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

doi:10.21203/rs.3.rs-3292429/v1

Download PDF

Research Article

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

https://doi.org/10.21203/rs.3.rs-3292429/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

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.

Paroxysmal kinesigenic dyskinesia

KCNJ10

Kir4.1

cerebellum

Paroxysmal kinesigenic dyskinesia (PKD), the most common type of paroxysmal dyskinesia, is characterized by transient and recurrent dystonic or choreoathetoid attacks precipitated by sudden voluntary movements.^{1, 2} Primary PKD is attributed to genetic factors, and approximately one-third of PKDs are caused by mutations in proline-rich transmembrane protein 2 (PRRT2), identified in 2011 as the first causative gene for PKD.^{3, 4} In addition, PRRT2 is not only the cause of PKD but also the reason for several other paroxysmal neurological diseases, indicating the shared mechanism between PKD and other episodic disorders.^5–10 Moreover, multiple genes related to episodic neurologic disorders have also been reported to cause PKD attacks, including PNKD, SLC2A1, KCNMA1, DEPDC5, KCNA1, SCN8A, and CHRNA4.^11–15 Although previous findings support the hypothesis that PKD is probably related to the disturbance or imbalance in neural excitability,^16–18 the underlying mechanisms remain unclear. It has been well accepted that the basal ganglia-thalamo-cortical circuit is highly associated with PKD due to its clinical manifestation.¹⁹ However, increasing evidence from both clinical and mechanistic studies has highlighted the role of the cerebellum in PKD.^{17, 20, 21}

No genetic cause is found in more than 60% of PKD individuals,^{1, 5} unceasing efforts have been made to seek causative genes for PKD. Using the GRIPT (Gene Ranking, Identification and Prediction Tool) tool, we identified the TMEM151A gene as the second major gene responsible for PKD, accounting for approximately 4.80% of PRRT2-negative patients.²² Here, we apply a similar strategy of case-control analysis on whole-exome sequencing (WES) data in large-scale samples of PKD from the China Paroxysmal Dyskinesia Collaborative Group (CPDCG) to map a novel candidate gene for PKD.

We found that rare variants of potassium inwardly rectifying channel subfamily J member 10 (KCNJ10) clustered in both PRRT2- and TMEM151A-negative PKD patients, while in the control group, variants in KCNJ10 seldom occurred, suggesting that mutations in the KCNJ10 gene are potentially associated with PKD. Subsequently, we further investigated the electrophysiological changes in KCNJ10 variants and the relationship between Kir4.1 and PKD in Kcnj10-deficient mice.

Participants, data collection, and patient consent

In total, 522 PKD probands without PRRT2 or TMEM151A mutations were enrolled in our study. All probands were from the dataset of the China Paroxysmal Dyskinesia Collaborative Group (CPDCG). Whole-exome sequencing (WES) was applied to 476 probands. In addition, the WES database containing 1600 samples of non-paroxysmal disorders (WESctrl_1600) was used to compare the frequency difference of candidate genes between PKD and non-paroxysmal diseases. Another 46 PRRT2- and TMEM151A-negative PKD probands and unaffected individuals (n = 500) serving as matched healthy controls were subsequently examined for variants in candidate genes by Sanger sequencing.

The PKD patients’ clinical features were evaluated by at least two neurologists, and the clinical diagnostic criteria are detailed as follows¹: First, patients should meet both core symptoms: 1) Kinesigenic triggers and attacks presenting as dystonia, chorea, ballism, or a combination; 2) No loss of consciousness during attacks. In addition, supportive evidence such as the presence of aura, an attack duration < 1 min, a positive result of a high-knee exercise test or a good response to low-dose voltage-gated sodium channel blockers could further confirm the diagnosis, while PKD patients with secondary aetiologies were excluded. Detailed clinical information on demographics, family history, features of attacks and response to treatment were collected via in-person interviews and clinical questionnaires.

This study was approved by the ethics committee of the Sixth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China and was registered at http://www.chictr.org.cn (registration number: ChiCTR2100050834). All participants or their legal guardians signed written informed consent.

Whole-exome sequencing and bioinformatics

Genomic DNA was extracted using a standard phenol/chloroform extraction protocol. WES was performed on 476 both PRRT2- and TMEM151A-negative PKD probands using SureSelect v6 reagents for capturing exons and the HiSeq X Ten platform for sequencing. Alignment to the human genome assembly hg19 was carried out, followed by recalibration and variant calling. Then, we built a large allele frequency database containing 5000 virtual samples based on the gnomAD database (https://gnomad.broadinstitute.org/). Referring to the method (GRIPT) described by Wang et al. in 2018, ²³ we used this collection of data as a control group to filter the common human genome variants. The minor allele frequency (MAF) was set at < 0.01%. The Rare Exome Variant Ensemble Learner (REVEL) scores were applied to annotate the variants, which was considered an ensemble method for predicting the pathogenicity of missense variants by integrating all the analysis tools. To filter benign variants as much as possible, we set the cut-off of REVEL at 0.3. It should be noted that REVEL could not evaluate the pathogenesis of loss-of-function (LOF) variants; therefore, we set a default value of 0.9 for these variants. Among the genes filtered through GRIPT, we compared the frequency of rare variants in these genes between 476 PKD patients and WESctrl_1600. Genes with much higher frequency in PKD patients were selected, and we further focused on those whose expression is predominantly in the central nervous system, especially motor-related regions, including the cortex, basal ganglia, thalamus and cerebellum.

Sanger sequencing of the candidate gene

Sanger sequencing was performed to determine the presence of the variants filtered out by GRIPT analysis. Furthermore, we sequenced all the exons and flanking introns of the candidate genes with direct polymerase chain reaction using customized primers in 46 other PRRT2- and TMEM151A-negative PKD probands as well as 500 neurologically normal controls. Afterwards, the frequency and predicted pathogenesis of all the detected variants detected via Sanger sequencing were evaluated with various population databases and software, including REVEL, the 1000 Genomes Project (http://www.internationalgenome.org), the Single Nucleotide Polymorphism Database (http://www.ncbi.nlm.nih.gov/projects/SNP), the gnomAD database, the Human Gene Mutation Database (http:www.hgmd/.cf.ac.uk/ac/index.php), MutationTaster (http://www.mutationtaster.org), PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2), and SIFT (http://sift.jcvi.org). Among these tools, the 1000 Genomes Project, the Single Nucleotide Polymorphism Database and the gnomAD database are used to evaluate the frequency of the variants. SIFT and PolyPhen-2 are applied aiming to predict the pathogenicity of missense variants. The algorithm of SIFT is based on the evolutional conservatism of sequence while PolyPhen-2 is based on mixed features of sequence and structure. REVEL is a tool of comprehensive evaluation of missense variants. As SIFT, PolyPhen-2 and REVEL are not applicable of assessing nonsense or frameshift variants, we additionally use MutationTaster to evaluate the pathogenicity of these types of variants. The primers for Sanger sequencing of the exons in KCNJ10 are listed in Supplementary Table 1b.

