Infantile-Onset Paroxysmal Movement Disorder and Episodic Ataxia Associated with a TBC1D24 Mutation

Abstract

Mutations that disrupt the TBC1D24 presynaptic protein have been implicated in various neurological disorders including epilepsy, chronic encephalopathy, DOORS (deafness, onychodystrophy, osteodystrophy, mental retardation, and seizures) syndrome, nonsyndromic hearing loss, and myoclonus. We present the case of a 22-month-old male with infantile-onset paroxysmal episodes of facial and limb myoclonus. The episodes were linked to biallelic variants in exon 2 of the TBC1D24 gene that lead to amino acid changes (c.304C >T/p.Pro102Ser and c.410T > C/p.Val137Ala), each variant being inherited from a parent. Follow-up imaging in adolescence revealed widened right cerebellar sulci. We discuss the evolving landscape of TBC1D24 associated phenotypes; this case adds to a growing body of evidence linking this gene to movement disorders in children.

Keywords

TBC1D24 - genetics - infancy - cerebellum - episodic ataxia - myoclonus

Introduction

TBC1D24 encodes a member of the family of Tre2-Bub2-Cdc16 (TBC) domain-containing proteins.[1] This family helps to coordinate proteins of the Ras superfamily, including Rab proteins and GTP (guanosine triphosphate) ases, in the regulation of exo- and endocytosis. Mutations in TBC1D24 are associated with an expanding list of neurological disorders.[2] This list includes various epilepsies and epileptic encephalopathies, DOORS (deafness, onychodystrophy, osteodystrophy, mental retardation, and seizures) syndrome, several nonsyndromic deafness syndromes, and several newly-described conditions involving myoclonus with and without seizure activity. This report focuses on an unusual neurological presentation associated with new TBC1D24 variants.

Case Study

This healthy and developmentally normal boy was born after an uneventful pregnancy and term birth. Family history was positive for childhood seizures or paroxysmal events of unknown significance in his maternal cousin, grandfather, and great-grandfather, none of which lasted into adulthood. The child's maternal great-grandparents were consanguineous.

The patient first presented at our institution at the age of 22 months with a 2-month history of paroxysmal movements, mostly precipitated by fever or fatigue. These movements were characterized by facial myoclonus involving a single eyelid and generalizing to the entire face ([Video 1]), occasionally accompanied by alternating upper and lower limb tremors. These episodes could last up to 45 minutes and responded rapidly to 5 mg intrarectal diazepam. Repeated electroencephalograms (EEGs) and a brain magnetic resonance imaging (MRI) performed at 21 months were normal. No episodes were captured during the initial EEGs; however, by 30 months, these paroxysmal episodes had evolved into left sided ataxic episodes involving sudden flexion of the left wrist, tremor of the left upper limb, confusion, and diminished speech. The diminished speech was difficult to assess given the patient's age and was thought to be a form of aphasia, but could not be clearly distinguished from mutism or anarthria. One of these events was recorded on EEG and correlated to a physiological mu rhythm in the right central–parietal region that was present prior to, during, and shortly after the episode. The episodes could not be provoked in the patient.

Video 1

Video of upper and lower limb tremors during clinic visit around 22 months of age.

An extensive neurometabolic workup ([Table 1]) including blood, urine, and cerebrospinal fluid (CSF) analyses revealed no abnormalities ([Table 2]). The patient was started on a regimen of gabapentin and clobazam as maintenance therapy, along with clonazepam or rectal diazepam as rescue treatment. He responded well initially with a significant decrease in the frequency of episodes. Over the following 2 years, however, the abnormal movements resumed, occurring up to multiple times per week. The introduction of acetazolamide at 6.25 mg/kg/day administered twice a day led to a substantial improvement in the frequency of the attacks.

