We describe the clinical, neurophysiologic, and genetic features of a new, large family with familial cortical myoclonic tremor and epilepsy (FCMTE).
We describe the clinical, neurophysiologic, and genetic features of a new, large family with familial cortical myoclonic tremor and epilepsy (FCMTE).
Reliable clinical information was obtained on the 127 members. Thirty-one collaborative individuals were assessed by a detailed clinical interview and a complete neurologic examination. A polygraphic study was conducted in 15 patients, back-averaging analysis and somatosensory evoked potentials with C-reflex study in four. The genetic study investigated 30 subjects with microsatellite markers at three loci on chromosomes 8q (FCMTE1), 2p (FCMTE2), and 5p (FCMTE3).
The pedigree included 25 affected members (M/F: 9/16). We studied 16 of the 19 living affected members (M/F: 5/11; mean age 47.8 years). Cortical myoclonic tremor (CMT) was associated with generalized seizures in 10 patients (62.5%). The mean age at onset of CMT and seizures was 28.1 and 33.8 years, respectively. Four patients (25%) reported a slow progression of CMT, with severe gait impairment in one. Psychiatric disorders of variable severity recurred in 37.5% of cases. Rhythmic bursts at 7–15 Hz were recorded in all 11 affected members tested. Additional neurophysiologic investigations disclosed a cortical origin of myoclonus in all patients tested. Generalized epileptiform discharges were recorded in 25% of cases, and a photoparoxysmal response in 31%. Genetic analysis established linkage to the FCMTE2 locus on chromosome 2p11.1-2q12.2 (OMIM 607876) and narrowed the critical interval to a 10.4 Mb segment. Haplotype analysis in the present family identified a founder haplotype identical to that previously observed in families from the same geographic area.
This study confirms evidence of a founder effect in Italian families and reduces the number of positional candidate genes in the FCMTE2 locus to 59, thereby contributing to future gene identification by Next Generation Sequencing approaches.
Familial cortical myoclonic tremor and epilepsy (FCMTE) is a rare autosomal dominant syndrome of adult onset primarily characterized by cortical myoclonic tremor (CMT), a form of cortical rhythmic myoclonus. This is variably associated with rare generalized tonic–clonic seizures (GTCS; 18.2–100%) or prolonged myoclonic seizures (myoclonic status), often elicited by common triggering factors (photic stimulation, sleep deprivation, and emotional stress). CMT manifests with a rhythmic low amplitude distal jerking exacerbated by posture maintenance with superimposed asynchronous myoclonia.
Neurophysiologically, CMT shows the features of cortical reflex myoclonus (Ikeda et al., 1990). Transcranial magnetic stimulation (TMS) findings (low motor threshold and shorter contralateral silent period with short-interval intra-cortical inhibition significantly reduced) suggest that motor cortex hyperexcitability results from a reduced inhibition of γ-aminobutyric acid (GABA)ergic interneurons (Guerrini et al., 2001; Van Rootselaar et al., 2007) that could be secondary to a dysfunction of cortex inhibitory input from the cerebellum, as suggested by pathologic (Van Rootselaar et al., 2004) and metabolic findings (Striano et al., 2009).
