NeONATAL epilepsies / EIEE

> Cette page en françaisEpilepsiesnostravaux_fr.html
OUR TEAMEquipe1_gb.html

our RESULTS & projects

> BackEpilepsies1_gb.html


>> Pathogenic variants in the PCDHGC4 gene cause a new neurodevelopmental syndrome.

This is a description of 19 patients and the description of a new autosomal recessive syndrome where epilepsy is present.

>> Epilepsy and mGluR7 dysfunction: a joint work from the SEED project.

In the framework of the H2020 SEED project (see elsewhere on these pages), we have identified a pathogenic variant in the mGluR7 glutamate receptor in a large tunisian consanguineous family. This contribution allows to better describe the clinical picture associated with mGluR7 variants.

>> A series of 18 new cases of CACNA1A patients.

The pediatric neurology team of Angers has assembled a cohort of 18 new patients, allowing a better description of CACNA1A related disorder.

>> Expansion of the genotypic and phenotypic spectrum of WASF1-related neurodevelopmental disorder.

In this collaborative work, we describe 6 new patients carrying pathogenic variants in the WASF1 gene and describe milder forms.

>> KCNH1 epilepsy in non-classical forms.

There are cases of KCNH1 epilepsies that do not correspond to the classical clinical pictures for the disease, in particular because they lack gingival or nail manifestations.

>> SYNGAP1 and visual sensitive epilepsy.

This work, supervised by the center of reference for rare epilepsies, shows how SYNGAP1-associated seizures are triggered by eye closure.


>> For the first time, a role for KCNQ2 in the development of locomotor networks.

In collaboration with Institut des Neurosciences de La Timone, our KCNQ2 mouse model contributes to demonstrate a role of KCNQ2 to set the speed of locomotion.

>> The first mouse model for KCNQ2-related developmental and epileptic encephalopathy.

We have developed a new mouse model that reproduces the DEE phenotype caused by KCNQ2 mutations, introducing recurrent KCNQ2 variant into the mouse genome.

>> A series of 17 patients allows to better define FHF1 associated epileptic encephalopathies.

This collaborative study allows a better description of the electro-clinical phenotype, to study genotype-phenotype relationships and to offer a better prognosis to the affected individuals and their families.

>> New cases of PIGA mutations increase the knowledge of the associated epileptic phenotype.

This study refines the clinical and paraclinical data for new patients affected by an epileptic encephalopathy phenotype and carrying a germline PIGA mutation.


>> Mutations in the KMT2E gene cause a new syndrome, including epileptic encephalopathy in some cases.

That international collaborative study describes 38 patients and shows that missense variants in the KMT2E gene are associated with epileptic encephalopathy.

>> Clinical study of 19 patients carrying a pathogenic variant in the SCN8A gene.

Mathieu Milh and Julien Denis have studied these cases and describe two orthogonal modes of onset for epilepsy in these patients.

>> Publication of 20 patients carrying  a pathogenic variant in the WWOX gene.

That study allows to refine the phenotype of these patients and offers a review of published cases.

>> Publication of 37 patients carrying a pathogenic variant in the IQSEC2 gene.

That study offers a very detailed description of the male and female phenotypes for IQSEC2 and discusses the importance of the dosage of the gene product for proper brain development.

>> Variants in the components of the NALCN-UNC80-UNC79 complex cause a large phenotypic spectrum.

The paper describes 29 patients presenting pathogenic variants in that complex and allows to refine the associated phenotypes.

>> A link between congenital ataxias and epileptic encephalopathies.

In collaboration with numerous french colleagues, we show that there is a link between congenital ataxias and epileptic encephalopathies in a series of patients.


>> Description of new pathogenic variants in the ATP1A3 gene.

We have contributed to the identification of new variants in the ATP1A3 gene in patients suffering from epileptic encephalopathy and dystonic fits.

>> Abnormal UBA5 function can cause epileptic encephalopathy.

Using exome sequencing, we have identified a pathogenic variant in the ubiquitin-like modifier activating enzyme 5 (UBA5) gene. That gene is involved in post-traductional modifications. Functional analysis of the variant, performed in collaboration with the laboratory of Eva Liebau, show that the variant we identified almost completely abolishes the UBA5 protein activity.

>> Our cohort of patients affected by early onset epilepsy reaches 800 cases !

Thanks to our participation to the french network for the study of genetic epilepsies (EpiGene) and the participation of many families, we are now studying a cohort of 800 patients suffering from early onset epilepsy (seizure onset before the age of 6 months). Next generation sequencing techniques are used to identify the genetic basis of these diseases and to better understand the molecular mechanisms causing epilepsy in these patients. Our final goal remains the same: to known these epilepsies in more detail in order to develop new treatments.


