JLE

Epileptic Disorders

MENU

Age-dependent epileptic encephalopathy associated with an unusual co-occurrence of ZEB2 and SCN1A variants Volume 22, issue 1, February 2020

Video

  • Age-dependent epileptic encephalopathy associated with an unusual co-occurrence of ZEB2 and SCN1A variants

Figures


  • Figure 1

  • Figure 2

Mowat-Wilson syndrome (MWS) (OMIM #235730) is characterized by peculiar facial features associated with intellectual disability/global developmental delay (ID/GDD), and variable structural anomalies including agenesis of the corpus callosum, microcephaly, hypospadias, cardiac defects, Hirschsprung disease (HSCR) and eye defects (Garavelli et al., 2009; Yamada et al., 2014; Ivanoski et al., 2018). In addition, epilepsy and language impairment are the main features of the syndrome. MWS is caused by de novo heterozygous variants in the ZEB2 gene on chromosome 2 and its haploinsufficiency appears to be the primary pathogenetic mechanism (Wakamatsu et al., 2001). Although more than 300 individuals have been reported so far, the variability in phenotype, which may be categorised into subgroups with only small numbers of patients, makes it challenging to establish genotype-phenotype correlations (Ivanoski et al., 2018).

Here, we report a patient with a severe neurological phenotype of MWS associated with a newly described intestinal defect and a novel heterozygous frame-shift de novo variant in the ZEB2 gene, as well as an additional novel heterozygous missense variant in the SCN1A gene inherited from his healthy father. This report aims to contribute to better diagnosis of the disorder and help establish possible genotype-phenotype correlations.

Case study

This study was approved by the ethics committee of Palermo 1 University Hospital.

The detailed clinical evaluation of the patient was performed after obtaining written informed consent for publication from the patient's parents. The patient, a 10-year-old girl, was the third offspring born to healthy Italian unrelated parents. Her family history was remarkable for intellectual disability, psychiatric disorders, and epilepsy in her paternal mother and two great aunts. She was born at term by elective Caesarean section following an uneventful pregnancy. APGAR scores were 8-10, head circumference 33 cm (10th centile), birth weight 3,240 g (50th centile), and height 51 cm (75th centile). At birth, she was diagnosed with anal atresia and rectoperineal fistula corrected by anoplasty and, later at six months of age, by posterior anal relocation. At three months of age, cardiac assessment revealed patent ductus arteriosus and apical ventricular septal defect with small left-to-right shunting.

Her developmental milestones were delayed: head control, sitting without support, and walking were achieved at 3, 18, and 36 months, respectively. At 14 months, she was referred to our department due to developmental delay. On admission, neurological and psychomotor examination revealed microcephaly (42 cm [<3th centile]), mild hypotonia, normal deep tendon reflexes, difficulty in unsupported sitting, poor spontaneous motility, inability to grasp objects, poor eye contact, absence of social smiling, repetitive sticking-out of the tongue, under-reaction to pain, and a high level of sensory seeking. In addition, the patient showed craniofacial dysmorphic features including a broad nasal bridge, hypotelorism, a thin upper lip, a high arched palate, a narrow forehead, and four spots (variable in size) on the left groin and thigh (see video sequence).

The EEG displayed a symmetric irregular posterior 5-6-Hz activity associated with isolated spike-wave discharges mainly in both frontal regions (figure 1).

At three years of age, the patient first experienced atypical absence seizures with head oscillation and drop attacks with a frequency of 2-3/die, associated, during drowsiness, with abnormal interictal EEG displaying isolated or short series of recurring and generalized 3-Hz spike-wave complex discharges (figure 1). In addition, the child showed unsteady posture, worsening of uncertain walking, drooling, restlessness, irritability, self-injury, and poor responsiveness to environmental stimuli. She could speak only a few meaningful sounds, used in association with a restricted conventional repertoire of gestures, to communicate with other individuals, but she understood simple sentences. The clinical course, to date, is characterized by alternating remission and relapse of an electroclinical pattern, partially refractory to common antiepileptic drugs. Indeed, an increase in seizure frequency was constantly associated with increased EEG abnormalities (leading to continuous spike-wave activity during slow sleep [CSWS]) and a worsening of the above-mentioned behaviour and motor abnormalities (worsening of ataxic-like gait and speech language, drooling, irritability, and self-injury). In addition, the relapse of seizures was adversely influenced by febrile illness, suggesting a phenotype with fever-triggered seizures. However, a combination of levetiracetam and ethosuximide was the most effective treatment leading to a long period of seizure freedom. Brain MRI performed at two and four years of age showed a mild unmodified thinning of the corpus callosum isthmus (figure 2).

Comprehensive genetic testing was performed during the clinical course. A next-generation sequencing panel showed a new heterozygous missense variant (c.25C>G) corresponding to within the N-terminal domain of the SCN1A gene (NM_001165963.1), resulting in substitution of a highly conserved amino acid (p.Pro9Ala). This variant, not reported in the ExAC database, was inherited from her healthy father. Bioinformatics assessment of the significance of the SCN1A missense mutation using two different algorithms, PolyPhen2 (http://genetics.bwh.harvard.edu/pph2/) and Mutation Taster (http://www.mutationtaster.org/), predicted a possible deleterious effect on protein function (probably damaging with a score of 0.989 based on PolyPhen2, and disease-causing with a score of 27 based on Mutation Taster).

