Epileptic encephalopathies are a heterogeneous group of epilepsy syndromes. Using next-generation sequencing, Hernandez et al. identify six de novo GABR mutations – including novel GABRA5 and GABRA1 mutations – in patients with early-onset epileptic encephalopathies. The mutations impair GABAA receptor function and biogenesis, and are likely to alter GABAergic network excitability.
Keywords: GABAA receptors, encephalopathy, GABRA1, GABRA5, GABAergic synapses
Abstract
We performed next generation sequencing on 1696 patients with epilepsy and intellectual disability using a gene panel with 480 epilepsy-related genes including all GABAA receptor subunit genes (GABRs), and we identified six de novo GABR mutations, two novel GABRA5 mutations (c.880G>T, p.V294F and c.1238C>T, p.S413F), two novel GABRA1 mutations (c.778C>T, p.P260S and c.887T>C, p.L296S/c.944G>T, p.W315L) and two known GABRA1 mutations (c.335G>A, p.R112Q and c.343A>G, p.N115D) in six patients with intractable early onset epileptic encephalopathy. The α5(V294F and S413F) and α1(P260S and L296S/W315L) subunit residue substitutions were all in transmembrane domains, while the α1(R112Q and N115R) subunit residue substitutions were in the N-terminal GABA binding domain. Using multidisciplinary approaches, we compared effects of mutant GABAA receptor α5 and α1 subunits on the properties of recombinant α5β3γ2 and α1β3γ2 GABAA receptors in both neuronal and non-neuronal cells and characterized their effects on receptor clustering, biogenesis and channel function. GABAA receptors containing mutant α5 and α1 subunits all had reduced cell surface and total cell expression with altered endoplasmic reticulum processing, impaired synaptic clustering, reduced GABAA receptor function and decreased GABA binding potency. Our study identified GABRA5 as a causative gene for early onset epileptic encephalopathy and expands the mutant GABRA1 phenotypic spectrum, supporting growing evidence that defects in GABAergic neurotransmission contribute to early onset epileptic encephalopathy phenotypes.
Introduction
Epileptic encephalopathies are a heterogeneous group of epilepsy syndromes associated with severe recurrent seizures and neurodevelopmental impairment (Helbig and Tayoun, 2016). Early onset epileptic encephalopathies (EOEE) are invariably associated with seizure onset occurring during the first few years of life, cognitive regression and intellectual disability (Nieh and Sherr, 2014). Twenty-five to 40% of them have no known acquired cause, and their EOEE syndrome is thought to have a genetic aetiology (McTague et al., 2016; Axeen and Olson, 2018). Next-generation sequencing data have indicated that sporadic de novo mutations are a major cause of these disorders (Hamdan et al., 2017). Further, more than 30 genes have been strongly established as causative for genetic epileptic encephalopathies (Helbig et al., 2016, 2017). De novo pathogenic mutations in the GABAA receptor α1 subunit gene (GABRA1) have been associated with a broad phenotypic range of severe epileptic encephalopathies including Ohtahara syndrome, infantile spasms, myoclonic astatic epilepsy (Doose syndrome) and Dravet syndrome (Johannesen et al., 2016; Kodera et al., 2016). In addition, the epilepsy phenotypic spectrum of de novo mutations in GABAA receptor β3 and γ2 subunit genes (GABRB3 and GABRG2, respectively) include Lennox-Gastaut syndrome, infantile spasms and Dravet syndrome (Allen et al., 2013; Hamdan et al., 2014; Hernandez et al., 2017a, b; Le et al., 2017; Moller et al., 2017; Shen et al., 2017). Conversely, no association of GABRA5 and EOEE has been established.
GABAA receptor α subunits show high levels of structural conservation with β and γ subunits (Hibbs and Gouaux, 2011; Miller and Aricescu, 2014; Du et al., 2015; Laverty et al., 2019). All six de novo mutations identified in our cohort produced residue changes clustered within or in close proximity to the GABA-binding pocket (GABRA1: c.G335A, p.R112Q and c.A343G, p.N115D) or transmembrane domains forming the ion channel pore of the receptor (GABRA5: c.G880T, p.V294F and c.C1238T, p.S413F; GABRA1: c.T778C, p.P260S and c.T887C, p.L296S; c.G944T, p.W315L) (Table 1). Indeed, we and others have reported mutations in other GABRs that change residues in the N-terminal domain as well as the pore region of GABAA receptors (Huang et al., 2014; Hernandez et al., 2016, 2017a, b; Janve et al., 2016; Ishii et al., 2017; Shen et al., 2017). The de novo GABRA1 p.R112Q and p.N115D residue substitutions have been detected recurrently in unrelated patients within the previous cohort studies (Carvill et al., 2014; Johannesen et al., 2016; Kodera et al., 2016). Clustering of de novo mutations in GABRA that code for canonical domains of GABAA receptor α subunits, which are closely related to the gating of the channel, might result in impaired inhibitory synaptic transmission. We hypothesized that common synaptic defects that are likely to modulate GABAergic network excitability are altered in patients carrying EOEE-associated GABRA5 and GABRA1 mutations.
Table 1.
Pathogenicity prediction of GABRA5 and GABRA1 missense mutations in six individuals with EOEE
| Gene | SNPs | Protein substitution | Sequence conservation | Structural domain | PolyPhen-2 | SIFT | MutationTaser | ExAC | Recurrence |
|---|---|---|---|---|---|---|---|---|---|
| GABRA5 | c.880G>T | p.V294F | Conservative | Pore (M2) | Probably damaging | Deleterious | Disease causing | NF | Novela |
| GABRA5 | c.1238C>T | p.S413F | Non-conservative | Intracellular (N-tail M4) | Benign | Tolerated | Polymorphism | NF | Novel |
| GABRA1 | c.335G>A | p.R112Q | Non-conservative | Binding (β3 strand) | Benign | Tolerated | Disease causing | NF | Six casesb |
| GABRA1 | c.343A>G | p.N115D | Non-conservative | Binding (α3 helix) | Probably damaging | Tolerated | Disease causing | NF | Two casesc |
| GABRA1 | c.778C>T | p.P260S | Identical | Pore (M1) | Probably damaging | Deleterious | Disease causing | NF | Noveld |
| GABRA1 | c.887T>C | p.L296S | Non-conservative | Pore (M2) | Probably damaging | Deleterious | Disease causing | NF | Novel |
| GABRA1 | c.944G>T | p.W315L | Non-conservative | Pore (M3) | Probably damaging | Tolerated | Disease causing | NF | Novel |
aOne unrelated case with severe epilepsy and developmental delay were described by Butler et al. (2018) with the de novo p.Val294Leu variant.
bSix unrelated cases reported by Carvill et al. (2014) (two cases), Kodera et al. (2016) (one case), Johannesen et al. (2016) (two cases) and the present study (one case).
cTwo unrelated cases reported by Johannesen et al. (2016) (one case) and the present study (one case).
dTwo unrelated cases with Ohtahara and West syndromes were described by Kodera et al. (2016) with the de novo p.Pro260Leu substitutions.
NF = not found.
Materials and methods
Patient phenotypes
A cohort of 1969 patients with epilepsy and intellectual disability were screened by a gene panel targeting 480 epilepsy-related genes that included all GABAA receptor subunit genes (GABRs) (Hernandez et al., 2017b). The patients were recruited from the Department of Pediatrics at the Peking University First Hospital from 2006 to 2015. Clinical work-up of all patients showed no definite perinatal brain injury, no hypoxia, and ischaemia, infection of the CNS or cranial trauma. No evidence of typical inherited metabolic disorders or specific neurodegenerative disorders based on clinical features, neuroimaging or blood/urinary metabolic diseases screening were found. Additionally, normal routine karyotyping and the detection of chromosome sub-telomeric rearrangements with multiplex ligation-dependent probe amplification (MLPA) showed no abnormalities, strongly suggesting a genetic aetiology. Six de novo GABRA1 and GABRA5 mutations were identified in six patients who were clinically diagnosed as having epilepsy and severe intellectual disability of unknown origin and diagnosed as EEOE (Tables 2 and 3). The parents in the six trios had no epilepsy or any related history. All genomic DNA used in the experiments was extracted from peripheral leucocytes. The study was approved by the Ethics committee of Peking University First Hospital, and written informed consents were obtained from all patients or their legal guardians.
Table 2.
Clinical features in two individuals with EOEE and GABRA5 missense mutations
| Patient 1 | Patient 2 | |
|---|---|---|
| Inheritance | De novo | De novo |
| Amino acid change | V294F | S413F |
| Sex | Male | Male |
| Diagnosis | EOEE | EOEE |
| Current age | 3 years, 10 months | 7 years |
| Age at onset | 4 months | 3 months |
| Seizure types | Focal seizures with generalization, febrile seizures, and status epilepticus | Focal and tonic seizures. Epileptic spasms |
| EEG | Generalized slow waves in background activity. Multifocal sharp waves during attack period | Spams in clusters with focal seizures and hypsarrhythmia |
| Motor development | Severe delay | Severe delay |
| Cognitive outcome | Severe intellectual disability | Severe intellectual disability |
| MRI findings | Frontotemporal atrophy and thin corpus callosum | Cortical atrophy and thin corpus callosum |
Table 3.