Construction of plasmids

Fusion plasmid constructs of human cDNA encoding full-length KCNJ10 (NM_002241) with GFP at the N-ter were generated by Genecreate Technology (Wuhan, China). The wild-type (WT) KCNJ10 gene was synthesized by overlapping PCR, during which the fragment was synthesized with overlapping primers. Then, the target fragment was cloned into the pcDNA3.1 vector by Gibson assembly. Finally, the exact sequence was validated through direct Sanger sequencing. The mutations c.436C > T, c.539G > A, c.889C > T, c.883C > T, c.910_912del, c.724A > G, c.776T > C, c.1028G > A, c.554C > A, c.321_322del, c.811C > T, c.422delC, c.191G > A, c.1042C > T and c.1105G > A were introduced into the WT constructs by site-directed mutagenesis.

Cell culture transfection

HEK 293T cells were obtained from the Cell Bank of Chinese Academy of Sciences (www.cellbank.org.cn) and maintained in Dulbecco’s Modified Eagle Medium (DMEM) with 10% foetal bovine serum (FBS) and 1% penicillin/streptomycin (PS) at 37°C in a humidified incubator with 5% CO2. One day before transfection, cells were plated at 150,000 cells per well in 6-well culture dishes. The next day, cells were transfected with 2.5 µg of GFP control plasmid DNA or KCNJ10-GFP wild-type (KCNJ10-WT) or mutant plasmid DNA using Lipofectamine 3000 transfection reagent (Invitrogen).

Immunofluorescence

HEK 293T cells transfected with the respective expression constructs were washed in PBS and fixed using 4% paraformaldehyde (PFA) for immunofluorescence analysis. Cells were blocked with 10% normal donkey serum (NDS) and 0.3% Triton X-100 in 0.01 M PBS (PBST) for 60 min, incubated with primary antibodies against chicken anti-GFP (1:1000, ab13970, Abcam) and rabbit anti-Kir4.1 (Kcnj10) (APC-035, Alomone Labs) in blocking solution at 4°C overnight and incubated with Alexa Fluor 488 or 594 secondary antibodies (1:1000, Life). DAPI (40,6-diami-dino-2-phenylindole) (1:10,000, Life) was used for nucleic acid staining. Images were taken with a Zeiss 710 confocal microscope.

Electrophysiological recordings

HEK-293T cells were transfected with mutant plasmid or control vector for 24 h. Glass pipettes with a resistance of 3–4 MΩ were filled with intracellular solution containing (in mM) 125 K-gluconate, 15 KCl, 8 NaCl, 10 HEPES, 0.2 EGTA, 3 Na2-ATP, and 0.3 Na-GTP and pH set to 7.3. The external solution contained (in mM) 150 NaCl, 10 glucose, 10 HEPES, 2 CaCl₂, 5 KCl, and 1 MgCl₂, with the pH adjusted to 7.3 with Tris-base. BaCl₂ (100 µM) was used to block Kir4.1 channel currents in all experiments. All the cells were visualized with an upright epifluorescence microscope (BX51WI; Olympus, Tokyo, Japan) equipped with differential interference contrast optics and an infrared CCD camera (optiMOS, Q IMAGING; Olympus, Tokyo, Japan). Whole-cell recordings were made from every mutant and wild type with a MultiClamp 700B amplifier (Molecular Devices, Sunnyvale, CA, USA). Signals were low-pass filtered at 2 kHz and sampled at 20 kHz using Digidata 1550A (Molecular Devices) in all experiments.

Mouse model

All animal experiments were conducted according to protocols approved by the Animal Care & Welfare Committee of Shanghai Jiao Tong University School of Medicine. Kir4.1^flox/flox mice (Stock No: 026826) and Rosa26-mGFP mice (Stock No: 007676) were purchased from Jackson Laboratories, USA. C57BL/6J mice were purchased from Charles River (Beijing Vital River Laboratory Animal Technology). All animals were housed in the laboratory animal facility in the Department of Laboratory Animal Science at Shanghai Jiao Tong University School of Medicine.

Real-time quantitative PCR (RT‒qPCR)

The whole brain, forebrain or cerebellum tissues were rapidly dissected on ice and homogenized in TRIzol Reagent (Tsingke) at 4°C. RNA was then extracted and dissolved in nuclease-free water to a final concentration of 1000 ng/µl. Total RNA was reverse transcribed using HiScript III RT SuperMix (R323-01, Vazyme), followed by RT-qPCR using the ArtiCanATM SYBR qPCR kit (TSE501, Tsingke) and an ABI StepOne PlusTM thermocycler (Applied Biosystems), according to the recommendations of the manufacturer. Dissociation curve analysis was carried out after PCR amplification to confirm the absence of nonspecific amplification products and primer dimers. The results of real-time PCR were normalized to control values. Information on the primers used is listed in Supplementary Table 1a.

Immunohistochemistry

Mice were euthanized by barbiturate overdose and perfused transcardially with 0.01 M PBS followed by 4% PFA. Brains were removed, post-fixed overnight and dehydrated in 30% sucrose. Samples were then embedded in optimal cutting temperature compound (OCT) and stored at -80°C before sectioning. Tissues with regions of interest were cut by a cryostat (CM1950, Leica) at a thickness of 30 µm. Sections were rinsed with 0.01 M PBS 3 times for 10 to 15 min, after which they were blocked with 4% NDS in PBST at room temperature (RT) for 2 hours. Then, the samples were incubated with primary antibodies with 1% NDS in PBST at 4°C overnight. After PBST rinsing for 3 changes, the samples were reacted with fluorescent dye-conjugated secondary antibodies with 1% NDS in PBST with DAPI (1:1000, D9542, Sigma‒Aldrich) at RT for 2 hours and then rinsed three times before mounting with antifade mounting medium (0100-01, Sothern Biotech). Images were taken with an Olympus SLIDEVIEW™ VS200 Slide Scanner and Nikon C2 confocal microscope.

Primary antibodies included rabbit anti-Kir4.1 (1:800, APC-035, Alomone Labs), mouse anti-GFAP (1:500, MAB360, Sigma‒Aldrich) and goat anti-GFP (1:500, ab190289, Abcam). Secondary antibodies conjugated to Alexa Fluor 488 (AF488), AF568 and AF647 were 1:1000 diluted, including AF488 donkey anti-goat (705-545-003, Jackson), AF568 donkey anti-rabbit (A10042, Invitrogen) and AF647 donkey anti-mouse (715-605-151, Jackson).

