Background

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by social and communication impairment [1].

The pathogenesis of ASD is not fully clarified. It seems that it is caused by genetic factors affecting neural connectivity, consequently impairing social communication development and leading to stereotypies and restricted interests [1, 2]. It has been hypothesized that the heterogeneous clinical presentation of ASD results from the interaction between multiple genes or gene combinations as well as epigenetic factors and exposure to environmental modifiers [3].

Next-generation sequencing technologies have greatly enhanced the identification of new genes associated with ASD. Recently, ANK2 gene has also been pointed out as a candidate gene for the etiology of ASD [4,5,6].

With this study, the authors aim to report the detection of a de novo heterozygous ANK2 variant in a child with ASD and epilepsy, providing a detailed clinical description.

Case presentation

We report the case of a female patient, born after a non-eventful gestation. Prenatal ultrasounds were described as normal. She was born via cesarean at 39 weeks, with Apgar score 9/10 and appropriate birth size. Newborn screening tests and perinatal period were unremarkable. She is the only child of non-consanguineous parents and family history was non-contributory.

At 2 months old, she was admitted to the Pediatric ward due to focal clonic seizures of her left lower limb, lasting approximately 10 s, which did not cease by pressing the limb and had been occurring either during sleep or awake, two to four times a day, for at least 1 week. Laboratory blood tests including blood count, inflammatory parameters (C-reactive protein), muscle enzymes, glucose and electrolytes did not show relevant abnormalities. Electroencephalography was performed during hospital stay and revealed normal background with interictal and ictal epileptiform discharges restricted to the midline central and right central regions (Cz-P4). No structural brain anomalies were identified by neuroimaging. She was initially treated with levetiracetam 10 mg/kg/dose twice daily and frequency of seizures was reduced to one/two times a day. Levetiracetam was progressively increased to 30 mg/kg/dose, but she maintained the same periodicity of seizures. On the third week of treatment, antiseizure medication was changed to phenobarbital (3 mg/kg/day in 3 divided doses). During the following week, she presented three brief episodes of increased muscle tone in the left lower limb, lasting approximately five seconds. Since then, no more episodes of seizure were described.

At 2 years old, she was referred to the Pediatric Neurodevelopment department since she was unable to achieve the language developmental milestones for her age. She displayed an abnormal gait pattern (toe walking) since she was 20 months old. She presented difficulties on establishing eye contact, following simple instructions or responding consistently to her name. She also struggled on joint attention and symbolic play. Her speech was characterized mostly by jargon with no intelligible words. She was unable to name body parts or objects in pictures. Her mother reported an obsession with water and keyboards. According to her school daycare educator, she engaged in a more passive play with her peers. Her daycare educator also reported some difficulty with specific textures while playing. There was no history of unusual eating habits.

On physical examination, she displayed some peculiar facial features including a large mouth with widely spaced teeth. Head growth evaluation was normal, including her current head circumference. Neurological examination was unremarkable aside from toe walking. Auditory evoked potentials were performed and displayed normal results. Diagnosis of ASD was confirmed using the criteria from the Diagnostic and Statistical Manual of Mental Disorders, fifth edition [1]. She was started on multimodality interventions, namely speech-language and occupational therapies (with sensory integration approach). During her follow-up, she made progresses regarding social interaction and autonomy, though she maintained an important speech delay. At the age of 3 years, electroencephalography was repeated and did not show abnormalities; at this point she began a progressive withdrawal of phenobarbital, without recurrence of seizures. She only began to speak clearly in simple sentences at the age of four. On her latest evaluation in the Pediatric Neurodevelopment Department, at the age of seven, she revealed a very good memory capacity, she could read simple words and showed a favorable improvement in verbal and non-verbal interactions. Subjectively, she appears to have a normal cognitive function. She currently attends the first grade of elementary school, and she has interest and good performance in Mathematics according to her teacher.

Additional evaluation included array comparative genomic hybridization (aCGH) and Angelman syndrome methylation analysis which were normal. A clinical exome sequencing (CES) was performed, and a heterozygous variant of uncertain significance was identified in the ANK2 gene (NM_001148.5): c.3412C > T p.(Arg1138Ter) (Fig. 1A). The genetic variant was then searched in her parents and was negative in both, suggesting a de novo variant (Fig. 1B).

Fig. 1
figure 1

A Next-generation sequencing (NGS) identified the heterozygous variant c.3412C > T in the ANK2 gene. B Family pedigree and Sanger sequencing of the variant in the proband and parents, showing its presence in the patient and its absence in the mother and father

Full size image

The child was submitted to a formal cardiac evaluation, ruling out cardiovascular abnormalities.

Conclusions

Etiological assessment plays an important role in management of ASD, since it allows a more detailed information on prognosis, comorbidities and genetic counseling. In patients where clinical history, physical examination, aCGH and molecular genetic testing for fragile X syndrome are inconclusive, the next step on etiological assessment is usually a trio-based whole exome sequencing or CES [1]. In our patient, we decided to perform a CES for etiological enlightenment, unveiling a ANK2 gene variant.

It is vastly recognized in current literature that loss-of-function variants in ANK2 can result in a broad spectrum of clinical cardiac phenotypes including Long QT syndrome 4 and Ankyrin-B Syndrome (OMIM: 600919) [6]. According to animal model experiments, ANK2 encodes an ankyrin protein involved in synaptic stability. Studies in mice have shown that ANK2 pathogenic variants may increase axon branching and ectopic connectivity [7]. Due to its interactions with the voltage-gated CaV2.1 calcium channel, ANK2 variants had been associated to occurrence of seizures [8]. Recently, ANK2 gene has also been linked to the etiology of ASD [4,5,6]. However, this gene is currently not associated with neurodevelopmental disorders in the Online Mendelian Inheritance in Man (OMIM) database.

We recognize the challenge of assessing variant pathogenicity in candidate genes for ASD. The reported ANK2 variant was formally classified as a variant of uncertain significance. However, it was absent in parents and there was not family history, which favors its pathogenicity. Our study provides a detailed phenotypical description that is often lacking; thus, it can be added to other published reports correlating de novo ANK2 variants with ASD and neurological phenotypes (including seizures), some without cardiac impairment. Ji et al. reported a de novo heterozygous nonsense ANK2 variant in a 5-year-old female with seizures and neurodevelopmental impairment (severe global developmental delay and aggressive behavior) (4); she also displayed microcephaly, unlike our patient.

In summary, we report a de novo heterozygous ANK2 variant detected in a child with ASD and epilepsy. Additional research is required to better understand the association between ANK2 and ASD, as well as its implications on phenotypic spectrum.