RESULTS
Nine movement disorder conditions were included: Segawa disease, PARK-
Parkin, X-linked dystonia parkinsonism (XDP), dentatorubral-pallidoluysian atrophy (DRPLA), Woodhouse-Sakati syndrome (WSS), benign adult familial myoclonic epilepsy (BAFME), Kufor-Rakeb disease, tremulous dystonia associated with mutation of the calmodulin-binding transcription activator 2 (
CAMTA2) gene, and paroxysmal kinesigenic dyskinesia (PKD) (
Table 1).
- Segawa disease; DYT/PARK-GCH1; DYT5a
In 1970, a girl suffering from a walking difficulty that worsened toward the evening visited Dr. Segawa. A female cousin had the same symptoms. Dr. Segawa decided to try levodopa treatment—a drug that had been recently discovered at that time. In 1971, these two girls were reported in a Japanese journal as suffering from “hereditary basal ganglia disease with marked diurnal fluctuation” [
9]. When seeing a 51-year-old patient who was the grandmother of one proband with a clinical course of 43 years, Dr. Segawa noticed that their symptom was dystonia, not parkinsonism. Therefore, in 1975, Dr. Segawa reported a revised nomenclature of “hereditary progressive dystonia with marked diurnal fluctuation (HPD)” at the First International Dystonia Congress in New York [
10]. However, the disorder became known as “Segawa disease”.
Dr. Segawa classified the disease into two types: the postural and action types [
11]. Patients with the postural type show postural dystonia as the primary symptom throughout the disease course. The onset is at approximately 6 years of age, and the initial symptom is mostly unilateral pes equinovarus. With age, dystonia expands to other limbs and trunk muscles. Patients with the action type develop dystonic movement of one extremity or the neck (action retrocollis) in addition to a dystonic posture at approximately 8 to 10 years of age.
The marked and sustained response to levodopa without side effects suggested that this disease is a functional disorder of nigrostriatal (NS)-dopamine (DA) neurons [
10,
12-
16]. McGeer and McGeer have shown that tyrosine hydroxylase (TH) activity in the caudate nucleus decreases in an age-dependent manner [
17], giving a hint to the pathogenesis of the disorder [
12-
16]. This suggests that TH is decreased at the terminal of NS-DA neurons and follows the normal decrement [
12-
16]. The diurnal fluctuation of symptoms may be explained by the circadian oscillation of NS-DA neurons at the terminal (striatum) [
17] but not at the perikaryon (substantia nigra) [
18]. Sleep studies have confirmed a decrease in DA activity in the striatum [
10,
12-
16,
19-
21].
Dr. Segawa proposed [
19-
22] that the involvement of the direct striatal pathway (via D1 DA receptors) and the subsequent descending pathway from the basal ganglia to the midbrain and brain stem results in dystonia. In contrast, the action type results from the hypofunction of the nigro-subthalamic DA neurons [
19-
22] that innervate the subthalamic nucleus (STN) via D1 DA receptors in the STN [
23]. The net disinhibition of the thalamo-cortical pathway results in dystonia [
19-
22]. Neuropathological study findings have confirmed the proposed pathophysiology; TH activity is low in the striatum but not in the substantia nigra [
22]. Neuropathological and neurohistochemical studies have revealed TH activity and immunoreactive protein reduction in the striatum but not in the substantia nigra [
22,
24], and the subregional dorsoventral gradient of the DA decrement shows a marked decrease in the ventral area in contrast to the dorsal predominance in PD [
25]. This finding suggests that NS-DA neurons connecting to the D1-direct pathways are predominantly affected in Segawa disease [
14-
16,
19-
22].
In 1990, Fujita and Shintaku reported a decrease in biopterin and neopterin in cerebrospinal fluid and suggested a deficiency of guanosine-triphosphate-cyclohydrolase 1 (GCH1) as the cause of Segawa disease [
26]. In addition, Furukawa et al. [
27] analyzed the CSF of patients with Segawa disease, PD, and other disorders and showed a decrease in biopterin and neopterin only in patients with Segawa disease. Finally, in 1994, Dr. Ichinose discovered the
GTP cyclohydrolase1 gene to be the causative gene of Segawa disease [
28]. Since then, a few other gene mutations have been identified that cause similar symptoms that are responsive to low-dose levodopa, referred to as dopa-responsive dystonias [
29]. Parkinson’s disease symptoms can also be seen in some
GTP cyclohydrolase1 gene mutation carriers, suggesting that GCH1 deficiency may predispose patients to NS cell loss [
30].