To date, about 70 FCMTE families have been reported worldwide, mostly from Japan and Europe. Genetic studies disclosed four different genomic loci, FCMTE1, FCMTE2, FCMTE3, and FCMTE4, on chromosomes 8q23.3-q24.13 (Mori et al., 2011), 2p11.1-q12.2 (Crompton et al., 2012), 5p15.31-p15.1 (Depienne et al., 2010), and 3q26.32-3q28 (Yeetong et al., 2013), respectively, but failed to identify the underlying mutations. A founder effect has been documented in Japan, where six independent Japanese families were linked to the FCMTE1 locus (Mikami et al., 1999; Plaster et al., 1999; Ebihara et al., 2003). In Italy, the same FCMTE2 haplotype was shown to cosegregate with the disease phenotype in five apparently unrelated pedigrees originating from the province of Naples (Madia et al., 2008). These families and the Japanese ones present the “classical” FCMTE phenotype with adult onset and stable symptoms responsive to GABAergic agents (valproate and clonazepam) and other antimyoclonic drugs, such as levetiracetam (Striano et al., 2005). A more severe phenotype with earlier onset of intractable partial seizures and intellectual disability was first described in another Italian pedigree linked to the FCMTE2 locus (Guerrini et al., 2001). Partial seizures with visual symptoms, suggestive of an occipital focus, and a sensitivity to hypoglycemia, exercise, and vibration, have also been reported in a French pedigree linked to the FCMTE3 locus (Magnin et al., 2009; Depienne et al., 2010).
We recently observed an additional family with FCMTE originating from the Italian province of Caserta. We report herein the clinical and genetic studies performed to characterize this extended FCMTE pedigree.
The local ethical committee approved the study, and written informed consent was obtained from all patients who agreed to participate.
The family comprised 127 individuals. Reliable clinical information on the noncollaborative or deceased family members was obtained from the proband and her relatives assessed in loco (among them a physician) for the pedigree reconstruction. Among 109 living people, we directly evaluated 31 cooperative members with a detailed historical interview and a complete neurologic examination (NE). All the patients investigated in our department underwent a complete neurophysiological, neuropsychological, and neuroradiologic study, including video–electroencephalography (EEG) and simultaneous electromyography (EMG) monitoring, somatosensory evoked potentials (SEPs), C-reflex, and TMS. Details of the neurophysiologic tests are given in Data S1. Venus blood or saliva samples were collected from 30 individuals (three spouses) for DNA extractions and analysis. The 31 members studied were de-identified, and all their clinical and instrumental data were evaluated independently by three neurologists (PT, FB, IN) to establish the diagnosis. The required agreement among experts for the final diagnosis of FCMTE was 100%. When agreement was not reached, the case was resolved by discussion.
To test the hypothesis of linkage at the FCMTE2 locus and at the FCMTE1 and FCMTE3 loci we genotyped 11 affected individuals (III-4, III-8, III-20, III-22, III-24, III-27, III-34, IV-10, IV-21, IV-25, IV-26) at 22 microsatellite marker loci distributed across the three different critical regions. All these markers were selected from the Marshfield Genetic Map (http://bli.uzh.ch/BLI/Projects/genetics/maps/marsh.html), except for markers named FCMTE3-A, FCMTE3-B, FCMTE3-C, and FCMTE3-D, which were identified directly from the genomic sequence via the on-line tool Tandem Repeats Finder (http://tandem.bu.edu/trf/trf.html). Marker allele frequencies were set as equal and averaged male/female genetic centiMorgan positions were drawn from the Marshfield Genetic Map for all markers except markers FCMTE3-A, FCMTE3-B, FCMTE3-C, and FCMTE3-D, for which centiMorgan positions were approximated to those of the most proximal single nucleotide polymorphism (SNP) with a known position in the Marshfield Genetic Map. Initially, we performed exact linkage analysis with Merlin version 1.1.2 (Abecasis et al., 2002) to calculate single-point and multipoint logarithm of odds (LOD) scores. Penetrance was set at 0.001 for all individuals carrying the aa (homozygous for the wild-type allele) genotype, and at 0.999 for all the individuals carrying the aA or AA (heterozygous or homozygous for the mutant allele) genotype. Disease allele frequency was set at 0.0001.