Clinical features in four individuals with EOEE and GABRA1 missense mutations
| Patient 1 | Patient 2 | Patient 3 | Patient 4 | |
|---|---|---|---|---|
| Inheritance | De novo | De novo | De novo | De novo |
| Amino acid change | R112Q | N115D | P260S | L296S/W315L |
| Sex | Male | Male | Female | Female |
| Diagnosis | EOEE | EOEE | West syndrome | West syndrome |
| Current age | 19 years | 4 years, 5 months | 3 years, 1 month | 5 years, 9 months |
| Age at onset | 8 months | 12 months | 4 months | 8 months |
| Seizure types | Partial seizures, generalized tonic clonic seizures | Partial seizures, generalized tonic clonic seizures, febrile seizures. | Epileptic spasms | Epileptic spasms |
| EEG | Persistent diffuse slow background activity | Generalized discharges during seizures | Spams in clusters and hypsarrhythmia | Hypsarrhythmia or atypical hypsarrhythmia |
| Motor development | Nearly normal | Mild delay | Severe delay | Severe delay |
| Cognitive outcome | Severe intellectual disability | Severe intellectual disability | Severe intellectual disability | Severe intellectual disability |
| MRI findings | Cortical atrophya | Thin bilateral para-hippocampal gyrus | Thin corpus callosum | Frontotemporal atrophy |
aCT scan.
Next generation sequencing and variant validation analysis
The gene panel targeted 480 epilepsy-related genes found in several databases (OMIM, www.omim.org; HGMD, www.hgmd.cf.ac.uk; and EpilepsyGene, http://61.152.91.49/EpilepsyGene/) that were synthesized using the Agilent SureSelect Target Enrichment technique (Zhongguancun Huakang Gene Institute, China) (Hernandez et al., 2017b). We used the American College of Medical Genetics and Genomics (ACMG) guidelines to evaluate the pathogenicity of mutations before they were selected for functional study. Missense mutations were cross-validated according to known polymorphism databases including the 1000 Genomes project (Abecasis et al., 2012), ESP (Exome Variant Server), ExAC (Exome Aggregation Consortium) (Lek et al., 2016) and Ensembl (Aken et al., 2016) (Human_GRCh37/hg19). In addition to determining population frequency, the four GABRA1 and two GABRA5 variants identified in the patient were filtered for call quality and frequency in the Genome Aggregation Database (gnomAD). They were all absent from gnomAD, supporting their pathogenicity. More than 98% of targets had at least 30× (average 136×) coverage. Validation of suspicious mutations as well as segregation analysis in both parents of the six patients were performed by post hoc standard Sanger sequencing. Subsequent parental testing revealed the mutations to be de novo. Deleterious impact of missense mutations on protein structure and DNA sequence alterations were scored with PolyPhen-2 (Adzhubei et al., 2010), SIFT (Ng and Henikoff, 2001) and MutationTaster (Schwarz et al., 2010) (Table 1).
DNA constructs
The coding sequences of human α5 (NM_000810.3), α1 (NM_000806.5), β3 (NM_021912.4) and γ2L (NM_198904.2) GABAA receptor subunits and EGFP (LC008490.1) were cloned into pcDNA3.1(+) expression vectors (Invitrogen). Mutant α5 and α1 subunit cDNA constructs were generated using the QuikChange Site-Directed Mutagenesis Kit (Agilent) and confirmed by DNA sequencing. The haemagglutinin (HA) epitope was inserted between the fourth and fifth residue of the mature α5 subunit, a functionally silent position (Connolly et al., 1996).
Primary hippocampal neuronal culture
Rat hippocampal neurons were obtained from embryonic Day 17.5 embryos. Briefly, dissociated cells were plated at a density of 6.5 × 104 cells/cm2 onto 12-mm round coverslips in 24-well plates coated with poly-l-ornithine (0.5 mg/ml; Sigma). Hippocampal neurons were incubated at 37°C in a 5% CO2 incubator and maintained in serum-free Neurobasal™ medium (Gibco) supplemented with B27 supplement (Gibco), glutamine (Gibco) and penicillin/streptomycin (Gibco, 20 U/ml). For electrophysiology studies, hippocampal neurons were transfected at days in vitro (DIV) 7 with 2 µg of wild-type or mutant subunits and 0.5 µg of EGFP using X-tremeGENE™ 9 DNA transfection reagent, and recordings were obtained at DIV 14–19. For confocal studies, hippocampal neurons were transfected at DIV 10 using Lipofectamine® 2000 transfection reagent (Invitrogen) and imaging was obtained at DIV 12.
Non-neuronal cell culture
HEK293T cells (ATCC, CRL-11268) were cultured at 37°C in a humidified 5% CO2 incubator and maintained in Dulbecco’s modified Eagle medium (Invitrogen) supplemented with 10% foetal bovine serum (Life Technologies) and 100 IU/ml penicillin/streptomycin (Life Technologies). For western blot and surface biotinylation experiments, cells were transfected using polyethylenimine (PEI) reagent (40 kD, Polysciences) at a DNA:transfection reagent ratio of 1:2.5, and harvested 36 h after transfection. To express wild-type and mutant α5β3γ2 and α1β3γ2 receptors, 3 µg of subunit DNAs were transfected at a ratio of 1:1:1 into 6 cm dishes for most experiments except for whole-cell recording. For the mock-transfected condition, empty pcDNA3.1 vector was added to make a final DNA transfection amount of 3 μg. Whole cell recordings were obtained from HEK293T cells plated onto 12 mm coverslips at a density of 4 × 104 in 35 mm culture dishes (Corning) and transfected after 24 h with 0.3 μg cDNA of each α5, α1, β3 and γ2 subunit, and 0.05 µg of EGFP using X-tremeGENE™ 9 DNA transfection reagent (Roche Diagnostics). Recordings were obtained 48 h after transfection.
Immunocytochemistry and confocal microscopy
Coverslip-grown hippocampal neurons or HEK293T cells were fixed with Prefer (Anatech) and permeabilized with 0.2% Triton™ X-100 for 15 min to stain total proteins. The fixed/permeabilized cells were blocked for 2 h with 5% bovine serum albumin in phosphate-buffered saline and then stained with primary antibodies overnight, followed by incubation in Alexa 488-conjugated donkey anti-rabbit IgG antibodies and Cy3-conjugated donkey anti-mouse IgG antibodies. Primary antibodies used were as follows: rabbit monoclonal HA antibody (Cell Signaling), mouse monoclonal HA antibody (BioLegend), rabbit polyclonal MAP2 antibody (Cell Signaling), mouse monoclonal anti-gephyrin antibody (Synaptic Systems), mouse monoclonal anti-calnexin antibody (Abcam). Coverslips were mounted with ProLong® Gold antifade reagent (Thermo Fisher Scientific).
Confocal images were obtained using a Zeiss LSM 710 Meta inverted confocal microscope. Stained hippocampal neurons or HEK293T cells were excited with the 488 nm laser for the Alexa 488 fluorophore signal and the 543 nm laser for the Cy3 fluorophore signal. Co-localization analysis of hippocampal neurons was performed using the Coloc2 plugin in the open source image processing program Fiji (Schindelin et al., 2012). Both Pearson’s correlation coefficient (R) and Manders’ co-localization coefficient (MCC) were calculated (Shen et al., 2017).
Western blot and surface biotinylation of non-neuronal cells
Transfected HEK293T cells were collected in modified RIPA buffer [50 mM Tris (pH = 7.4), 150 mM NaCl, 1% NP-40, 0.2% sodium deoxycholate, 1 mM EDTA] and 1% protease inhibitor cocktail (Sigma). Collected samples were subjected to gel electrophoresis using 4–12% BisTris NuPAGE precast gels (Invitrogen) and transferred to PVDF-FL membranes (Millipore). Primary antibodies against human α5 (Millipore), α1 (NeuroMab), β3 (Novus Biologicals), and γ2 subunits (Millipore) were used to detect the subunits at a 1:500 dilution. Anti-Na+/K+ ATPase antibody (Developmental Studies Hybridoma Bank) at a 1:1000 dilution was used as a loading control. IRDye® (LI-COR Biosciences) conjugated secondary antibody was used at a 1:10 000 dilution in all cases. Membranes were scanned using the Odyssey Infrared Imaging System (LI-COR Biosciences). The integrated density value of bands was determined using the Odyssey Image Studio software (LI-COR Biosciences) (Hernandez et al., 2017a, b).
GABA-evoked currents of non-neuronal cells
Whole-cell recordings of wild-type and mutant GABAA receptor currents were obtained at room temperature from lifted HEK293T cells (Hernandez et al., 2017a, b). The external solution was composed of (in mM): 142 NaCl, 8 KCl, 10 D(+)-glucose, 10 HEPES, 6 MgCl2.6H2O, and 1 CaCl2 (pH 7.4, ∼326 mOsm). The internal solution consisted of (in mM): 153 KCl, 10 HEPES, 5 EGTA 2 Mg-ATP, and 1 MgCl2.6H2O (pH 7.3, ∼300 mOsm). GABA (1 mM) was applied for 4 s for measurements of current amplitude. The currents were recorded using an Axopatch 200B amplifier (Axon Instruments), low-pass filtered at 2 kHz using the internal 4-Pole Bessel filter of the amplifier, digitized at 10 kHz with Digidata 1550 (Axon Instruments) and stored for offline analysis as described previously.
