Essential Tremor, Seizures, and Periodic Paralysis 495

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Seizure 2000; 9: 493–497 doi: 10.1053/seiz.2000.0454, available online at http://www.idealibrary.com on

Clinical-molecular study of a family with essential tremor, late onset seizures and periodic paralysis

† † ¶ ‡ § † J. A. DOM´INGUEZ-MORAN´ , M. BARON´ , G. DE BLAS , L. M. ORENSANZ & A. JIMENEZ-ESCRIG´

† Servicio de Neurolog´ıa; ‡Neuroﬁsiolog´ıa and; §Dept. de Investigacion,´ Hospital Ramon´ y Cajal, Universidad de Alcala´ and; ¶Fundacion´ Hospital de Alcorcon,´ Madrid, Spain

Correspondence to: A. Jimenez-Escrig,´ Servicio de Neurolog´ıa, Hospital Ramon´ y Cajal, 28034 Madrid, Spain. E-mail: [email protected]

We report the clinical features of, and the molecular study performed on, a Spanish family with essential tremor (ET), late onset epilepsy and autosomal dominant hypokalemic periodic paralysis (hypoPP). The presence of hypoPP in this kindred suggested an ion channel as a candidate gene for ET. Our study identiﬁed an Arg528His CACNL1A3 mutation in patients with hypoPP, and excluded this mutation as the cause of tremor or epilepsy in this kindred. c 2000 BEA Trading Ltd

Key words: epilepsy; hypokalemic periodic paralysis; essential tremor; CACNL1A3.

INTRODUCTION These facts reinforce the use of a different method- ology, such as candidate gene mutation analysis, to Essential tremor (ET) is the most common move- evaluate ET pedigrees. In this approach, a particular ment disorder and one of the most frequent neuro- gene is hypothesized to be the locus for the disease, as logical diseases1, 2. The importance of genetic factors suggested for example by the cosegregation of the dis- in ET pathogenesis is unanimously endorsed, due to order under study with diseases of known genetic ori- its frequent familial presentation and to the absence gin. The subsequent molecular genetic study will re- of known environmental factors that could explain its veal if mutations in the candidate gene are present in pathogenesis. To date, two loci have been identiﬁed by patients but not in normal controls6. In the present linkage analysis in familial ET. The 3q13 locus was work, genes related to ion channel diseases could be reported in Iceland pedigrees3 and the 2p22-p25 locus considered as candidates for ET, given the good re- in one additional family4. These kindreds developed sponse some ET patients have to calcium channel an autosomal dominant form of ET, similar to the spo- blockers7, 8, or carbonic anhydrase inhibitors9, 10. radic form. These results have not been replicated in In this paper we report a Spanish family with as- other pedigrees. Thus, despite the high prevalence of sociated essential tremor, late onset complex par- this disease, available molecular genetic information tial seizures and hypokalemic periodic paralysis (hy- is scarce. poPP). These phenotypes could be explained by a The factors that challenge the detection of genes un- mutation in a single gene, e.g. a channelopathy. Fa- derlying ET include: incomplete and sometimes low milial hypoPP is due in most cases to mutations in penetrance; the presence of atypical clinical features; the CACNL1A3 gene, which encodes the α1 subunit the existence of individuals in the pedigree affected by of the dihydropyridine receptor11, 12. Also, epileptic sporadic, nongenetic causes of ET or phenocopies; and seizures have been seen in other channelopathies13–16, the wide range of age of onset within families5. and there are forms of malignant hyperthermia, a dis- These factors complicate the correct assignment of order allelic to familial hypoPP17, linked to a locus affectation status to a given subject. Among different in the same chromosomal region as familial ET18. kindreds, the existence of genetic heterogeneity, with We report the clinical features of this family and the different genes leading to the same or similar pheno- molecular study performed, to see if mutations in type, may obscure the detection of genetic linkage CACNL1A3 were responsible for part or all of the when multiple families are summed in linkage studies. neurological phenotypes described.

