Appendix e-1: Mutation analysis

Mutation analysis was performed according to standard protocols.Primers for DNA amplification and sequencing were designed to cover all coding regions of ATP1A3 including exon-intron boundaries based on sequence data from the GenBank database (GenBank accession numbers NM_001256213.1, NM_001256214.1, and NM_152296.4). We did direct dye terminator sequencing of PCR products with the BigDye Terminator Ready Reaction chemistry version 3.1 on an ABI PRISM 3130-Avant genetic analyzer (Life Technologies, Darmstadt, Germany). Sequence analysis was done with Lasergene software (DNASTAR, Madison, WI, USA). Nucleotide numbering of the coding sequence variants described is based on the respective reference cDNA sequence (NM_152296.4) with +1 corresponding to A of the ATG translation initiation codon. In 10 cases DNA from both parents could be analyzed. A previously not reported mutation was considered pathogenic if it occurred de novo and was not detected in 100 control alleles. Primer sequences were specified previously.1

1.Rosewich H, Thiele H, Ohlenbusch A, et al. Heterozygous de-novo mutations in ATP1A3 in patients with alternating hemiplegia of childhood: a whole-exome sequencing gene-identification study. Lancet Neurol 2012;11:764-773.

Appendix e-2: Clinical findings in new RDP and intermediate cases

Patient ATP1A3-41, a woman of presently 34 years of age, was the 1st of four children of non-consanguineous parents of Polish origin. There was no history of dystonia or parkinsonism in her family. She was sitting at 6 months, talking at 12 months and walking at 20 months of age. She always had problems with her balance and never learned to ride a bicycle. She showed excellent performance at school. At 9 years of age, in close context with a respiratory tract infection, she had abrupt onset of dystonic posturing of her left arm and to a lesser extent of her left leg as well as grimacing of the left side of her face when she wanted to present flowers to her teacher. Over the next few days she deteriorated and walking was no longer possible. She gradually recovered over a period of several months, with dystonia of her left arm remaining as the sole symptom. Two years after this first episode, on her first day at school after the holidays and again during a respiratory tract infection, she experienced acute asymmetric generalized dystonia with torticollis, dystonic gaping of her mouth and a clear rostrocaudal gradient of involvement, as head and arms were more severely affected than the legs. She had problems with swallowing and could not talk anymore. She was walking freely with bradykinesia and postural instability. There was little progression of her symptoms over the next few weeks, no improvement thereafter and almost no response to dopaminergic medication. Thus, this patient showed the classical evolution of symptoms in RDP with primary abrupt onset, preliminary improvement and second deterioration, triggered by physiological stressors. Therefore he is categorized as a new RDP patient here and included in the RDP section of table 4.

Patient ATP1A3-42, aged 20 years at present, was the 1st child of healthy parents of Caucasian origin. There was no history of dystonia or parkinsonism in his family. At age 14 he showed declining school performance and received methylphenidate for attention deficit disorder. At age 17 years, after strong physical exercise, he had abrupt onset of choreatic and ballistic movements of his left arm and leg, grimacing of the left side of his face, and slow, slurred speech. He was admitted to the local hospital where his movement disorder disappeared completely for about one hour after sedation with propofol, which he received to facilitate neuroimaging. However, dysarthria and asymmetric dystonia with bradykinesia recurred thereafter. Over the next year he showed slow recovery of his hyperkinesias, whereas dysarthria, bradykinesia and ataxia persisted. His speech would worsen with tiredness and improve again after a nap, and he experienced recurrent episodes of anarthria and loss of walking lasting several days triggered by physical or emotional stress. Taken together, this patient followed a baseline RDP course with superimposed AHC paroxysms, triggered by stressors and ameliorated by sleep. As RDP-like features are dominating this patient is categorized as a new RDP case for the purpose of this study and included in the RDP section of table 4, although he clearly shows an intermediate phenotype.

Patient ATP1A3-43, presently 21 years old, was the 4th of five children of a healthy mother and a father suffering from alcoholism. His siblings were healthy, and there was no history of dystonia or parkinsonism in his family. He had longstanding learning problems at school. He used to skip school and drink alcohol frequently. Around his 11th year of life facial tics were observed. At age 15 years he experienced transient dystonic posturing and functional impairment of his left hand. At 16 years of age he had abrupt onset of ataxia and dysarthria after drinking alcohol and falling on his head. For a week he was lying in bed at home, did not eat and could hardly drink. Thereafter he showed an unsteady gait with bradykinesia, postural instability and a slow, slurred speech. These features remained unchanged over the last five years. Thus, this patient showed a classical RDP course and is included as a new RDP case in table 3 and 4.

Patient ATP1A3-31, presently 18 years of age, was the 2nd child of healthy parents of Caucasian origin. Family history was unremarkable. He sat at 8 months, crawled at 10 months, and walked independently at 18 months of age. At 18 months of age and few days after an upper airway infection he experienced acute weakness and ataxia of his right arm while he was still able to walk. Within two weeks he recovered fully. A year later, again in context with a respiratory tract infection, he developed marked paresis of his right arm and mild functional impairment of his right leg as well as ataxia, with abrupt onset and slow recovery over a couple of weeks. From then on he frequently showed recurring episodes of flaccid hemiparesis strictly on the right side affecting predominantly the arm and hand, with intervals ranging from hours to a few weeks and duration of a few minutes to a few hours. These paroxysms were triggered by physical stress and accompanied by dysarthria and occasionally by retrocollis. Over the years the attacks changed from flaccid paresis to painful dystonic posturing of the right arm. He developed persistent mild ataxia with postural instability, bradykinesia and dysarthria or even anarthria as well as intellectual disability.