Adeno-associated virus (AAV) administration

For local knockout of Kcnj10 in the cerebellum by viral injection, either Kir4.1^+/flox::Rosa26-mGFP or WT mice were injected with AAV2/5-gfaABC1D-iCre-WPRE-pA (~ 10¹³ gc/ml, S0611-5, Taitool Bioscience). In brief, 6-week-old mice were anaesthetized with isoflurane (3% for induction and 1.5% for maintenance) and then placed on a stereotaxic apparatus (51730D, Stoelting), with a heating pad used to maintain body temperature. After subcutaneous injection of 2% lidocaine for perioperative analgesia and cleaning with 10% povidone iodine and 70% ethanol, a midline incision was made, and the skull was exposed and cleaned with cotton swabs. The small holes (~ 0.6 mm in diameter) for entry of pipettes were drilled on the skull using an electric cranial drill (SD-300, Yuyan). One microlitre of AAV2/5-gfaABC1D-iCre-WPRE-pA (~ 1013 gc/ml) was injected into the cerebellum vermis of mice at two sites (coordinates from bregma: 6.0 mm posterior, 0.0 mm lateral, 2.0 mm and 3.0 mm ventral) using a gastight syringe (7803-04, Hamilton) and glass pipettes at a slow rate (~ 100 nl/min) controlled by Quintessential Stereotaxic Injector (53311, Stoelting). Pipettes were left in place for at least 10 min. Animals were allowed to recover from anaesthesia on a heat pad after the injection was completed. Three weeks after the surgery, at least 3 mice were sacrificed for immunohistological analysis to verify the efficiency of virus transfection, and others were used for subsequent rotarod tests and in vivo multi-unit electrical recording.

Epidural KCl stimulation

Mice were anesthetized, placed on the stereotaxic apparatus as AAV administration. To induce dyskinetic postures in animals, KCl (1 M)-soaked cotton ball was applied on the local surface of the cerebellar cortex. ¹⁷ Then behaviour was monitored by an experimenter and recorded by a camera. The dyskinesia-attacked mice were expected to exhibit abnormal postures (such as an extended and stiff limb or/and tail, flatten posture, exaggerated truncal flexion, and truncal twisting). If the dyskinetic movements or postures appeared after stimulation and lasted for more than 5 s, the dyskinesia attack was recorded. The average occurrence of dystonia was calculated in each group (WT/Kcnj10-cKO).

Rotarod test

Mice were placed onto a rotating rod with auto acceleration from 4 to 40 rpm in 5 min and maintenance at 40 rpm for another 5 min (47650, UGO). A trial ended if the animal fell off the rods or gripped the device and spun around for 3 consecutive revolutions without attempting to walk on the rods. The latency and speed of the rod at the endpoint were recorded for each mouse in four repeated trials to evaluate motor coordination and motor learning ability.

In vivo multi-unit electrical recording

Mice were anaesthetized with 5% chloral hydrate. A small hole was drilled overlying the targeted brain regions (coordinates from bregma: 6.3 mm posterior, 1.5 mm lateral, 2.3 mm ventral from dura), and the dura was removed carefully. Neural signals were recorded from a multichannel data acquisition system (Zeus, Bio-Signal Technologies). Electrodes consisted of 16 individually insulated nichrome wires (35 µm inner diameter, impedance 300–900 Kohm; Stablohm 675, California fine wire). The implanted electrodes were secured with dental cement, and electrophysiological recording was initiated after recovery. Spikes were extracted with high-pass (300 Hz) filters and sampled at 30 kHz. Real-time spike sorting was performed using principal component analysis (PCA). Offline Sorter (Plexon) was used for spike sorting refinement before analysing data in NeuroExplorer (Nex Technologies).

Statistical analysis

The experimental data were analysed using GraphPad InStat3., Origin 8, and CorelDraw12. The data are illustrated as the mean ± standard error (mean ± s.e.m.). Unpaired Student’s t test was used for statistical analysis of the two groups of data with a normal distribution. For data not conforming to a normal distribution or with heterogeneity of variance, the Mann‒Whitney test was used to test the statistical significance. Two-way ANOVA was used for data on a quantitative dependent variable at multiple levels of two categorical independent variables. In the electrophysiological experiment, “n” represents the number of cells recorded. p < 0.05 indicates a significant difference.

Heterozygous variants in the KCNJ10 gene are associated with PKD

A total of 13 rare heterozygous variants in KCNJ10 were detected in 14 probands through exome sequencing (Fig. 1A), including c.191G > A (p.Trp64X), c.321_322del (p.Val109Glyfs*15), c.422delC (p.Pro141Hisfs*57), c.436C > T (p.Leu146Phe), c.539G > A (p.Arg180His), c.554C > A (p.Ala185Glu), c.724A > G (p.Thr242Ala), c.776T > C (p.Val259Ala), c.811C > T (p.Arg271Cys), c.883C > T (p.Gln295X), c.889C > T (p.Arg297Cys), c.910_912del (p.Glu304del) and c.1028G > A (p.Arg343His). In addition, 2 out of 46 PRRT2- and TMEM151A-negative PKD probands were detected with KCNJ10 variants {c.1042C > T (p.Arg348Cys) and c.1105G > A (p.Gly369Ser)} by Sanger sequencing (Fig. 1A). In contrast, only 9 variants with low frequency were detected in WESctrl_1600 (16/522 versus 9/1600, 3.07% versus 0.56%). No pathogenic variants in KCNJ10 were found in any of the 500 healthy controls.

Clinical and genetic findings of PKD patients with heterozygous KCNJ10 variants

Sixteen probands with KCNJ10 variants were from 14 sporadic cases and 2 PKD families (Fig. 1A). The brother (II:1) in Family 1 and the father (I:1) in Family 2 experienced PKD attacks in adolescence as well. KCNJ10 variants were co-segregated with patients in these 2 families. Family 2 was a typical autosomal dominant family while the co-segregation analysis of Family 1 showed that the asymptotic father of the two affected brothers was the carrier of the variant c.883C > T. In addition, the co-segregation analysis in partial sporadic cases showed that c.436C > T and c.554C > A were inherited from the father of the proband while c.321_322del was maternally inherited. However, all the parents with variants were asymptomatic, which suggested that incomplete penetrance also existed in KCNJ10-related PKD.

