Two different presentations of de novo variants of CSNK2B: two case reports

Poirier–Bienvenu neurodevelopmental syndrome (POBINDS; OMIM #618732) is a recently described disorder characterized by hypotonia, seizures, and developmental delay [1]. POBINDS is caused by mutations in the CSNK2B gene (located at 6p21.33), which encodes the beta subunit (CK2β) of the casein kinase 2 enzyme (CK2). It has been reported that the CSNK2B gene is neither susceptible to missense mutations (Z = 3.83) nor loss of function (pLi 0.92; observed/expected = 0.08; 95% confidence interval 0.03–0.38). In fact, all cases reported have been subjacent to de novo mutations and caused variants that lead to loss of function [2].

Dysmorphic features other than those originally reported have recently been associated with craniodigital syndrome (CDS), a condition that can be distinguished from POBINDS [3]. However, these dysmorphic features are not fully characterized [4].

Th work aims to detail the clinical manifestations of patients with two CSNK2B variants presenting with distinct phenotypes, one with the phenotype closer to POBINDS and another with the phenotype closer to CDS. To the best of our knowledge, only two cases have described patients with similar dysmorphic features. We also discuss possible pathogenic mechanisms based on a literature review, emphasizing the necessity to better elucidate the clinical phenotype spectrum related to the CSNK2B gene.

After the initial report of POBINDS [1], two subsequent papers documented the medical history of three unrelated patients with marked impaired global development and complex partial seizures before 2 months of age [5, 6]. The refractory seizures observed in these patients included focal evolving to generalized tonic–clonic episodes, and epileptic encephalopathy. Epilepsy in POBINDS is also characterized by unaltered EEG and MRI. Being drug resistant, tonic–clonic seizures are the most frequent presentation [2, 7, 8]. The absence of a specific imaging pattern corroborated in both of our cases to a delayed diagnosis, as exemplified in Case #1 where sleep myoclonus was one of the differential diagnosis owing to the unaltered EEG.

Few in vivo studies have demonstrated the interaction of CK2 and voltage channels that could explain the epilepsy seen in these patients. Potassium channels are the most diverse group of the ion channel family. These channels are not only involved in shaping the action potential, but also in neuronal excitability and plasticity. One study using transfected Chinese hamster ovary cells expressing Kv3.1 (a potassium channel with a very high threshold) demonstrated the importance of its phosphorylation by CK2. When CK2 was inhibited, there was a decrease in the phosphorylation of the Kv3.1 channels, leading to the decline in its electric potentials [9, 10]. Although speculative, it might be related to the etiology of seizures found in these patients.

The most striking clinical feature of Case #2 was the dysmorphic features that were recently associated with CDS, which could be distinguished from POBINDS [3]. A recently published paper reported the case of a young male Japanese patient with a deletion of 6p21.33 (which encompasses the CSNK2B gene). The patient displayed very similar features to patient #2, including large low-set ears, downslanted palpebral fissures, flared eyebrows, wide-base nose, and flat philtrum with thin upper lip [11] (for further details, see [8] and Table 1). Collectively, these cases corroborate the loss of function as plausible mechanisms described for both POBINDS and CDS.

CK2 is a heterotrimeric enzyme consisting of two catalytic CK2α or CK2α′ subunits and two regulatory β subunits. It also contains several distinct domains. The three main domains are Asp/Glu-enriched (acidic), zinc-binding (where the dimerization of β subunit occurs), and α-subunit-interaction domains [12, 13]. Even though it is considered a regulatory subunit, CK2β is highly conserved. Previous studies have demonstrated that if this regulatory subunit is excessively synthesized, there is a trend to the formation of of CK2β dimers, which are, in turn, a prerequisite for the formation and correct function of the whole enzyme [12, 14].

It is believed that the zinc-binding domain might be a hotspot for pathogenic variants and that the α-subunit-interaction domain might be the one related to refractory myoclonic epilepsy [2, 15]. The closest zinc binding site in the variant-related protein from patient#1 is a cysteine in position 137. However, p.His165Arg, according to crystal structure, is in close interaction with the complementary beta subunit, as seen in Fig. 2. The dimers of CK2β are located at the core of the tetrameric CK2 complexes and we hypothesize that symptoms observed in patient #1 might be due to alterations in this beta-beta interaction, subsequently leading to a disruption in the CK2β dimerization.

Two de novo missense mutations (p.Asp32Asn and p.Asp32His) in the same CSNK2B codon, which is mutated in patient #2, have recently been associated with CDS [3, 7]. Due to the modification of the CK2 structure, the closest CK2 catalytic subunit would be position 32, which could exert a more severe dysmorphic phenotype. One study using whole transcriptome and whole phosphoproteome profiling demonstrated that variants in this protein amino acid position caused an up-regulation of CSNK2B gene expression at transcript and protein levels, resulting in impaired cross talk between α and β subunits of CK2 [3]. Both hypotheses, however, must be further demonstrated by functional studies. A definite association between phenotype and genotype in CSNK2B gene-related epilepsy has still not been described [2].

A hypothesis for the POBINDS-related “brain-centered symptoms” could be raised after preclinical research on the CK2 function in animal cell models. The CK2 subunit appears to be constitutively active, being a target for dopamine D1 receptors signaling pathway [1, 16]. The resulting alterations in the dopaminergic system (as per increased D1 receptor signaling and/or dopamine production) has also been found in patients with juvenile myoclonic epilepsy [8], possibly contributing to understanding of myoclonic epilepsy, as well as the excessive sweating observed in the patient in case #1.

Additionally, a study using GN11 cells as a model of immature migrating neurons, demonstrated that CK2β was involved in cell migration and microfilament/microtubule organization in neurons, thus playing a role in neurodevelopmental steps such as neural differentiation, neuritogenesis, and synaptic plasticity [15, 17]. In line with these findings, studies using β subunit CK2 knockout cell lines (KOβ) pointed out that the β subunit may serve as an anchorage point for other kinases, being involved in the myogenic commitment of C2C12 (myogenic cell lines in murines) by regulating MyoD expression independently from the catalytic subunits [15, 17]. In a C2C12 model, CK2β contributes to a variety of cell cycle processes such as proliferation and cellular transport in vitro. Collectively, these findings might contribute to the elucidation of the hypotonia presented by both patients. CK2 has also been implicated in organogenesis, playing a role in heart embryogenesis and limb bud differentiation via the Wnt signaling pathway. Further studies are warranted to determine the association of this role to the limb dysmorphisms found in the patient from the second case [4].

To date, 16 patients have been reported in the literature with POBINDS and 2 patients have been reported with craniodigital intellectual disability syndrome related to mutations in the CSNK2B gene. The present cases may assist in the diagnosis of POBINS when patients present with refractory epilepsy, hypotonia, and dysmorphic features. Further investigation is warranted to better understand the interaction of CK2 and the neuronal and neuromuscular systems. We suggest that screening of CSNK2B could be included in the most common gene panels for epilepsy.