Table 1 Neurometabolic workup

Specimen	Test	Result	Reference range
Blood
	Chemistry	Within normal limits
	Lactate	1.4 mmol/L	1.0–1.9
	Pyruvate	94 mol/L	50–105
	Lactate: pyruvate ratio	14.9	12–18
	Ammonium	40 mol/L	11–35
	Zinc	20.9 mol/L	10–23
	Ceruloplasmin	0.55 g/L	0.22–0.61
	Copper	21.7 mol/L	12.5–23.6
	Lipid panel	Within normal limits
	Erythrocytes	4.9 T/L	3.9–5.3
	Hemoglobin	133 g/L	115–135
	Platelets	506 G/L	168–392
	Leucocytes	11.4 G/L	5–14.5
	TSH	1.9 mU/L	0.4–4
	Nonesterified fatty acids (NEFA)	0.7 mmol/L	0.3–0.8
	Acylcarinitine profile	Unremarkable
	Complete amino acid profile	Glutamine at the upper limit of normal, otherwise unremarkable
	Antivoltage gated K+ channel antibody (anti-VGKC)	Negative
	Borrelia burgdorferi serologies (IgM, IgG)	Negative
CSF
	Appearance	Clear
	Erythrocytes	16 M/L
	Leucocytes	< 1 M/L
	Glucose	2.9 mmol/L	2.8–4.0
	Sodium	144 mmol/L	142–154
	Potassium	2.9 mmol/L	2.5–3.5
	Proteins	0.22 g/L	0.15–0.45
	Lactate	1.3 mmol/L	0.0–2.8
	Pyruvate	97 mol/L
	Biogenic amines	Biopterin at lower limit of normal, otherwise unremarkable
Urine
	Organic aciduria	negative
	Urinalysis	Normal

Abbreviations: CSF, cerebrospinal fluid; IgG, immunoglobulin G; IgM, immunoglobulin M; TSH, thyroid stimulating hormone.

Table 2 Results of variant prediction software

Prediction technique	Pro102Ser variant (maternal)	Val137Ala variant (paternal)
Polyphen-2	Probably damaging, score 1.00 (sensitivity = 0.0, specificity = 1.0)	Probably damaging, score 0.960 (sensitivity = 0.75, specificity = 0.95)
SIFT	Tolerated (score = 0.086, cut-off = 0.05)	Tolerated (score = 0.187, cut-off = 0.05)
PROVEAN	Neutral (score = 2.05, cut-off = 2.5)	Deleterious (score = 2.5, cut-off = 2.5
MutationTaster	Disease-causing	Disease-causing
CADD (PHRED-like, scaled) score	24.1	22.6

Abbreviations: CADD, Combined Annotation Dependent Depletion; PROVEAN, Protein Variant Effect Analyzer; SIFT, Sorting Intolerant from Tolerant.

During subsequent years, genetic investigations were performed. After sequencing of KCNA1, SLC2A1, CACNA1A, CACNB4, SLC1A3, FGF14, ADCY5, PRRT2, ATP1A3, NOL3, P2RX7, and PNKD all returned negative, the patient was found to have biallelic variants in exon 2 of the TBC1D24 gene that lead to amino acid changes (c.304C > T/p.Pro102Ser inherited from the mother; c.410T > C/p.Val137Ala inherited from the father). These variants were confirmed using Sanger's sequencing. The p.Pro102Ser variant is extremely rare in the ExAC database (1 out of 118,938 individuals), while the p.Val137Ala variant is entirely absent from polymorphism databases (including ExAC, 1,000 Genomes, Exome variant server [EVS]). The p.Pro102Ser variant was not found in the maternal grandfather who was known to have had childhood seizures. Other family members with childhood seizures were not available for DNA sequencing.

On the same drug regimen, the patient who was 1- year-old at the time of the discovery of these variants continued to experience these paroxysmal episodes once a month. A year later, he was briefly hospitalized at an outside institution for a prolonged myoclonic episode involving the left upper limb. An MRI at the time revealed a widening of the right-sided cerebellar sulci with gliosis seen as a hypersignal on T2-weighted imaging, along with enlarged Virchow–Robin spaces bilaterally ([Fig. 1]). After renewed discussions with the family, the patient was started on carbamazepine. On a regimen of 300 mg twice a day (13.2 mg/kg/day), he remained episode-free for 11 months. Shortly after the onset of a viral upper respiratory infection, he presented a breakthrough focal clonic seizure of the left arm evolving to a bilateral seizure, and, shortly thereafter, a similar event involving the right arm without generalization. Blood levels of carbamazepine were within normal ranges (6.8 mg/L, norm: 4–12). The doses of carbamazepine were left unchanged, and the patient has not presented further episodes since. He is currently a cognitively normal and socially well-adapted 14-year-old.