The detailed clinical manifestations of KCNJ10-related PKD patients are listed in Supplementary Table 3. The proportions of the main characteristics are shown in Table 1, and we further compared the genotype-phenotype correlations among patients with KCNJ10, PRRT2 and TMEM151A mutations. The relevant parameters of PRRT2-positive (PRRT2+) or TMEM151A-positive (TMEM151A+) PKD patients were quoted from our previously published research.^5,22 The mean onset age of all the patients with KCNJ10 variants was 13.56 ± 3.27 years, which is later than that of PRRT2 + patients (9.36 ± 4.42 years) but similar to that of TMEM151A + patients (12.93 ± 3.15 years). Four patients reported spontaneous remission of the disease, with a mean age of remission of 19.83 ± 0.64 years. The precipitating factors in KCNJ10-positive (KCNJ10+) patients were the same as those in patients with other genotypes, including sudden movement (100.00%), speed change (62.50%), emotional stress (68.75%), intentional movements (37.50%) and fatigue (18.75%). The majority of the patients (87.50%) experienced an aura before the onset of involuntary movements. Dystonia was still the most frequent type of attack (93.75%), followed by chorea (31.25%) and ballism (6.25%). The duration of attacks in KCNJ10 + patients, which mostly lasted within 10 seconds, was shorter than that in PRRT2 + patients. Face involvement was reported by 81.25% of patients. Most patients with KCNJ10 variants had a pure phenotype (87.50%), and approximately 56.25% (9/16) of patients were prescribed carbamazepine or oxcarbazepine as the first medical choice. Overall, patients with KCNJ10 variants still responded well to carbamazepine or oxcarbazepine, as 66.67% of patients experienced complete remission, and 22.22% had partial remission.

Table 1

Comparison of clinical manifestations of paroxysmal kinesigenic dyskinesia (PKD) patients with different genotypes
			PRRT2+ N (%)	TMEM151A+ N(%)	KCNJ10+ N (%)	P Value (P1) (PRRT2 vs. KCNJ10)	P Value (P2) (TMEM151A vs. KCNJ10)
Total			183	29	18
Probands			170	25	16
Sex (probands)	Male		123 (72.35)	18 (72.00)	16 (100)	0.037^b	0.056^b
Sex (probands)	Female		47 (27.65)	7 (28.00)	0 (0)
Form	Pure		124 (67.76)	25 (86.21)	14 (87.50)	0.174^b	1.000^b
Form	Complicated		59 (32.24)	4 (13.79)	2 (12.50)
Family history	Sporadic		83 (48.82)	19 (76.00)	12 (85.71)	0.008^a	0.686^c
Family history	Familial		87 (51.18)	6 (24.00)	2 (14.29)
Age at onset (y)			9.36 ± 4.42	12.93 ± 3.15	13.56 ± 3.27	< 0.001^d	0.533^d
Remission age (y)			21.01 ± 5.87	21.42 ± 2.33	19.83 ± 0.64	0.628^d	0.157^d
Aura		Yes	135 (75.00)	27 (93.10)	14 (87.50)	0.414^b	0.932^b
Aura		No	45 (25.00)	2 (6.90)	2 (12.50)
Triggers	Sudden movement		183 (100.00)	29 (100.00)	16 (100)
	Speed change	Yes	123 (67.96)	25 (86.21)	10 (62.50)	0.655^a	0.145^b
	Speed change	No	58 (32.04)	4 (13.79)	6 (37.50)
	Move intention	Yes	70 (39.33)	15 (51.72)	6 (37.5)	0.886^a	0.360^a
	Move intention	No	108 (60.67)	14 (48.28)	10 (62.5)
	Stress	Yes	97 (54.19)	19 (65.52)	11 (68.75)	0.262^a	0.826^a
	Stress	No	82 (45.81)	10 (34.48)	5 (31.25)
	Fatigue	Yes	32 (17.98)	10 (34.48)	3 (18.75)	1.000^b	0.441^b
	Fatigue	No	146 (82.02)	19 (65.52)	13 (81.25)
Attack	Dystonia	Yes	172 (93.99)	29 (100.00)	15 (93.75)	1.000^c	0.356^c
	Dystonia	No	11 (6.01)	0 (0.00)	1 (6.25)
	Chorea	Yes	89 (48.63)	8 (27.59)	5 (31.25)	0.182^a	1.000^b
	Chorea	No	94 (51.37)	21 (72.41)	11 (68.75)
	Ballism	Yes	10 (5.46)	3 (10.34)	1 (6.25)	1.000^c	1.000^b
	Ballism	No	173 (94.54)	26 (89.66)	15 (93.75)
Duration (sec)	≤ 10		65 (36.72)	17 (58.62)	13 (81.25)	0.003^c	0.304^c
	10 to 30		87 (49.15)	11 (37.93)	3 (18.75)
	≥ 30		25 (14.12)	1 (3.45)	0 (0)
Face involvement		Yes	109 (59.89)	25 (86.21)	13 (81.25)	0.092^a	0.992^b
Face involvement		No	73 (40.11)	4 (13.79)	3 (18.75)
Treatment effect	Complete Relief		91 (79.82)	14 (82.35)	6 (66.67)	0.191^c	0.550^c
	Incomplete Relief		21 (18.42)	3 (17.65)	2 (22.22)
	No response		2 (1.75)	0 (0.00)	1 (7.14)
Differences among PRRT2+, TMEM151A+, KCNJ10 + patients compared by ^aχ² test, ^bYates’ corrected χ² test, ^cFisher exact probability, and ^dtwo sample t-test. p value was set at 0.05.

Heterozygous KCNJ10 mutations lead to reduced Kir4.1 currents

All 15 mutants showed depolarized resting membrane potentials (RMPs) and increased membrane resistances (Rms) to different degrees compared with the wild type (Supplementary Table 2). Accordingly, apparent reductions in macroscopic K⁺ currents as well as Kir4.1-sensitive currents were observed in all the mutants (Fig. 2). Furthermore, according to the different degrees of decreased currents, the 15 mutants were categorized into 3 groups: complete blockade of Kir4.1 (bare currents were detected, including c.191G > A, c.321_322del, c.422delC, c.436C > T, c.554C > A, c.776T > C, c.883C > T, c.889C > T, c.910_912del), partial blockade of Kir4.1 (approximately 80% of the currents were detected, including c.539G > A, c.811C > T, c.1028G > A and c.1105G > A) and slight blockade of Kir4.1 (approximately 20% of the currents were blocked, including c.724A > G and c.1042C > T) (Fig. 2I).

The expression pattern of Kir4.1 in mice

To explore the expression pattern of Kir4.1 in the central nervous system, we measured the level of mouse Kcnj10 mRNA by real-time PCR. We found that in the whole brain, Kcnj10 was relatively low during the embryonic period, markedly increased during postnatal stages, peaked at postnatal day 30 (P30), and declined in adulthood (Fig. 3A). Similar expression trends were observed in the cerebellum (Fig. 3B) and forebrain (Fig. 3C). Immunofluorescence analyses in P30 mice confirmed that Kcnj10 mostly expressed in astrocytes (Fig. 3D) and was ubiquitously expressed in the central nervous system, with high levels in the cerebellar cortex, olfactory bulb, thalamus, hypothalamus, pons, cortex and hippocampus (Fig. 3E). Similar to mRNA level, the protein level of Kir4.1 also peaked at P30 during development, confirmed by immunofluorescence in the cerebellum and cortex of P7, P30, P56 and P84 mice (Fig. 3F-H).

