Christianson syndrome (CS) is induced by sudden abnormal discharge of brain neurons, and is characterized by recurrent seizures due to hypersynchronous discharge (Liu et al., 2018). Symptoms of CS often begin in childhood, with a complex clinical manifestation that is associated with social and learning disabilities. Moreover, severe epilepsy comprises a group of devastating neurological disorders characterized by frequent epileptic seizures associated with developmental delay or regression (McTague et al., 2016). Each type of epilepsy has specific clinical manifestations and distinct underlying pathophysiologic or genetic mechanisms. It is believed that identifying the underlying cause of epilepsy or specific electroclinical syndrome is central to understand the natural history and optimal treatment of the condition (Wirrell, 2016).

To date, multiple genes including Na⁺/H⁺ exchanger protein 6 (NHE6, also known as SLC9A6) have been linked to CS (McTague et al., 2016). NHE6 plays a role in both excitatory presynaptic and postsynaptic functions (Deane et al., 2013). Additionally, NHE6 is highly expressed in the brain, and is thought to participate in the targeting of intracellular vesicles and recycling of synaptic vesicles. Therefore, mutations leading loss of function in SLC9A6 protein may damage the development of neurons (Mignot et al., 2013). Furthermore, SLC9A6 has been identified as a pathogenic gene (Ikeda et al., 2020) and its pathogenic variations can lead to an Angelman-like syndrome (AS-like, known as CS) (Cardon et al., 2016), which is characterized by epilepsy, intellectual disability, microcephaly, ophthalmoplegia, craniofacial dysmorphism and progressive cerebellar atrophy (Padmanabha et al., 2017).

Childhood epilepsy, especially the seizures, is manifested in the context of brain development and can change over time. CS in children presents a wide range of treatment challenges that are unique to this age group. Therefore, it is important to distinguish a de novo mutation caused childhood epilepsy or hereditary childhood epilepsy, and effective therapies for different forms of epilepsy should be targeted to the individual child according to the type of seizures. In this study, we identified a novel and de novo SLC9A6 frameshift mutation c.1548_1549insT in the patient with tonic-clonic seizures and generalized developmental delay by whole exome sequencing (WES). We also summarized the pathogenic variants reported previously and widen the spectrum of SLC9A6 mutations.

CS is a rare but increasingly diagnosed neurodevelopmental and degenerative form of X-linked intellectual disability disorder. NHE6 regulates recycling endosomal pH homeostasis and trafficking (Petitjean et al., 2020). The ion transporter is abundantly expressed in the central nervous system (CNS), which may explain why the neural phenotype of CS is so complex and diverse. Therefore, the identification of NHE6-dependent specific receptors helps to clarify the underlying mechanism of neuronal dysfunction in CS. Endosomal acidification as a result of a non-functional or absent NHE6 caused by SLC9A6 gene mutation may affect the transport of vesicles containing AMPA receptors to and from the postsynaptic membrane (Shepherd and Huganir, 2007). Brain-derived neurotrophic factor (BDNF)/tropomyosin receptor kinase B (TrkB) neurotrophin signaling pathway are necessary for correct dendrite development in many neurons in the CNS (Barford et al., 2017). Coupled with accelerated degradation of TrkB, this over-acidification can lead to a disruption of endosomal BDNF/TrkB signaling, resulting in the failure of neuronal axons and dendritic branches (Fuster and Alexander, 2014).

In addition, microtubule-associated proteins (MAP) also play a significant role in neurodevelopment and growth. Tau protein is the main microtubule associated protein of mature neurons. The spatiotemporal expression of different tau isoforms is a characteristic during brain development (Avila et al., 2004), suggesting that the regulation of tau isoforms is important during the formation of the brain (Lee et al., 1989). Moreover, deposition of tau may be mediated by the interaction with the mutant SLC9A6 protein (Garbern et al., 2010). After using a conservative genome-wide correction, it was found that the low expression of SLC9A6 was significantly related to increased tau deposition (Pescosolido et al., 2019). The most obvious pathological event in several neurodegenerative diseases is the aggregation of tau subtypes into the filamentous content of neurons, resulting in hyperphosphorylation of tau proteins (Buée et al., 2000). The c.1548_1549insT pathogenic variant in the SLC9A6 gene causes a frameshift starting at codon Leucine 517, which changes this amino acid to a proline residue, and creates a premature stop codon at position 5 of the new reading frame, denoted p.Leu517fs*5 which cause premature termination of NHE6 protein synthesis and may greatly reduce the stability of tau proteins. It can be further speculated that epilepsy and developmental delay in our patient may be due to the inability of the mutant SLC9A6 to compensate in those tissues.

