Paroxysmal Kinesigenic Dyskinesia Caused by 16p11.2 Microdeletion and Related Clinical Features.

BACKGROUND AND OBJECTIVES: Isolated paroxysmal kinesigenic dyskinesia (PKD) is mainly caused by PRRT2 variants and TMEM151A variants. Patients with proximal 16p11.2 microdeletion (16p11.2MD) (including PRRT2) often have neurodevelopmental phenotypes, whereas a few patients have PKD. Here, we aimed to identify 16p11.2MD in patients with PKD and describe the related phenotypes.
METHODS: Whole-exome sequencing and bioinformatics analysis of copy number variant (CNV) were performed in patients with PKD carrying neither PRRT2 nor TMEM151A variant. Quantitative PCR and low-coverage whole-genome sequencing verified the CNV.
RESULTS: We identified 9 sporadic patients with PKD and 16p11.2MD (∼535 kb), accounting for 9.6% (9/94) of our patients. Together with 9 previously reported patients with PKD and 16p11.2MD, we found that 16p11.2MD was de novo in 11 of 12 tested patients and inherited from a parent in the other patient. And 80% (12/15) of these patients had a mild language delay, 64.3% (9/14) had compromised learning ability, 42.9% (6/14) had a mild motor delay, and 50% (6/12) had abnormal neuroimaging findings. No severe autism disorders were observed. DISCUSSION: Mild developmental problems may be overlooked. A detailed inquiry of developmental history and CNV testing are necessary to distinguish patients with 16p11.2MD from isolated PKD.

Paroxysmal kinesigenic dyskinesia (PKD; MIM #128200) is a rare disorder manifesting as recurrent and brief episodes of choreoathetosis or dystonia triggered by sudden voluntary movements. Variants in the gene PRRT2 located on chromosome region 16p11.2 were identified to cause PKD in 2011.[1] To date, PRRT2 variants account for 77%–93% of familial PKD and 21%–45% of sporadic PKD.[2] TMEM151A variants were recently identified to cause PKD.[3] The proximal ∼600 kb breakpoints 4 and 5 16p11.2 microdeletion (16p11.2MD) (29.6–30.2 Mb–hg19, including PRRT2) could be a cause of PKD accompanied by other clinical features.[4] The 16p11.2MD is one of the most common etiologies of neurodevelopmental disorders, including motor/language developmental delay, autism spectrum disorder (ASD), intellectual disability (ID), congenital dysmorphism, and/or obesity, with an incidence of 0.25%–2.9%.[5] To date, more than 400 16p11.2MD carriers have been reported, among which 9 cases have been reported to have PKD.[5,6] Here, we report 9 new patients with PKD and 16p11.2MD.

Methods

Standard Protocol Approvals, Registrations, and Patient Consents

This study was approved by the ethics committee of the Second Affiliated Hospital of Zhejiang University School of Medicine, and all participants provided informed consent.

Clinical and Genetic Analysis

Patients were enrolled between January 2015 and July 2021. Whole-exome sequencing and bioinformatics analysis of copy number variants (CNVs) were performed in patients who were negative for both PRRT2 and TMEM151A variants. For details, please see eMethods.

Data Availability

The original data that support the findings are available from the corresponding author on reasonable request.

Results

Among 94 enrolled patients, 9 sporadic patients (9.6%) were identified with 16p11.2MD (eTable 1, links.lww.com/NXG/A513). Quantitative PCR confirmed deletions in PRRT2, which were de novo in 4 patients (P1, P5, P6, and P9), inherited from a father in P8, and could not be tested in parents of the other 4 patients (Figure 1A). Paternity was confirmed by microsatellite analysis (eFigure 1). Low-coverage whole-genome sequencing further confirmed 16p11.2MD in all 9 patients and the father of P8, with a span of ∼535 kb (chr.16:29649997-30185157, GRCh 37) containing 30 genes from SPN to MAPK3 (Figure 1B). The father with 16p11.2MD had obesity (body mass index 33.41 kg/m2) with a head circumference of 61 cm, but denied PKD and developmental disorders, and neurologic examinations were normal. Variants in other paroxysmal dyskinesia-related genes were not detected in these 9 patients.