For treatment, levodopa shows marked and sustained effects without any side effects (
Supplementary Video 1 in the online-only Data Supplement) [
12-
16], but the effect of levodopa may fluctuate in some patients with the action type [
11]. Anticholinergics were used before the discovery of levodopa [
31], which showed beneficial effects on dystonia but not on tremors. Tetrahydrobiopterin (BH4) may be used for patients who are compound heterozygotes with early-onset cases [
32].
- PARK-Parkin; PARK2
Nasu et al. [
33] first reported, based on only clinical features, the cases of four patients with familial juvenile Parkinson’s disease in 1958, with two subsequent reports quickly following this [
34,
35]. Then, in 1973, based on these Japanese case reports, Yamamura et al. [
36] summarized the clinical features of the disorder as one genetic entity when he described the cases of a familial group of patients with early-onset parkinsonism marked by sleep benefit that subsided as the disease progressed. He coined the disorder “paralysis agitans of early onset with marked diurnal fluctuation of symptoms” [
36]. Clinical progression was slow, and all patients responded to trihexyphenidyl treatment. One patient treated with levodopa showed a dramatic improvement in her symptoms.
The discovery of the
parkin gene as the responsible genetic link followed the report of a family with young-onset parkinsonism, thought to be associated with a polymorphism of the
manganese superoxide gene (
MnSOD) [
37]. The gene locus of
manganese superoxide dismutase has been mapped to the long arm of chromosome 6. Shimoda-Matsubayashi and her colleagues were unable to identify a pathogenic mutation [
37]. However, this information provided important clues for the identification of a new causative gene for young-onset PD. In 1998, Kitada et al. [
38] identified a novel causative gene,
parkin, using the Keio BAC (bacterial artificial chromosome) library. The
parkin cDNA consists of a ubiquitin-like domain and two Really Interesting New Gene (RING) finger motifs. As neuropathologic examination of patients with
parkin mutations showed a lack of Lewy bodies, the group hypothesized that the function of this gene might be involved in proteolytic systems, such as the ubiquitin‒proteasome pathway. Subsequently,
parkin was shown to encode the E3 ubiquitin ligase, which is directly linked to the ubiquitin‒proteasome system [
39]. Parkin protein is involved in mitophagy, a type of autophagy, in collaboration with PTEN-induced putative kinase 1 (PINK1) [
40]. This information suggests that both molecules share a common pathway, mitophagy, that is responsible for the removal of abnormal mitochondria.
Parkin-mutation-induced astrocytic alteration, especially reduced astrocytosis, suggests the possibility of an astrocyte-related nonautonomous cell death mechanism for dopaminergic neurons in the brains of patients with
parkin mutations [
41]. Recent evidence has shown that PINK1-Parkin functions as the core machinery for mitophagy in not only neurons but also astrocytes, and deficiency in this pathway causes young-onset PD [
42].
A prominent clinical feature of
parkin-linked PD, especially in very early-onset patients, is dystonia, with a prevalence of up to 85% in those with disease onset of 20 years or younger. Patients are especially susceptible to dopa-induced dyskinesia and motor function fluctuation (
Supplementary Video 2 in the online-only Data Supplement) [
38]. More widespread genetic testing of the
parkin gene has widened the clinical spectrum. For example, Klein et al. [
43] reported the presence of
parkin deletions in a family with adult-onset, tremor-dominant parkinsonism. More recently, a “benign tremulous parkinsonism” phenotype was proposed [
44]. The patients had juvenile-onset tremors for many years before the appearance of parkinsonism and had a dramatic and sustained response to anticholinergic monotherapy [
44]. Initially,
parkin-associated PD was thought to be peculiar to Japan; however, it is now known to be found throughout the world [
45].