Fig. 1 Axial T2-weighted MRI at (A) 21 months and (B) 11 years of age. The latter MRI shows a widening of the sulci of the right cerebellum, with gliosis, indicated by the arrow. MRI, magnetic resonance imaging.

Discussion

Our patient's clinical presentation involves a complex movement disorder associated with compound heterozygous variants of TBC1D24. These variants both occur in functional domains, near clusters of other known pathogenic variants, and involve changes in conserved amino acid sequences. While it is difficult to definitely ascertain that these variants are pathogenic, evidence from bioinformatic analyses suggest that these variants are likely not benign. The p.Pro102Ser variant was predicted to be damaging by Polyphen-2[3] and MutationTaster[4] but not by SIFT[5] and PROVEAN.[6] The p.Val137Ala variant was predicted to be damaging by Polyphen-2, MutationTaster, and PROVEAN but not by SIFT. Using CADD,[7] both variants were found to be among the top 1% of most deleterious substitutions possible in the human genome. Combining these insights from bioinformatics with the clinical insight that many compound heterozygous variants in TBC1D24 have been associated with a clinical picture of infantile-onset paroxysmal myoclonus similar to that seen in our patient, we consider these variants as likely pathogenic.[8] Nerve conduction studies and electromyography (EMG) with back-averaging, which could have further characterized the movement disorder, were discussed but could not be performed due to logistical difficulties.

Thanks to advances in genetic sequencing, there has been a remarkable increase in the number of symptoms and phenotypes associated with TBC1D24 mutations. These include an important number of epilepsies and epileptic encephalopathies, DOORS syndrome, several nonsyndromic deafness syndromes, and clinical syndromes involving multifocal myoclonus as a central feature.[2] Epileptic syndromes in the setting of TBC1D24 mutations are phenotypically heterogeneous and, according to a recent review, no clear genotype–phenotype correlation has yet emerged.[2] In addition to phenotypic pleiotropy in unrelated patients with the same TBC1D24 mutations, most patients studied to date are compound heterozygotes, which makes the study of individual mutations difficult.[2] Several distinct clinical entities featuring epilepsy have been reported. The spectrum ranges from infantile epilepsy without long-term neurodevelopmental sequelae, to severe forms of infantile encephalopathy with epilepsy, resulting in profound intellectual disability and early death. The most common clinical profile of TBC1D24-related epilepsy seems to be that of treatment-refractory infantile-onset myoclonic epilepsy.[2]

Multifocal myoclonus with and without EEG correlation has been reported in patients with TBC1D24 mutations. Myoclonic attacks were described in two families with TBC1D24-related epilepsy.[9] [10] Doummar et al found novel TBC1D24 mutations in a patient with cortical multifocal myoclonus, gait ataxia, and mild intellectual disability. Their patient was also found to have a cerebellar sequela on T2-weighted imaging, similar to that observed in our patient.[11] Another report described new TBC1D24 mutations in siblings with multifocal myoclonus, cerebellar atrophy, and signaling abnormalities yet without EEG correlation.[12] Finally, a Chinese boy was reported to have paroxysmal episodes of focal myoclonus affecting eyelids and limbs, without EEG correlation, and was found to have novel compound heterozygous mutations of TBC1D24.[13] These episodes were precipitated by fever or fatigue and were associated with ataxia and mild intellectual disability.

Our case adds to the understanding of the myoclonus phenotype associated with TBC1D24 variants by presenting a unique association of myoclonus, tremors, and possible aphasia without intellectual disability or EEG abnormalities. It also adds to a list of cases involving cerebellar involvement with or without clinically significant ataxia and cerebellar syndromes.[11] [12] [14] [15] While the exact clinical significance of these cerebellar findings remains unclear, the increasing number of reports of cerebellar anomalies on imaging in patients with TBC1D24-related disorders is certainly worth noting.