Given the prominent role of the cerebellum in the pathogenesis of PKD, we generated Kcnj10-cKO mice with conditional heterozygous knockout of Kir4.1 in cerebellar astrocytes. Specifically, we crossed Kir4.1^flox/flox mice with Rosa26-mGFP mice. AAV2/5-gfaABC1D-iCre-WPRE-pA was then stereotaxically injected into the cerebellar vermis of Kir4.1^+/flox::Rosa26-mGFP (Kcnj10-cKO) and WT mice as controls (Fig. 4A). Three weeks after injection, to assess the transfection efficiency, immunohistochemical analyses were performed and revealed that 97.68 ± 0.67% (mean ± s.e.m., n = 3) of cerebellar astrocytes of Kcnj10-cKO coexpressed GFP (Fig. 4B, C) with ablation of Kir4.1 (Fig. 4D).

Because Kir4.1 channel in astrocyte are crucial for the maintenance of extracellular K⁺ concentration and thus influence the neural excitability, we applied K⁺ stimulation in the cerebellar cortex of Kcnj10-cKO mice to induce the dyskinetic postures in animals. The results showed that stiff lower limb and tail was induced in Kcnj10-cKO mice (occurrence%=66.7%, n = 6), while no abnormal dystonic posture was induced in WT mice (n = 6) (Fig. 4M, N). The induced dystonic phenotype suggested that impaired cerebellar Kir4.1 channel is associated with the abnormal motor function in mice.

The rotarod test was performed to assess the motor function of cerebellar Kcnj10-cKO mice, and the test was repeated four times for each mouse (Fig. 4E). Both the latency and the speed of the rod at the point of falling were significantly reduced in the Kcnj10-cKO group compared to the controls (Fig. 4F, G), indicating a deficiency of motor coordination in cerebellar Kir4.1-knockout mice. In addition, Kcnj10-cKO mice also showed poorer motor learning ability, as they failed to show improvement during repeated trials (Fig. 4H, I).

In the cerebellum, the deep cerebellar nuclei (DCN) play an essential role in motor coordination. Since poor balance was observed in Kcnj10-cKO mice, we subsequently investigated DCN neuronal firing in Kcnj10-cKO mice compared with WT mice (Fig. 4J), and the results showed that the average firing rate of the DCN neurons in Kcnj10-cKO mice was increased (WT: 41 recorded neurons from 4 mice ; Kcnj10-cKO: n = 62 recorded neurons from 4 mice) (Fig. 4K, L), indicating that hyperexcitability occurred in DCN neurons due to an elevated K⁺ level in the extracellular space in Kir4.1-deficient mice.

PKD is representative of paroxysmal dyskinesia, and it is widely accepted that primary PKD is attributed to genetic factors, but more than 60% of PKD patients are not identified with certain mutations.¹ In our previous study, we identified TMEM151A as the second causative gene for PKD by means of GRIPT.²² Here, we applied the same strategy to analyse the WES data of both PRRT2- and TMEM151A-negative patients and eventually filtered out and focused on the KCNJ10 gene. Because PKD is a rare disease with incomplete genetic penetrance, the high frequency of KCNJ10 variants in PKD patients suggests its potential role in the pathogenesis of PKD.

KCNJ10 (NM_002241.5), located on chr 1q23.2, consists of 2 exons and encodes 379 amino acids. Its protein product is the inwardly rectifying potassium (Kir) 4.1 channel. The primary structure of the Kir4.1 subunit is composed of two transmembrane (TM) regions, an extracellular pore-forming region, which has a -G-Y-G- signature sequence acting as an ion filter, and intracellular N- and C-terminal domains (Fig. 1B).²⁴ In this study, a total of 15 KCNJ10 variants (4 nonsense/frameshift and 11 missense) were identified, and most of them (11 mutations) cluster in the N-terminal cytoplasmic domain (Fig. 1B), while mutation c.436C > T (p.Leu146Phe) is in the transmembrane domain and two frameshift mutations {c.321_322del (p.Val109Glyfs*15) and c.422delC (p.Pro141Hisfs*57)} are in the extra-membrane domain. It was initially known that homozygous and compound heterozygous KCNJ10 mutations are associated with sensorineural deafness, ataxia, impaired intellectual development, and electrolyte imbalance (SeSAME syndrome) and enlarged vestibular aqueducts.²⁵ Frequent mutations of the KCNJ10 gene in patients with SeSAME syndrome can cause marked inhibition of Kir4.1 channels.²⁴ To validate whether the variants in this study lead to the dysfunction of Kir4.1, we investigated the electrophysiological changes of the mutants, and the results illustrated that all the mutations caused the decreased Kir4.1 current to different degrees. Moreover, most mutations (13/15) profoundly reduced the Kir4.1 current (≥ 80% Kir4.1-sensitive currents blocked), indicating that the mutations identified in PKD patients significantly disturbed the function of Kir4.1. All the patients with KCNJ10 presented typical clinical features of PKD and it is hard to conclude the relationship between the degree of the Kir4.1 blockage and the severity of phenotype in this study. However, the clinical and genetic heterogeneity of PKD is widely accepted as one mutation could cause different phenotype varying from asymptomatic carriers to complicated form of PKD. ^1,5 The underlying reason for the heterogeneity is unclear but more collected data in the following study would assist to clarify the mechanism.

However, whether Kir4.1 defects are related to PKD needs further exploration. Therefore, we investigated the expression pattern of Kcnj10, and the results showed that Kcnj10 mRNA was relatively low during the embryonic period, significantly increased during postnatal stages, peaked at adolescence (P30), and finally declined in adulthood. In 2015, Moroni et al. disclosed a gradual increment of Kir4.1 expression in the somatosensory cortex of rats during postnatal to P30.²⁶ In the present study, we detected an increment before P30 but a progressive decrease during P30 ~ P84 of Kir4.1 expression in both cerebellum and cortex of mice by quantifying the immunoreactivity of Kir4.1 immuno-positive cells in cerebellum and cortex, which is accordance with the mRNA expression pattern during qPCR. The finding is consistent with the natural course of PKD, and the other two specific proteins related to PKD, Prrt2 and Tmem151a, also share the same tendency of expression.