Regarding treatment, most patients with
parkin mutations respond favorably to levodopa treatment even at a low dose, albeit with a propensity to develop motor complications. In addition, it appears that deep brain stimulation (DBS) is a good option for
parkin mutation carriers [
46,
47]. Recently, improvements in symptoms with apomorphine infusion has been reported [
48].
- XDP; Lubag disease; DYT-TAF1; DYT3
XDP came to medical attention in 1969 when Dr. George Viterbo described the cases of five males labeled as having “dystonia musculorum deformans” from the Island of Panay [
49,
50]. This finding resulted in an epidemiological survey of patients on the island [
51]. In 1975, 28 cases of “torsion dystonia of Panay” were presented at the Second Dystonia Meeting in New York City and, in 1976, were published in a landmark paper [
49]. All patients were male. There was no male-to-male transmission, so sex-linked transmission was inferred, which was previously undescribed at that time [
49]. Since then, an XDP Study Group and Registry has been formed in the Philippines, and an alliance with the local government and Health Department has been forged, with a system put in place to identify, refer and study suspected cases. After 25 years of follow-up, the natural history of the disease has become much clearer [
50,
51].
The disease usually progresses in 3 phases. The disease starts with the dystonic phase, where patients typically present with focal dystonia that can involve any part of the body [
50,
51]. Focal dystonia generalizes in 84% of patients within 5 years [
50]. Patients will usually have severe, nearly constant axial dystonia, retrocollis and prominent oromandibular dystonia, such as jaw opening and tongue protrusion (
Supplementary Video 3 in the online-only Data Supplement). These features prevent the patients from assuming any comfortable position, but they can still move about in a bizarre fashion. This dystonic phase lasts between 2 and 7 years; however, as early as the 2nd year, parkinsonian features may start to appear. At the 7th to 10th year of illness, dystonia reduces, and parkinsonism becomes more prominent. This is referred to as the combined dystonia-parkinsonian phase. By the 15th year of illness, called the parkinsonian phase, the predominant picture is Parkinson-like. Not all patients reach this stage, as most patients die within 10 years. Less than 6% present with parkinsonism initially, who usually also have a later age at onset and a more benign course (
Supplementary Video 4 in the online-only Data Supplement) [
52,
53]. In addition, non-motor features can significantly contribute to a patient’s disability. Cognition (including executive, attentional or visuospatial domains), mood, and sensory (pain) or autonomic (increased sweating) functions may be affected [
52,
54]. Most female XDP carriers are asymptomatic. Symptomatic females may have focal dystonia, chorea, focal tremors and parkinsonism, which are generally less severe. Detailed genealogical studies have traced almost all cases to mothers with roots on Panay Island.
Magnetic resonance imaging (MRI) findings of markedly hyperintense putaminal rims are a constant feature in all phases of the disease. Caudate head and putaminal atrophy are most prominent in the combined and predominantly parkinsonian stages [
55]. White matter and pallidal abnormalities are also described (
Figure 1) [
56].
X-linked inheritance was confirmed by subsequent linkage disequilibrium analysis narrowing the gene location to a < 350 kb region in Xq13.1 [
57,
58]. Nolte and his colleagues [
59] described disease-specific changes (DSCs) and a potential multiple transcript system (MTS) in the distal portion of the
Taf1 gene, and Makino et al. [
60] reported SVA (short interspersed nuclear element, variable number of tandem repeats, and Alu composite) retrotransposon insertion in intron 32 of the
Taf1 gene.
The SVA element in XDP includes a hexameric DNA repeat expansion, (CCCTCT)n, the length of which was found to be inversely correlated with the age of disease onset [
61,
62]. Moreover, the longer the hexameric repeats, the more prominent the dystonic component, whereas shorter repeats were more frequently associated with parkinsonian features [
61,
62].
In general, the response of XDP symptoms to oral medical therapy has been poor and inconsistent [
63]. However, patients have shown improvements with regular botulinum toxin injections and bilateral pallidal DBS, with dystonia responding to DBS better than parkinsonism (
Supplementary Video 5 in the online-only Data Supplement) [
64-
69].