GABAA receptor-mediated miniature inhibitory postsynaptic currents
Miniature inhibitory postsynaptic currents were recorded from DIV 14–21 cultured hippocampal neurons (Hernandez et al., 2017a). Whole-cell patch-clamp recordings were obtained at room temperature (22–23°C) at a holding potential of −60 mV. Hippocampal neurons were perfused with an extracellular solution containing 145 mM NaCl, 3 mM KCl, 1.5 mM CaCl2,1 mM MgCl2.6H2O, 10 mM glucose, and 10 mM HEPES (pH 7.4, ∼320 mOsm). The internal solution consisted of 135 mM CsCl, 2 mM MgCl2.6H2O, 10 mM EGTA, 5 Mg-ATP, and 10 mM HEPES (pH 7.2, ∼300 mOsm). GABAA receptor-mediated mIPSCs were recorded by blocking both action potentials and excitatory neurotransmission. Thus, tetrodotoxin (TTX, 0.5 μM), d(−)-2-amino-5-phosphonopentanoate (AP5, 40 μM), and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 μM) were added the extracellular solution.
Statistical analysis
Numerical data were reported as mean ± standard error of the mean (SEM). Statistical analysis was performed using GraphPad Prism (GraphPad Software 6.07). Statistically significant differences were taken as P < 0.05 using one-way ANOVA followed by Dunnett’s multiple comparison test.
Data availability
The data supporting the findings of this study are available within the article and its Supplementary material.
Results
Patient phenotypes with de novo GABRA5 missense mutations found in cases with EOEE
The clinical features of the two patients with GABRA5 mutations are summarized in Table 2, and their representative EEG and brain MRI images presented in Fig. 1A and B. Both patients were diagnosed with EOEE. The seizures began within the first 4 months of life. Seizure semiology at onset was described as partial seizures with secondary generalization, febrile seizures and status epilepticus in Patient 1, and epileptic spasms and tonic seizures in Patient 2. A variety of EEG abnormalities were found in Patient 1 including generalized slow-wave epileptiform discharges and multifocal sharp waves (Fig. 1A). Patient 2 showed some symptoms of Ohtahara syndrome at an early age, which were evolving towards symptoms of infantile spasms and West syndrome at a later age. Figure 1B showed an abnormal interictal pattern on EEG (i.e. hypsarrhythmia at 5 months) found in Patient 2 at 5 months old. Brain MRIs showed cortical atrophy and thinning of the corpus callosum (Fig. 1A and B). Seizures remained intractable as of last follow up despite combination therapy with antiepileptic drugs for both patients. Developmentally, these patients had severe intellectual disability, were nonverbal and had severe motor disabilities.
Figure 1.
Novel de novo GABRA5 and GABRA1 mutations were identified in individuals with EOEE. Segregation analysis (left), representative EEGs (middle) and brain MRIs (right) are presented for patients with two novel GABRA5 mutations (A and B) and two novel GABRA1 mutations (C and D) that were identified in four trios (Patients 1 and 2, Table 2; Patients 3 and 4, Table 3). On the left of the panels, the filled symbols indicate probands, and the arrows on Sanger chromatograms indicate the missense nucleotide. Four representative EEGs are presented. The top EEG demonstrates generalized slow waves in background activity (A) and typical hypsarrhythmia (B). The lower two EEG traces show more spasms in clusters and hypsarrhythmia (C) and atypical hypsarrhythmia (D). All the patients had common abnormalities in their brain MRIs (right). The four patients had cortical atrophy, and thin corpus callosum as shown (open arrows). All images are presented at the level of head of caudate. On the left for each trio, square symbols indicate males, circles indicate females, filled symbols indicate the affected index case and the open symbols indicate the unaffected parents.
Patient phenotypes with de novo GABRA1 missense mutations found in cases with EOEE
The clinical features of the four patients with GABRA1 mutations were summarized in Table 3. Representative EEG and brain MRI images of Patients 3 and 4 are shown in Fig. 1C and D. The age of onset of epilepsy was within the first year of life in all four patients (range 4 to 12 months of age). Seizure semiologies at onset were described as partial seizures with secondary generalized tonic-clonic seizures in Patients 1 and 2, epileptic spasms in Patient 3 (Fig. 1C), and myoclonic seizures in Patient 4 (Fig. 1D). These patients were examined frequently to determine if they fit criteria for a specific EEG syndrome diagnosis (i.e. hypsarrhythmia). Patients 1 and 2 were diagnosed with EOEE, and Patients 3 and 4 eventually had features of West syndrome when multiple individual spasms were observed. Epilepsy outcome was variable, with Patient 1 eventually becoming seizure free for nearly 1 year after a conventional antiepileptic drug combination of pentobarbital, valproic acid, clonazepam and phenytoin. Patient 3 responded poorly to antiepileptic drugs, but epileptic spasms were reduced by a combination of adrenocorticotropic hormone and MgSO4. In contrast, the seizures in Patients 2 and 4 remained intractable despite combination therapy with antiepileptic drugs. Developmentally, all four individuals had severe intellectual disability, were nonverbal, and had different degrees of motor disabilities, ranging from nearly normal (Patient 1) to mild (Patient 2) or to severe motor delay (Patients 3 and 4). A variety of interictal epileptiform discharges were observed on EEGs in this cohort including persistent diffuse slow background activity (Patient 1), generalized discharges (Patient 2), hypsarrhythmia (Patients 3 and 4), and bilateral multi-spikes (all patients). Brain MRIs (Patients 2–4) and CT scan (Patient 1) were abnormal in all patients (Fig. 1C and D shows only Patients 3 and 4).
Altered synaptic distribution and gephyrin clustering of two novel EOEE mutant α5 subunits in primary hippocampal neurons
To determine whether GABAA receptors containing EOEE-associated mutant α5(V294F) or α5(S413F) subunits had defective synaptic localization in dendrites, we expressed HA-fusion-proteins encoding wild-type α5 subunits or mutant α5(V294F) or α5(S413F) subunits in hippocampal neurons (DIV 12). We used double immunostaining for MAP2 and HA to demonstrate that α5 subunit-containing GABAA receptors formed clusters at GABAergic synapses in dendrites of cultured hippocampal neurons (Fig. 2A). On dendrites we observed a marked decrease of puncta containing HA in neurons expressing mutant α5(V294F)HA subunits compared with those expressing wild-type α5 subunits (Fig. 2A). In contrast, in neurons expressing mutant α5(S413F)HA subunits, the dendrites displayed numerous synaptic contacts containing HA, similar to those observed in neurons expressing wild-type α5 subunits.
Figure 2.
Altered synaptic distribution and gephyrin clustering of mutant α5 subunit-containing GABAA receptors in hippocampal neurons in cell culture. (A) Hippocampal neurons in cell culture (DIV 10) were transfected with wild-type (wt) α5HA, subunits or mutant α5(V294F)HA or α5 (S413F)HA subunits and were fixed 48 h later. Neurons were stained with anti-HA antibodies (red) and MAP2 (green). Staining patterns were characterized using confocal microscopy. The right of the panel contains enlarged images of the neurites in white boxes. (B) Representative mISPC traces recorded at −60 mV in cultured hippocampal neurons over-expressing mutant α5(V294F)HA or α5(S413F)HA subunits or wild-type α5HA subunits are presented. Bottom bar graphs represent summarized mISPC data. Values are expressed as mean ± SEM (see text). One-way ANOVA with Dunnett’s post-test was used to determine significance compared to the wild-type condition. (C) Gephyrin clustering patterns were revealed by confocal microscopy. Neurons were stained with anti-HA (green) and anti-gephyrin (red) antibodies. The bottom of the panel contains enlarged images of the neurites in white boxes. (D) Statistical analyses of gephyrin clustering were performed using MCC. Values are expressed as mean ± SEM (see text). One-way ANOVA with Dunnett’s post-test was used to determine significance compared to the wild-type condition.
We then asked whether the reduction of dendritic expression of mutant α5(V294F) subunit-containing receptors altered the efficacy of GABAergic inhibition of hippocampal neurons (Fig. 2B). We measured GABAA receptor-mediated mIPSCs in neurons transfected with wild-type α5 subunits and those transfected with mutant α5(V294F) or α5(S413F) subunits recorded in the presence of TTX, AP5, and CNQX to block voltage-gated sodium channels and glutamate receptors, respectively.
As expected, mean mIPSC amplitude in hippocampal neurons transfected with mutant α5(V294F)HA subunits was significantly decreased (11.5 ± 1.6 pA, P = 0.0001, n = 5) relative to that of neurons transfected with wild-type α5 subunits (25.2 ± 1.7 pA, n = 5). Further, mean mIPSC amplitude in hippocampal neurons transfected with mutant α5(S413F) subunits was significantly reduced (18.5 ± 0.4 pA, P = 0.0086, n = 5) relative to hippocampal neurons transfected with wild-type α5 subunits (25.2 ± 1.7 pA) (Fig. 2B, bottom left).
The decay time of mIPSCs in neurons transfected with mutant α5(V294F) subunits was decreased (28.30 ± 4.02 ms, P = 0.0001, n = 5), but the decay time of mIPSCs in neurons transfected with mutant α5(S413F) subunits (72.11 ± 5.17 ms, P = 0.1608, n = 5) was similar to the decay time of mIPSCs in neurons transfected with wild-type α5 subunits (82.96 ± 3.20 ms, n = 5) (Fig. 2B, bottom middle).
The rise time of mIPSCs in neurons transfected with mutant α5(S413F) subunits was increased (5.54 ± 0.81 ms, P = 0.0347, n = 5), but the rise time of mIPSCs in neurons transfected with mutant α5(V294F) subunits (3.32 ± 0.61 ms, P = 0.9608, n = 5) was similar to that of mIPSCs in neurons transfected with wild-type α5 subunits (3.11 ± 0.41 ms, n = 5) (Fig. 2B, bottom right). These results demonstrated that the decreased dendritic puncta caused by mutant α5(V294F) subunits was associated with reduced GABAergic inhibition in hippocampal neurons. In contrast, the unaltered dendritic localization in neurons transfected with mutant α5(S413F)HA subunits did not explain the reduced GABAergic inhibition, and thus, other mechanisms must be responsible.
