytoVision software (Applied Imaging, San Jose, California, USA).

Sequencing of CASK

The coding region of the CASK gene (27 exons; GenBank accession no. NM_003688) was amplified from genomic DNA. Primer sequences are available on request. Amplicons were directly sequenced using the ABI BigDye Terminator Sequencing Kit (Applied Biosystems, Darmstadt, Germany) and an automated capillary sequencer (ABI3130; Applied Biosystems). Sequence electropherograms were analysed using Sequence Pilot software (JSI medical systems, Kippenheim, Germany). Where mutations were shown to have arisen de novo, we verified relationships by genotyping both parents and the patient at 10 microsatellite loci.

CASK transcript analysis

A fresh venous blood sample (2.5 ml) of patient 8 was collected into a PAXgene Blood RNA Tube (PreAnalytiX; Qiagen, Hilden, Germany). Total RNA was extracted by use of the PAXgene Blood RNA Kit (Qiagen) according to the manufacturer's instructions; 1 μg of RNA was reverse transcribed into cDNA (Superscript II; Invitrogen, Karlsruhe, Germany) using random hexanucleotides (Invitrogen) according to the manufacturer's protocol. Reverse transcriptase PCR (RT-PCR) fragments were obtained by the use of forward primer CASK-ex1F (5′-GTTTTCGAAGCCCTCCACGCTGCG-3′) and reverse primer CASK-ex4R (5′-GCTTCACTGTACACAAAACCAGC-3′) according to standard PCR protocols. PCR products were cloned into pCR2.1 TOPO TA Cloning Vector (Invitrogen). Escherichia coli clones were subjected to colony PCR and PCR products from individual clones were sequenced.

CASK copy number analysis

Real time quantitative PCR on genomic DNA was carried out using Rotor Gene RG-3000 (Qiagen) in 100 μl tubes (72-well rotor) with a final reaction of 20 μl. All reactions were prepared with 10 μl of 2x SYBR Green JumpStart Taq ReadyMix (Sigma-Aldrich, Taufkirchen, Germany) and 400 nM forward and reverse primers. Primers for real time experiments were designed using the web based ‘qPCR Primer & Probe Design’ tool provided by MWG-Eurofins ( and primer sequences are available on request. The exon 4 of the CFTR gene was used as an internal reference. A total of 30 ng of DNA was used as template for each sample that was analysed in duplicate. Thermal cycling conditions included a pre-run of 5 min at 95°C. Cycle conditions were 40 cycles at 95°C for 30 s, 58°C for 30 s, and 72°C for 45 s. Relative quantification of exon copy numbers on genomic DNA was carried out using the comparative threshold cycle (ddCt) method: the starting copy number of exons in patients was determined in comparison with the known copy number of the calibrator sample (healthy male control), using the following formula: ddCt = [dCt CASK (CASK patient) − dCt CFTR (CASK patient) − (dCt CASK (healthy male) − dCt CFTR (healthy male))]. The relative exon copy number was calculated by the expression 2^-(ddCt ± SE)—that is, of about 2 for a diploid sample (female) and about 1 for a haploid sample (male).

X chromosome inactivation assay

Quantitative examination of the methylation pattern at the AR locus was performed as previously described.12

Computational analyses

Splice site prediction of one intronic variant identified in CASK was calculated by using the online tools SpliceView, the Berkeley Drosophila Genome Project (BDGP) (, and the NetGene2 server (


The results of molecular and brain imaging studies, and clinical data of the 20 new cases with CASK mutations, are summarised in table 1.

Table 1

Clinical data from 20 individuals with CASK mutations

Molecular results

Identification of deletions and duplications covering CASK by molecular karyotyping