- DRPLA
The name DRPLA was first proposed by Smith et al. [
70] in 1958. He reported the cases of patients with an unusual form of cerebellar ataxia associated with choreoathetosis whose pathological study showed the atrophy of the dentatorubral and pallidoluysian systems. The name was based on only the pathologically affected sites of lesions. Thereafter, several case reports of DRPLA were published in Japan based on only pathological findings. In 1972, 14 years later, Naito et al. [
71] reported a dominantly inherited form of myoclonic epilepsy in Japanese individuals. It was at this time when the present disease entity of DRPLA, an autosomal dominantly inherited disease with epilepsy and ataxia, was established. Then, in 1978, the clinical picture from Japan was reported in an English journal [
72]. In 1982, based on these accumulated clinical descriptions, the present disease concept of DRPLA was established by Naito and Oyanagi [
73]. Their report fully described the clinical and pathological features of DRPLA, all of which are now accepted as typical features. These features comprise progressive myoclonic epilepsy, choreoathetosis, and dementia with an autosomal dominant form of inheritance. In some cases, mainly cerebellar ataxia or choreoathetosis (usually adult-onset) or epilepsy and dementia (usually childhood- or juvenile-onset) are shown in the same family, demonstrating intrafamilial heterogeneity (
Figure 2 and
Supplementary Video 6 in the online-only Data Supplement). In 1984, Iizuka et al. [
74] also proposed the classification of DRPLA into three types based on their own experience and previously reported cases. Their concept is broader than the current DRPLA concept. It is assumed that some of these types included other diseases. In 1988, Takahashi et al. [
75] demonstrated the common characteristics of combined degeneration of the dentatorubral and pallidoluysian systems and the pathological diversity of the disease. Viewing the process to establish the DRPLA concept, we can trace the amazing footsteps of Japanese clinicians and neuropathologists who, at a time when there were no search engines such as PubMed, carefully reviewed the literature to uncover unknown disease groups.
DRPLA has been reported more frequently in individuals of Japanese ancestry and less frequently in individuals from other Asian countries, such as Singapore, Korea and China [
76]. DRPLA is also reported less frequently in individuals of European ancestry [
76]. In 1989, an African-American family living around the Haw River in North Carolina, USA, with a five-generation history of ataxia, seizures, chorea, progressive dementia, and death within 15 to 25 years, was reported as having “Haw River syndrome” [
77]. In 1994, Haw River syndrome was confirmed to be DRPLA through genetic analysis [
78].
In 1990, Kondo et al. [
79] showed that DRPLA is a disease distinct from Huntington disease by linkage analysis using a single marker. In the early 1990s, various neuromuscular diseases were determined to be caused by an increase in the number of repetitive sequences in the gene [
80]. These discoveries revealed that repeat lengths are unstable from generation to generation and that there is a close relationship between clinical severity and repeat length. In June 1993, Li et al. [
81] isolated several gene fragments that met the criteria for increased repetition, but they failed to find these repetitive sequences in genetic neuropsychiatric disorders. In 1994, two Japanese groups simultaneously found that the CAG repeat sequence in CTG-B37, reported by Li et al. [
81], was increased in patients with DRPLA [
82,
83], leading to the identification of the gene for the disease. The results of the genetic analysis showed that the clinical and pathological variability between generations could be explained by the degree of repeated sequences. The exact function of the isolated gene and therapeutic methods have not yet been established. However, model mice carrying the full-length genomic sequence of human
Atrophin-1 (the causative gene of DRPLA), including the promoter and the expanded CAG repeat sequence, have been created in Japan [
84]. We hope that a treatment method for this disease will also be developed.
- WSS
The name WSS is derived from the two doctors who, in 1983 [
85], described the cases of six patients from two consanguineous Saudi Arabian families who had an adolescent-onset autosomal recessive disorder, with characteristic facial appearance, hypogonadism, mental abnormalities, alopecia or sparse hair, sensorineural deafness, diabetes mellitus and electrocardiogram (ECG) abnormalities (
Figure 3). All female patients had failed sexual maturation. Males had testicular failure [
85].
