Heterozygous KCNJ10 variants affecting Kir4.1 channel cause paroxysmal kinesigenic dyskinesia

Background Paroxysmal kinesigenic dyskinesia is the representative form of paroxysmal dyskinesia, and its mechanism is unclear. Although paroxysmal kinesigenic dyskinesia is mostly attributed to genetic factors, more than 60% of paroxysmal kinesigenic dyskinesia cases are of uncertain mutations. We searched for novel genetic causes of paroxysmal kinesigenic dyskinesia and explored the corresponding pathophysiology. Methods A cohort of 476 probands with primary paroxysmal kinesigenic dyskinesia of uncertain genetic causes were enrolled for whole exome sequencing. Gene Ranking, Identification and Prediction Tool, a method of case-control analysis,was applied to identify the candidate genes. Another 46 probands were subsequently screened with Sanger sequencing. Whole-cell patch-clamp recording was applied to verify the electrophysiological impact of the identified variants. Amouse model with cerebellar heterozygous knockout of the candidate gene was generated via adeno-associated virus injection, and dyskinesia-like phenotype inducement and rotarod tests were performed. In vivo multiunit electrical recording was applied to investigate the change in neural excitability in knockout mice. Results Heterozygous variants of potassium channel inwardly rectifying subfamily J member 10 ( KCNJ10 ) mainly clustered in patients withparoxysmal kinesigenic dyskinesia compared with the control groups. Fifteenvariants were detected in 16 out of 522 probands (frequency = 3.07%). Patients with KCNJ10 variants tended to have a later onset age and shorter duration of attacks than patients with proline-rich transmembrane protein 2 mutations. Inwardly rectifying potassium channel 4.1 (Kir4.1) is highly expressed in the cerebellum of mice,and its expression pattern is consistent with the natural course of paroxysmal kinesigenic dyskinesia. Further electrophysiological recordings revealed that all the variants identified in patients led to different degrees of reduction in Kir4.1 currents, and mice with heterozygous conditional knockout of Kcnj10 in the cerebellum presented dystonic posture with epidural KCl stimulation in cerebellum, as well as poor motor coordination and motor learning ability in rotarod tests. The firing rate of deep cerebellar nuclei was significantly elevated in Kcnj10 -cKO mice, indicating abnormal hyperexcitability in the Kir4.1-deficient mouse model. Conclusion We identified heterozygous mutations of KCNJ10 as a novel genetic cause of paroxysmal kinesigenic dyskinesia. Based on the findings in the present study, we suppose that the impaired function of Kir4.1 might lead to defective homeostatic maintenance of extracellular potassium and glutamate levels and thus cause abnormal neuronal excitability. The findings elucidated the pathogenesis of paroxysmal kinesigenic dyskinesia, thoughadditional efforts are needed to reveal the role of Kir4.1 in movement disorders.