High resolution molecular karyotyping was performed on a clinical basis on eight female patients with microcephaly and severe developmental delay (DD) or ID. We identified submicroscopic deletions covering (parts of) CASK in six patients and duplications in two (figure 1). The smallest deletion was identified in patient 24. She carried a de novo ∼60 kb deletion including CASK exons 6–8, confirmed by FISH with fosmids G248P80427H9 and G248P80557B1 (figure 1). Patient 27 had a de novo deletion of ∼479 kb in Xp11.4 covering CASK exons 1 and 2. This copy number loss was confirmed by real-time quantitative PCR (figure 1 and data not shown). A minimal deletion of ∼2 Mb in Xp11.3p11.4 was detected in patient 7 and confirmed by FISH with BAC clones CTD-3115N12 and RP11-244N4 (figure 1 and data not shown). The deletion contains at least exons 1–12 of CASK, and the distal breakpoint could be mapped between exons 12 and 17 (figure 1). Neither parent carried the deletion (data not shown). A de novo ∼2.1 Mb deletion in Xp11.3p11.4 was also identified in patient 22. The deletion comprising CASK exon 1 was confirmed by FISH using RP5-879N19 (Xp11.3) (data not shown). Two other submicroscopic deletions of ∼4.5 Mb and ∼4.24 Mb (both in Xp11.3p11.4) were found in patients 15 and 23, respectively. FISH with BAC clones CTD-3115N12 and RP11-760P24 yielded only one signal on metaphase spreads of patient 15 (data not shown), indicating that this de novo deletion encompasses at least CASK exons 1–9 (figure 1). The deletion in patient 23, which covers the complete CASK gene, was confirmed with BAC RP11-204C16 (figure 1 and data not shown). This BAC gave two signals on metaphase spreads of patient 23's mother, while the father was not available (data not shown). We detected two de novo duplications including part of CASK in patients 21 and 25 (figure 1). Both duplications were confirmed by quantitative PCR (figure 2). The intragenic ∼215 kb duplication in patient 21 includes exons 2–8, while the duplication of ∼300 kb in patient 25 encompasses at least exons 1–13 (figure 2).

Figure 1

Deletions and duplications including (part of) CASK. Physical map of the Xp11.4 region. The exon–intron structure of the four genes NYX, CASK, GPR34, and GPR82 is shown: horizontal bars represent exons and vertical lines introns. Arrowheads indicate the 5′→3′ orientation of the genes. Exon numbers of the CASK gene are given. BAC (RP11 human BAC library and CTD CalTech human BAC library), PAC (RPCI human PAC library 1), and fosmid clones (F- [G248P8]; WIBR-2 human fosmid library) used for confirmation of submicroscopic deletions are indicated by red and blue bars, respectively, and names are given. Horizontal black lines depict deletions identified in patients 7, 15, 18, 22–24, 26 and 27, whereas grey lines represent duplications detected in patients 21 and 25. Dotted lines indicate that the deletion/duplication breakpoints have not been fine mapped.

Figure 2

Relative quantification of copy numbers of CASK exons 2, 3, 5–13, and 15 (Ex2, Ex3, Ex5-Ex13, and Ex15), two regions of intron 1 (In1a and In1b), and a region 1400 bp upstream of CASK exon 1 (upEx1) by real-time PCR on genomic DNA of patients 20, 21, 25, and 26. Values of CASK exons 7 and 8 in patient 20 (light blue bars) and of two amplicons located in intron 1 plus the one upstream of CASK exon 1 in patient 26 (blue bars) are comparable to that of a haploid sample. In contrast, values of exons 2–8 in patient 21 (yellow bars) and exons 1–13 in patient 25 (green bars) correspond to three copy numbers. As quantitative PCR for CASK exon 14 failed, exon 15 was analysed in patient 25 yielding a value comparable to that of a diploid sample.

Sequence analysis of CASK

Sequence analysis of the 27 coding exons of CASK was performed on 12 patients with DD/ID, microcephaly, and hypoplasia of the cerebellum and the pons. Ten heterozygous mutations were found in nine patients: the four nonsense mutations c.379C>T (p.E127X) in patient 16, c.316C>T (p.R106X) in patient 19, c.1639C>T (p.Q547X) in patient 28, and c.2074C>T (p.Q692X) in patient 29; the three splice mutations c.831+2T>G in patient 12, c.1668+1G>A in patient 13, and c.430-2A>T in patient 30; and the 1 bp deletion c.68delT (p.F23SfsX18) in patient 11 (table 1). In patient 8, two sequence variations—the splice mutation c.173-2A>C in intron 2 and the missense mutation c.174T>A (p.D58E) in exon 3—were identified (table 1). Both sequence changes destroy a BglII restriction site. By subsequent restriction of the genomic PCR product harbouring exon 3 and adjacent intronic sequences with BglII we obtained two restricted fragments (from the wild-type allele) and an unrestricted fragment (from the mutant allele; data not shown). These data indicate that both point mutations occurred on the same allele and are likely the result of a small rearrangement on one CASK allele of patient 8. In silico analysis using the splice site prediction programs ‘Splice View’, ‘NetGene2’, and ‘Berkeley Drosophila Genome Project’ revealed that the c.173-2A>C mutation disrupts the splice acceptor site and most probably results in altered splicing (data not shown). Next, we analysed CASK transcripts by generating an RT-PCR product with primers located in exons 1 and 4 and obtained an amplicon with the expected size of 409 bp in patient 8 (figure 3A). A second PCR product of ∼300 bp was only observed in the patient and not in control individuals (figure 3A). Subcloning and direct sequencing of both amplicons of patient 8 demonstrated correct splicing of exon 2 to exon 3 in the 409 bp amplicon (data not shown), while exon 2 was spliced to exon 4 in the 303 bp amplicon (figure 3B). Skipping of exon 3 leads to a frameshift in the CASK mRNA. In this scenario, the missense mutation c.174T>A (p.D58E) in exon 3 on the same allele has no pathogenic relevance. Five of the nine CASK alterations were de novo, while the parents of patients 16 and 28–30 were not available.