>> Heterogeneity of FHF1 related phenotype: novel case with early onset severe attacks of apnea, partial mitochondrial respiratory chain complex II deficiency, neonatal onset focal seizures without neurodegeneration.

We report the case of a child prospectively followed in our institution for a severe, neonatal onset epilepsy presenting with severe attacks of apnea that were not initially recognized as seizure since they were not associated with any abnormal movement and since interictal EEG was normal. Recording of attacks using prolonged video-EEG recording allowed to confirm the diagnosis of epileptic seizures. Using whole exome sequencing we found a de novo heterozygous, missense mutation of FHF1 (p.Arg52His, NM_004113), a mutation that has been very recently described in 7 patients with an early onset epileptic encephalopathy. The initial workup showed a partial deficit of the complex II of the respiratory chain in muscle and liver. The prospective follow-up demonstrated that 2 drugs seemed to be more effective than the others: sodium blocker carbamazepine, and serotonine reuptake blocker fluoxetine. GABAergic drugs seemed to be ineffective. No drug aggravated the epilepsy. This case report contributes to the description of an emerging phenotype for this condition

>> Early-onset epileptic encephalopathies with “suppression-burst” EEG pattern: genetic characterization of a European cohort of 70 patients

Early-onset epileptic encephalopathies (EOEE) are rare and severe neurodevelopmental disorders, characterized by motor seizures occurring before three month of life associated with a very abnormal EEG pattern who can be described as being “suppression-burst” (SB), where bursts of paroxysmal activity alternate with periods of electric silence or flattened tracing. There is a strong and heterogeneous genetic component to these diseases. We studied 70 patients with a SB EEG pattern, which is the most characteristic and easily identifiable for clinicians. Genetic explorations were first performed in a hospital setting (for known genetic causes) and secondly, in a research framework. Trio whole exome sequencing analyses were performed for negative cases. We identified 30 genetic abnormalities. The diagnostic rate was about 40%. An overview of the 70 patients highlights the implication of KCNQ2 and STXBP1, which are the most frequently mutated genes (20% and 7%, respectively). Other mutations were identified in several genes, at a lower frequency: SCN2A, KCNT1, ARX, GABRB3, GABRG2, ALGT11. The vast majority of mutations was heterozygote and arose de novo. Interestingly, no patient had a mutation in SCN8A gene, which is frequently involved in other EOEE. Deciphering molecular basis of EOEE is the first step to better understanding these severely diseases. Progresses in the field of diagnosis, coupled with development of personalized medicine could modify the quality of life of the patients and their families.

This work was supported by Inserm, Aix Marseille University, Marseille University Hospital and the JED Foundation.


>> We have created the first mouse model for Ohtahara syndrome.

Ohtahara syndrome is the most severe form of intractable epilepsy. We curently have 80 Ohtahara syndrome patients in our epilepsy cohort. Pathogenic variations in the KCNQ2 gene, encoding the Kv7.2 potassium channel subunit, are the major cause of Ohtahara syndrome.

We have engineered a transgenic knock-in mouse with the KCNQ2 variation causing the most severe phenotype in our cohort. This is the first mouse model for Ohtahara syndrome based on the knock-in of a human variation. The preliminary characterization of the phenotype of the knock-in animals reveals that they are likely to reproduce precisely the human condition.

We now will characterize the model in more detail and use it to test pharmacological hypotheses. We hope to be able to provide a better understanding of the molecular and cellular mechanisms causing one of the most severe epilepsy phenotype and to generate new data to develop therapeutic approaches for this devastating and currently intractable condition.


>> The JED Foundation awards a grant to our project entitled «Deciphering molecular basis of early onset epileptic encephalopathies to treat them better».

There is a strong genetic component for early onset epileptic encephalopathies for which dozens of causes have been identified. At present, the best laboratories conclude on a genetic cause in about 30% of patients after sequencing several major genes. An important amount of work thus remains to be performed to decipher the molecular basis of these conditions, and we want to do it in the first objective of the current project by taking advantage of: 1- one of the largest European cohorts of patients (currently 540 cases), 2- available high-throughput sequencing technologies in our laboratory, 3- our expertise in molecular genetics and pediatric neurology.

We also hope to develop new therapeutic solutions able to treat the seizures and to preserve as much as possible the neurological development of the affected children, using models developed in our laboratory.


>> Specific mutations in the KCNQ2 enhance channel activity rather than reducing it.

