Abstract
Objective: To use genetic linkage analysis to localize a gene for paroxysmal kinesigenic dyskinesia (PKD) in a three generation African-American kindred.
Background: PKD is a rare autosomal dominant disorder characterized by episodic choreiform or dystonic movements that are brought on or exacerbated by voluntary movement. There are individuals with the clinical features of PKD but with no family history of the disease, but whether these sporadic cases represent spontaneous mutations of PKD or have a distinct condition is unknown.
Methods: A genome-wide linkage scan of polymorphic microsatellites at 25 cM resolution was performed to localize a gene for PKD in one African-American kindred. Pairwise multipoint linkage analyses were performed at different penetrance estimates.
Results: Evidence for linkage of the kinesigenic form of paroxysmal dyskinesia to chromosome 16 was obtained. A maximum lod score of 4.40 at θ = 0 was obtained with D16S419. Critical recombinants place the PKD gene between D16S3100 and D16S771.
Conclusions: A paroxysmal kinesigenic dyskinesia (PKD) locus lies within an 18 cM interval on 16p11.2-q11.2, between D16S3100 and D16S771. A gene for infantile convulsions with paroxysmal choreoathetosis has also been mapped to this region. These two regions overlap by approximately 6 cM. These two diseases could be caused by different mutations in the same gene or two distinct genes may lie within this region.
Patients with paroxysmal kinesigenic dyskinesia (PKD) develop sudden attacks of dystonia or chorea in response to voluntary movements. The abnormal movements characteristically begin between 6 and 16 years of age, and about half of the affected individuals have a family history of the disorder.1 The frequency of the paroxysms tends to increase by adulthood in many patients. An autosomal dominant pattern of inheritance has been described.2-6 Neither the biochemical defect responsible for the disease nor the genetic location of susceptibility alleles, the penetrance of disease alleles, or the extent of genetic heterogeneity is known.
PKD can be distinguished from other forms of par-oxysmal movement disorders by its induction by voluntary movements, the consistent improvement on various anticonvulsant medications, its often familial pattern, and the lack of other disorders known to cause dystonia or chorea. However, although PKD is considered to be a distinct clinical entity, the terminology used to describe it and related disorders has been inconsistent. The term paroxysmal kinesigenic choreoathetosis was used by Kertesz et al. to underscore the frequency of chorea as the disorder’s dominant movement, and various other terms have been suggested.1,7 However, chorea, athetosis, and dystonia all seem to occur in patients and families with otherwise typical clinical manifestations of PKD, and Dimirkiran et al. recently designated the disorder as PKD.8
Several attempts had been made to classify familial paroxysmal dyskinesia. Demirkiran classified paroxysmal dyskinesia into four broad categories: 1) kinesigenic (PKD), which is defined by the onset of attacks induced by sudden voluntary movement; 2) nonkinesigenic (PNKD), in which attacks occur spontaneously; 3) exertion-induced (PED), where attacks are induced by prolonged physical activity; and 4) hypnogenic (PHD), which is defined as sudden involuntary movements that occur during sleep.8 These categories were further subdivided according to the duration of the attacks. The most frequently described familial form of the disease is kinesigenic.
Most cases of PKD have been classified as familial, although a smaller number have been described as acquired or symptomatic and have been associated with other neurologic diseases such as MS.9,10
Two independent studies described mapping of a locus for autosomal dominant forms of nonkinesigenic familial paroxysmal dyskinesia (FPD1) to similar regions on chromosome 2q31-36. 11,12 A gene for autosomal dominant paroxysmal choreoathetosis/spasticity (CSE) has been mapped to a 2 cM region on chromosome 1p, close to a cluster of potassium channel genes.13 Another study described mapping of sporadic focal dystonia to chromosome 18p.14
To understand the molecular basis of PKD, the defective gene needs to be isolated. We identified a large African-American kindred with the disease and performed a genome-wide scan with highly polymorphic microsatellites at 25 cM resolution. Parametric and nonparametric linkage analyses localized the gene in this family to chromosome 16p11.2-q11.2. The nonkinesigenic form of the disease has been linked to chromosome 2q,11,12 reinforcing the conjecture that the two diseases are distinct entities. The extent of genetic heterogeneity of these diseases needs to be defined by an analysis of additional families. Infantile convulsions with paroxysmal dyskinesia (ICCA syndrome) has also been mapped to this region on chromosome 16 in four families from northwestern France15 and in one family of Chinese origin,16 suggesting that different alterations in a single gene may be responsible for both syndromes or that there are a cluster of genes for epilepsy within this region.