Similar cases were subsequently described in Kuwait by AlAwadi et al. [
86] in 1985 and in Turkey by Gül et al. [
87] in 2000. Due to the lack of full phenotypic manifestations, the cases by Al-Awadi et al. [
86] were not considered WSS by the authors at the time. In 2007, Al-Semari and Bohlega published an extensive phenotypic description and laboratory and MRI findings of WSS in 26 affected patients from 12 consanguineous families in Saudi Arabia, including the first six patients described by Woodhouse and Sakati [
88]. In addition to alopecia, hypogonadism, deafness and diabetes mellitus, 17 patients developed either progressive generalized or focal dystonia. All patients had low insulin-like growth factor 1 (IGF-1) levels. Brain MRI showed low signal intensities in the basal ganglia with white matter hyperintensities.
WSS was concluded to be most likely a multisystemic disorder and may in fact have been described earlier by Crandall et al. [
89] in 1973 in three siblings presenting with alopecia, hypogonadism and sensorineural deafness, prior to the description by Woodhouse and Sakati [
85], and may also include the syndrome described by Al-Awadi et al. [
86] WSS has since been described in other ethnic groups, including Italian [
90], Indian [
91], Turkish [
87], Pakistani [
92], Tunisian [
93] and Portuguese ethnicities [
94], with variable and less severe phenotypes as well as variable neuroimaging features. More recently, a detailed report of an additional 58 patients from Qatar [
95] was published suggesting that WSS is perhaps more common among Arab individuals and may be a reflection of the endogamous marital practices of the Arabic community.
A comprehensive review in 2008 categorized WSS features into endocrine manifestations (alopecia, hypogonadism, diabetes mellitus and thyroid dysfunction), progressive neurological syndrome (dystonia, mental retardation, sensorineural deafness) and unspecified abnormalities such as ECG abnormalities, keratoconus, camptodactyly and acanthosis nigricans [
96].
In a recent review, two distinct phenotypic spectra were reported [
97]. Type 1 had onset in early adolescence, a more disabling and rapidly progressive disease course, and more frequent occurrence of intellectual disability (
Supplementary Video 7 in the online-only Data Supplement). Type 2 presented in late adolescence with a milder disease course, less severe disability, and relatively preserved intellectual abilities (
Supplementary Video 8 in the online-only Data Supplement). Neurological involvement was present in 81.5% of these patients, with dystonia being the most common associated symptom, followed by mental impairment and sensorineural deafness in 30% [
97].
The causative mutation was identified in 2008 as a single base pair deletion (
c.436delC) in the gene
C2orf37, now known as
DCAF17, which encodes a nucleolar protein [
98]. Mutations causing a single base pair deletion in exon 3 of the
DCAF17 gene were discovered to be the cause of WSS in Pakistani families [
99]. WSS is characterized by variable interfamilial and intrafamilial phenotypic heterogeneity, including a milder phenotype with normal intellectual abilities and an absence of extrapyramidal syndrome. Similarly, other phenotypes of WSS caused by the
C2orf37 mutation were described in seven patients from Qatar, in which patients presented without diabetes mellitus, sensorineural hearing loss or dystonia [
99].
Common brain imaging findings of WSS include a small pituitary gland, iron deposition in the globus pallidus and nonenhancing white matter hyperintensities in the periventricular and frontoparietal regions (
Figure 4). Less common findings include restricted diffusion in the splenium of the corpus callosum and prominent perivascular spaces [
100]. To date, more than 122 cases have been described, mostly affecting Arab families [
95,
101,
102].
- BAFME
BAFME is also known as familial adult myoclonic epilepsy (FAME); familial cortical myoclonic tremor with epilepsy (FCMTE); and autosomal dominant cortical tremors, myoclonus, and epilepsy (ADCME).