Mutations in the KCNQ2 gene encoding the voltage-gated potassium channel subunit Kv7.2 cause early onset epileptic encephalopathy (EOEE). Most mutations have been shown to induce a loss of functi
on or to affect the subcellular distribution of Kv7 channels in neurons. Here, we investigated the functional consequences and subcellular distribution of the p.V175L mutation located in the third transmembrane segment of Kv7.2 (Kv7.2V175L) and found in a patient presenting EOEE with suppression-burst. We observed that the mutation produced a 25-40 mV hyperpolarizing shift of the conductance-voltage relationship of both the homomeric Kv7.2V175L and heteromeric Kv7.2V175L/Kv7.3 channels compared to wild-type channels and a 10 mV hyperpolarizing shift of Kv7.2V175L/Kv7.2/ Kv7.3 channels in a 1:1:2 ratio mimicking the patient situation. Mutant channels also displayed faster activation kinetics and an increased current density. However, the p.V175L mutation did not affect the protein expression of Kv7 channels and its localization at the axon initial segment. We conclude that p.V175 is a gain of function mutation. This confirms previous observations showing that mutations having opposite consequences on M channels can produce EOEE. These findings alert to the fact that drugs aiming to increase Kv7 channel activity may have adverse effects in this disease.


>> Epileptic patients with de novo STXBP1 mutations: key clinical features based on 24 cases.

Mutations in the syntaxin binding protein 1 gene (STXBP1) have been associated mostly with early onset epileptic encephalopathies (EOEEs) and Ohtahara syndrome, with a mutation detection rate of approximately 10%, depending on the criteria of selection of patients. The aim of this study was to retrospectively describe clinical and electroencephalography (EEG) features associated with STXBP1-related epilepsies to orient molecular screening.

We screened STXBP1 in a cohort of 284 patients with epilepsy associated with a developmental delay/intellectual disability and brain magnetic resonance imaging (MRI) without any obvious structural abnormality. We reported on patients with a mutation and a microdeletion involving STXBP1 found using array comparative genomic hybridization (CGH).

We found a mutation of STXBP1 in 22 patients and included 2 additional patients with a deletion including STXBP1. In 22 of them, epilepsy onset was before 3 months of age. EEG at onset was abnormal in all patients, suppression-burst and multifocal abnormalities being the most common patterns. The rate of patients carrying a mutation ranged from 25% in Ohtahara syndrome to <5% in patients with an epilepsy beginning after 3 months of age. Epilepsy
improved over time for most patients, with an evolution to West syndrome in half. Patients had moderate to severe developmental delay with normal head growth. Cerebellar syndrome with ataxic gait and/or tremor was present in 60%.

Our data confirm that STXBP1 mutations are associated with neonatal-infantile epileptic encephalopathies. The initial key features highlighted in the cohort of early epileptic patients are motor seizures either focal or generalized, abnormal initial interictal EEG, and normal head growth. In addition, we constantly found an ongoing moderate to severe developmental delay with normal head growth. Patients often had ongoing ataxic gait with trembling gestures. Altogether these features should help the clinician to consider STXBP1 molecular screening.



>> A new mechanism for early onset epileptic encephalopathy: mutations in KCNQ2 can alter the localization of potassium channels in neurones.

Mutations in the KCNQ2 gene encoding the voltage-dependent potassium M channel Kv7.2 subunit cause either benign epilepsy or early onset epileptic encephalopathy (EOEE). It has been proposed that the disease severity rests on the inhibitory impact of mutations on M current density. In this work, we have analyzed the phenotype of 7 patients carrying the p.A294V mutation located on the S6 segment of the Kv7.2 pore domain. We investigated the functional and subcellular consequences of this mutation and compared it to another mutation associated with a benign epilepsy and affecting the same residue. We report that all the patients carrying the p.A294V mutation presented the clinical and EEG characteristics of EOEE.
In transfected cells mimicking the patients' heterozygous genotype, the current density is reduced similarly for the mutation causing the severe phenotype or the mild phenotype. We found that in neurons, the p.A294V mutation induced a mislocalization of heteromeric mutant channels to the somato-dendritic compartment, while the benign mutation did not affect the localization of the heteromeric channels to the axon initial segment. The p.A294V mutation does not exert a dominant-negative effect on wild-type subunits but alters the preferential axonal targeting of heteromeric Kv7 channels. Our data suggest that the disease severity is not necessarily a consequence of a strong inhibition of M current and that additional mechanisms such as abnormal subcellular distribution of Kv7 channels could be determinant.

To access the article : Abidi et al. 2015 - Neurobiology of Disease



>> Early-onset epileptic encephalopathy as the initial clinical presentation of WDR45 deletion in a male patient.

