The other sodium channel gene. The week before Christmas, the Kiel group identified its first patient with SCN2A encephalopathy. At the same time, a questionably benign SNP in the same gene is haunting our Israel Epilepsy Family Project. Time to review the mysterious SCN2A gene that initially entered the scene as a candidate for a rare, benign familial epilepsy syndrome – only to return as one of the most prominent genes for autism, intellectual disability, and epileptic encephalopathies to date.

Pyramids. The SCN2A protein is one of the sodium channels on excitatory neurons in the Central Nervous System. In contrast to SCN1A, which is primarily located on inhibitory neurons, SCN2A resides at the axon initial segment, where the excitatory and inhibitory postsynaptic potentials are integrated and translated into action potentials. Accordingly, the SCN2A protein is located at a crucial bottleneck for signal transduction in neurons. Alterations to the function of the SCN2A protein may therefore have widespread consequences with respect to both the excitability of neurons and the overall function of the Central Nervous System.

BFNIS. Mutations in SCN2A were first described in patients with Benign Familial Neonatal-Infantile Seizures (BFNIS). BFNIS is one of the three autosomal dominant benign epilepsy syndromes starting in the first year of life. Benign Familial Neonatal Seizures (BFNS) due to mutations in KCNQ2 or KCNQ3 and Benign Familial Infantile Seizures (BFIS) with mutations in PRRT2 represent the other two syndromes in this group. Seizures in these familial syndromes are largely limited to the first year of life, and the outcome is usually good. However, some patients may suffer from persisting seizures and neuropsychiatric comorbidities such as intellectual disability. “Neonatal-Infantile” in BFNIS refers to the observation that individuals with neonatal seizures and individuals with infantile seizures can be found in families carrying SCN2A mutations. However, overlapping phenotypes of neonatal and infantile seizures in individual patient are exceptions. BFNIS is a rare familial syndrome, and the SCN2A gene was regarded a relatively exotic epilepsy gene – until it was rediscovered in autism, intellectual disability, and epileptic encephalopathies.

Rediscovery. The comeback of SCN2A occurred with the advent of trio exome sequencing studies. With the possibility of querying much of the coding sequence of the human genome without prior hypotheses, these studies make it possible to identify de novo mutations, i.e. genetic alterations that are present in the child, but absent in both parents. In rapid succession, SCN2A was found in studies of patients with autism and intellectual disability. For some time, it was even one of the few genes in these studies that showed overall significance in combined cohorts; it basically rose above the level of genomic noise. In contrast to BFNIS, SCN2A mutations in autism and intellectual disability can be truncating mutations. In parallel, autism or autistic features are prominent features of patients with deletions of SCN2A. This indicates that the autism phenotype is due to a loss of function in SCN2A, in contrast to some of the BFNIS mutations that may result in a gain of function. Simply put, a gain of function increases the overall excitability of pyramidal cells, resulting in hyperexcitability. The strict age-dependent expression of the SCN2A protein may then explain the strict age-dependence of the seizures. In addition to BFNIS, more severe SCN2A phenotypes were already described quite early. However, the full recognition of SCN2A encephalopathy as a distinct entity occurred only during the last two years.

Epileptic encephalopathies. As of early 2014, probably more than 30 patients with SCN2A encephalopathy have been reported in the literature. Even though the genotype/phenotype correlation for SCN2A encephalopathies is a little unclear, some initial patterns emerge; as suggested by Nakamura and colleagues, mutations in the linker regions that connect transmembrane segments of the channel may result in Ohtahara Syndrome, a severe, early-onset encephalopathy with a suppression-burst EEG. In contrast, transmembrane mutations may either result in Infantile Spasms or Lennox-Gastaut Syndrome. This is the case in the two patients with identical mutations (R853Q) in the Epi4K study. The same mutation has also been described in West Syndrome in a different study, suggesting that this site might be a mutational hotspot. Transmembrane mutations may also lead to unclassified early-onset epileptic encephalopathy that is sometimes accompanied by dystonia. Interestingly, the antiepileptic drug lamotrigine may lead to a striking reduction of seizures in these patients.

Function. Functional studies suggest that the effects on channel function may be more pronounced in encephalopathy-related mutations as compared to BFNIS mutations. However, it is not entirely clear yet whether the functional consequence is a loss or gain of functional. Notably, truncation mutations or deletions have not been reported in epileptic encephalopathies, suggesting that a simple loss of channel function is insufficient to explain the phenotype. However, the reason why alterations in SCN2A may lead to a broad range of phenotypes and whether similar or distinct functional consequences underlie the various phenotypes remains unclear. A gain of function mechanism in epileptic encephalopathies might represent an interesting target for future drug developments to allow for precision medicine based on the causative genetic finding. Either way, SCN2A has become one of the most prominent genes for various neurodevelopmental conditions. We will surely learn more about the phenotypes and disease mechanisms in the near future.