It is well known that α5 subunit-containing GABAA receptors are clustered at synapses by interacting with gephyrin (Christie and de Blas, 2002; Jacob et al., 2005; Serwanski et al., 2006). We analysed the synaptic localization of wild-type α5 subunits and mutant α5(V294F)HA and α5(S413F)HA subunits in hippocampal neurons by comparing their co-localization with gephyrin (Fig. 2C and D). First, we observed an inverse relationship in the number of clusters co-localized with gephyrin and the type of α subunit-containing GABAA receptors. Endogenous wild-type α1 and exogenous wild-type α5 subunit-containing receptors were both co-localized with gephyrin, but the clusters were larger with α1 than with α5 subunit-containing receptors (α1-MCC = 0.72 ± 0.01, n = 3; α5-MCC = 0.25 ± 0.02, n = 12, P = 0.0001). When comparing the gephyrin clusters with mutant subunit expressing receptors, the mutant α5(V294F)HA subunits presented more clusters than wild-type α5 subunits (MCC = 0.41 ± 0.03, n = 11, P = 0.0001), and the mutant α5(S413F) subunits had significantly more gephyrin clusters than the wild-type α5 subunits (MCC = 0.41 ± 0.03, n = 11, P = 0.0001) or mutant α5(S413F) subunits (MCC = 0.62 ± 0.02, n = 18, P = 0.0001). The high level of gephyrin clustering found in α5(S413F)HA subunit expressing neurons was not different from that of endogenous α1 subunit-containing receptors (P = 0.155). These data suggest aberrant redistribution of α5 subunit-containing GABAA receptors at the synapses of both mutant α5(V294F)HA and α5(S413F)HA subunit-expressing neurons.
De novo EOEE mutant α5 subunits trafficked to cell membranes as functional receptors
We observed a differential distribution of mutant α5(V294F)HA and α5(S413F)HA GABAA receptor subunits in somatodendritic compartments of hippocampal neurons (Fig. 2A). While mutant α5(S413F)HA subunits were localized primarily on the dendrites, mutant α5(V294F)HA subunits accumulated in the soma. These results suggested impaired biogenesis of mutant α5(V294F)HA subunits. Further, we sought to determine whether the differences in the somatodendritic distribution of mutant α5(V294F)HA subunit-containing receptors were due to surface trafficking defects that would result in some subunit retention in the endoplasmic reticulum (ER). To assess this, we performed confocal microscopy staining of calnexin, an ER marker that exhibits a typical perinuclear and reticular distribution suggestive of an ER distribution (Fig. 3A), to co-localized mutant α5HA subunits that did not form trafficking-competent receptors and were trapped in the ER. To assess ER retention, permeabilized HEK293T cells were immunolabelled with anti-HA and anti-calnexin antibodies. We found that mutant α5(V294F)HA subunits predominantly localized to the ER as evidenced by their co-localization with calnexin (MCC = 0.68 ± 0.02, n = 3, P = 0.0003), in contrast to mutant α5(S413F)HA subunits that spread outside the ER (MCC = 0.46 ± 0.03, n = 3, P = 0.4115) and showed a distribution similar to wild-type α5HA subunits (MCC = 0.42 ± 0.01, n = 3), suggesting trafficking to the cell surface. Although these results indicated that there was a defect in the trafficking of GABAA receptors containing mutant α5(V294F)HA subunits to the surface membrane, it did not totally exclude the possibility that a fraction of receptors were trafficked to the surface membrane and remained functional, as shown in hippocampal neurons (Fig. 2B).
Figure 3.
Mutant α5 subunits associated with de novo EOEE and co-expressed with β3 and γ2 subunits were trafficked to the cell surface as functional receptors in HEK293T cells. (A) Wild-type α5HA subunits or mutant α5(V294F)HA or α5(S413F)HA subunits were co-expressed with β3 and γ2 subunits in HEK293T cells. The transfected cells were permeabilized, and wild-type and mutant α5HA subunits were labelled with anti-HA antibodies (green). The ER was visualized with anti-calnexin antibody (red). White boxes on the merged images depict the enlarged area shown in the images to the right. Also shown are DAPI nuclear counterstaining (blue) and merge of the staining. (B) Representative GABA-evoked current traces obtained following rapid application of 1 mM GABA for 4 s to lifted HEK293T cells voltage clamped at −20 mV are shown at the top of the panel, and concentration-response curves obtained for wild-type α5 and mutant α5(V294F) and α5(S413F) subunit-containing α5β3γ2 GABAA receptors are shown at the bottom of the panel. (C and D) Surface (C) and total (D) levels of wild-type α5HA or mutant α5(V294F)HA and α5(S413F)HA subunits co-expressed with β3 and γ2 subunits in HEK293T cells. Band intensity of surface and total expressed α5, β3 and γ2 subunits were normalized to the ATPase signal, and the summarized data are shown in the bar graphs. Values are expressed as mean ± SEM (see text). One-way ANOVA with Dunnett’s post-test was used to determine significance compared to the wild-type condition.
To shed light on the functional properties of the receptors expressed on the surface membrane, we measured macroscopic GABA-evoked currents in transfected HEK293T cells (Fig. 3B). Similar to the results shown above, transfection of cells with mutant α5(V294F)HA and α5(S413F)HA subunit-containing GABAA receptors decreased GABA-evoked currents to different extents. While α5(V294F)HA subunits decreased currents by ∼39% (5241 ± 289 pA, n = 15, P = 0.0001), α5(S413F) subunits decreased currents by ∼24% (6498 ± 278 pA, n = 22, P = 0.0009) relative to wild-type currents (8532 ± 88 pA, n = 6) (Fig. 3B). Concomitantly, the sensitivity to Zn2+ inhibition was increased in GABAA receptors containing mutant α5(V294F)HA subunits (25 ± 4%, n = 10, P = 0.0005). In contrast, GABAA receptors containing mutant α5(S413F)HA subunits had a smaller fractional Zn2+ inhibition (14 ± 1%, n = 16, P = 0.3238), similar to that of GABAA receptors containing wild-type α5 subunits (9 ± 2%, n = 8). Unexpectedly, neither mutant α5(V294F)HA nor mutant α5(S413F)HA subunit altered GABA potency, and receptors containing either mutant subunit had GABA concentration-response curves with EC50’s similar to those of GABAA receptors containing wild-type (WT) α5 subunits [EC50(V294F) = 4.33 µM, P = 0.6870, n = 6; EC50(S413F) = 5.84 µM, P = 0.8128, n = 6; EC50(WT) = 5.77 µM, n = 15].
Taken together these results demonstrated that the GABRA5 EOEE-associated mutations decreased GABA-evoked current amplitudes in both neuronal and non-neuronal cells due to impaired biogenesis of receptors leading to decreased or altered expression of surface receptors. To determine to what extent the mutant α5(V294F) and α5(S413F) subunits assembled with β3 and γ2 subunits and trafficked to cell surface membranes, we co-expressed wild-type α5HA subunits or mutant α5(V294F)HA or α5(S413F)HA subunits with β3 and γ2 subunits in HEK293T cells and evaluated surface levels of wild-type and mutant α5 subunits by surface biotinylation (Fig. 3C). Compared to co-expressed wild-type α5 subunits, we found that surface levels of co-expressed mutant α5(V294F)HA subunits were reduced (0.59 ± 0.04, P = 0.0012, n = 4), but no changes were found for mutant α5(S413F)HA subunits (0.99 ± 0.08, P = 0.963, n = 4) (Fig. 3C, top). Furthermore, mutant α5(S413F)HA subunits had a dominant negative effect by decreasing the trafficking of partnering β3 (0.65 ± 0.06, P = 0.0003, n = 4) and γ2 (0.65 ± 0.06, P = 0.0106, n = 4) subunits to the surface (Fig. 3C, middle and bottom). Mutant α5(S413F)HA subunits had a dominant negative effect solely on β3 subunits (0.83 ± 0.01, P = 0.0423, n = 4), with no significant changes in γ2 subunits (0.86 ± 0.09, P = 0.3928, n = 4).
To determine to what extent mutant α5(V294F) and α5(S413F) subunits reduced biogenesis of GABAA receptors, whole-cell lysates were analysed by western blot (Fig. 3D). Total levels of mutant α5(V294F)HA (0.59 ± 0.16, P = 0.0611, n = 4), and co-expressed β3 (0.69 ± 0.08, P = 0.0489, n = 4) and γ2 (0.69 ± 0.07, P = 0.0429, n = 4) subunits were all reduced; in contrast, expression of mutant α5(S413F)HAβ3γ2 subunits did not differ from those of wild-type subunits (α5 1.05 ± 0.05, P = 0.7668; β3 1.01 ± 0.08, P = 0.8874; γ2 0.87 ± 0.09, P = 0.5320, n = 4). Similar findings were found in hippocampal neurons transfected with α5(V294F) and α5(S413F) subunits (data not shown).
