Epilepsy with eyelid myoclonias and Sotos syndrome features in a patient with compound heterozygous missense variants in APC2 gene

Abstract

Epilepsy with eyelid myoclonias, originally depicted by Jeavons in 1977, is a reflex epilepsy characterized by jerking of the eyelids with or without absences precipitated by eye closure or by light (eyelid myoclonia, EM), eye closure-induced EEG paroxysms and photosensitivity. Childhood-onset, female predominance and a normal development are typical features, though a mild intellectual disability has been reported. Sotos syndrome is a disorder characterized by a distinctive facial appearance, learning disability and overgrowth in childhood with macrocephaly, caused by heterozygous pathogenic variants or deletions in NSD1 gene. Generalized and focal seizures have been reported in up to 25 % of patients, though EM was never documented. Here we report the novel association of Epilepsy with EM and Sotos syndrome features in a patient with two likely pathogenic missense variants in APC2 gene. The patient is a 27-year-old woman, with an unremarkable family history. During her childhood, the association of early developmental delay, intellectual impairment (IQ 65), macrocephaly and dysmorphisms (dolichocephalic head shape, broad nasal bridge, narrow palate, brachydactyly, see Fig. 1A) raised the suspicion of Sotos syndrome. DHPLC and MLPA analyses of NSD1 gene did not highlight molecular defects. Karyotyping, CGH-array (8 × 60 K) and Southern blotting of FMR1 locus resulted negative. At the age of nine, she had a first paroxysmal event described as prolonged loss of consciousness and post-ictal vomiting. Since then, she experienced brief episodes characterized by upward deviation of the eyes and slight retropulsion of the head, possibly associated with brief awareness impairment, several times a day. She was treated with valproate and then levetiracetam with incomplete benefit. At the age of 21, the patient was referred to our Epilepsy Centre for the transition to adult care and was recruited for the Epi25 project (https://epi-25.org). The neurological examination was unremarkable. EEG showed normal background activity, spontaneous and eye closure-induced generalized spike and polyspike-and-wave discharges, and photosensitivity. Discharges were associated with EM, upward deviation of the eyes, and very rarely upper limb myoclonias (Fig. 2). Ictal impairment of awareness was not demonstrated. Brain MRI performed with a dedicated high-resolution epilepsy protocol was unremarkable. Whole-exome sequencing detected two novel missense variants in APC2 and a canonical splice site variant in GABRG2. The c.6620C > T (p.Pro2207Leu) and c.1063 G > A (p.Val355Ile) in APC2 resulted of paternal and maternal inheritance respectively, the last one was also identified in her unaffected brother (Fig. 1B,C); the c.1249-1G > T, in GABRG2, inherited from her father, was also identified in her brother. Genes associated with Epilepsy with EMs (CHD2, SYNGAP, KIAA2022, RORF, SLC1A2, TWIST1, DHFR, SLC6A1, SERPINI1, HCN1, KCNB1, GABRB3) were analysed and pathogenic variants were excluded. Add-on therapy with ethosuximide achieved an apparent complete seizure control after two years of follow-up, while subclinical generalized discharges on EEGs persisted, though less frequent. We reported the association of Epilepsy with EM and Sotos syndrome features in a patient with two compound heterozygous variants in APC2 gene. Previously, a homozygous frameshift pathogenic variant in the same gene was reported in two Egyptian siblings with Sotos-like phenotype including intellectual disability, macrocephaly and dilated brain ventricles on MRI, representing the first and only association between APC2 and Sotos syndrome. No history of seizures was reported at the time of the study, when patients were eight and ten years old, respectively [[1]Almuriekhi M. Shintani T. Fahiminiya S. Fujikawa A. Kuboyama K. Takeuchi Y. et al.Loss-of-function mutation in APC2 causes sotos syndrome features.Cell Rep. 2015; 10: 1585-1598https://doi.org/10.1016/j.celrep.2015.02.011Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar]. Moreover, bi-allelic loss of function variants in APC2 were demonstrated in 12 patients with generalized and myoclonic seizures, lissencephaly, subcortical heterotopia and global developmental delay [[2]Lee S. Chen D.Y. Zaki M.S. Maroofian R. Houlden H. Di Donato N. et al.Bi-allelic loss of human APC2, encoding adenomatous polyposis coli protein 2, leads to Lissencephaly, subcortical heterotopia, and global developmental delay.Am J Hum Genet. 