Extended approximate linkage and haplotype analysis were performed exclusively for the FCMTE2 locus. Nineteen additional individuals, for a total of 30 family members, were included in the study, and four additional microsatellite marker loci (D2S337, D2S2181, D2S2159, and D2S2222) were genotyped to obtain a map of 13 markers. We used Mega2 version 4.5.5 (Mukhopadhyay et al., 2005) to prepare file formats for haplotype and linkage analysis with Simwalk2 version 2.91 (Sobel & Lange, 1996). Disease allele frequency and penetrances were set as in the initial analysis. Allele frequencies were calculated on a sample of 50 apparently unrelated control individuals originating from the province of Naples. Averaged male/female genetic centiMorgan positions were drawn from the deCODE Genetic Map (www.decode.com), because some of the additional markers were absent in the Marshfield Genetic Map.
Finally, we genotyped 12 markers in two affected individuals from two different families in which a founder haplotype was described (Madia et al., 2008), to verify that the disease-linked chromosome in the present family was the same observed in the other families from the province of Naples.
The pedigree reconstruction showed 25 affected members (six deceased) over four generations, without consanguineous marriages, in which FCMTE segregates with an autosomal dominant inheritance (Fig. 1). Among the 31 members assessed directly, 16 were considered affected (Table 1) and six uncertain.
|Pts||Sex/Age (years)||Clinical features||Neurophysiology||NE myoclonus severity +/++/+++||Therapy|
|CMT onset years/progression||Seizures type/onset (years)||Other||Interictal EEG/photosensitivity||g-SEP/C-reflex||BA|
|III-4||F/58||40/↔||GTCS/40||CI, headache||na||na||na||++ FR||CNZ, L-dopa|
|III-8||F/62||20/↑||GTCS/37||Headache||G/+||na||na||+++ (LL)||VPA, BZP|
|III-11||F/56||30/↔||–||Psych (PA, GAD)||na||na||na||+++||CDD|
|III-16||F/51||30/↑||–||CI||Burst of diffuse slow W and intermixed L post S/na||na||na||+++ (E), FR||–|
|III-20||F/64||40/↑||M/63||Migraine||Burst of diffuse sharply contoured slow W > ant/na||na||+||+++||CDD (AED ceased)|
|III-22||F/55||unk||M, GTCS/unk||–||na/na||na||na||+||CNZ (VPA ceased)|
|III-24||M/57||unk||GTCS/30||–||Burst of diffuse slow W/+||na||na||±||VPA|
|III-27||F/65||unk/↑||M, GTCS/29||Psych (D), CI||G/−||+/+||na||+++ (LL, O) FR||SNRI (VPA ceased)|
|III-34||F/60||unk||–||–||Bursts of diffuse slow W and S > R ant/na||na||na||+||–|
|IV-3||F/35||20/↔||–||Psych (PA, SA)||N||na||na||+||– (SSRI ceased)|
|IV-10||F/44||30/↔||M/42||Psych (D)||Diffuse slow W and intermixed S > ant/+||na||+||+++ (LL, O)||PZP, Trazodone|
|IV-13||M/39||30/↔||–||–||Runs of diffuse theta activity/na||na||na||+||–|
|IV-21||M/29||19/↔||M, GTCS/24||Migraine||G/+||−/+||+||++ (↑on CBZ)||CBZ (→VPA)|
|IV-25||F/34||30/unk||GTCS/34||Psych (PA, D)||G/+||+/+||na||+||LEV, PZP, SNRI|
|IV-26||M/22||20/↔||–||Psych (PA)||Not specific/na||na||na||+ (LL)||PZP, TCA, SSRI|
All 16 affected members studied (M/F = 5/11; mean age of 47.8 years; range: 22–65) have CMT and 10 patients (62.5%) experienced additional infrequent generalized seizures. CMT was the first symptom in most cases, with a mean age at onset of 28.1 years (mean calculated on 11 individuals; range 19–40). Only in one case (III-15) did the seizures precede CMT by a few years. This patient had a moderate-severe intellectual disability evident since preschool age and attributed to a questionable perinatal injury. Since the age of 5 years he has also experienced infrequent GTCS treated with valproate (VPA).