The cases of patients considered to be affected by BAFME have been reported in Japan before the establishment of the clinical entity of BAFME in the 1990s based on familial, nonprogressive epileptic clinical features and autosomal dominant inheritance [
103-
105]. Case reports can be traced back to the 1920s (summarized in Inazuki 1990) [
106,
107]. In 1979, Naito and Kaji [
108] tried to classify the conditions of 169 individuals from 39 families with dominantly inherited myoclonus epilepsy and identified that approximately 15%–20% of affected patients had adult-onset, non-progressive clinical features. Inazuki et al. [
106] summarized the detailed clinical features in 1990. In the same year, Ikeda et al. [
109] revealed giant somatosensory evoked potential (SEP) recorded on the contralateral primary sensory cortex and C reflex after the stimulation of the median nerve. They also detected myoclonus-related cortical activity by the jerk-locked backaveraging method. They named the involuntary movement cortical tremors and concluded that these tremors represented a variant of cortical reflex myoclonus [
109]. Yasuda independently described their findings from a detailed clinical and electrophysiological study of large families in 1991 [
110].
Currently, the accepted core clinical features of BAFME are as follows [
111]: 1) an autosomal dominant mode of inheritance with high penetrance, 2) minimally progressive cortical tremors (tremulous myoclonus) (
Supplementary Video 9 in the online-only Data Supplement) and infrequent seizures, and 3) electrophysiological study findings of cortical hyperexcitability underlying the cortical tremors. The onset of cortical tremors often precedes the onset of seizures. Whereas the disease is sometimes recognized as virtually non-progressive, there are electrophysiological studies showing progression by age [
112,
113].
Genetic linkage studies were conducted, and most affected Japanese families and an affected Chinese family was shown to be linked to 8q24 (BAFME1) [
114-
118]. Outside East Asia, similar diseases have been identified, and linkage analysis revealed additional independent loci on the centromeric region of chromosome 2 (BAFME2) [
119-
124], chromosome 5 (BAFME3) [
125], and chromosome 3 (BAFME4) [
126]. In 2018, Ishiura et al. [
127] discovered pentanucleotide repeat expansions in intron 4 of
SAMD12 as the cause of BAFME1. The lengths of the repeat expansions and their instability over generations explained the anticipation phenomenon observed in BAFME families [
127,
128]. The pathogenic repeat expansions include expansions of TTTTA and TTTCA repeats, while a short TTTTA repeat is registered in the human reference sequence. In situ hybridization analysis revealed RNA foci containing UUUCA repeats in autopsied brains [
127]. These observations raised the possibility that an RNA-mediated mechanism, particularly by UUUCA repeats, is strongly involved in the pathogenesis of BAFME [
127]. The study also revealed two families without repeat expansions in
SAMD12; instead, the researchers identified repeat expansions including TTTCA and TTTTA repeats in the introns of
TNRC6A and
RAPGEF2 in these families and designated the diseases as BAFME6 and BAFME7, respectively [
127]. BAFME1 is also found in Sri Lanka, India, and Thailand [
129-
131]. Founder haplotypes were identified in the families [
129,
131,
132], suggesting that repeat expansions rarely occurred in human history.
After the discovery of pentanucleotide repeats in three genes in BAFME1, BAFME6, and BAFME7, BAFMEs linked to chromosomes 2, 3, and 5 were discovered to be caused by expansions of TTTCA and TTTTA repeats in
STARD7 (BAFME2) [
133],
YEATS2 (BAFME4) [
134], and
MARCHF6 (BAFME3) [
135], respectively, as predicted previously [
127]. These findings emphasize a strong repeat motif-phenotype correlation in BAFME [
136]. This finding that expansions of identical repeat motifs lead to similar diseases strongly supports the theory that the expression of RNA molecules containing these repeats is involved in the pathogenesis of BAFME.
Patients with BAFME are usually treated with anticonvulsants such as sodium valproate, clonazepam, levetiracetam, piracetam, zonisamide, and primidone. Epilepsy in BAFME is generally easily controlled. The effect of low-dose perampanel on cortical tremors was also reported recently [
137].