1059–1311/00/070493 + 05 $35.00/0 c 2000 BEA Trading Ltd

494 J. A. Dom´ınguez-Moran´ et al.

analysis, as previously described11, 12, 19. This procedure is feasible in both cases because the Arg528His Familial hypoPP I1 I2 mutation (CGC CAC) causes the loss of a BbvI site, while the Arg1239His→ mutation (CGT CAT) creates ET and seizures a new NlaIII restriction site. The procedure→ involved II1 II2 PCR ampliﬁcation, overnight enzyme digestion, 16% polyacrylamide gel separation, and ethidium bromide staining. Table 1 shows PCR primer pairs, annealing CACNL1A3 temperatures and product lengths, along with restric- 77 tion enzymes and fragment lengths after digestion, for normal and mutated DNA. Arg528His 44

Bbv-I 33 RESULTS AND CASE REPORTS

Pedigree overview Fig. 1: Top: Kindred pedigree and phenotypes (familial hy- poPP; ET and seizures). Below: 12% polyacrylamide gel showing results of CACNL1A3 Arg528His mutation analysis The kindred included four members in two consecu- after digestion with the restriction enzyme BbvI. Patients I-1 tive generations. The pedigree diagram in Fig. 1 shows and II-2 have an abnormal 77 bp band consistent with a het- its pattern of inheritance and clinical manifestations, erozygote state for this mutation. along with restriction enzyme analysis results. Clini- cal data are summarized in Table 2. Detailed case his- tories are presented below. Two individuals in the first PATIENTS AND METHODS generation had late onset seizures and ET, one with hypoPP. In the second generation one individual had Pedigree overview hypoPP, while his sister was not affected, and neither have developed seizures or tremor so far. The average The following data are from the kindred pedigree as age at onset was 15 years old for hypoPP, 57 years shown in Fig. 1. All individuals were directly evalu- old for seizures and 68 years old for ET. The ET phe- ated by at least one of us. Evaluation consisted of a notype was identical to that seen in sporadic cases, neurological history and physical examination to as- with asymmetric involvement initially, mild cogwheel certain affection status. We performed laboratory stud- rigidity and neither akinesia nor postural instability. ies including plasma glucose, creatinine, hepatic func- Rate of progression was slow. Seizures were complex tion, erythrocyte sedimentation rate, creatine kinase partial or generalized. They were controlled with med- and thyroid status. The diagnosis of ET was based on ication (phenytoin or phenobarbitone) and tremor par- the presence of postural and kinetic tremor unilaterally tially responded to phenobarbitone. Dementia, pyra- or bilaterally, anytime during the study period and on midal, cerebellar or autonomic disturbances were not the absence of any secondary causes. HypoPP was di- found. Laboratory investigations were normal in all agnosed when there was a typical history of transient subjects. In spite of the small size of this pedigree, paralytic attacks. Hypokalemia during attacks was ob- hypoPP seemed to be transmitted with an autosomal served but not considered a prerequisite. Age at on- dominant pattern. It was not possible to assign a mode set of ET seizures or hypoPP was defined as the age of inheritance for ET and seizures as they were present when an individual first complained of these neurolog- only in the first generation. However, in view of a pos- ical symptoms. ET patients were videotaped according itive history in one of the parents (not included, as de- to a standard protocol. ceased), and to the absence of known consanguinity, ET was also assumed to be an autosomal dominant.

Molecular genetic study Case reports Blood samples were collected after informed con- sent. Total genomic DNA was isolated from blood Case I-1. leukocytes by hypotonic lysis, followed by standard phenol–chloroform extraction and ethanol precipita- This 80-year-old man presented attacks of periodic tion. Arg528His and Arg1239His mutations in the paralysis caused by stress or exercise since his teens, CACNL1A3 gene, reported to exist in patients with once every two to six months, until his ﬁfties. Since familial hypoPP, were studied by restriction enzyme then he has had no more attacks. When 51 years old Essential tremor, seizures, and periodic paralysis 495

Table 1: Methods: Restriction enzyme assay.

Mutation Primers PCR annealing PCR product Enzyme Fragments length (bp) Normal Mutated

Arg528His 50-GGAGATCCTGCTGGTGGAGTCG-30 58 ◦C 77 bp BbvI 44, 33 77, – 50-TCCTCAAGGAGGCGGATGCAG-30

Arg1239His 50-TCCAGCGCCTTCTTCCGCCT-30 Touch down: 81 bp NlaIII 48, 33 48, 28, 5 5 -CAGGAGGGTTCGCACTCCTT-3 58 C 50 C 0 0 ◦ → ◦

Table 2: Clinical manifestations.