Two novel GABRA1 mutations associated with EOEE caused overall reduction of GABAergic synapses of hippocampal neurons
Following the same approach as described above, we sought to determine the synaptic distribution of the mutant α1(P260S) and α1(L296S/W315L) subunits. Thus, we expressed HA fusion proteins encoding wild-type α1HA or mutant α1(P260S)HA or α1(L296S/W315L)HA subunits in hippocampal neurons cultured for DIV 12 (Fig. 4A). We found major decreases in HA-stained dendritic puncta for both mutant α1(P260S)HA and α1(L296S/W315L)HA subunits transfected into hippocampal neurons compared with wild-type α1HA subunit-containing neurons, which displayed a clear pattern of HA-stained synaptic contacts around the dendrites (Fig. 4A). As expected, reduction of dendritic expression of mutant α1(P260S)HA and α1(L296S/W315L)HA subunits significantly decreased the efficacy of GABAergic inhibition of hippocampal neurons (Fig. 4B). In comparison to wild-type α1 subunit-containing neurons (40.6 ± 2.4 pA, n = 5), the amplitudes of mIPSCs in neurons expressing mutant α1(P260S) (29.9 ± 1.9 pA, P = 0.0069, n = 5) or α1(L296S/W315L) (29.3 ± 1.9 pA, P = 0.0046, n = 5) subunits were reduced (Fig. 4B, bottom left). In addition, transfection of either mutant α1(P260S) or α1(L296S/W315L) subunits resulted in faster mIPSC decay times (90.39 ± 4.37 ms, P = 0.0425, n = 5; 69.93 ± 2.44 ms, P = 0.0002, n = 5) and slower rise times (3.59 ± 1.10 ms, P = 0.1793, n = 5; 7.23 ± 1.15 ms, P = 0.0014, n = 5) relative to those recorded in hippocampal neurons transfected with wild-type α1 subunits (decay time, 106.6 ± 5.8 ms; rise time, 1.23 ± 0.41 ms, n = 5) (Fig. 4B, bottom middle and right).
Figure 4.
Two novel EOEE mutant α1 subunits caused an overall reduction of GABAergic IPSCs in hippocampal neurons in cell culture. (A) Hippocampal neurons were transfected with wild-type (wt) α1HA subunits or mutant α1(P260S)HA or α1(L296S/W315L)HA subunits and fixed 48 h later. Neurons were stained with anti-HA (red) and anti-MAP2 (green) antibodies. Staining patterns were characterized using confocal microscopy. The right of the panel contains enlarged images of the neurites in white boxes. (B) Representative mISPCs recorded at −60 mV in cultured hippocampal neurons over-expressing wild-type α1HA subunits or mutant α1(P260S)HA or α1(L296S/W315L)HA subunits are presented. Bottom graphs represent summarized mISPC data. Values are expressed as mean ± SEM (see text). One-way ANOVA with Dunnett’s post-test was used to determine significance compared to the wild-type condition. (C) Wild-type α1HA subunits or mutant α1(P260S)HA or α1(L296S/W315L)HA subunits were co-expressed with β3 and γ2 subunits in HEK293T cells. The transfected cells were permeabilized, and wild-type and mutant α1HA subunits were labelled with anti-HA antibodies (green). The ER was visualized with anti-calnexin antibody (red). White boxes on the merged images depicted the enlarged area shown in the images to the right. Also shown was DAPI nuclear counterstaining (blue) and the merge of the stainings. (D) Representative GABA-evoked current traces obtained following rapid application of 1 mM GABA for 4 s to lifted HEK293T cells voltage clamped at −20 mV are shown at the top of the panel. Concentration-response curves were obtained for wild-type α1 or mutant α1(P260S) or α1(L296S/W315L) subunit-containing α1β3γ2 GABAA receptors. (E and F) Surface (E) and total (F) levels of wild-type α1 or mutant α1(P260S) or α1(L296S/W315L) subunits co-expressed with β3 and γ2 subunits in HEK293T cells are presented. Band intensity of surface and total expressed α1, β3 and γ2 subunits were normalized to the ATPase signal, and the summarized data are shown in the bar graphs. Values are expressed as mean ± SEM (see text). One-way ANOVA with Dunnett’s post-test was used to determine significance compared to the wild-type condition.
Next, we sought to determine whether the preferential distribution of mutant α1(P260S) and α1(L296S/W315L) subunits in the somata of hippocampal neurons (Fig. 4A) was due impaired biogenesis of GABAA receptors containing mutant α1 subunits, which could result in the retention of mutant α1 subunits in the ER (Fig. 4C). We found that both mutant α1(P260S) (MCC = 0.73 ± 0.04, n = 3, P = 0. 0010) and α1(L296S/W315L) (MCC = 0.66 ± 0.02, n = 3, P = 0. 0045) subunits were substantially co-localized with calnexin in the ER (α1 wild-type MCC = 0.45 ± 0.03, n = 3). Although these results suggest that a fraction of the receptors was retained in the ER, functional data showed that a significant fraction of GABAA receptors containing mutant α1 subunits were effectively trafficked to the membrane forming functional receptors (Fig. 4D). Macroscopic peak GABA-evoked currents recorded from HEK293T cells co-transfected with β3, γ2 and mutant α1(P260S) or α1(L296S/W315L) subunits were both decreased by ∼60% (3430 ± 255 pA, n = 29, P = 0.0001; 3913 ± 118 pA, n = 19, P = 0.0001), relative to wild-type currents (9071 ± 273 pA, n = 31). In addition, currents recorded from co-transfected cells with mutant α1(P260S) subunits had unchanged GABA potency and had GABA concentration-response curves with EC50’s similar to those of wild-type α1β3γ2 GABAA receptors [EC50(P260S) = 19.9 µM, P = 0.8020, n = 29; EC50(WT) = 25.6 µM, n = 31]. Conversely, currents recorded from co-transfected cells with the mutant α1(L296S/W315L) subunit had a ∼5-fold increase in GABAA receptor potency (EC50 = 5.03 µM, P = 0.0001, n = 19). To determine whether the differences found in GABA potency were due to an increase in the fraction of binary αβ receptors of mutant α1(L296S/W315L) GABAA receptors, the fractional Zn2+ inhibition was measured. Whereas receptors containing mutant α1(P260S) subunits displayed a fractional Zn2+ inhibition that resembled that of wild-type α1β3γ2 receptors (7 ± 1%, n = 7, P = 0.8567; 8 ± 1%, n = 16, respectively), currents recorded from receptors containing mutant α1(L296S/W315L) subunits had significantly increased Zn2+ sensitivity (34 ± 2%, n = 7, P = 0.0001).
Two novel EOEE mutant α1 subunits decreased both surface and total levels of α1 subunits
We co-expressed β3, γ2 and either wild-type α1, mutant α1(P260S) or mutant α1(L296S/W315L) subunits in HEK293T cells and assessed surface and total expression levels of β3, γ2 and α1 subunits (Fig. 4E and F). Co-transfection of mutant α1(P260S) subunits with β3 and γ2 subunits decreased surface α1 subunit levels (0.55 ± 0.04, n = 4, P = 0.0001) (Fig. 4E, top), with no effect on surface β3 subunit levels (0.94 ± 0.09, n = 4, P = 0.9799) (Fig. 4E, middle) or surface γ2 subunit levels (0.87 ± 0.04, n = 4, P = 0.3162) (Fig. 4E, bottom). Co-transfection of mutant α1(L296S/W315L) subunits with β3 and γ2 subunits decreased surface α1 subunit levels (0.36 ± 0.03, n = 7, P = 0.0001) (Fig. 4E, top) and decreased surface β3 subunit levels (0.81 ± 0.03, n = 7, P = 0.0353) (Fig. 4E, middle), with no effect on surface γ2 levels (0.85 ± 0.02, n = 7, P = 0.0764) (Fig. 4E, bottom).
Co-transfection of mutant α1(P260S) subunits with β3 and γ2 subunits decreased total α1 subunit levels (0.69 ± 0.04, n = 5, P = 0.0028) (Fig. 4F, top), with no effect on total β3 subunit levels (1.08 ± 0.04, n = 5, P = 0.9491) (Fig. 4F, middle) or total γ2 subunit levels (1.02 ± 0.04, n = 5, P = 0.9979) (Fig. 4F, bottom). Co-transfection of mutant α1(L296S/W315L) subunits with β3 and γ2 subunits decreased total α1 subunit levels (0.33 ± 0.06, n = 5, P = 0.0001) (Fig. 4F, top), with no effect on total β3 subunit levels (0.80 ± 0.14, n = 5, P = 0.7042) (Fig. 4F, middle) or total γ2 subunit levels (0.90 ± 0.15, n = 5, P = 0.9822) subunits (Fig. 4F, bottom). Similar findings were found in hippocampal neurons transfected with α1(P260S) and α1(L296S/W315L) subunits (data not shown).