- Kufor-Rakeb syndrome; PARK-ATP13A2; PARK9
Kufor-Rakeb syndrome was first described by Najim Al-Din et al. [
138] in 1994 as a juvenile-onset, autosomal recessive, progressive pallido-pyramidal syndrome in a Jordanian family. In November 1992, a patient from the Kufor-Rakeb village in northern Jordan visited his clinic with unique clinical features. Najim Al-Din and two other doctors visited the community and examined 87 family members. Of these, five siblings born to consanguineous parents had neurological symptoms. All of them had normal birth histories and development milestones until the end of the 1st decade, when they started to develop slowness of movement, predominantly affecting their walking at approximately 13 years (range 12–16 years). The older two affected patients had poor intellectual ability. All had typical mask-like facies and developed progressive supranuclear gaze palsy and moderate to severe corticospinal tract involvement, in addition to marked parkinsonism (akinetic-rigid type, without tremors). The disease was described to be rapidly progressive, and the patients were bedridden within 2 years [
138]. Patients were started on levodopa, which resulted in dramatic improvement in parkinsonism and improvement in walking in the younger siblings [
138]. The cognitive difficulties, spasticity and supranuclear gaze palsy progressed over time, with the eventual development of levodopa-induced dyskinesia associated with narrowing of the therapeutic window, cognitive decline, and the appearance of facial-faucial mini myoclonus and visual hallucinations [
139]. Brain MRI showed pallidal, pyramidal and generalized cerebral atrophy [
138].
In 2001, in a genetic linkage analysis study of the Jordanian family performed by Hampshire et al. [
140], the syndrome was mapped to the 9 cM region of chromosome 1p36, which was assigned as PARK9. The gene responsible was later identified by Ramirez et al. [
141] as the
ATP13A2 gene encoding lysosomal type 5-P type ATPase through genetic analysis of a large Chilean kindred presenting with autosomal recessive juvenile-onset parkinsonism and dementia similar to the originally described KuforRakeb syndrome. This mutation in the
ATP13A2 gene was also verified in the original Jordanian family [
141]. In the Chilean kindred, symptoms were observed in compound heterozygote carriers of
ATP13A2 gene mutations, and there were T2-weighted hypointensities in the caudate and lenticular nuclei on brain MRI, suggestive of brain iron accumulation [
142]. The Chilean family had 2 different missense mutations (one in the father and one in the mother), whereas in the Jordanian family, there was a 22-bp duplication in all affected family members leading to a frameshift and stop codon after 236 extraneous amino acids leading to a loss of function [
141]. Since then, other phenotypes have been described in other populations, including juvenile-onset parkinsonism [
143], ataxia-myoclonus [
144], action myoclonus and seizures [
145], ataxia and polyneuropathy [
146], neuronal ceroid lipofuscinosis type 12 [
147], complicated hereditary spastic paraplegia type 78 (HSP78) [
148], juvenile-onset amyotrophic lateral sclerosis [
149] and prominent psychiatric and behavioral manifestations [
150]. The
ATP13A2 gene mutation is now included as a rare cause of dystonia-parkinsonism151 and neurodegeneration with brain iron accumulation (
Supplementary Video 10 in the online-only Data Supplement) [
151,
152].
- CAMTA2 tremulous dystonia; Bohlega syndrome
This syndrome was first described by Bohlega et al. [
153] in 1995 in two unrelated Saudi Arabian families, with a clinical presentation of juvenile onset, myoclonic tremor syndrome, and spasticity associated with cerebral white matter changes. The disease is autosomal recessive and has an onset age in the first decade of life. In the first family, the syndrome began at an average age of 7 years, with a slow frequency tremor (4–6 Hz) affecting the arms at rest. The arm tremor was worsened by posture and action, and there was a side-to-side (no-no) head tremor associated with focal or segmental craniocervical dystonia. In the second family, the tremor was jerky and involved both hands, with mild orolingual and facial twitches, consistent with myoclonus. In both families, the disease was slowly progressive, with the presence of spasticity later in the disease course. Patients were followed up for 24 years with no change in the movement disorder, there were no changes in cognition or intellectual ability, and patients were able to lead a relatively normal life, except for some mild gait difficulties later in adulthood that were due to spasticity (
Supplementary Video 11 in the online-only Data Supplement). Accelerometric recording from an outstretched hand showed 4-Hz tremors, and brain MRI showed bilateral confluent white matter changes not characteristic of any of the classic leukodystrophies and with no change over time [
153,
154]. The results of screening for lysosomal and metabolic enzyme deficiencies, including arylsulfatase levels for metachromatic leukodystrophy, were negative in both families.