Mutations in SLC9A6 seem to cause clinical symptoms that reflect developmental and progressive pathophysiology (Pescosolido et al., 2014). Slc9a6 knockout mouse models revealed that loss of functions of SLC9A6 leads to altered endosomal-lysosomal function and cholesterol accumulation in some neuronal populations in a manner that is similarly toprimary lysosomal storage diseases (Sinajon et al., 2016). Moreover, Slc9a6^−/− mutant female mouse and Slc9a6–/0 mutant male mouse showed an epileptic phenotype and reduced seizure threshold (Gilfillan et al., 2008). One of the causes of cognitive and language impairment in children with SLC9A6 mutations may be the failure of axonal and dendritic branching, leading to impaired neuronal connectivity (Roxrud et al., 2009).

Disruption of SLC9A6 should be considered in male patients with a non-15q11-13-related AS phenotype, particularly when X-linked inheritance in the family is suspected. Moreover, CS and AS share some common clinical features, and part of the patients with AS-like phenotypes carried mutation in SLC9A6 gene. Supplementary Table S2 summarizes the main clinical phenotypes of SLC9A6 mutations. AS-like phenotype was observed in the related families described by case 1 and case 2. Developmental delay, microcephaly, seizures, language retardation were present in most patients. Intellectual disability, behavior disorder and ataxia are seen in several cases. The phenotype of patients in the table is consistent with clinical diagnosis, but genotype and phenotype correlations were confirmed to be negative. Subsequently, SLC9A6 conserved domain was predicted (Figure 3). Total of 28 pathogenic variants were reported in ClinVar Database. However, 22 out of 28 (78.57%) pathogenic variants in SLC9A6 located in predicted functional domains, which include frameshift mutations, missense mutations and splicing mutations. And CS was the most reported clinical phenotype in these variations. Although 6 variants (21.43%,6/28) located in the non-domain region, 5 of them with CS located in non-domain region. It might imply that there was no significant correlation between the type of SLC9A6 gene variant and the clinical phenotype and protein domain.

Based on the information of SLC9A6 gene variation and clinical phenotype included in ClinVar Database (SLC9A6 [gene]–ClinVar–NCBI (nih.gov)), we summarized the variants that were only caused by SLC9A6 mutations (Supplementary Table S3): 1) All frameshift mutations (N = 13) of SLC9A6 are pathogenic variations. The clinical outcomes of partial variation were related to CS and no phenotype was reported for the remaining variants. 2) Seven cases of single nucleotide variants (SNVs) were pathogenic and one case were likely pathogenic, which were also associated with CS. Most SNVs in SLC9A6 were not pathogenic, but the phenotypes of most patients with SNVs were typical CS. The frameshift mutations seem to be more serious than the SNVs. Taken together, no matter the type of SLC9A6 gene mutation is frameshift mutations or SNVs, the clinical phenotype mainly characterized by CS and developmental disorder. Moreover, the complexity of the neuronal arbors in CS patients was reduced regardless of the nature of the mutation and the deficiency may contribute to postnatal microcephaly in CS (Lizarraga and Ma, 2021). The great majority of CS patients with SLC9A6 mutations appear to be result in loss of function of the relevant proteins, such as putatively protein-truncating due to early frameshift, nonsense or splicing mutations, while some missense or intra-frame deletions may be residual protein (Pescosolido et al., 2014). The most recent research indicates that most CS mutations lead to loss of protein through nonsense-mediated decay (NMD) mechanisms, the loss and expression of wild-type SLC9A6 in cells with an SLC9A6 nonsense mutation could rescue the neuronal arborization defect; But cells with recurrent missense mutation (e.g., c.1148G > A, p.G383D) in the SLC9A6 gene that resulted in a residual, nonfunctional NHE6 protein could not be rescued (Lizarraga and Ma, 2021).