Figure 3

Molecular analysis of CASK mutations in patients 8 and 20. (A) The c.173-2A>C mutation in patient 8 caused skipping of CASK exon 3. RT-PCR products were amplified from cDNA of patient 8 (P8) and two healthy individuals (C1 and C2) with primers located in CASK exons 1 and 4. The normal 404 bp product is present in patient 8 and two controls; however, an additional amplicon of ∼300 bp was only amplified from the cDNA of patient 8. (B) In the upper part, part of sequence profiles of CASK exon 2–intron 2 and intron 3–exon 4 junctions are shown. In the lower part, representative sequence profile of a cloned PCR product revealed splicing of CASK exon 2 to exon 4 indicating skipping of exon 3 by the c.173-2A>C mutation in patient 8. (C) Sequence analysis of a deletion specific junction fragment in patient 20. Schematic representation of part of the CASK gene (top). CASK exons are indicated by grey boxes and introns by black lines. Boxes surrounded by dashed lines represent deleted exons. Partial DNA sequence of intron 6 and intron 8 is indicated above the exon–intron structure of CASK. The 5 bp motif present at the deletion junction in patient 20 is framed and the size of the deletion is given. Part of the DNA sequence electropherogram of the deletion specific junction fragment (below). At the junction, an overlap (AGATC) was identified. Sequence homology between the Alu repeats in intron 6 and intron 8 of CASK is shown below the sequence profile.

Copy number analysis of CASK by FISH and real-time quantitative PCR

By CASK sequence analysis, we did not identify a mutation in patients 18, 20, and 26. To detect a possible microdeletion of CASK in these patients, serial FISH analyses with BAC, PAC, and fosmid clones covering the complete gene were performed (figure 1). BAC clone CTD-3115N12 yielded only one signal on metaphase spreads of patient 18, similar to the five fosmid clones G248P82722E4, G248P80319B6, G248P82570C9, G248P80427H9, and G248P80557B1 (figure 1 and data not shown). Thus, patient 18 was shown to have a ∼120 kb microdeletion covering at least CASK exons 6–12. Neither of her parents carried the deletion (data not shown). In patients 20 and 26, no deletion could be detected by FISH. Next, we used real-time PCR for comparative quantification of copy numbers of selected CASK exons, intron 1, and a DNA region upstream of CASK on genomic DNA of the two patients. The relative copy number of CASK exons 7 and 8 in patient 20 and of amplicons surrounding CASK exon 1 in patient 26 was comparable to a haploid sample (figure 2). The de novo deletion of ∼100 kb in patient 26 was confirmed by FISH with fosmids G248P86259A10 and G248P80448B4 (figure 1 and data not shown). From the DNA of patient 20, we amplified a 376 bp breakpoint-spanning PCR product (data not shown). Direct sequencing revealed an 11 218 bp deletion comprising 3351 bp of CASK intron 6, exon 7 (176 bp), intron 7 (4715 bp), exon 8 (123 bp), and 2853 bp of intron 8 (figure 3C). At the junction, a 5 bp overlap (AGATC) was detected (figure 3C). We identified Alu repeats next to the breakpoint (data not shown) suggesting non-allelic homologous recombination (NAHR) as the likely mechanism for the rearrangement in patient 20. The junction fragment could not be amplified from the DNA of patient 20's parents, indicating de novo occurrence (data not shown). We confirmed all parental identities where mutations had arisen de novo.