Although the mechanism of PKD is not yet well elucidated, it has long been established that this disease is related to the disturbance or imbalance in neural excitability based on its clinical features and electrophysiological studies¹⁹. Moreover, recent studies have highlighted that abnormal excitability in the cerebellum is involved in PKD.^{17, 20, 21} Because of the high expression of Kcnj10 in the cerebellum, we further explored the motor ability of cerebellar Kcnj10-cKO mice. Poor balance and deficits in motor learning ability were obvious in Kcnj10-cKO mice, revealing that the dysfunction of cerebellar Kir4.1 could disturb motor function in mice. Although Kir4.1 is mostly expressed in astrocytes in the nervous system, it plays a crucial role in maintaining the structural and functional integrity of the brain, including formation of the blood‒brain barrier, regulation of synaptogenesis, and maintenance of metabolic, ionic, and neurotransmitter homeostasis.²⁴ Among these functions, spatial potassium (K⁺) buffering by astrocytes is an essential system for controlling extracellular K⁺ concentration and neuronal excitability.^{27, 28} It has been validated that the inhibition of Kir4.1 channels by gene mutations, expressional suppression, or pharmacological treatments could compromise the spatial buffering of K⁺, resulting in increased extracellular K⁺ and depolarized RMPs.²⁴ In addition, Kir4.1 channel-mediated spatial K⁺ buffering is functionally linked to excitatory amino acid transporters (EAATs) and plays an important role in glutamate homeostasis in tripartite synapses.^28–31 Kir4.1 channel blockade can cause a depolarization of astrocytes, which in turn inhibits astrocytic glutamate uptake via EAAT2.^30–32 Elevated levels of extracellular K⁺ and glutamate are both key factors for the hyperexcitation of neurons. In accordance with the theory, we induced the dystonic posture in cerebellar Kcnj10-cKO mice by applying high K⁺ concentration on local surface of the cerebellar cortex. The inducement of the phenotype not only verified that defect Kir4.1 might be associated with the attack of PKD, but also indicated that the abnormal excitability is related with the impaired spatial K⁺ buffering by astrocyte. Further aberrant DCN neuronal firing was observed in cerebellar Kir4.1-deficient mice in the present study. In the cerebellum, the DCN plays an essential role in motor coordination,³³ and its dysfunction has been implicated in animal models of dyskinesia.^34–37 The increased rate of DCN firing in this study suggests hyperexcitability of the cerebellum, which is involved in the mechanism of PKD. In Prrt2 mutant mice, similar electrophysiological abnormalities in the DCN have also been observed.¹⁷ Therefore, we speculate that the dysfunction of Kir4.1 caused by heterozygous KCNJ10 variants might be associated with abnormal hyperexcitability in the cerebellum, which is involved in the pathogenesis of PKD. In other words, cerebellar Kir4.1 probably acts as a negative modulator of neural excitability in PKD. However, a more detailed mechanism of Kir4.1 in the regulation of cerebellar neurons in PKD remains to be discovered.

In this study, KCNJ10 variants accounted for approximately 3.07% (16/522) of PRRT2- and TMEM151A-negative PKD probands. Its frequency in overall PKD patients is lower than that of PRRT2 and TMEM151A mutations.^{5, 22} Although DNA analysis was not conducted in all the parents of the probands, the current results of co-segregation analysis showed that incomplete penetrance is apparent in KCNJ10-related PKD, which is also common in PKD with PRRT2 and TMEM151A mutations. Due to the limited parents enrolled, the exact incomplete penetrance could not be obtained. However, PRRT2 mutations are identified in one-third of PKD patients and the prevalence of PRRT2 mutations in familial PKD is much higher than in sporadic cases,⁵ indicating that incomplete penetrance is more common in sporadic cases and other genotype except PRRT2. Following DNA analysis would be continued to figure out the accurate incomplete penetrance in PKD patients with KCNJ10 mutations and the underlying reason for the high frequency of incomplete penetrance in PKD. Regarding clinical manifestations, patients with KCNJ10 mutations are more similar to those carrying TMEM151A mutations than PRRT2 + patients.^{5, 22}KCNJ10 + patients have predominantly sporadic cases, tending to have a later onset age and a short duration of attacks. Other phenotypic spectra of the KCNJ10 + patients, including triggers, attack form, occurrence of facial involvement and aura, were similar to the patients with PRRT2 and TMEM151A mutations in our previous studies.^{5, 22, 38, 39} Nevertheless, as patients with KCNJ10 mutations are still limited, more subjects are needed to determine the definite phenotype-genotype correlation. Moreover, an increasing number of studies have reported the relationship between brain diseases and Kir4.1 dysfunction despite classical SeSAME, including depressive disorders, epileptic diseases, Huntington’s disease, autism spectrum disorders, Parkinson’s disease, Alzheimer’s disease, amyotrophic lateral sclerosis, and neuropathic pain.^{24, 40} In the present study, 3 mutations (c.889C > T/p.Arg297Cys; c.811C > T/p.Arg271Cys; c.1042C > T/p.Arg348Cys) were previously documented in other phenotypes (SeSAME syndrome; digenic non-syndromic hearing loss; seizure susceptibility, respectively).^{25, 41, 42} However, all the 3 documented variants were reported in cases with homozygous or compound heterozygous mutations, while heterozygous KCNJ10 mutations are the cause for PKD. In general, the symptoms in diseases with autosomal recessive inheritance are more severe than that in disorders with autosomal dominant inheritance. SeSAME syndrome is a complicated syndrome composed of several symptoms including seizures, sensorineural deafness, ataxia, impaired intellectual development, and electrolyte imbalance. In contrary, KCNJ10-related PKD is much less severe in manifestation compared with SeSAME. The different type of inheritance and phenotype indicate that impaired Kir4.1 in different degree could result in a wide range of clinical phenotypes and PKD would be a new spectrum and milder phenotype of Kir4.1-associated channelopathy.

Overall, we identified heterozygous KCNJ10 mutation as a novel genetic cause of PKD through the GRIPT method, broadening the known spectrum of PKD genotypes as well as Kir4.1-related channelopathy phenotypes. KCNJ10 mutations account for 3.06% of PRRT2- and TMEM151A-negative PKD, following the proportion of PRRT2 and TMEM151A mutations in PKD patients. The phenotype of KCNJ10-related PKD is much more similar to that of patients carrying TMEM151A variants, as KCNJ10 mutations are more common in sporadic cases and are prone to present a shorter duration of attacks and to have a later onset age. All the mutations identified in the present study led to reduced Kir4.1 currents to different degrees. Dystonic posture could be induced in heterozygous Kcnj10-cKO mice by high K⁺ concentration in cerebellum and poor motor coordination and abnormal cerebellar electrophysiological changes were also observed. Based on the findings in the present study, we suppose that the dysfunction of cerebellar Kir4.1 might result in defective homeostatic maintenance of extracellular potassium and glutamate levels and thus lead to abnormal neuronal excitability, which is involved in the mechanism of PKD. The identification of the KCNJ10 gene for PKD elucidated the pathogenesis of PKD, although additional efforts are needed to further reveal the whole picture of the role of Kir4.1 in movement disorders.