Recently, in 2017, the candidate gene underlying this unique dystonia syndrome proposed by the authors as “Bohlega syndrome” was identified as the
CAMTA2 (Calmodulin Binding Transcription Activator 2) gene, encoding the CAMTA2 protein, a member of the calmodulin—binding transcription activator protein family. A common domain structure is shared in this family of proteins, including a transcription activation domain, a DNA-binding domain, and a calmodulin-binding domain [
154]. CAMTA2 has six protein-coding isoforms (CAMTA2 ENSG00000108509) ranging from 178 to 1241 amino acids; however, the expression pattern and function of each isoform remain unknown [
154]. CAMTA2 is expressed in the heart and brain. Diseases associated with CAMTA2 include retinitis pigmentosa and cardiac valvular dysplasia [
154]. At this time, there have been no new reports of this syndrome in any other populations outside these Saudi Arabian families. Further replication studies with large dystonia cohorts in other ethnicities are needed to confirm the role of
CAMTA2 in patients with dystonia.
- PKD; PxMD-PRRT2; PKD-PRRT2
‘Familial paroxysmal choreoathetosis’ was first proposed by Mount and Reback [
155] to describe the condition of a 23-year-old white man, who showed infantile-onset intermittent choreodystonic attacks that lasted several hours and could be induced by alcohol, coffee, and fatigue, and several of his affected family members in an autosomal dominant pattern of inheritance. He probably referred to what we now know as paroxysmal non-kinesigenic dyskinesia (PNKD). Andrew Kertesz, in 1967, introduced the term “kinesigenic”, which means “induced by movement” [
156]. He highlighted the differences between the cases where patients had short attacks precipitated by movements, thus differentiating them from the cases described by Mount and Reback [
156]. In 1973, Toyokura discovered that Shuzo Kure, a Japanese psychiatrist, had originally described the disorder that is now termed PKD in 1892 [
157].
The original description, supported by hand-traced high-speed photographs (
Figure 5), to be the first case of PKD was reported in the Japanese literature in 1892 by Shuzo Kure, who provided details of the attacks in a 23-year-old man that were triggered by sudden movement that brought an odd sensation as a kind of sensory aura that was rapidly followed by peculiar, purposeless, irregular involuntary movements of the legs of a very short duration, sometimes spreading to the right side of the body [
157,
158]. Although Kure labeled the condition as atypical Thomsen’s disease (myotonia congenita), noting the lack of long-lasting muscle contractions during the attack and percussion myotonia (hence, “atypical”), his clinical descriptions on the clinical manifestations as we currently recognize as PKD remain valid, except for a lack of being hereditary. In His image showed a chair placed behind the patient denoting attacks brought about by quickly rising from the chair (
Figure 5) [
157,
158].
Studies have recognized several clinical manifestations of paroxysmal dyskinesias that are inclusive of dystonia, chorea, athetosis, ballism, or any form of dyskinesias that may occur in isolation or as a combination and suddenly develop after precipitants of attacks of a variable duration, resulting in the most widely adopted classification into three subtypes based on triggering factors, namely, PKD, PNKD and paroxysmal exercise-induced dyskinesia (PED) [
159,
160]. Further genetic discoveries in these subtypes, where mutations of the
PRRT2 [
161,
162],
PNKD or
MR-1 [
163,
164], and
SLC2A1 [
165] genes account for the major primary forms of PKD, PNKD, and PED cases, respectively, have expanded our understanding that there is a marked pleiotropy of mutations in such genes with a still expanding clinical spectrum (i.e., hemiplegic migraine, ataxia, and myotonia in complicated PKD), and not all patients clinically presenting with PKD, PNKD, or PED have mutations in these genes [
166].
PRRT2 mutations can be associated with a spectrum of disorders from benign familial infantile epilepsy and infantile convulsion to PKD. This and the prevalence of PKD, suggesting that it is more common in Asians, have been reviewed elsewhere by the TF [
167].