In summary, our case report expands the spectrum of manifestations associated with TBC1D24 mutations. The complex movement disorder described in our otherwise healthy patient could represent a suggestive feature. Further and larger studies, both clinical and biological, are needed to confirm the role of TBC1D24 in childhood-onset movement disorders and seizures.

Conflict of Interest

None.

Acknowledgments

Our gratitude goes to all pediatric neurology and movement disorder experts who gave their valuable input on various aspects of our patient's clinical presentation before we were able to reach a genetic diagnosis. These include Prof. Jean Aicardi, Paris; Prof. Emilio Fernandez-Alvarez, Barcelona; Prof. Bernard Echenne, Montpellier; Prof. Thierry Deonna, Lausanne; Dr. Wang Qing, Cleveland; Dr. Agathe Roubertie, Montpellier; and Dr. Diane Doummar, Paris.

• References

• 1 Mucha BE, Hennekam RCM, Sisodiya S, Campeau PM. TBC1D24-related disorders. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25719194 . Accessed April 5, 2019
• 2 Balestrini S, Milh M, Castiglioni C. , et al. TBC1D24 genotype-phenotype correlation: Epilepsies and other neurologic features. Neurology 2016; 87 (01) 77-85
• 3 Adzhubei I, Jordan DM, Sunyaev SR. Predicting functional effect of human missense mutations using PolyPhen-2. Curr Protoc Hum Genet 2013; Chapter 7 (Unit 7): 20
• 4 Schwarz JM, Rödelsperger C, Schuelke M, Seelow D. MutationTaster evaluates disease-causing potential of sequence alterations. Nat Methods 2010; 7 (08) 575-576
• 5 Vaser R, Adusumalli S, Leng SN, Sikic M, Ng PC. SIFT missense predictions for genomes. Nat Protoc 2016; 11 (01) 1-9
• 6 Choi Y, Chan AP. PROVEAN web server: a tool to predict the functional effect of amino acid substitutions and indels. Bioinformatics 2015; 31 (16) 2745-2747
• 7 Kircher M, Witten DM, Jain P, O'Roak BJ, Cooper GM, Shendure J. A general framework for estimating the relative pathogenicity of human genetic variants. Nat Genet 2014; 46 (03) 310-315
• 8 Fischer B, Lüthy K, Paesmans J. , et al. Skywalker-TBC1D24 has a lipid-binding pocket mutated in epilepsy and required for synaptic function. Nat Struct Mol Biol 2016; 23 (11) 965-973
• 9 Corbett MA, Bahlo M, Jolly L. , et al. A focal epilepsy and intellectual disability syndrome is due to a mutation in TBC1D24. Am J Hum Genet 2010; 87 (03) 371-375
• 10 Guven A, Tolun A. TBC1D24 truncating mutation resulting in severe neurodegeneration. J Med Genet 2013; 50 (03) 199-202
• 11 Doummar D, Mignot C, Apartis E. , et al. A novel homozygous TBC1D24 mutation causing multifocal myoclonus with cerebellar involvement. Mov Disord 2015; 30 (10) 1431-1432
• 12 Ngoh A, Bras J, Guerreiro R. , et al. Tbc1d24 mutations in a sibship with multifocal polymyoclonus. Tremor Other Hyperkinet Mov (N Y) 2017; 7: 452
• 13 Li WH, Zhou SZ, Zhang LM. , et al. [Novel compound heterozygous TBC1D24 mutations in a boy with infantile focal myoclonic epilepsy and literature review]. Zhonghua Er Ke Za Zhi 2017; 55 (01) 50-53
• 14 Campeau PM, Kasperaviciute D, Lu JT. , et al. The genetic basis of DOORS syndrome: an exome-sequencing study. Lancet Neurol 2014; 13 (01) 44-58
• 15 Afawi Z, Mandelstam S, Korczyn AD. , et al. TBC1D24 mutation associated with focal epilepsy, cognitive impairment and a distinctive cerebro-cerebellar malformation. Epilepsy Res 2013; 105 (1,2): 240-244