Variants in the WD repeat 45 (WDR45) gene in human Xp11.23 have recently been identified in patients suffering from neurodegeneration with brain iron accumulation, a genetically and phenotypically heterogeneous condition. WDR45 variants cause a childhood-onset encephalopathy
accompanied by neurodegeneration in adulthood and iron accumulation in the basal ganglia. They have been almost exclusively found in females, and male lethality was suggested. In this article, we describe a male patient suffering from a severe and early neurological phenotype, initially presenting early-onset epileptic spasms in clusters associated with an abnormal interictal electroencephalography showing slow background activity, large amplitude asynchronous spikes and abnormal neurological development. This patient is a carrier of a 19.9-kb microdeletion in Xp11.23 containing three genes, including WDR45. These findings reveal that males with WDR45 deletions are viable, and can present with early-onset epileptic encephalopathy without brain iron accumulation

To access the article : Abidi et al. 2015 - European Journal of Human Genetics



>> Variable clinical expression in patients with mosaicism for KCNQ2 mutations.

Mutations in the KCNQ2 gene, encoding a potassium channel subunit, were reported in patients presenting epileptic phenotypes of varying severity. Patients affected by benign familial neonatal epilepsy (BFNE) are at the milder end of the spectrum, they are affected by early-onset epilepsy but their subsequent neurological development is usually normal. Mutations causing BFNE are often inherited from affected parents. Early infantile epileptic encephalopathy type 7 (EIEE7) is at the other end of the severity spectrum and, although EIEE7 patients have early onset epilepsy too, their neurological development is impaired and they will present motor and intellectual deficiency. EIEE7 mutations occur de novo. Electrophysiological experiments suggested a correlation between the type of mutation and the severity of the disease but intra- and interfamilial heterogeneity exist. Here, we describe the identification of KCNQ2 mutation carriers who had children affected with a severe epileptic phenotype, and found that these individuals were mosaic for the KCNQ2 mutation. These findings have important consequences for genetic counseling and indicate that neurological development can be normal in the presence of somatic mosaicism for a KCNQ2 mutation.

To access the article : Milh et al. 2015 - American Journal of Medical Genetics A



>> Mutations in the TBC1D24 gene cause different types of early onset epilepsy.

In collaboration with Gaétan Lesca in Lyon on one side, and Diane Doummar in Paris on the other side, we have identified several mutations in patients presenting various epileptic phenotypes: familial infantile myoclonic epilepsy or cortical myoclonus with cerebellar ataxia.

To access the articles :

Poulat et al. 2015 - Epilepsy Research

Doummar et al. 2015 - Movement Disorders



>> The French National Research Agency awards a grant to our project on KCNQ2 biology and related epilepsies for the period 2015-2017.

Our project starts with one of the largest cohort of EOEE patients in the world to end with the preclinical evaluation of candidate drugs to treat this devastating disease and improve human health. Meanwhile, our experiments will contribute to improve the knowledge of normal and mutant potassium channel biology and to understand the mechanisms leading to epilepsy and to the severe neurological dysfunction caused by KCNQ2 mutations. We believe that this is a unique opportunity to simultaneously provide new insights into basic mechanisms of neuronal excitability and the neurobiology of potassium channels, and to test therapeutic strategies in a domain where not much is offered to the affected families. The project will also generate a large amount of new tools and models that will be useful in the long term for all those interested by potassium channel, whether they are clinical neurologists, molecular geneticists or basic neuroscientists.

This project is collaboration between our group, the group of Laurent Aniksztejn (INMED, Marseille) and Jérôme Devaux (CRN2M, Marseille).



>> Mutations in SLC13A5 cause autosomal-recessive epileptic encephalopathy with seizure onset in the first days of life.

In collaboration with Laurence Faivre and Jean-Baptiste Rivière, we ascertained two multiplex families (including one consanguineous family) consistent with an autosomal-recessive inheritance pattern of EE. All seven affected individuals developed subclinical seizures as early as the first day of life, severe epileptic disease, and profound developmental delay with no facial dysmorphism.Screening of 68 additional unrelated individuals with early-onset epileptic encephalopathy for SLC13A5 mutations led to identification of one additional subject with compound heterozygous mutations of SLC13A5 and a similar clinical presentation as the index subjects. These findings underline the value of careful clinical characterization for genetic investigations in highly heterogeneous conditions such as EEP and further highlight the role of citrate metabolism in epilepsy.

To access the article : Thevenon et al. 2014 - American Journal of Human Genetics



>> Mutations in the TBC1D24 gene are causing familial malignant migrating partial seizures of infancy (MMPSI).