X chromosome inactivation analysis

Analysis of the X chromosome inactivation (XCI) pattern identified random XCI (up to 79:21) in 11 out of 16 females with a CASK alteration and slightly skewed XCI (80:20–89:11) in three patients (table 1). Patient 15 had a highly skewed XCI (≥90:10) (table 1). Patient 13 was not informative at the AR locus (table 1).

Brain imaging studies

Brain imaging studies (MRI in 23 individuals and CT in patient 9) and clinical data (see below) were reviewed in a total of 25 CASK mutation positive individuals, including the 20 novel patients described here and five published patients (patients 1–4 in Najm et al1 and case 4 in Froyen et al,2 numbered as patient 9 in figure 4). Brain imaging consistently showed pontocerebellar hypoplasia with a dilated fourth ventricle, but of remarkably varying degrees (figure 4). The cerebral cortex presented with simplified gyration (mildly reduced number and complexity of gyri) in the frontal region in eight of 22 cases, whereas in 14 of 22 cases providing sufficient information, the cerebrum appeared normal with mild dilatation of the lateral ventricles in a few cases. The corpus callosum was intact with normal appearance in all cases (22/22). The pons was moderately to severely flattened in one of 23 cases, appeared mildly or subtly (very mildly) hypoplastic in 15 of 23 and six of 23 cases, respectively, and was normal in the remaining one of 23. The cerebellar hypoplasia affected the vermis and cerebellar hemispheres proportionally. Of the 24 cases, the cerebellum was mildly hypoplastic in 10 cases, moderately hypoplastic in 11, and severely hypoplastic in three cases. The fourth ventricle was enlarged in 18 of 23 cases and the size appeared normal in five of 23 cases. In at least two cases this was due to the small size of the posterior fossa with verticalisation of the tentorium. In one patient an arteriovenous malformation was observed (patient 30). In patient 22, original scans were not available but brainstem and cerebellar hypoplasia was present by report of CT imaging. No other brain anomalies were identified.

Figure 4

Selected axial, coronal, and sagittal MR images (reconstructed CT images in patient 9) in 20 patients with CASK mutations. In the first row axial images show mildly reduced number and complexity of the frontal gyri in patients 16, 19, 26, and 27 (white arrowheads). The coronal images show hypoplastic, flattened cerebellar hemispheres with proportionally reduced size of the vermis. The sagittal images in the third row show low forehead indicative for microcephaly, intact corpus callosum in all cases, different grades of pontine hypoplasia (white arrows) ranging from nearly normal (patient 7), mild to moderate-severe (patients 18, 20, 26), and different grades of vermis hypoplasia ranging from nearly normal (patients 7 and 21), moderate (patients 8 and 15) to severe (patient 18).

Clinical data

Clinical data of a total of 25 female patients could be studied (20 individuals reported here, patients 1–4 in Najm et al1 and case 4 in Froyen et al2). All patients were female, aged between 10 months and 21 years at last follow-up, with a mean age of 6 years 2 months. With the exception of two individuals (patient 15 in this study and patient 3 in Najm et al1), all showed severe psychomotor retardation or ID, depending on their age. The majority never attained the ability to walk: only four out of 21 patients of at least 1 year of age learnt to walk without support at ages ranging from 18 months to 4½ years. Patient 15 could speak two-word sentences at the age of 4 years, but all other patients had no speech or produced vowels, syllables or <10 words. Fourteen out of 22 patients showed general or axial hypotonia, whereas 10 of 22 exhibited general or peripheral hypertonia, sometimes progressing to spasticity. Movement disorders were seen infrequently (patients 16, 22, 26, 30), and one patient showed mild bulbar involvement (patient 19). Eight patients suffered from seizures of various types with onset at 1 to 8 years.

Eight out of 23 patients had sensorineural hearing loss. Nineteen out of 24 patients exhibited various ophthalmologic anomalies with optic nerve hypoplasia (5/19) or optic disc pallor (4/19), and strabismus (5/19) being the most common ones; two girls had a colobomatous defect.