In addition, whole-exome sequencing showed a novel heterozygous de novo frame-shift variant, c.2264dupA p.Asp755fs, in the ZEB2 gene (NM_014795.3), not reported in the ExAC database (for more details see supplementary data online).

Discussion

The ZEB2 protein is a complex transcription factor that encompasses multiple functional domains that interact with the regulatory regions of target genes. In particular, the ZEB2 plays a key role in several processes associated with nervous system development. It was found to be highly expressed in all brain regions of the human foetal telencephalon, involving several neuronal populations including pyramidal neurons of the hippocampus, cortico-spinal neurons, dopaminergic neurons in the brainstem (Nishizaki et al., 2014), and cerebellum Bergmann glia precursors (He et al., 2018). Recently, ZEB2 has been shown to regulate the differentiation of interneuron progenitors, migrating from the medial ganglionic eminence to cortical layers by repression of the Nkx2-1 homeobox transcription factor (McKinsey et al., 2013). It has therefore been suggested that the deficit in ZEB2, resulting in decrease in cortical interneurons associated with increased striatal GABAergic interneurons, may induce an imbalance of inhibition/excitation, leading to epileptic seizures, the main component of Mowat-Wilson syndrome (Cordelli et al., 2013). In addition, our patient has a novel SCN1A variant (c.25C>G; p.Pro9Ala) inherited from her unaffected father.

Whereas we are only just starting to identify genotype-phenotype correlations for Kv7.2 channelopathies (Piro et al., 2019), SCN1A mutations, resulting in reduction of NaV1.1-mediated sodium currents in GABAergic interneurons, are associated with a spectrum of epilepsy syndromes, ranging from relatively mild phenotypes with febrile seizures plus to severe myoclonic epilepsy of infancy, also known as Dravet syndrome (DS) (Claes et al., 2001). Although the father of the presented patient carries a SCN1A variant (c.25C>G), and has never had any epileptic seizure, we have been cautious in classifying the variant as benign. Indeed, previous studies have highlighted that SCN1A variants may have a low penetrance and a familial recurrence that affects the descendants of non-symptomatic parents (Gataullina and Dulac, 2017). Furthermore, the neurodevelopmental profiles of ZEB2 and SCN1A mutations show some overlap, making it difficult to establish the role of the SCN1A variant in our patient's phenotype. In addition, this unusual co-occurrence underlines similarities with some features of Dravet syndrome including ataxic gait, language, cognitive, and behaviour deficit, hyperthermia sensitivity and a distinct profile of language deficit with significant impairment of expressive language development but relatively good receptive language (Battaglia et al., 2013; Gataullina and Dulac, 2017).

To date, there is no precise genotype-phenotype relationship for Mowat-Wilson syndrome or Dravet syndrome as their respective genotypes are associated with significant phenotypic variability. In recent years, however, intrafamilial and clinical heterogeneity have been widely discussed in the literature and the results seem to point to a relationship between loss of function of Nav1.1 channels in GABAergic interneurons and the spectrum of severity of Nav1.1 -associated epilepsy syndromes (Catterall et al., 2010; Zuberi et al., 2011; Nissenkorn et al., 2019). On the other hand, a recent collaborative international study carried out to further characterize the phenotype, natural history, and genotype-phenotype correlation of 87 patients with a genetic diagnosis of MWS revealed that some retention of ZEB2 protein function may lead to a milder clinical phenotype. Moreover, a mild phenotype may be associated with mutations resulting in a loss of function of ZEB2 protein (Ivanovski et al., 2018). Remarkably, the two genes affected by mutation in our patient regulate the differentiation and functioning of GABAergic interneurons, of which the action potential firings (of at least 75%) are supplied by sodium current of Nav1.1 channels. In addition, based on studies to investigate the phenotype of patients with SCN1A and ZEB2 mutation, authors hypothesize that other genetic and environmental factors may be responsible for the phenotypic variability.

Our findings therefore suggest that the SCN1A variant, c.25C>G, together with the ZEB2 mutation, may have influenced the severity of our patient's neurological phenotype, playing a role in the dysfunction of interneurons. Furthermore, the lack of structural brain abnormalities on MRI in our patient supports the hypothesis that the electroclinical pattern is caused by cortical interneuron dysfunction (Cordelli et al., 2013). Clearly, further studies will be needed to confirm this hypothesis. Furthermore, the clinical features of our patient included anal atresia (not previously reported in literature) which may reflect the role of ZEB2 in the regulation of target genes leading to different congenital defects, further expanding the spectrum of phenotypes associated with ZEB2-related disorders.

Supplementary data

Supplementary data and summary didactic slides are available on the www.epilepticdisorders.com website.

Acknowledgements and disclosures

We thank the child and her parents who participated in this study.

We thank Dr. E. Gennaro for expert advice in performing and interpreting the NGS analysis.

None of the authors have any conflict of interest to declare.