PKD

paroxysmal kinesigenic dyskinesia

KCNJ10

potassium channel inwardly rectifying subfamily J member 10

Kir4.1

inwardly rectifying potassium channel 4.1

SeSAME

seizures, sensorineural deafness, ataxia, mental retardation, and electrolyte imbalance

PRRT2

proline-rich transmembrane protein 2

TMEM151A

transmembrane protein 151A

GRIPT

Gene Ranking, Identification and Prediction Tool.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.

Author contributions

Xiaojun Huang analyzed and interpreted the data, and was a major contributor in writing the manuscript. Xin Fu and Jingying Wu analyzed and interpreted the data, and also participated in writing the manuscript. Xiaoqi Hong assisted in the electrophysiological experiments. Ziyi Li interpreted the clinical data of enrolled patients. Lan Zheng, Qing Liu, Shendi Chen, Beisha Tang, Yuwu Zhao, Xiaorong Liu, Xunhua Li, Xiaoli Liu, Zaiwei Zhou, Li Wu, Kan Fang, Ping Zhong, Mei Zhang, Xinghua Luan participated in the clinical evaluation of recruited patients. Wotu Tiananalyzed and interpreted the data, and revised the manuscript. Xiaoping Tong provided critical comments and suggestions in electrophysiological experiments, and further revised the manuscript, Li Cao designed and conceptualized the study, and revised the manuscript. All authors read and approved the final version of the paper.

Acknowledgements

We appreciate the cooperation of all the patients and their families. We are indebted to all the members from the China Paroxysmal Dyskinesia Collaborative Group (CPDCG) for their contributions. In addition, we thank Dr Wu Jingchuan from the Department of Neurosurgery, Huashan Hospital, and the innovative research team of high-level local universities in Shanghai for their support (SHSMU-ZDCX20211901).

Funding

Prof. Cao received funds from the National Natural Science Foundation of China (No. 81870889) and the Training Program for Research Physicians of Innovative Translational Ability (SHDC2022CRD037).

Prof. Tong received funds from the Ministry of Science and Technology China Brain Initiative (2022ZD0204702), the National Natural Science Foundation of China (82271466, 31970904).

Dr. Wotu Tian received funds from the National Natural Science Foundation for Young Scholars of China (No. 82201398), the China Postdoctoral Science Foundation (No. 2022M712117), and the Shanghai Pujiang Program (22PJD052).

Competing interests

The authors report no competing interests.

Ethics approval and consent to participate

This study was approved by the Shanghai Jiao Tong University School of Medicine Affiliated Sixth People’s Hospital Foundation Ethical Committee. Written informed consent was obtained from all individual participants or their caregivers.

Consent for publication

Not applicable.