All patients had significant microcephaly (occipitofrontal circumference (OFC) ≤3.4 SD, up to −10 SD) which was of prenatal or, most commonly, of early postnatal origin and evident in all cases during the first year of life. At birth, nine newborns showed an OFC of less than −2 SD, the OFC of six babies corresponded to −2 SD, and the birth OFC of the other 10 babies was still within the normal range. Fifteen out of 24 patients also showed postnatal growth retardation of varying degree, whereas birth length had been in the normal range for all of them. Other congenital anomalies (fusion of kidneys, mitral valve incompetence, pectus excavatum, hypoplastic toenails) were only seen in a minority of patients, with no particular anomaly occurring recurrently.

The facial phenotype of 23 patients could be studied. Apart from microcephaly, the following features were noticed: upslanting palpebral fissures (2), hypertelorism or telecanthus (4), hypotelorism (1), epicanthus (4), broad nasal bridge (14), prominent nasal bridge (6), broad or bulbous nasal tip (8), small or short nose (4), long philtrum (5), small jaw (10), and large ears (12) (figure 5).

Figure 5

Facial appearance of 13 female patients with CASK loss-of-function mutations. Numbers refer to patient numbers. Frontal view of the patients is shown in (A) and lateral view in (B). Apart from microcephaly, a prominent and/or broad nasal bridge and tip, a small chin and large ears were noticed in the majority of patients. In some cases, the nose was short or small, and the philtrum long (photographs submitted with written consent from the patients' legal guardians for publication in print and online).


We present molecular data of 20 new female patients and clinical data of a total of 25 female individuals with a CASK alteration. The mutations most likely lead to a CASK null allele: 11 submicroscopic copy number alterations, including nine deletions of ∼11 kb to 4.5 Mb and two duplications, all covering (part of) CASK, four splice mutations, four nonsense mutations, one missense mutation, and one 1 bp deletion. The intragenic duplication comprising exons 2–8 could interfere with splicing and lead to aberrant CASK transcripts, as demonstrated for the 0.2 Mb intragenic duplication in patient 9 described by Hayashi et al.4 If the duplicated fragment of the second duplication containing at least exons 1–13 was inserted in reverse orientation within the CASK gene, the integrity of the gene could be disturbed resulting in a CASK null allele. We could not establish a correlation between the genotype and specific clinical features or the degree of brain anomalies. In particular, the phenotype of patients with larger deletions including multiple genes other than CASK was indistinguishable from that caused by mutations only affecting CASK. Likewise, Hayashi et al found similar phenotypes of ID with MICPCH caused by different CASK loss-of-function mutations, including two deletions of 1.1 and 3.0 Mb.4

The XCI pattern was random or slightly skewed in the majority (14/15) of females. The only two patients with a highly skewed XCI (patient 15 and case 4 in Froyen et al2) harboured a larger deletion in Xp11.3p11.4. Remarkably, patient 15 with the largest deletion comprising CASK seems to have the mildest clinical manifestation, which may be explained by highly skewed XCI. Skewed XCI can result from a cell selection process downstream of X inactivation.13 We hypothesise that pronounced skewing of XCI in the two patients with a large Xp11.3p11.4 deletion can be attributed to loss of other genes in the deleted interval, while loss of CASK itself has no deleterious effect on proliferation and/or cell division of leucocytes as demonstrated by random XCI in all patients with an intragenic CASK mutation (table 1).

The 25 patients clinically studied had a remarkably similar phenotype. Brain imaging consistently demonstrated diffuse brainstem and cerebellar hypoplasia of a remarkably varying degree, proportionally affecting the vermis and cerebellar hemispheres. The cerebral cortex showed simplified gyration in the frontal region in some cases, whereas it appeared normal in the majority of patients, with mild dilatation of the lateral ventricles in some of them. The corpus callosum (CC) was normal in all cases. These data corroborate the recent finding by Takanashi et al that a normal CC together with a low cerebrum/CC ratio may be an imaging clue to specifically recognise females with a CASK mutation.14 Moreover, these findings significantly help in distinguishing CASK related diffuse brainstem and cerebellar hypoplasia from other disorders. The proportionate involvement of cerebellar vermis and hemispheres in most girls with CASK mutations may differentiate them from children with pontocerebellar hypoplasia types 2 and 4 (PCH2 and PCH4), which have disproportionate involvement. The latter consists of flat cerebellar hemispheres and small but less severely affected vermis that give a dragonfly-like appearance to the cerebellum.15 16 In only a few females with CASK alterations are the cerebellar hemispheres more severely affected than the vermis—for example, in patient 29. These findings are in line with a recent paper on de novo CASK mutations in three females; one was reported to have a dragonfly-like pattern of PCH and two had a proportionate hypoplasia of cerebellar hemispheres and vermis.17 PCH2 and PCH4 are caused by mutations in three different genes, namely TSEN54, TSEN34, and TSEN2.15 Clinically, PCH2 is characterised by progressive microcephaly, failure of motor development, central visual impairment, and dystonia or dyskinesia. In PCH type 4, which was previously known as olivopontocerebellar hypoplasia, an earlier prenatal onset with C-shaped inferior olives and often early postnatal demise is seen.16 In particular, movement disorders may distinguish PCH2 from the CASK associated phenotype as they are less common in patients with a CASK mutation. In addition, spasticity shortly after birth and epilepsy may lead to the diagnosis of PCH2,16 18 since spasticity develops later and seizures are not very frequent in patients with CASK mutation (eight of 25 cases), while the presence of sensorineural hearing loss may prompt CASK testing.