Patients and methods.
Clinical evaluation.
This study was approved by the Institutional Review Board for Human Research of the University of Texas Southwestern Medical Center, and each subject or a parent gave informed consent.
The proband was an 11-year-old boy who was referred for evaluation of possible epilepsy. His attacks consisted of 15- to 20-second spasmlike movements, which were typically induced by voluntary movement such as walking or running. These episodes began about a year before evaluation, and the family sought medical attention because the movements became more frequent and painful. His episodes usually started in the right leg with what the patient described as “quivering” and then spread to the right arm and sometimes the left arm as well.
His neurologic examination between attacks was normal. One of the investigators (E.S.R.) witnessed a typical attack; it consisted of sudden-onset dystonia of the right arm and leg brought on by efforts to stand and run across the room. There was no alteration of consciousness, and the episode lasted for approximately 30 seconds. Another episode occurred during an EEG, and the test remained normal.
A diagnosis of PKD was made, and his dystonic episodes promptly disappeared after starting carbamazepine 100 mg each evening (2.6 mg/kg/day). Although this carbamazepine dose eliminated the visible dystonic episodes, he still complained of subjective tightness of his extremities in response to movement. These feelings resolved when the carbamazepine dose was increased to 200 mg per night. The dystonia recurred several years later when medication was briefly withdrawn but resolved once more on carbamazepine. The patient is now 18 years of age, and his dystonia remains controlled on carbamazepine.
Numerous family members also have PKD (figure 1), although evidently none was diagnosed with PKD until after the proband’s diagnosis. A sister, uncle, and a second cousin of the proband were subsequently seen as patients, and all but one (Patient 9) of the other affected family members as well as many of the unaffected subjects were seen by one of the authors (E.S.R.) at a family reunion. There each person’s history was reviewed, an abbreviated neurologic examination was performed, and blood specimens were collected from patients and cooperative first-degree relatives.
Figure 1. Pedigree of an African-American kindred with autosomal dominant paroxysmal dyskinesia. Circles represent female subjects; squares represent male subjects. Black circles or squares represent affected individuals with PKD; unaffected individuals are not shaded. Chromosome 16 loci are indicated. Reconstructed haplotypes using these loci are shown below each individual. Black bars correspond to the proposed affected haplotype. < indicates a meiotic recombination event.
Ascertainment of disease status was difficult in a few instances. Both the severity and age at onset varied significantly between affected family members. Symptoms were not usually recognized before 5 years of age, and the PKD in this family tends to become less severe or even resolve by adulthood. An episode of dystonia was not actually witnessed by an investigator except for the proband’s. Thus, we had to rely on the sometimes inconsistent recollections of the individuals or family members to confirm a diagnosis of a disorder that had sometimes resolved by the time of questioning. In light of these difficulties, it is not surprising that we could not always pinpoint the age at onset for each person or be certain of the exact nature of the abnormal movements (i.e., whether an affected person had choreoathetosis, dystonia, or both). Nevertheless, the majority of those affected started to have the abnormal movements after 6 years of age and before adulthood, and based on their description of the movements, dystonia seemed to be more common than choreoathetosis. Most reported abnormal movements of the extremities rather than the trunk or face. Individual patients might have either unilateral or bilateral movements; some individuals’ movements occurred consistently on one side or the other, whereas in others the predominant side seemed to fluctuate.
One woman had only mild dystonic episodes until adulthood, when her attacks became more frequent and severe during her pregnancy, and after delivery she required carbamazepine for her dystonia. Young children without symptoms are shown in the pedigree as unaffected, although some of them may yet become symptomatic. Most of the adults were listed on the basis of a history of the typical episodes by that individual and confirmed by a relative, although some of them are no longer symptomatic.
DNA analysis.