At age 15 years one of the common paroxysms of right hemidystonia and dysarthria was for the first time accompanied by a flaccid paresis of the left side. Whereas dysarthria and right hemidystonia regressed within days, a left hemiparesis with spasticity and dystonia persisted. Over the next 3 years he showed fixed left hemidystonia with a rostrocaudal gradient of involvement and triggered hemidystonic episodes of his right side with dysarthria. Obviously this patient´s features combined characteristics of AHC on his right side with a course typical of RDP on his left side. For the purpose of this study we categorized this patient as AHC and included him in the AHC section of Table 3, although he clearly shows an intermediate phenotype.

Supplementary Table e-1 Meta-analysis of paroxysmal and non-paroxysmal manifestations in AHC patients with ATP1A3 mutations
Reference / Paroxysmal features
Hemiplegic / hemidystonic episodes / Onset triggered by stressors / Seizures / Dystonic episodes / Episodes of abnormal ocular movements / Bulbar symptoms / Episodes of autonomic dysfunction / Disappearance on falling asleep / Paroxysmal respiratory disturbances / Rostro-caudal gradient
This study / 15/15 (100%) / 11/15 (73%) / 7/14 (50%) / 14/15 (93%) / 12/14 (81%) / 11/15 (73%) / 5/10 (50%) / 15/15 (100%) / 1/9 (11%) / 11/11 (100%)
4 / 79/79 (100%) / na / 47/78 (60%) / 47/76 (62%) / 73/79 (92%) / na / 56/79 (71%) / na / na / na
6 / 24/24 (100%) / 24/24 (100%) / 20/24 (83%) / 23/23 (100%) / 23/24 (96%) / 17/24 (71%) / 18/24 (75%) / 24/24 (100%) / 8/24 (33%) / 24/24 (100%)
5 / 10/10 (100%) / na / 6/10 (60%) / na / na / na / na / na / 6/10 (60%) / na
21 / 4/4 (100%) / 4/4 (100%) / 0/4 (0%) / 4/4 (100%) / 1/4 (25%) / 4/4 (100%) / na / na / na / na
Overall / 132/132 (100%) / 39/43 (90%) / 80/130 (62%) / 88/118 (75%) / 109/121 (90%) / 32/43 (74%) / 79/113 (70%) / 39/39 (100%) / 15/43 (35%) / 35/35 (100%)
Reference / Non-paroxysmal features
Developmental impairment / Best motor function
Unaided
sitting walking / Muscular hypotonia / Choreoathetosis / Ataxia / Dysarthria / Progression of non-paroxysmal features
This study / 15/15 (100%) / 3/15 (20%) / 12/15 (80%) / 11/15 (73%) / 3/13 (23%) / 10/14 (71%) / 1/10 (100%) / 3/9 (33%)
4 / 78/79 (99%) / na / na / na / 33/75 (44%) / 59/73 (81%) / na / na
6 / 23/24 (96%) / 1/25 (4%) / 16/25 (64%) / 18/23 (78%) / 15/25 (60%) / 22/25 (88%) / 23/25 (92%) / 9/25 (36%)
5 / 9/10 (90%) / 1/10 (1%) / 4/10 (40%) / na / 2/10 (20%) / na / na / 5/10 (50%)
21 / 4/4 (100%) / na / na / na / 2/4 (50%) / 2/4 (50%) / 1/4 (25%) / na
Overall / 129/132 (98%) / 5/50 (10%) / 32/50 (64%) / 29/38 (76%) / 55/127 (43%) / 93/116 (80%) / 34/39 (87%) / 11/44 (25%)
na Information not available
Supplementary Table e-2 Unique and overlapping clinical features of RDP and AHC
Symptoms / RDP / AHC
This study / Previous reports* / overall / This study / Previous reports # / overall
Overlapping clinical symptoms / Abrupt onset of dystonia/hemiplegia / 3/3 (100%) / 46/46 (100%) / 49/49 (100%) / 15/15 (100%) / 38/38 (100%) / 53/53 (100%)
Rostrocaudal gradient of involvement / 3/3 (100%) / 45/46 (98%) / 48/49 (98%) / 10/10 (100%) / 24/24 (100%) / 23/34 (100%)
Prominent bulbar findings / 3/3 (100%) / 38/43 (88%) / 41/42 (89%) / 10/14 (71%) / 21/28 (75%) / 31/46 (67%)
Trigger factors / 3/3 (100%) / 23/37 (62%) / 26/40 (65%) / 9/14 (64%) / 28/28 (100%) / 37/42 (88%)
Unique clinical symptoms / Onset < 18 month / 0/3 (0%) / 1/42 (2%) / 1/45 (2%) / 15/15 (100%) / 114/116 (98%) / 130/131 (99%)
Onset > 18 month / 3/3 (100 %) / 44/45 (98%) / 47/48 (98%) / 0/15 (0%) / 2/116 (2%) / 2/131 (2%)
Polyphasic disease course / 1/3 (33%) / 2/46 (4%) / 3/49 (6%) / 15/15 (100%) / 115/115 (100%) / 130/130 (100%)
Mono - triphasic disease course / 2/3 (67%) / 44/46 (95%) / 46/49 (94%) / 0/15 (0%) / 0/115 (0%) / 0/130 (0%)
Rapid recovery of symptoms / 0/3 (0%) / 0/46 (0%) / 0/49 (0%) / 14/15 (93%) / 112/112 (100%) / 126/127 (99%)
Slow or absent recovery of symptoms / 3/3 (100%) / 46/46 (100%) / 49/49 (100%) / 1/15 (7%) / 0/112 (0%) / 1/127 (1%)
References: *3, 7, 8, 13-26; #4, 5, 6, 21