Case # Age Sex Tremor Seizures Periodic paralysis CACLN1A3 (age at onset) (age at onset) (age at onset) mutation I-1 80 M (71) (51) (15) Arg528His I-2 84 M + (74) + (64) —+ — II-1 45 F —+ +— — — II-2 42 M — — (16) Arg528His + M male; F female. = = he began to have episodes of impaired awareness and Case II-2. motor automatism that were treated with phenobarbitone and phenytoin with good response. The medica- This 42-year-old man had periodic episodes of gener- tion was withdrawn 5 years later and the ictal episodes alized paralysis since 15 years of age, caused by stress recurred. At 71 years old he was first evaluated by us, or exercise. Two of the episodes were observed in our complaining of tremor in both upper limbs that inter- hospital and mild hypokalemia was recorded. There fered with writing and eating. An EEG showed focal was no evidence of epilepsy or tremor, at the time of left sharp wave discharges and a contrast-enhanced assessment. cranial CT was normal. Tremor progressed slowly, and, at the time of this report, 9 years later, it impaired writing and daily activities. Propranolol ameliorated Mutation analysis the tremor but was discontinued because of associated adverse effects. Nimodipine did not improve his NlaIII enzyme analysis showed normal results in all tremor. the individuals, excluding an Arg1239His mutation. On the other hand, after digestion with BbvI, cases I-1 and II-2 showed a 77 bp band along with the normal Case I-2. 33 and 44 bp bands, concordant to a heterozygote state for the Arg528His mutation. Cases I-2 and II-1, did not This 84-year-old man had a history of complex partial have this mutation. These results confirmed that famil- and generalized seizures, since he was 64. An EEG ial hypoPP in this kindred was inherited in an autoso- showed bilateral temporal slow wave discharges, and mal dominant fashion, due to this CACNL1A3 gene a cranial CT was normal. Since he was 74, he had mild mutation, with complete penetrance. The Arg528His tremor in his left upper limb. Ten years later, seizures mutation did not segregate with the ET trait (Fig. 1). were under good control with phenobarbitone, and tremor did not impair his usual daily activities. ET involved mainly the left forearm, in certain postures DISCUSSION at rest, but could also be seen while maintaining pos- ture and in the finger-nose manoeuvre. At 79 years he Familial hypoPP is a disorder of muscle excitabil- had two episodes of transient ataxia and diplopia, eval- ity characterized by transient attacks of hypotonic uated elsewhere as vertebrobasilar ischemic attacks. paralysis with variable severity and duration, and as- Another cranial CT at that time was normal. He did sociated mild hypokalemia, that occur at intervals not have episodes of periodic paralysis. of days to years20, 21. The main provocative factors are rest after exercise and carbohydrate-rich meals. Case II-1. The most frequent age of onset is within the second decade of life, and it usually remits at older This 45-year-old woman had neither periodic paralysis ages. It is an autosomal dominant disease with in- nor epileptic seizures or tremor. complete penetrance19. Late in the course of the dis- 496 J. A. Dom´ınguez-Moran´ et al. ease these patients may develop a vacuolar myopathy, there is a form, allelic with hypoPP18, in which an al- not related to the frequency or intensity of episodes. teration in the dyhidropyridine receptor causes an ab- It is a voltage-gated muscle calcium channel disor- normal interaction with the ryanodine receptor in the der. Initial genetic studies linked this disease to a electrical–mechanical coupling of muscle contraction. locus in chromosome 1q22; later, it was discovered Taking all above considerations together, genes that to correspond to mutations in the CACNL1A3 gene could link ET and/or epileptic seizures to familial which encodes the α1 subunit of the skeletal mus- hypoPP in the kindred presented include those cod- cle L-type calcium channel, named the dyhidropyri- ing the L-type calcium channel. Among these, the dine receptor11. Three point mutations have been iden- CACNL1A3 gene is a good candidate, since it is re- tified in this gene, to date: Arg528His, Arg1239His sponsible for most cases of familial hypoPP but, as and Arg1239Gly. The first one is the most usual, stated above, mutations have not been linked so far to while the last one is very uncommon11, 12. Never- systemic diseases. We therefore performed a molecu- theless these mutations do not explain all familial hy- lar analysis