DNA was isolated from whole blood using standard protocols. 17 Semi-automated genotyping was performed with the ABI377 using fluorescently tagged primers (TET, HEX, or FAM) obtained from Research Genetics Weber panel version 8 (Huntsville, AL). PCR reactions were performed in volumes of 10 μL containing 40 to 50 ng of genomic DNA, 0.15 mM of MgCl, PCR buffer (Perkin Elmer, Foster City, CA), dNTPs (200 μM each), primers (1.25 pM each), 0.5 U of Taq polymerase, and 1.67 mM spermidine. Reactions were performed in a Perkin-Elmer 9600 thermal cycler with the following parameters: 5-minute initial denaturation at 95 °C, then 28 cycles (15 seconds at 95 °, 15 seconds at 55 °, 30 seconds at 72°C), with a final extension of 10 minutes at 72 °C. After amplifications, PCR reactions for a single individual were pooled if alleles were of sufficiently different length to be resolved by electrophoresis (approximately 20 base pairs apart). Each pool was thoroughly mixed, TAMRA-500 standard and loading dye added, and then denatured (3 minutes 95 °C). Pooled alleles were separated by electrophoresis in a 36-cm 5% denaturing (FM urea) Sequagel acrylamide gel (National Diagnostics, Atlanta, GA) at 3 kV for 1.5 to 3 hours with an ABI PRISM 377 DNA Sequencer (Applied Biosystems, Foster City, CA). After electrophoresis, analysis was performed with the GeneScan (Ver 2.1) software (Applied Biosystems). After gel runs, alleles were assigned with Genotyper software (Applied Biosystems) and downloaded to linkage analysis programs.
Linkage analysis.
Pedigree data was entered with CYRILLIC (Cherwell Scientific) into a format suitable for linkage analysis. Linkage analysis was performed assuming an autosomal dominant mode of inheritance with penetrance estimates of 0.80 and 0.90 using the FASTLINK version of LINKAGE.18,19 Analyses were performed using a disease allele frequency of 0.01%.
Results.
Linkage analysis.
A genome-wide linkage scan was performed at an average resolution of 25 cM using a subset of the Weber v8 panel of polymorphic microsatellites (Research Genetics). The initial analyses were performed with individuals numbered 1 through 45, excluding Individual 9 (see figure 1). With 80% penetrance, the highest two-point lod score obtained as a result of this scan was 3.67 (0.01% disease allele frequency) at D16S753 with a corresponding θ value of 0.001. We then performed fine-structure mapping with additional markers. Strongest evidence for linkage was obtained with two loci on chromosome 16: D16S753 and D16S419. A maximum two-point lod score of 3.76 at θ = 0.001 was obtained with D16S419.
We later acquired DNA from individuals numbered 9, plus 46 through 54. Two point lod scores with corresponding θ values are shown in the table. A maximum two-point lod score of 4.40 at θ = 0.001 was obtained with D16S419.
Two-point lod scores for chromosome 16 loci and paroxysmal kinesigenic dyskinesia
Haplotype analysis showed critical meiotic-recombination events in Individual 32 between D16S3100 and D16S261; and between D16S419 and D16S771 placing the disease gene in this family in a 18 cM region between D16S3100 and D16S771 (figures 1 and 2⇓). It is unfortunate that the diagnosis of Individual 53 is uncertain because if he were affected, the recombination event between D16S753 and D16S3396 would further narrow the PKD region.
Figure 2. Idiogram of chromosome 16 indicating the relative position of loci within the paroxysmal kinesigenic dyskinesia (PKD) and infantile convulsions with paroxysmal choreoathetosis (ICCA) intervals. The genetic location of loci flanking these intervals is shown.
Discussion.
The current study localizes a gene for the kinesigenic form of paroxysmal dyskinesia to human chromosome 16 in a large kindred. Szepetowski et al.15 reported linkage of familial infantile convulsions with paroxysmal choreoathetosis to the pericentromeric region of chromosome 16 (16p12-q21) in four French families. They obtained a maximum two-point lod score for D16S3133 of 6.76 at a recombination fraction of 0 and a maximum multipoint lod score for the same locus of 7.06. Meiotic recombination events place the disease gene in these families between D16S401 and D16S517, which overlaps our PKD critical region by ∼6 cM between D16S769 and D16S517 (see figure 2).
A gene for ICCA syndrome has also been mapped to 16p12-q12 in one Chinese family.16 Clinical heterogeneity was observed between the French and Chinese families, and certain unaffected individuals in both French and Chinese families carried the disease haplotype. Patients in these families have seizures as infants, and paroxysmal movements are not movement induced, suggesting that a different gene or a different allele of the same gene may be responsible for PKD.