Figure 1

Genetic Findings and Brain Images

(A) The copy number of PRRT2 detected by qPCR in 9 patients with PKD and available parents. C1 and C2 are controls, P1–P9 represent patients, 1 F/M is the father/mother of patient 1, and so on. The relative copy number of PRRT2 was calculated with CopyCaller software v2.1, and subject C1 was selected as the calibrator sample with a copy number of 2. (B) Black arrows indicate the 16p11.2 microdeletions detected with low-coverage whole-genome sequencing. The associated genes were generated with the NCBI Genome Browser (GRch37.p13) and displayed under the graphic structure of chromosome 16. (C) MRI of patient 3 showing a lesion at the left edge of the cerebellum (white arrows, left: T2 weighted, right: T1 FLAIR). (D) MRI of patient 7 showing an arachnoid cyst at the cerebellar vermis (white arrows, left: T2 weighted, right: T1 FLAIR). FLAIR = fluid-attenuated inversion recovery; PKD = paroxysmal kinesigenic dyskinesia.

The clinical data of the 9 newly identified patients (P1–P9) with PKD and 16p11.2MD and the previously published 9 cases (P10–P18) were summarized in Table 1 (average onset age of PKD, 9.6 years; males/females/unknown, 14/3/1). Features of kinesigenic attacks were typical in all patients except that durations extended to several minutes in 3 patients (P1, P6, and P13).

Table 1

Clinical and Genetic Features of Patients With PKD And 16p11.2 Microdeletion

Regarding the 16p11.2MD-related phenotype, 80% (12/15) of these patients had mild speech delay, and 4 patients developed a lisp. Learning abilities were compromised in 64.3% (9/14) of patients, among whom 2 received additional educational support, and 7 had a poor academic performance. In addition, mild motor delays were present in 42.9% (6/14) of patients. The average age of walking was 16.50 ± 5.95 months in P1–P9. Uncommonly, ASD, subtle dysmorphism, and obesity were each found in 1 patient. Notably, brain MRI revealed anomalies in 50% (6/12) of patients. The cerebellum was frequently affected, including suspected dermoid cysts (P3, Figure 1C), arachnoid cysts (P7, Figure 1D, P11), and cerebellar atrophy (P10). Neuroimaging findings were usually normal in isolated PKD.

Discussion

The developmental delay in patients with PKD and 16p11.2MD is similar but milder than that in other 16p11.2MD carriers. In other deletion carriers, language delay was present at high rates (83%), intellectual ability varied with ∼30% falling in the normal range and 10.3%–28.1% having ID, and motor delay was common (37.6%–57.1%) with an evaluated mean age of walking of 20.5 ± 8.6 months.[5,7] Brain MRI abnormalities in 16p11.2MD patients mostly are related to posterior fossa and/or craniocervical junction,[7] consisted with our findings.

To our knowledge, neurodevelopmental disorders have been rarely reported in patients with PKD carrying heterozygous PRRT2 frameshift/missense variants. A few patients with biallelic PRRT2 variants were reported to have cognitive problems.[4] Neurodevelopmental phenotypes in 16p11.2MD are related to other genes in this region. For example, MAPK3 is a key component of MAPK/ERK pathway associated with ASD-related features in mice models.[5] Deficiency of the KCTD13 ortholog in zebrafish resulted in macrocephaly.[5] And both Kctd13-deficient and Taok2-deficient mice displayed cognitive problems.[5]

The 16p11.2MD is characterized by a broad spectrum of neurodevelopmental phenotypes with varying penetrance.[5] Here, 16p11.2MD accounted for 9.6% of patients with PKD carrying neither PRRT2 nor TMEM151A variant. Developmental features of 16p11.2MD may be mild; therefore, a detailed inquiry of developmental history and CNV testing are necessary to distinguish these patients from isolated PKD. The WES-based CNV testing is cost-effective while has limitation in detecting small CNVs (<100 kb). High-resolution chromosomal microarray could be a precise tool.