JLE

Epileptic Disorders

MENU

ATP6V1B2-related epileptic encephalopathy Volume 22, numéro 3, June 2020

Vidéo

  • ATP6V1B2-related epileptic encephalopathy

Illustrations


  • Figure 1

  • Figure 2

  • Figure 3

  • Figure 4

Tableaux

ATP6V1B2 (ATPase, H+ transporting, lysosomal VI subunit B, isoform 2), that encodes for one of the subunits of the cross-membrane proton pump, is responsible for keeping the lysosomal pH acidic (Marshansky et al., 2014). This activity is necessary for adequate functioning of several lysosomal hydrolases. Two de novo variants of this gene have been associated with Zimmermann-Laband 2 (ZLS2, OMIM # 616455): a missense variant, p.(Arg485Pro) (Kortüm et al., 2015), clinically characterized by intellectual disability, a dysmorphic face, gingival hyperplasia, hypertrichosis, and nail aplasia or hypoplasia; and the p.(Glu374Gln) variant, reported by Popp et al. (2017), in a patient with microcephaly and intellectual disability, who had three seizures. A third de novo variant in ATP6V1B2 was associated with dominant deafness onychodystrophy (DDOD, OMIM # 124480): nonsense p.(Arg506*) (Yuan et al., 2014; Menendez et al., 2017), leading to congenital deafness and hypoplastic nails, associated with malformation of fingers and toes. The morphological changes similar to those seen in human DDOD have been reproduced in a murine model of this variant (Zhao et al., 2019). Recently, Shaw et al. (2019) reported the variant, p.(Leu398Val), in ATP6V1B2 in a multigenerational family with six evaluated individuals showing epilepsy of variable severity, intellectual disability (present in three individuals), dystrophic nails (in two individuals), and gingival hyperplasia (found in three individuals). Epileptic encephalopathy, as seen in our patient, has not been reported in any ATP6V1B2-related phenotypes. Another unrelated gene, KCNH1, has been associated with two conditions associated with epilepsy as one of the symptoms: Temple-Baraitser syndrome (OMIM # 6118216) and Zimmermann-Laband 1 syndrome (ZLS1, OMIM # 135500). Both of these conditions are phenotypically similar to ZLS2, except for epilepsy and, occasionally, microcephaly, seen only in KCNH1-related syndromes (Bramswig et al., 2015; Mastrangelo et al., 2016; Mégarbané et al., 2016). A similar phenotype was recently reported with monoallelic de novo variants of KCNN3 (Potassium Channel, Calcium-Activated, Intermediate/Small Conductance, Subfamily N, Member 3, OMIM * 602983) (Bauer et al., 2019) and KCNK4 (Potassium Channel, Subfamily K, Member 4, OMIM * 605720) (Bauer et al., 2018), expanding the genetic heterogeneity of ZLS. Epilepsy was seen in only two of three reported cases with KCNK3 variants(table 1). Herein, we report a patient with a novel de novo nonsense variant of ATP6V1B2 associated with severe epileptic encephalopathy and microcephaly, accompanied by all of the other expected clinical features associated with ZLS2. This study was approved by the institutional review board, and the patient's family consented to this publication.

Case study

A two-year-old female was born at term via Caesarean section after an uncomplicated pregnancy, in good general health condition (birth weight: 3.570 g, percentile 25-50, z score:-0.4; OFC = 35 cm, percentile 25, z score:-0,7). She was the first child of non-consanguineous parents. Infantile spasms began at four months and tended to occur mostly in clusters upon awakening. The patient was treated with several antiepileptic drugs, including vigabatrin, levetiracetam, oxcarbazepine, clonazepam, topiramate, valproic acid, and phenobarbital, which failed to control her seizures. She currently has two or three isolated tonic seizures per day, usually upon awakening. She has microcephaly (OFC = 43.5 cm, z score <-2), with a metopic suture of neural crest origin, and frontal narrowing; she has small and wide-spaced teeth, with remarkable gingival hypertrophy. Hypertrichosis (mostly frontal and affecting the upper and lower limbs), hypoplastic distal phalanges of fingers and toes, and small toenails are also present (figure 1). She has profound developmental delay and became unable to support her head, sit alone, roll over onto her back, or grasp objects. Her vision and hearing are apparently normal. Brain magnetic resonance imaging (MRI) showed encephalic volume reduction. Karyotyping, chromosomal microarray analysis, metabolic work-up, hand X-ray, and brainstem evoked potential were all normal. At 22 months of age, a video electroencephalogram (video-EEG) showed disorganized background activity (excessive delta and theta waves during wakefulness and sleep) and epileptiform discharges characterized by bilateral anterior multispikes. (figures 2, 3). Four asymmetric tonic seizures were recorded during sleep. The seizures were electrographically characterized by suppression of background activity, lasting for three seconds, followed by rhythmic delta activity over the bilateral anterior regions mixed with sharp waves (figure 4). Whole-exome sequencing (WES) detected a nonsense variant [(Chr8:20,077,842A>T; c.1465A>T ENST0000027390 p.(Lys489*)] in ATP6V1B2. This variant, which leads to a premature protein truncation, is not found in population databases (1000 Genomes, GnomAD) and has not been reported in the literature or other databases (ClinVar, HGMD). The presence of this variant was confirmed by Sanger sequencing in the patient (but not in the parents), occurring as a de novo event. According to the American College of Medical Genetics (ACMG) criteria, this variant is classified as pathogenic (PVS1, PM2 and PM6) (Richards et al., 2015). Consistent with GnomAD, ATP6V1B2 is highly intolerant to loss-of-function (pLI=0.99), a finding that supports the pathogenicity of the p.(Lys489*) variant detected in our patient. Additionally, it is located in exon 14, the last exon of ATP6V1B2, and might escape mRNA nonsense-mediated decay, leading to translation of an abnormal truncated protein, shortened by 23 amino acids. No other potentially pathogenic variants in genes known to be associated with epileptic encephalopathy were identified by WES.

Discussion

Only three de novo pathogenic variants have been reported in ATP6V1B2: the two missense variants, c.1454G>C p.(Arg485Pro) (Körtum et al., 2015) and p.(Glu374Gln) (Popp et al., 2017), leading to ZLS2, and the recurrent nonsense variant c.1516C>T p.(Arg506*), reported in four patients with DDOD (Menendez et al., 2017; Yuan et al., 2014). The nonsense p.(Lys489*) variant reported in our patient is the fourth known pathogenic de novo change detected in ATP6V1B2, leading to a more complex phenotype, which combines developmental and physical features of ZLS2 with infantile-onset epileptic encephalopathy and microcephaly, a characteristic that might be seen in KCNH1-associated epilepsy. A summary of the clinical presentation of all ATP6V1B2 patients reported is presented in table 2. The nonsense variants p.(Lys489*) and p.(Arg506*) might escape nonsense-mediated decay, as they are located in exon 14, the last exon of the gene. Nevertheless, they are associated with significantly different phenotypes, with a much more severe presentation for the p.(Lys489*) variant. One might speculate that protein function is more severely affected by the loss of the last 23 amino acids for p.(Lys489*) compared to the loss of six amino acids for p.(Arg506*). Epilepsy was unusually severe in our patient, with onset at infancy and refractoriness to several antiepileptic drugs. Profound developmental delay associated with post-natal microcephaly was also part of the clinical presentation, as well as characteristic dysmorphic features, including gingival hypertrophy, hirsutism, and onychodystrophy. ATP6V1B2 should be included in the expanding list of genes associated with early-infantile epileptic encephalopathy.

Acknowledgements and disclosures

We are grateful to the unrestricted collaboration of the patient's family.

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