The core clinical phenotype in patients with a CASK mutation studied here consisted of severe DD/ID, mild and variable congenital microcephaly, and severe postnatal microcephaly (OFC −3.5 to −10 SD) by 1 year of age, often associated with postnatal growth retardation, and (axial) hypotonia with or without hypertonia of extremities, possibly progressing to spasticity. Optic nerve hypoplasia and/or other eye abnormalities, and sensorineural hearing loss were found in some cases. By evaluating clinical photographs of females with a CASK alteration, a recognisable facial phenotype emerged, including prominent and/or broad nasal bridge and tip, small or short nose, long philtrum, small chin, and/or large ears.

The severe and distinct brain malformation phenotype in females caused by loss-of-function mutations in CASK is different from the milder phenotype caused by hypomorphic CASK alleles described so far. The CASK sequence change c.83G>T (p.R28L) was found in three males of an Italian family with clinical findings suggestive of FG syndrome (ID, relative macrocephaly, congenital hypotonia, constipation, and behavioural anomalies) and a normal MRI. Two carrier females were unaffected.10 This mutation leads to partial skipping of exon 2 and results in an out-of-frame CASK transcript.10 Additional sequence alterations of CASK have been reported in male patients from six families with mild to severe ID, with or without congenital nystagmus, microcephaly, and/or dysmorphic facial features: five missense mutations (p.Y268H, p.P396S, p.D710G, p.W919R, p.Y728C) and the splice mutation c.2521-2A>T resulting in an in-frame deletion of several amino acids in CASK.9 11 Carrier females in these families were unaffected or mildly affected showing nystagmus, tremor and/or mild ID. Only some of the affected males had undergone brain imaging, which showed cerebellar hypoplasia and pachygyria in one of them; further information was not available.9 Together, these data suggest that highly deleterious mutations in CASK in the heterozygous state are consistently associated with a severe phenotype in females, while hypomorphic point mutations cause a milder phenotype, including X-linked nystagmus and variable ID, particularly in males. In these cases, the CASK protein may still fulfil some functions.

In summary, we describe a very consistent phenotype in females with a CASK null allele, comprising diffuse brainstem and cerebellar hypoplasia of variable degree with normal appearing corpus callosum, severe microcephaly of prenatal or early postnatal origin, severe ID, and possibly sensorineural hearing loss. Developmental features, brain malformations, and the emerging facial phenotype should allow clinical recognition of appropriate girls and females with severe ID and prompt genetic testing of CASK.


We are grateful to the patients and their families who contributed to this study. We thank Inka Jantke for skilful technical assistance and Birte Lübker and Angela Fritsch for help with FISH experiments.


View Abstract


  • UM and KK contributed equally.

  • Funding This work was supported by grants from the Werner Otto-Stiftung (WOS 3/76 to GU) and the Deutsche Forschungsgemeinschaft (KU 1240/5-1 to KK). This work was part of the German Mental Retardation Network (MRNET) funded by the German Ministry of Research and Education as part of the National Genome Research Network NGFNplus (project reference numbers 01GS08162, 01GS08164, 01GS08168).

  • Competing interests None declared.

  • Patient consent Parental consent obtained.

  • Ethics approval The clinical data and samples were obtained with informed consent, including consent to use the photographs in this report, under protocols approved by Institutional Review Boards at all participating institutions.

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

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.