Three de novo GABRA1 mutations were relatively frequently found in cases with EOEE
Previously, five unrelated cases carrying the GABRA1 mutation c.G335A, p.R112Q (Carvill et al., 2014; Johannesen et al., 2016; Kodera et al., 2016), and one case carrying the GABRA1 mutation c.A343G, p.N115D (Johannesen et al., 2016) were found in EOEE, but functional studies have not been reported for either of these mutations. We report two additional cases carrying these mutations (Tables 1 and 3). An aspect that draws attention to these two mutations, and unlike the previous ones described above, these mutations were found in the N-terminal region of GABAA receptors, which means that they are likely in part of the GABA binding domain. We determined the functional consequences of the α1(R112Q) and α1(N115D) subunit amino acid substitutions by measuring macroscopic GABA-evoked currents from cells co-transfected with wild-type or mutant α1 subunits and β3 and γ2 subunits (Fig. 5A, top). Unexpectedly, no changes were found in peak GABA-evoked current amplitudes recorded from receptors containing mutant α1(R112Q) or α1(N115D) subunits (8167 ± 346 pA, n = 17, P = 0.2122; 7878 ± 625 pA, n = 19, P = 0.0668), relative to wild-type receptor currents (9071 ± 273 pA, n = 31). Since these mutations are located in the GABA binding domain, it is likely that these mutations disrupted coupling of GABA binding to channel gating leading to altered GABA potency. Accordingly, both mutant α1(R112Q) and α1(N115D) subunits caused a 2–5-fold decrease in GABAA receptor potency (EC50(R112Q) = 48.3 µM, n = 10, P = 0.0235; EC50(N115D) = 126 µM, n = 10, P = 0.0001) relative to that of wild-type receptors (EC50 = 25.6 µM, n = 31) (Fig. 5A, bottom left). Moreover, consistent with the functional data, none of the mutations reduced surface or total levels of α1 subunits (surface: R112Q, 1.04 ± 0.09, n = 4, P = 0.9261; N115D, 1.03 ± 0.03, n = 7, P = 0.9419; total: R112Q, 0.99 ± 0.06, n = 5, P = 0.9997; N115D, 1.03 ± 0.07, n = 5, P = 0.9657), β3 subunits (surface: R112Q, 0.93 ± 0.09, n = 4, P = 0.9485; N115D, 1.03 ± 0.03, n = 7, P = 0.9512. total: R112Q, 1.04 ± 0.04, n = 5, P = 0.9961; N115D, 0.98 ± 0.13, n = 5, P = 0.9999) or γ2 subunits (surface: R112Q, 0.92 ± 0.05, n = 4, P = 0.7917; N115D, 0.96 ± 0.04, n = 7, P = 0.9936; total: R112Q, 1.08 ± 0.05, n = 5, P = 0.9081; N115D, 1.05 ± 0.07, n = 5, P = 0.9840) subunits (Fig. 5C and D).
Figure 5.
Three de novo mutations in GABRA1 found in several cases with EOEE cause divergent effects in GABAA receptor function. (A) Representative GABA-evoked current traces obtained following rapid application of 1 mM GABA for 4 s to lifted HEK293T cells voltage clamped at −20 mV are shown at the top of the panel. Concentration-response curves were obtained for wild-type (wt) α1 or mutant α1(R112Q), α1(N115D), α1(P260L), α1(L296S) or α1(W315L) subunit-containing α1β3γ2 GABAA receptors are shown at the bottom of the panel. (B) Wild-type α1HA subunits or mutant α1HA(L296S) or α1HA(W315L) subunits were co-expressed with β3 and γ2 subunits in HEK293T cells. α1HA subunits were labelled with anti-HA antibody (green), and the ER was visualized with anti-calnexin antibody (red). White boxes on the merged images depict the enlarged area shown in the images to the right. Also shown are DAPI nuclear counterstaining (blue) and merge of the staining. (C and D) Surface (C) and total (D) levels of wild-type and mutant α1(R112Q), α1(N115D), α1(P260L), α1(L296S) and α1(W315L) subunits co-expressed with β3 and γ2 subunits in HEK293T cells are shown. Band intensity of surface and total expressed α1, β3 and γ2 subunits were normalized to the ATPase signal, and the summarized data are shown in the bar graphs. Values are expressed as mean ± SEM (see text). One-way ANOVA with Dunnett’s post-test was used to determine significance compared to the wild-type condition.
Further, two unrelated cases with Ohtahara and West syndromes were found with the de novo GABRA1 p.P260L mutation (Kodera et al., 2016), which is at the same position of the mutant α1(P260S) subunit described above. Similar to the mutant α1(P260S) subunit (Fig. 4D), co-transfection of mutant α1(P260L) subunits with β3 and γ2 subunits decreased macroscopic GABA-evoked currents by ∼60% (2762 ± 174 pA, n = 14, P = 0.0001) (Fig. 5A), did not alter GABA potency (38.3 µM, P = 0.5444, n = 6), and did not alter fractional Zn2+ inhibition (7 ± 1%, n = 6, P = 0.9999). Conversely, the mutant α1(P260L) subunit did not affect surface or total expression of α1 subunits (surface: 0.85 ± 0.08, n = 4, P = 0.1868; total: 0.80 ± 0.04, n = 5, P = 0.1007), β3 subunits (surface: 0.91 ± 0.06, n = 4, P = 0.7861; total: 1.04 ± 0.04, n = 5, P = 0.9949) or γ2 subunits (surface: 0.94 ± 0.04, n = 4, P = 0.9458; total: 1.06 ± 0.07, n = 5, P = 0.9503) subunits (Fig. 5C and D, left).
In addition, we studied the mutations that are part of the compound mutant α1(L296S/W315L) subunit to gain insight into whether there were differences in the functional and/or expression properties of receptors with α1 subunits containing these single point mutations. Resembling the compound mutant, receptors containing either mutant α1(L296S) or α1(W315L) subunit reduced GABA-evoked currents to ∼70 to 50% (4702 ± 396 pA, n = 14, P = 0.0001; 6150 ± 311 pA, n = 14, P = 0.0001) relative to control currents (9071 ± 273 pA, n = 31) (Fig. 5A, top). However, mutant α1(L296S) subunits did not change GABAA receptor potency (10.9 µM, P = 0.0764, n = 10), while mutant α1(W315L) subunits had GABA concentration-response curves with EC50’s similar to those found with the compound α1(L296S/W315L)β3γ2 subunits (7.02 µM, P = 0.0001, n = 8) (Fig. 5A, bottom right).
Comparison of fractional Zn2+ inhibition demonstrated that receptors containing mutant α1(L296S) or α1(W315L) subunits had an increased Zn2+ sensitivity (16 ± 3%, n = 9, P = 0.0018; 13 ± 1%, n = 8, P = 0.0748) relative to control receptors (8 ± 1%, n = 16), which suggested trafficking deficiency of α1β3γ2 receptors containing mutant subunits to the cell surface and increase in trafficking of α1β3 Zn2+ sensitive receptors to the cell surface. Therefore, HEK293T cells were permeabilized and immunolabelled with anti-HA and anti-calnexin antibodies to assess retention of mutant α1(L296S) and α1(W315L) receptors in the ER. Similar to the ER retention shown by mutant α1(L296S/W315L) subunits (Fig. 4C), mutant α1HA(L296S) subunits co-localized with calnexin. In contrast, mutant α1HA(W315L) subunits spread outside the ER (Fig. 5B). In addition, only mutant α1(L296S) receptors decreased surface (L296S, 0.40 ± 0.04, n = 5, P = 0.0001; W315L, 0.93 ± 0.03, n = 5, P = 0.8637) and total (L296S, 0.37 ± 0.02, n = 5, P = 0.0001; W315L, 0.91 ± 0.05, n = 5, P = 0.9143) α1 subunit levels (Fig. 5C and D, right). Neither of the mutant α1(L296S) and α1(W315L) receptors reduced surface (L296S, 0.93 ± 0.04, n = 5, P = 0.9485; W315L, 0.92 ± 0.05, n = 5, P = 0.8580) or total (L296S, 0.87 ± 0.10, n = 5, P = 0.9518; W315L, 0.96 ± 0.11, n = 5, P = 0.9999) β3 subunit levels or surface (L296S, 0.87 ± 0.07, n = 5, P = 0.2217; W315L, 1.01 ± 0.02, n = 5, P = 0.9940) or total (L296S, 0.86 ± 0.07, n = 5, P = 0.8309; W315L, 1.01 ± 0.08, n = 5, P = 0.9995) γ2 subunit levels.
Discussion
Several studies have described the association of EOEE with de novo mutations in neurotransmitter receptors, synaptic proteins, and voltage- and ligand-gated ion channels; however, little is known about the molecular mechanisms that lead to the disease. Interestingly, despite the multiplicity of targets that have been associated with the disease, ultimately all of them lead to an imbalance in neuronal excitability. Undoubtedly, differences in expression level and signalling pathways, and specificity in brain regions create the broad spectrum of symptoms that characterize the phenotypic diversity of EOEE. Mutations in GABRs are frequently associated with genetic epilepsy syndromes (Kang and Macdonald, 2016). More than 60 reported sporadic de novo mutations in GABRA1, GABRB1–3, and GABRG2 have been found to be linked to the severe EOEEs (Ohtahara syndrome, infantile spasms, Doose syndrome and Dravet syndrome). To a certain extent, different mutations in the same GABR produce different epilepsy syndromes commonly associated with haploinsufficiency and hyperexcitability (Allen et al., 2013; Carvill et al., 2014; Srivastava et al., 2014; Papandreou et al., 2015; Reinthaler et al., 2015; Zhang et al., 2015; Johannesen et al., 2016; Kodera et al., 2016; Lien et al., 2016; Baldridge et al., 2017; Hamdan et al., 2017; Hernandez et al., 2017a, b; Ishii et al., 2017; Moller et al., 2017; Shen et al., 2017; Heyne et al., 2018).
Most of the GABAA receptor α1, β(1,2,3) and γ2 subunit residue substitutions associated with severe epilepsy phenotypes occur more frequently in the pore-forming domain of the receptor critical for GABAA receptor activation (Bianchi et al., 2001; Bianchi and Macdonald, 2002; Althoff et al., 2014; Laverty et al., 2019; Masiulis et al., 2019), where they are more likely to cause impaired receptor function (Janve et al., 2016; Hernandez et al., 2017a, b; Ishii et al., 2017; Shen et al., 2017).