2019; 105: 844-853https://doi.org/10.1016/j.ajhg.2019.08.013Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar]. The features of our case encompass those reported in previous works, depicting a new phenotype, also including Epilepsy with EM. Despite a well-defined electroclinical description, the recognition of Jeavons syndrome as a distinct nosological entity is still a matter of issue, since its features could occur in several epileptic disorders [[3]Striano S. Capovilla G. Sofia V. Romeo A. Rubboli G. Striano P. et al.Eyelid myoclonia with absences (Jeavons syndrome): a well-defined idiopathic generalized epilepsy syndrome or a spectrum of photosensitive conditions?.Epilepsia. 2009; 50: 15-19https://doi.org/10.1111/j.1528-1167.2009.02114.xCrossref PubMed Scopus (86) Google Scholar]. A family history of seizures or idiopathic generalized epilepsy is common, suggesting shared genetic etiologies. We excluded pathogenic variants in genes reported in association with EM so far, and we therefore speculate that it is an additional feature of the complex phenotype of our patient. We already described the genetic findings in a paper reporting the results of WES analysis in 71 patients with developmental and epileptic encephalopathies (Patient 8) [[4]Minardi R. Licchetta L. Baroni M.C. Pippucci T. Stipa C. Mostacci B. et al.Whole-exome sequencing in adult patients with developmental and epileptic encephalopathy: it is never too late [published online ahead of print, 2020 Jul 29].Clin Genet. 2020; (10.1111/cge.13823)https://doi.org/10.1111/cge.13823Crossref PubMed Scopus (5) Google Scholar]. The proband harbored two missense variants in APC2: the p.Pro2207Leu involves a well-conserved residue across the species, is absent in gnomAD database, and is predicted as pathogenic by the main principal prediction tools, while the p.Val355Ile, even if predicted as likely benign by computational evidence, has an allele frequency of 5.04e-5 (less than 0.0001 threshold for recessive genes) and was never found in homozygous state. Since these two variants were found in compound heterozygosity, and the proband’s phenotype highly overlaps the one of the patients described by Almuriekhi, we classified them as likely pathogenic according to the ACMG guidelines. APC2 localizes to actin and microtubule fibers and was demonstrated to be a downstream effector of NSD1 in neurons, as NSD1 knockdown in embryonic mouse brain downregulated endogenous APC2 expression and impaired the migration and laminar positioning of cortical neurons as observed also in Apc2-/- mice. Moreover, Apc2-deficient mice showed abnormal head shape and brain structure, learning disabilities, and seizures [[1]Almuriekhi M. Shintani T. Fahiminiya S. Fujikawa A. Kuboyama K. Takeuchi Y. et al.Loss-of-function mutation in APC2 causes sotos syndrome features.Cell Rep. 2015; 10: 1585-1598https://doi.org/10.1016/j.celrep.2015.02.011Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar]. In our case, as in Almuriekhi, cortical malformations were absent, possibly because these missense variants did not fully inactivate the protein. The additional variant identified in GABRG2 was identified in two healthy family members of our proband and has been recently proposed to affect susceptibility to febrile seizure (HGMD Accession number: CS1721924). Even though we cannot exclude its possible impact on modulating the epileptic phenotype, we ruled out a major role of this variant. In conclusion, we reported the first case of Epilepsy with EM and Sotos-like phenotype with compound heterozygous missense variants in APC2. Moreover, this is the third case from a second unrelated family with a Sotos-like phenotype and no defects of neuronal migration. We suggest considering APC2 in gene panel testing for Epilepsy with EM, specifically when associated with macrocephaly and intellectual disability. Further studies on larger case series are needed to determine the full phenotypic spectrum associated with APC2 disease-causative variants, and to clarify its significance in the complex genetic background of generalized epilepsies.