The clinical manifestations of PKD vary even among affected members of the same family, but overall our family has similar clinical features to the individuals described in the literature. As in our family, the abnormal movements of PKD typically begin between 6 and 16 years of age, although both earlier and later age at onset have been described. Goodenough et al.1 noted that 46 of 64 (72%) patients in their series started to have abnormal movements between 6 and 16 years of age. Similarly, Tan et al.20 recorded the onset of symptoms between 6 and 16 years in 13 of 15 patients (87%); symptoms started after 16 years in the 2 remaining patients in this series. As in our patients, the extremities are more often involved than the trunk, neck, or face. 1,20 Although the early descriptions emphasize choreoathetosis as the dominant movement of PKD, more recent series indicate that dystonia or a mixed pattern with both choreoathetosis and dystonia may be more common.8,20 Although we were unable to ascertain the specific nature of the dyskinesia in some of our patients, most of them described dystonia.
In most instances, PKD is clinically distinct from epilepsy despite its typical responsiveness to antiepileptic drugs such as carbamazepine and phenytoin.20,21 Our proband had an EEG performed during a PKD attack that was normal. The patients treated with carbamazepine in our family, like others described in the literature, seemed to respond to much lower medication doses than typically required for epileptic seizures.21 Tan et al.20 identified 5 patients with epilepsy among 64 PKD patients pulled from their clinic and from the literature, so there could be some as yet undefined overlap between epilepsy and PKD.20
Other studies11,12 localized a gene for autosomal dominant nonkinesigenic paroxysmal dyskinesia to chromosome 2q31-36. One family was of Italian descent, the other of Polish descent. This disease is distinct from the kinesigenic form in that patients’ dystonia occurred at rest and alcohol or caffeine provoked attacks. The fact that the kinesigenic and nonkinesigenic forms of this disease map to different locations reinforces their classification as different disease entities. As in our 16-linked family, penetrance in both families was high but not 100%. In the family of Italian descent, one member (8 years of age) harbored the disease haplotype but had not exhibited symptoms of the disease. Similarly, three of our family members described as unaffected (Patients 4, 19, and 47) have the affected haplotype on chromosome 16. This could be because the disease may go undiagnosed if the symptoms are mild or brief. Alternatively, environmental influences or modifier genes could be contributing factors in the clinical manifestation of the disease.
Mutations in ion-channel genes are probably responsible for several paroxysmal cerebral disorders.22-24 A locus for a form of paroxysmal choreoathetosis with spasticity has been mapped to chromosome 1p13 near a cluster of potassium-channel genes. The phenotype for this disease is clearly different to PKD: Some of the patients had nonvariable spasticity; most of them were mentally impaired and had slowing on EEG; attacks lasted longer; and patients had episodic ataxia and responded to acetazolamide. Other loci for autosomal episodic ataxias exist on chromosome 19p1325-27 and on 12p13, where a mutation in a potassium-channel gene (KCNA1) causes episodic ataxia/myokymia.28,29 Mutations in other ion-channel genes have been associated with hyperkalemic periodic paralysis, paramyotonia congenita and acetazolamide-responsive myotonia congenita (SCNA4),30-32 and hypokalemic periodic paralysis (CACNA1S).33,34 Mutations in a novel ion-channel gene on chromosome 16 may be responsible for PKD and/or ICCA. The γ-subunit of a sodium channel35 is located in the region of interest on chromosome 16. Also in this region are a sodium/glucose co-transporter (SLC5A2),36 a cerebellin 1 precurser,37 a neurotransmitter transporter,38 and the Batten/Spielmeyer-Vogt disease gene (CLN3).39
Refinement of the PKD interval with additional linked families is important. Additional families with PKD will also provide information on the extent of genetic heterogeneity. Once the region is refined, mutational analysis of candidate genes mapping to the interval will be required to isolate a gene responsible for PKD.
Acknowledgments
Supported in part by a grant from the Crystal Charity Ball Center for Pediatric Neuroscience Research of the University of Texas Southwestern Medical Center.
Acknowledgment
The authors thank Ross Wilson for technical help and Robert Barnes for help with the statistical analysis.
- Received July 14, 1999.
- Accepted October 29, 1999.
References
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.