We have demonstrated common impairments caused by de novo GABR mutations including impaired receptor biogenesis (impaired gene transcription, subunit translation, subunit folding, receptor assembly, receptor cell surface trafficking and surface stability), impaired agonist binding and impaired channel gating (Janve et al., 2016; Hernandez et al., 2017a, b; Ishii et al., 2017; Shen et al., 2017). In the present study, we demonstrated that mutations in GABRA5 and GABRA1 found in cases with EOEE were accompanied by a reduced amplitude of GABAA receptor mediated mIPSCs and altered synaptic clustering and distribution in hippocampal neurons. Our data suggested a decrease in the number of active postsynaptic receptors that could weaken synaptic efficacy. As GABAA receptor clustering at inhibitory synapses is likely to modulate network excitability during brain development (Petrini et al., 2014), our results suggested that a ‘hyperexcitability syndrome’ can be caused additionally by altered distribution of mutant receptors at postsynaptic sites.
Synaptic strength is tuned by efficient clustering and distribution of GABAA receptors composed of different subunits at inhibitory synapses, which in turn are highly regulated by postsynaptic scaffolding proteins (Essrich et al., 1998; Jacob et al., 2005; Tretter et al., 2008; Luscher et al., 2011; Choii and Ko, 2015). Gephyrin was described as the major candidate protein for synaptic α1, α2, α3, β2/3 and γ2 GABAA receptor subunit clustering, although α5 subunits were found to be recruited by gephyrin (Christie and de Blas, 2002; Serwanski et al., 2006). Further, previous studies found that gephyrin knockout mice had loss of synaptic GABAA receptor clusters containing α2, α3, β2/3 and γ2, subunits, but α1 and α5 subunits were unaltered (Kneussel et al., 1999), while GABAA receptor γ2 subunit knockout mice had reduced gephyrin-dependent clustering of postsynaptic GABAA receptors (Essrich et al., 1998). Thus, it is expected that α1, α2, α3, β2/3 and γ2 GABAA receptor subunits solely contribute to the postsynaptic gephyrin-dependent clustering. Conversely, clustering of α5 subunits depends on the location of these receptors, either at synaptic or extrasynaptic sites (Brunig et al., 2002; Christie and de Blas, 2002; Serwanski et al., 2006). Accordingly, GABAA receptors containing α5 subunits at synapses are clustered by gephyrin (Christie and de Blas, 2002; Serwanski et al., 2006) and at extrasynaptic sites by radixin (Loebrich et al., 2006). We found that EOEE-associated mutant α5(V294F) and α5(S413F) subunits significantly increased synaptic recruitment of the scaffold protein gephyrin at synapses in hippocampal neurons. In contrast, EOEE-associated mutant α1(P260S) and α1(L296S/ W315L) subunits were lost from synaptic sites and were localized almost exclusively in the soma. Our data suggested that clustering of GABAA receptors containing mutant α1 and α5 subunits at synapses represented two active functional pools of receptors that were mobilized and clustered at GABAergic synapses in a gephyrin-dependent manner and were likely to modulate network excitability in EOEE.
For the most part, all patients carrying pathogenic de novo mutations in GABRA1 have been reported to have a wide phenotypic range of severe epileptic encephalopathies (Johannesen et al., 2016; Kodera et al., 2016). Remarkably, we found that there is a recurrence of the same mutations in different subunits in the structural domains of the receptor linked to the specific functions, which are commonly found between the N-terminal extracellular domain and the pore that are critical for GABAA receptor activation (Bianchi et al., 2001; Bianchi and Macdonald, 2002; Laverty et al., 2019; Masiulis et al., 2019). Thus, the α1 subunit substitutions P260S and P260L occurred at a homologous position in the γ2 subunit (P282S) found in patients with epileptic encephalopathies (Shen et al., 2017). In addition, the α1 subunit substitution (N115D) also occurred at the homologous position in β3 subunits (N110D) found in a patient with infantile spasms (Allen et al., 2013; Janve et al., 2016). In the same way, the α5 subunit substitution (V294F) also occurred at the homologous position of the α1 subunit (V287L) found in a patient with EOEE (Kodera et al., 2016), and a recurrent variant in the same position of V294 but replaced by a leucine (V294L) was found in a case with delayed motor and cognitive development (Butler et al., 2018). Therefore, identification of patients carrying de novo pathogenic mutations in GABRA5 is remarkable. We previously described GABAA receptor gating defects in three rare mutations in GABRA5 (c.610G>A, p.V204I, c.838T>C, pW280R, c.1358C>T, p.P453L) found in individuals with sporadic genetic epilepsy (Klassen et al., 2011; Hernandez et al., 2016). Later, similar findings were reported in an exome-based case-control study of rare variants in GABRA5 in cases clinically evaluated for genetic generalized epilepsy (May et al., 2018). Another study reported two rare missense mutations in GABRA5 (c.610G>A, p.V204I, c.338G>C, pG113A) in autism spectrum disorder cases (Zurek et al., 2016).
This was a deep functional study of GABRA5 and GABRA1 de novo mutations by use of primary hippocampal neurons and non-neuronal cells. We found that GABRA5 and GABRA1 mutations impaired GABAA receptor function and biogenesis, and GABAA receptor immobilization and accumulation by gephyrin at the synapse. Overall these findings demonstrate the recurrence of multiple mutations and variants in common structural ‘hot spots’ in the receptor that are likely to cause impairment of GABAergic function. Our studies reveal new molecular mechanisms that might lead to the neurological dysfunction found in patients who carry mutations in GABRA5 and GABRA1, complementing the prevailing GABAergic channelopathy paradigm in EOEE.
Supplementary Material
Acknowledgements
We thank Taoyun Ji, MD, PhD and Ming Liu, MD for reviewing select portions for their deep knowledge and invaluable insight concerning of the electroencephalographic and neuroimaging data of the cases.
Glossary
Abbreviations
- EOEE
early onset epileptic encephalopathy
- ER
endoplasmic reticulum
- GABR
GABAA receptor subunit genes
- HA
haemagglutinin
- MCC
Manders’ co-localization coefficient
- mIPSCs
miniature inhibitory postsynaptic currents
Funding
This work was supported by the NIH RO1 NS 33300 grant to R.L.M.
Competing interests
The authors declare no competing financial interests.
References
- Abecasis GR, Auton A, Brooks LD, DePristo MA, Durbin RM, Handsaker RE, et al. An integrated map of genetic variation from 1,092 human genomes. Nature 2012; 491: 56–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P, et al. A method and server for predicting damaging missense mutations. Nat Methods 2010; 7: 248–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aken BL, Ayling S, Barrell D, Clarke L, Curwen V, Fairley S, et al. The Ensembl gene annotation system. Database 2016; 2016: baw093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Allen AS, Berkovic SF, Cossette P, Delanty N, Dlugos D, Eichler EE, et al. De novo mutations in epileptic encephalopathies. Nature 2013; 501: 217–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Althoff T, Hibbs RE, Banerjee S, Gouaux E. X-ray structures of GluCl in apo states reveal a gating mechanism of Cys-loop receptors. Nature 2014; 512: 333–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Axeen EJT, Olson HE. Neonatal epilepsy genetics. Sem Fetal Neonatal Med 2018; 23: 197–203. [DOI] [PubMed] [Google Scholar]
- Baldridge D, Heeley J, Vineyard M, Manwaring L, Toler TL, Fassi E, et al. The Exome Clinic and the role of medical genetics expertise in the interpretation of exome sequencing results. Genet Med 2017; 19: 1040–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bianchi MT, Haas KF, Macdonald RL. Structural determinants of fast desensitization and desensitization-deactivation coupling in GABAA receptors. J Neurosci 2001; 21: 1127–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bianchi MT, Macdonald RL. Slow phases of GABA(A) receptor desensitization: structural determinants and possible relevance for synaptic function. J Physiol 2002; 544(Pt 1): 3–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Brunig I, Scotti E, Sidler C, Fritschy JM. Intact sorting, targeting, and clustering of gamma-aminobutyric acid A receptor subtypes in hippocampal neurons in vitro. J Comparat Neurol 2002; 443: 43–55. [DOI] [PubMed] [Google Scholar]
- Butler KM, Moody OA, Schuler E, Coryell J, Alexander JJ, Jenkins A, et al. De novo variants in GABRA2 and GABRA5 alter receptor function and contribute to early-onset epilepsy. Brain 2018; 141: 2392–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carvill GL, Weckhuysen S, McMahon JM, Hartmann C, Moller RS, Hjalgrim H, et al. GABRA1 and STXBP1: novel genetic causes of Dravet syndrome. Neurology 2014; 82: 1245–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Choii G, Ko J. Gephyrin: a central GABAergic synapse organizer. Exp Mol Med 2015; 47: e158. [DOI] [PubMed] [Google Scholar]
- Christie SB, de Blas AL. alpha5 Subunit-containing GABA(A) receptors form clusters at GABAergic synapses in hippocampal cultures. Neuroreport 2002; 13: 2355–8. [DOI] [PubMed] [Google Scholar]
- Connolly CN, Krishek BJ, McDonald BJ, Smart TG, Moss SJ. Assembly and cell surface expression of heteromeric and homomeric gamma-aminobutyric acid type A receptors. J Biol Chem 1996; 271: 89–96. [DOI] [PubMed] [Google Scholar]