Our team has identified in 2012, with the support of the Rare Disease Foundation and its program for high-throughput sequencing,
mutations in a family suffering from malignant migrating partial seizures in two affected children. The mutations affect the TBC1D24 gene, a gene important for the development of neuronal arborization. In collaboration with the Italian team of Anna Fassio, specialist of that gene, we demonstrated that the mutations identified in our patients severely disturb the function of the gene TBC1D24 and impact the maturation of neuronal circuits.

To access the article : Milh et al. 2013 - Human Mutation



>> Mutations of the KCNQ2 gene are causing the disease in almost a quarter of neonates suffering from early onset epileptic encephalopathy.

We studied 71 neonates with severe epilepsy without cerebral morphological abnormalities and have identified a neomutation of the KCNQ2 gene in 16 of them (23%). Mutations in the KCNQ2 gene are responsible for a very mild form of epilepsy called benign familial neonatal seizures (BFNS). In our series, almost all patients began their epilepsy very early (first week of life), with stormy motor seizures. However, unlike BFNS cases, the background rhythm of the interictal EEG was abnormal and the evolution of the disease was generally severe: 15 of 16 children had delayed acquisitions, 11 did not walk before the age of 3 years and only two of them have acquired language !

To access the article : Milh et al. 2013 - Orphanet Journal of Rare Disease



>> Epileptic and nonepileptic features in patients with STXBP1 mutations

Our objective was to refine the phenotype associated with STXBP1 aberrations in early onset epileptic syndromes. To this aim, we screened 52 patients with early onset epilepsy (first seizure observed before the age of 3 months), no cortical malformation on brain magnetic resonance imaging (MRI), and negative metabolic screening. All patients were screened through multiple video-electroencephalography (EEG) recordings for a time period spanning from birth to their sixth postnatal month. Subsequently, patients had standard EEG or video-EEG recordings.

We found 5 novel STXBP1 mutations in patients for whom video-EEG recordings could be sampled from the beginning of the disease. All patients with a mutation displayed Ohtahara syndrome, since most early seizures could be classified as epileptic spasms and since the silent EEG periods were on average shorter than bursts. Each patient displayed frequent nonepileptic movement disorders that could easily be mistaken for epileptic seizures. These movements could be observed as early as the neonatal period.

These results confirm that STXBP1 is a major gene to screen in cases of Ohtahara syndrome, since it is mutated in >10% of the Ohtahara patients within our cohort. This gene should particularly be tested in the case of a surprising evolution of the patient condition if epileptic seizures and EEG paroxysmal activity disappear and are replaced by fast rhythms after the end of the first postnatal year.

To read the corresponding article :  Milh et al. 2011 - Epilepsia



>> GRIN2A mutations EIEE patients

N-methyl-D-aspartate (NMDA) receptors mediate excitatory neurotransmission in the mammalian brain. These receptors are composed of several subunits, including two glutamate-binding subunits (NR2A and NR2B). The NR2A subunit is encoded by the GRIN2A gene and the NR2B subunit by the GRIN2B gene.

In collaboration with the group of Kerstin Kutsche and Lionel van Maldergem, we discovered mutations in these two genes in patients affected by intellectual disability and/or epilepsy. For GRIN2A, we contributed to the identification of a mutation in an EIEE patient. This specific mutation strongly alters the function of the NMDA receptor.

These findings suggest that disturbances in the neuronal electrophysiological balance during development result in variable neurological phenotypes depending on which NR2 subunit of NMDA receptors is affected..

To read the corresponding article :  Endele et al. 2010 - Nature Genetics


>> Identification of new EEIE genes

Our strategy for this project is summarized in the figure below. We are excluding all cases with abnormal MRI or abnormal metabolic assessment.

On the cases with a normal MRI and normal metabolic parameters, we start by genotyping the most frequently mutated genes (ARX, STXBP1, SLC25A22, GRIN2A, KCNQ2 etc.) depending on the familial and clinical history.

If these analysis are normal, we proceed to high resolution comparative genomic hybridization (720K arrays) in search for inframicroscopic chromosomal rearrangements.

If this analysis is negative, we keep the biological sample for additional research projects. Thanks to a grant from the Rare Disease Foundation, an exome sequencing project was performed in perfectly phenotyped and genotyped patients (without known causes) and continues as more cases are available.

If we identify a mutation in a gene or a small chromosomal rearrangement, we are studying the segregation of this anomaly in the family of the patient. We do not study inherited variants.

We have already identified mutations and de novo microscopic chromosomal rearrangements in the available cohort. These cases are currently being studied in more detail. Given that this condition is very rare, we are looking for collaborations (see our dedicated page).


Patient EESB


Metabolic screen






Genotyping of the most frequent causes


     Whole genome CGH


  Research project




de novo