of CACNL1A3 mutations that could ac- poPP cases. count for hypoPP in this family. At the same time, the There are no reports of other neurological disorders molecular analysis would help to confirm or exclude in familial hypoPP. However, the family we report pre- the hypothesis of a single monogenic disease respon- sented hypoPP, along with late-onset seizures and ET, sible for the phenotypes in this kindred. only manifest in the first generation studied to date. The molecular analysis of CACNL1A3 gene muta- The late-onset age of tremor and seizures in this kin- tions performed by us showed that case I-2, with ET dred could explain the absence of these manifestations and late onset seizures, did not have the Arg528His in the second generation individuals. Also, as familial mutation that segregated with the hypoPP trait. The hypoPP may have an incomplete penetrance, case I-2 coexistence of seizures and ET of similar character- could be an asymptomatic carrier of this trait. More- istics and age of onset in cases I-1 and I-2 makes it over, as referred above, both ET and hypoPP seemed very unlikely that there is a phenocopy in case I-2. to be transmitted with an autosomal dominant pattern. This family thus presented hypoPP caused by an All these facts suggest that a mutation in a single gene Arg528His mutation in the CACNL1A3 gene in two with variable penetrance could be the cause of these individuals, and unlinked familial ET and late onset phenotypes. seizures in others. Moreover, calcium channel antag- Both clinical and genetic factors point to an ion onists did not relieve the tremor. In conclusion, these channel disorder in the family presented: on clinical results do not support an L-type calcium channel as a grounds, this interpretation agrees with several reports candidate gene for ET in the family presented. of a positive response in ET to calcium-receptor an- tagonists, which bind to the dyhidropyridine receptor, an L-type voltage-gated receptor7, 8. Genetic jus- REFERENCES tification for our hypothesis includes evidence that relate epilepsy or tremor to channelopathies. Spe- 1. Louis, D. E. and Ottman, R. How familial is familial tremor: the genetic epidemiology of essential tremor. Neurology 1996; cific epileptic syndromes are associated with muta- 46: 1200–1205. tions in the gene for the β1 subunit of voltage-gated 2. Larson, T. and Sjogren, T. Essential tremor: a clinical and ge- sodium channel SCN1B13, and in the potassium chan- netic population study. Acta Psychiatria Neurologica Scandi- nel genes KCNQ2 and KCNQ314, 15. In addition, a navica 1960; 36 (Suppl. 144): 1–176. disturbance in the ATP-sensitive potassium channel 3. Gulcher, J. R., Jonsson, I. D., Kong, A., Kristjanson, K., Frigge, M. L., Karanson, A. et al. Mapping of a familial essen- system has been found in patients with familial hy- tial tremor gene, FET1, to chromosome 3q13. Nature Genetics poPP20. Also, individuals with mutations in the hu- 1997; 17: 84–87. man voltage-gated potassium channel Kv1.1, respon- 4. Higgins, J. J., Pho, L. T. and Nee, L. E. A gene (ETM) for es- sible for episodic ataxia type 1, sometimes have gener- sential tremor maps to chromosome 2p22-p25. Movement Dis- order 1997; 12: 859–864. alized or secondarily generalized seizures, during the 5. Louis, D. E., Marder, K., Cote, L., Pullman, S., Ford, B., 16 course of the disease , and a mouse knockout for Wilder, D. et al. Differences in the prevalence of essential Kv1.1 has a lethal epilepsy phenotype23. Moreover, tremor among elderly African Americans, whites, and Hispan- CACNA1A gene mutations, responsible for episodic ics in Northern Manhattan, NY. Archives of Neurology 1995; 52: 1201–1205. ataxia type 2 and migraine in humans, cause seizures 6. Strachan, T. and Read, A. P. Human Molecular Genetics. Ox- 24 in mice . Tremor mimicking essential tremor can be ford, BIOS Scientific Pub Ltd, 1996. seen in both episodic ataxia type 1 and type 216, 25. 7. Biary, N., Bahou, Y., Sofi, M. A., Thomas, W. and al Deeb, Finally, one of the loci associated with familial ET is S. M. The effect of nimodipine on essential tremor. Neurology in the same chromosomal region as a form of malig- 1995; 45: 1523–1525. 3, 17 8. Garc´ıa-Ruiz, P. J., Garc´ıa de Yebenes,´ J. and Jimenez-Jim´ enez,´ nant hyperthermia (3q13) . This disease is usually J. Effect of nicardipine on essential tremor: brief report. Clini- caused by alterations in the ryanodine receptor, but cal Neuropharmacology 1993; 16: 456–459. Essential tremor, seizures, and periodic paralysis 497