- Du J, Lu W, Wu S, Cheng Y, Gouaux E. Glycine receptor mechanism elucidated by electron cryo-microscopy. Nature 2015; 526: 224–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Essrich C, Lorez M, Benson JA, Fritschy JM, Luscher B. Postsynaptic clustering of major GABAA receptor subtypes requires the gamma 2 subunit and gephyrin. Nat Neurosci 1998; 1: 563–71. [DOI] [PubMed] [Google Scholar]
- Hamdan FF, Myers CT, Cossette P, Lemay P, Spiegelman D, Laporte AD, et al. High rate of recurrent De Novo mutations in developmental and epileptic encephalopathies. Am J Hum Genet 2017; 101: 664–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hamdan FF, Srour M, Capo-Chichi JM, Daoud H, Nassif C, Patry L, et al. De novo mutations in moderate or severe intellectual disability. PLoS Genet 2014; 10: e1004772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Helbig I, Tayoun AA. Understanding genotypes and phenotypes in epileptic encephalopathies. Mol Syndromol 2016; 7: 172–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Helbig I, von Deimling M, Marsh ED. Epileptic encephalopathies as neurodegenerative disorders. Adv Neurobiol 2017; 15: 295–315. [DOI] [PubMed] [Google Scholar]
- Helbig KL, Farwell Hagman KD, Shinde DN, Mroske C, Powis Z, Li S, et al. Diagnostic exome sequencing provides a molecular diagnosis for a significant proportion of patients with epilepsy. Genet Med 2016; 18: 898–905. [DOI] [PubMed] [Google Scholar]
- Hernandez CC, Klassen TL, Jackson LG, Gurba K, Hu N, Noebels JL, et al. Deleterious rare variants reveal risk for loss of GABAA receptor function in patients with genetic epilepsy and in the general population. PLoS One 2016; 11: e0162883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hernandez CC, Kong W, Hu N, Zhang Y, Shen W, Jackson L, et al. Altered channel conductance states and gating of GABAA receptors by a pore mutation linked to Dravet syndrome. eNeuro 2017a; 4(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hernandez CC, Zhang Y, Hu N, Shen D, Shen W, Liu X, et al. GABA A Receptor coupling junction and pore GABRB3 mutations are linked to early-onset epileptic encephalopathy. Sci Rep 2017b; 7: 15903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Heyne HO, Singh T, Stamberger H, Abou Jamra R, Caglayan H, Craiu D, et al. De novo variants in neurodevelopmental disorders with epilepsy. Nat Genet 2018; 50: 1048–53. [DOI] [PubMed] [Google Scholar]
- Hibbs RE, Gouaux E. Principles of activation and permeation in an anion-selective Cys-loop receptor. Nature 2011; 474: 54–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Huang X, Hernandez CC, Hu N, Macdonald RL. Three epilepsy-associated GABRG2 missense mutations at the gamma+/beta- interface disrupt GABAA receptor assembly and trafficking by similar mechanisms but to different extents. Neurobiol Dis 2014; 68: 167–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ishii A, Kang JQ, Schornak CC, Hernandez CC, Shen W, Watkins JC, et al. A de novo missense mutation of GABRB2 causes early myoclonic encephalopathy. J Med Genet 2017; 54: 202–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jacob TC, Bogdanov YD, Magnus C, Saliba RS, Kittler JT, Haydon PG, et al. Gephyrin regulates the cell surface dynamics of synaptic GABAA receptors. J Neurosci 2005; 25: 10469–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Janve VS, Hernandez CC, Verdier KM, Hu N, Macdonald RL. Epileptic encephalopathy de novo GABRB mutations impair GABAA receptor function. Ann Neurol 2016; 79: 806–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Johannesen K, Marini C, Pfeffer S, Moller RS, Dorn T, Niturad C, et al. Phenotypic spectrum of GABRA1: from generalized epilepsies to severe epileptic encephalopathies. Neurology 2016; 87: 1140–51. [DOI] [PubMed] [Google Scholar]
- Kang JQ, Macdonald RL. Molecular pathogenic basis for GABRG2 mutations associated with a spectrum of epilepsy syndromes, from generalized absence epilepsy to Dravet syndrome. JAMA Neurol 2016; 73: 1009–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Klassen T, Davis C, Goldman A, Burgess D, Chen T, Wheeler D, et al. Exome sequencing of ion channel genes reveals complex profiles confounding personal risk assessment in epilepsy. Cell 2011; 145: 1036–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kneussel M, Brandstatter JH, Laube B, Stahl S, Muller U, Betz H. Loss of postsynaptic GABA(A) receptor clustering in gephyrin-deficient mice. J Neurosci 1999; 19: 9289–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kodera H, Ohba C, Kato M, Maeda T, Araki K, Tajima D, et al. De novo GABRA1 mutations in Ohtahara and West syndromes. Epilepsia 2016; 57: 566–73. [DOI] [PubMed] [Google Scholar]
- Laverty D, Desai R, Uchanski T, Masiulis S, Stec WJ, Malinauskas T, et al. Cryo-EM structure of the human α1β3γ2 GABAA receptor in a lipid bilayer. Nature 2019; 565: 516–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Le SV, Le PHT, Le TKV, Kieu Huynh TT, Hang Do TT. A mutation in GABRB3 associated with Dravet syndrome. American journal of medical genetics Part A 2017; 173(8): 2126–31. [DOI] [PubMed] [Google Scholar]
- Lek M, Karczewski KJ, Minikel EV, Samocha KE, Banks E, Fennell T, et al. Analysis of protein-coding genetic variation in 60,706 humans. Nature 2016; 536: 285–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lien E, Vatevik AK, Ostern R, Haukanes BI, Houge G. A second patient with a De Novo GABRB1 mutation and epileptic encephalopathy. Ann Neurol 2016; 80: 311–2. [DOI] [PubMed] [Google Scholar]
- Loebrich S, Bahring R, Katsuno T, Tsukita S, Kneussel M. Activated radixin is essential for GABAA receptor alpha5 subunit anchoring at the actin cytoskeleton. EMBO J 2006; 25: 987–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Luscher B, Fuchs T, Kilpatrick CL. GABAA receptor trafficking-mediated plasticity of inhibitory synapses. Neuron 2011; 70: 385–409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Masiulis S, Desai R, Uchanski T, Serna Martin I, Laverty D, Karia D, et al. GABAA receptor signalling mechanisms revealed by structural pharmacology. Nature 2019; 565: 454–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- May P, Girard S, Harrer M, Bobbili DR, Schubert J, Wolking S, et al. Rare coding variants in genes encoding GABAA receptors in genetic generalised epilepsies: an exome-based case-control study. Lancet Neurol 2018; 17: 699–708. [DOI] [PubMed] [Google Scholar]
- McTague A, Howell KB, Cross JH, Kurian MA, Scheffer IE. The genetic landscape of the epileptic encephalopathies of infancy and childhood. Lancet Neurol 2016; 15: 304–16. [DOI] [PubMed] [Google Scholar]
- Miller PS, Aricescu AR. Crystal structure of a human GABAA receptor. Nature 2014; 512: 270–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Moller RS, Wuttke TV, Helbig I, Marini C, Johannesen KM, Brilstra EH, et al. Mutations in GABRB3: From febrile seizures to epileptic encephalopathies. Neurology 2017; 88: 483–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ng PC, Henikoff S. Predicting deleterious amino acid substitutions. Genome Res 2001; 11: 863–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nieh SE, Sherr EH. Epileptic encephalopathies: new genes and new pathways. Neurotherapeutics 2014; 11(4): 796–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Papandreou A, McTague A, Trump N, Ambegaonkar G, Ngoh A, Meyer E, et al. GABRB3 mutations: a new and emerging cause of early infantile epileptic encephalopathy. Dev Med Child Neurol 2015; 58: 416–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Petrini EM, Ravasenga T, Hausrat TJ, Iurilli G, Olcese U, Racine V, et al. Synaptic recruitment of gephyrin regulates surface GABAA receptor dynamics for the expression of inhibitory LTP. Nat Commun 2014; 5: 3921. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Reinthaler EM, Dejanovic B, Lal D, Semtner M, Merkler Y, Reinhold A, et al. Rare variants in gamma-aminobutyric acid type A receptor genes in rolandic epilepsy and related syndromes. Ann Neurol 2015; 77: 972–86. [DOI] [PubMed] [Google Scholar]
- Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 2012; 9: 676–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schwarz JM, Rodelsperger C, Schuelke M, Seelow D. MutationTaster evaluates disease-causing potential of sequence alterations. Nat Methods 2010; 7: 575–6. [DOI] [PubMed] [Google Scholar]
- Serwanski DR, Miralles CP, Christie SB, Mehta AK, Li X, De Blas AL. Synaptic and nonsynaptic localization of GABAA receptors containing the alpha5 subunit in the rat brain. J Comparat Neurol 2006; 499: 458–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shen D, Hernandez CC, Shen W, Hu N, Poduri A, Shiedley B, et al. De novo GABRG2 mutations associated with epileptic encephalopathies. Brain 2017; 140: 49–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Srivastava S, Cohen J, Pevsner J, Aradhya S, McKnight D, Butler E, et al. A novel variant in GABRB2 associated with intellectual disability and epilepsy. Am J Med Genet Part A 2014; 164a: 2914–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tretter V, Jacob TC, Mukherjee J, Fritschy JM, Pangalos MN, Moss SJ. The clustering of GABA(A) receptor subtypes at inhibitory synapses is facilitated via the direct binding of receptor alpha 2 subunits to gephyrin. J Neurosci 2008; 28: 1356–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang Y, Kong W, Gao Y, Liu X, Gao K, Xie H, et al. Gene mutation analysis in 253 Chinese children with unexplained epilepsy and intellectual/developmental disabilities. PLoS One 2015; 10: e0141782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zurek AA, Kemp SW, Aga Z, Walker S, Milenkovic M, Ramsey AJ, et al. alpha5GABAA receptor deficiency causes autism-like behaviors. Ann Clin Transl Neurol 2016; 3: 392–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are available within the article and its Supplementary material.