Cao L, Huang X, Wang N, Wu Z, Zhang C, Gu W, et al. Recommendations for the diagnosis and treatment of paroxysmal kinesigenic dyskinesia: an expert consensus in China. Transl Neurodegener Feb. 2021;16(1):7.
Bruno MK, Hallett M, Gwinn-Hardy K, Sorensen B, Considine E, Tucker S, et al. Clinical evaluation of idiopathic paroxysmal kinesigenic dyskinesia: new diagnostic criteria. Neurol Dec. 2004;28(12):2280–7.
Wang JL, Cao L, Li XH, Hu ZM, Li JD, Zhang JG, et al. Identification of PRRT2 as the causative gene of paroxysmal kinesigenic dyskinesias. Brain Dec. 2011;134(Pt 12):3493–501.
Chen WJ, Lin Y, Xiong ZQ, Wei W, Ni W, Tan GH, et al. Exome sequencing identifies truncating mutations in PRRT2 that cause paroxysmal kinesigenic dyskinesia. Nat Genet Nov. 2011;20(12):1252–5.
Huang XJ, Wang SG, Guo XN, Tian WT, Zhan FX, Zhu ZY, et al. The Phenotypic and Genetic Spectrum of Paroxysmal Kinesigenic Dyskinesia in China. Mov Disord Aug. 2020;35(8):1428–37.
Erro R, Bhatia KP. Unravelling of the paroxysmal dyskinesias. J Neurol Neurosurg Psychiatry Feb. 2019;90(2):227–34.
Kim SY, Lee JS, Kim WJ, Kim H, Choi SA, Lim BC, et al. Paroxysmal Dyskinesia in Children: from Genes to the Clinic. J Clin Neurol Oct. 2018;14(4):492–7.
Nobile C, Striano P. PRRT2: a major cause of infantile epilepsy and other paroxysmal disorders of childhood. Prog Brain Res. 2014;213:141–58.
Heron SE, Dibbens LM. Role of PRRT2 in common paroxysmal neurological disorders: a gene with remarkable pleiotropy. J Med Genet Mar. 2013;50(3):133–9.
Ebrahimi-Fakhari D, Moufawad El Achkar C, Klein C et al. PRRT2-Associated Paroxysmal Movement Disorders. In: Adam MP, Ardinger HH, Pagon RA, eds. GeneReviews((R)). Seattle (WA); 1993.
Wang HX, Li HF, Liu GL, Wen XD, Wu ZY. Mutation Analysis of MR-1, SLC2A1, and CLCN1 in 28 PRRT2-negative Paroxysmal Kinesigenic Dyskinesia Patients. Chin Med J (Engl). May 5 2016;129(9):1017–1021.
Tian WT, Huang XJ, Mao X, Liu Q, Liu XL, Zeng S, et al. Proline-rich transmembrane protein 2-negative paroxysmal kinesigenic dyskinesia: Clinical and genetic analyses of 163 patients. Mov Disord Mar. 2018;33(3):459–67.
Yin XM, Lin JH, Cao L, Zhang TM, Zeng S, Zhang KL, et al. Familial paroxysmal kinesigenic dyskinesia is associated with mutations in the KCNA1 gene. Hum Mol Genet Feb. 2018;15(4):757–8.
Gardella E, Becker F, Moller RS, Schubert J, Lemke JR, Larsen LH, et al. Benign infantile seizures and paroxysmal dyskinesia caused by an SCN8A mutation. Ann Neurol Mar. 2016;79(3):428–36.
Jiang YL, Yuan F, Yang Y, Sun XL, Song L, Jiang W. CHRNA4 variant causes paroxysmal kinesigenic dyskinesia and genetic epilepsy with febrile seizures plus? Seizure. Mar 2018;56:88–91.
Fruscione F, Valente P, Sterlini B, Romei A, Baldassari S, Fadda M et al. PRRT2 controls neuronal excitability by negatively modulating Na + channel 1.2/1.6 activity. Brain. Apr 1 2018;141(4):1000–1016.
Lu B, Lou SS, Xu RS, Kong DL, Wu RJ, Zhang J, et al. Cerebellar spreading depolarization mediates paroxysmal movement disorder. Cell Rep Sep. 2021;21(12):109743.
Mo J, Wang B, Zhu X, Wu X, Liu Y. PRRT2 deficiency induces paroxysmal kinesigenic dyskinesia by influencing synaptic function in the primary motor cortex of rats. Neurobiol Dis Jan. 2019;121:274–85.
Li ZY, Tian WT, Huang XJ, Cao L. The Pathogenesis of Paroxysmal Kinesigenic Dyskinesia: Current Concepts. Mov Disord Jan 31 2023, Epub ahead of print.
Ekmen A, Meneret A, Valabregue R, Beranger B, Worbe Y, Lamy JC, et al. Cerebellum Dysfunction in Patients with PRRT2-Related Paroxysmal Dyskinesia. Neurol Mar. 2022;8(10):e1077–89.
Binda F, Valente P, Marte A, Baldelli P, Benfenati F. Increased responsiveness at the cerebellar input stage in the PRRT2 knockout model of paroxysmal kinesigenic dyskinesia. Neurobiol Dis May. 2021;152:105275.
Tian WT, Zhan FX, Liu ZH, Liu Z, Liu Q, Guo XN, Zhou ZW, et al. TMEM151A Variants Cause Paroxysmal Kinesigenic Dyskinesia: A Large-Sample Study. Mov Disord Mar. 2022;37(3):545–52.
Wang J, Zhao L, Wang X, Chen Y, Xu M, Soens ZT, et al. GRIPT: a novel case-control analysis method for Mendelian disease gene discovery. Genome Biol Nov. 2018;26(1):203.
Ohno Y, Kunisawa N, Shimizu S. Emerging Roles of Astrocyte Kir4.1 Channels in the Pathogenesis and Treatment of Brain Diseases. Int J Mol Sci Sep 23 2021;22(19).
Scholl UI, Choi M, Liu T, Ramaekers VT, Häusler MG, Grimmer J, et al. Seizures, sensorineural deafness, ataxia, mental retardation, and electrolyte imbalance (SeSAME syndrome) caused by mutations in KCNJ10. Proc Natl Acad Sci U S A Apr. 2009;7(14):5842–7.
Moroni RF, Inverardi F, Regondi MC, Pennacchio P, Frassoni C. Developmental expression of Kir4.1 in astrocytes and oligodendrocytes of rat somatosensory cortex and hippocampus. Int J Dev Neurosci. 2015;47(Pt B):198–205.
Walz W. Role of astrocytes in the clearance of excess extracellular potassium. Neurochem Int Apr. 2000;36(4–5):291–300.
Kofuji P, Newman EA. Potassium buffering in the central nervous system. Neuroscience. 2004;129(4):1045–56.
Olsen ML, Sontheimer H. Functional implications for Kir4.1 channels in glial biology: from K + buffering to cell differentiation. J Neurochem Nov. 2008;107(3):589–601.
Djukic B, Casper KB, Philpot BD, Chin LS, McCarthy KD. Conditional knock-out of Kir4.1 leads to glial membrane depolarization, inhibition of potassium and glutamate uptake, and enhanced short-term synaptic potentiation. J Neurosci Oct. 2007;17(42):11354–65.
Kucheryavykh YV, Kucheryavykh LY, Nichols CG, Maldonado HM, Baksi K, Reichenbach A, et al. Downregulation of Kir4.1 inward rectifying potassium channel subunits by RNAi impairs potassium transfer and glutamate uptake by cultured cortical astrocytes. Glia Feb. 2007;55(3):274–81.
Chever O, Djukic B, McCarthy KD, Amzica F. Implication of Kir4.1 channel in excess potassium clearance: an in vivo study on anesthetized glial-conditional Kir4.1 knock-out mice. J Neurosci Nov. 2010;24(47):15769–77.
Heck DH, De Zeeuw CI, Jaeger D, Khodakhah K, Person AL. The neuronal code(s) of the cerebellum. J Neurosci Nov. 2013;6(45):17603–9.
Fremont R, Calderon DP, Maleki S, Khodakhah K. Abnormal high-frequency burst firing of cerebellar neurons in rapid-onset dystonia-parkinsonism. J Neurosci Aug. 2014;27(35):11723–32.
Fremont R, Tewari A, Angueyra C, Khodakhah K. A role for cerebellum in the hereditary dystonia DYT1. Elife. Feb 15 2017;6.
Isaksen TJ, Kros L, Vedovato N, Holm TH, Vitenzon A, Gadsby DC, et al. Hypothermia-induced dystonia and abnormal cerebellar activity in a mouse model with a single disease-mutation in the sodium-potassium pump. PLoS Genet May. 2017;13(5):e1006763.
White JJ, Sillitoe RV. Genetic silencing of olivocerebellar synapses causes dystonia-like behaviour in mice. Nat Commun Apr. 2017;4:8:14912.
Huang XJ, Wang T, Wang JL, Liu XL, Che XQ, Li J, et al. Paroxysmal kinesigenic dyskinesia: Clinical and genetic analyses of 110 patients. Neurol Nov. 2015;3(18):1546–53.
Liu X, Ke H, Qian X, Wang S, Zhan F, Li Z, et al. Clinical and genetic analyses of 150 patients with paroxysmal kinesigenic dyskinesia. J Neurol Sep. 2022;269(9):4717–28.
Tong X, Ao Y, Faas GC, Nwaobi SE, Xu J, Haustein MD, et al. Astrocyte Kir4.1 ion channel deficits contribute to neuronal dysfunction in Huntington's disease model mice. Nat Neurosci May. 2014;17(5):694–703.
Buono RJ, Lohoff FW, Sander T, Sperling MR, O'Connor MJ, Dlugos DJ, et al. Association between variation in the human KCNJ10 potassium ion channel gene and seizure susceptibility. Epilepsy Res Feb. 2004;58(2–3):175–83.
Yang T, Gurrola JG 2nd, Wu H, Chiu SM, Wangemann P, Snyder PM, et al. Mutations of KCNJ10 together with mutations of SLC26A4 cause digenic nonsyndromic hearing loss associated with enlarged vestibular aqueduct syndrome. Am J Hum Genet May. 2009;84(5):651–7.

SupplementaryTable.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

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

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Participants, data collection, and patient consent

Whole-exome sequencing and bioinformatics

Sanger sequencing of the candidate gene

Construction of plasmids

Cell culture transfection

Immunofluorescence

Electrophysiological recordings

Mouse model

Real-time quantitative PCR (RT‒qPCR)

Immunohistochemistry

Adeno-associated virus (AAV) administration

Epidural KCl stimulation

Rotarod test

Statistical analysis

Results

Heterozygous variants in the KCNJ10 gene are associated with PKD

Clinical and genetic findings of PKD patients with heterozygous KCNJ10 variants

Heterozygous KCNJ10 mutations lead to reduced Kir4.1 currents

The expression pattern of Kir4.1 in mice

Discussion

Conclusion

Abbreviations

Declarations

References

Supplementary Files

Status:

Version 1