9. Busenbark, K., Pahwa, R., Hubble, J. and Koller, W. The ef- 18. Monnier, N., Procaccio, V., Stieglitz, P. and Lunardi, J. fect of acetazolamide on essential tremor: an open-label trial. Malignant-hyperthermia susceptibility is associated with a Neurology 1992; 42: 1394–1395. mutation of the α1-subunit of the human dyhidropyridine- 10. Busenbark, K., Pahwa, R., Hubble, J., Hopfensperger, K., sensitive L-type voltage-dependent calcium-channel receptor Koller, W. and Pogrebra, K. Double-blind controlled study of in skeletal muscle. American Journal of Human Genetics methazolamide in the treatment of essential tremor. Neurology 1997; 60: 1316–1325. 1993; 43: 1045–1047. 19. Sillen,´ A., Sorensen, T., Kantola, I., Friis, M. L., Gustavson, 11. Ptacek,´ L. J., Tawill, R., Griggs, R. C., Engel, A. G., Layzer, R. K. H. and Wadelius, C. Identiﬁcation of mutations in the B., Kwiecinski, H. et al. Dihydropyridine receptor mutations CACNL1A3 gene in 13 families of Scandinavian origin hav- cause hypokalemic periodic paralysis. Cell 1994; 77: 863–868. ing hypokalemic periodic paralysis and evidence of a founder 12. Jurkat-Rott, K., Lehmann-Horn, F., Elbaz, A., Heine, R., effect in Danish families. American Journal of Medicinal Ge- Gregg, R. G., Hogan, K. et al. A calcium channel mutation netics 1997; 69: 102–106. causing hypokalemic periodic paralysis. Human Molecular 20. Links, T. P., Smit, A. J., Molenaar, W. M., Zwarts, M. J. and Genetics 1994; 3: 1415–1419. Oosterhuis, H. J. G. H. Familial hypokalemic paralysis. Clin- 13. Wallace, R. H., Wang, D. W., Singh, R., Scheffer, I. E., George, ical, diagnostic and therapeutic aspects. Journal of Neurologi- A. L., Phillips, H. A. et al. Febrile seizures and generalized cal Sciences 1994; 122: 33–43. epilepsy associated with a mutation in the Na+-channel beta1 21. Fontaine, B., Lapie, P., Plassart, E., Tabti, N., Nicole, S., Re- subunit gene SCN1B. Nature Genetics 1998; 19: 366–370. boul, J. et al. Periodic paralysis and voltage-gated ion chan- 14. Singh, N. A., Charlier, C., Stauffer, D. et al. A novel potassium nels. Kidney Int 1996; 49: 9–18. channel gene, KCNQ2, is mutated in an inherited epilepsy of 22. Gregg, R. G., Couch, F., Hogan, K. and Powers, P. A. Assign- newborns. Nature Genetics 1998; 18: 25–29. ment of the human gene for the alpha 1 subunit of the skeletal 15. Charlier, C., Singh, N. A., Ryan, S. G., Lewis, T. B., Reus, B. muscle DHP-sensitive Ca2+ channel (CACNL1A3) to chro- E., Leach, R. J. and Leppert, M. A pore mutation in a novel mosome 1q31-q32. Genomics 1993; 15: 107–112. KQT-like potassium channel gene in an idiopathic epilepsy 23. Smart, S. L., Lopantsev, V., Zhang, C. L. et al. Deletion of the family. Nature Genetics 1998; 18: 53–55. K(V)1.1 potassium channel causes epilepsy in mice. Neuron 16. Zuberi, S. M., Eunson, L. H., Spauchus, A., De Silva, R., 1998; 20: 809–819. Tolmies, J., Wood, R. C. et al. A novel mutation in the human 24. Terwindt, G. M., Ophoff, R. A., Haan, J., Sandkuijl, L. A., voltage-gated potassium channel gene (Kv1.1) associates with Frants, R. and Ferrari, M. D. the Dutch Migraine Genetics periodic ataxia type I and sometimes with partial epilepsy. Research Group. Migraine, ataxia and epilepsy: a challenging Brain 1999; 122: 817–823. spectrum of genetic determined calcium channelopathies. Eu- 17. Sudbrak, R., Procaccio, V., Klausnitzer, M. et al. Mapping of ropean Journal of Human Genetics 1998; 6: 297–307. a further malignant hyperthermia susceptibility locus to chro- 25. Stevanin, G., Durr, A., David, G. et al. Clinical and molecular mosome 3q13. American Journal of Human Genetics 1995; features of spinocerebellar ataxia type 6. Neurology 1997; 49: 56: 684–691. 1243–1246.