CMT showed a wide intrafamilial variability. Most patients had a small-amplitude CMT localized on the upper limbs only during posture maintenance or action, with minimal impairment of daily activities. Instead, about half the patients had frequent asynchronous, proximal myoclonias of higher amplitude superimposed on the postural, irregular oscillations of the extremities. The myoclonic jerks were often triggered by emotional stress, sleep deprivation, and, overall, intense photic stimulation. Lower limb involvement, together with the slow progression of CMT, was the principal factor of severe disability in these patients. In particular, the proband reported a progressive gait impairment with several falls and has been wheelchair bound since 64 years of age. The NE in this patient showed an impaired balance and slight deficit in coordination, without clear cerebellar signs or findings of peripheral polyneuropathy at the nerve conduction study.
In the 10 patients with associated generalized epileptic seizures, the mean age at onset of seizures was 33.8 years (range 5–63 years). Eight patients experienced GTCS, in some cases evolving from clusters of photoinduced myoclonias of rising intensity. Two patients experienced a few prolonged episodes of irregular, generalized myoclonic jerking provoked by photic stimulation and followed by a fall, without loss of consciousness. They have never experienced GTCS.
On the first evaluation, half the patients were on antiepileptic medications, mostly VPA with benefit, confirming the efficacy of this drug in FCMTE treatment. One patient, previously erroneously diagnosed with partial epilepsy and treated with carbamazepine, presented with a severe myoclonus, improved after switching to VPA.
One patient (III-4) was on L-dopa therapy. She had a left frontal stroke and presented with memory impairment and right hypertonia associated with bilateral upper limbs CMT. Neuroimaging was negative in all other cases.
Two other patients (III-16 and III-27) who complained of a memory impairment, made errors on temporal orientation, registration, and recall and showed frontal release signs at NE. A formal neuropsychological evaluation in patient III-27 showed some alterations in verbal reasoning and attention, with Mini Mental State Evaluation (MMSE) (Folstein et al., 1975) results within normal limits.
Six patients had a history of mood depression and/or anxiety, with panic attacks in three cases and a major depressive episode with a suicide attempt in one. All these patients were on psychiatric treatment only (except for patient IV-3 who discontinued antidepressant drugs during pregnancy). Moreover, in one patient with CMT and previous “provoked” GTCS, antiepileptic treatment was discontinued following a diagnosis of personality disorder with conversion. Two patients (III-22, IV-18) with GTCS and mild CMT were diagnosed with juvenile myoclonic epilepsy. The mean delay in diagnosis was 19.8 years (calculated on 14 patients).
Of the 30 individuals enrolled in the genetic study, 14 were affected (III-4, III-8, III-16, III-20, III-22, III-24, III-27, III-34, IV-3, IV-10, IV-13, IV-21, IV-25, IV-26) 10 unaffected (III-1, III-30, III-32, IV-2, IV-12, IV-16, IV-27 and the three spouses III-9, III-25, and III-33), and 6 of uncertain phenotype. We considered uncertain the cases with a negative or doubtful anamnesis and NE and without a polygraphic study available (III-28, III-29, IV-6, IV-22). A member of the last generation (V-1) was too young to define her clinical status with certainty. Another individual (IV-17) had a history of febrile seizures at the age of 2 years, and from 18 to 19 years of age experienced several prolonged episodes of diffuse, violent jerking, without consciousness impairment. His events were previously interpreted as nonepileptic. Our NE did not disclose CMT, and the EEG was negative. We could not perform further neurophysiologic investigations and considered this case uncertain.
A polygraphic study was conducted in 15 patients (eight studied in loco); 11 were affected. Surface EMG showed bursts of pseudorhythmic myoclonias at 7–15 Hz, without apparent cortical correlates in all the affected cases (Fig. 2). However, back-averaging of the EEG time-locked to myoclonus onset in the rectified first dorsal interosseus (FDI) EMG demonstrated a clear premyoclonic wavelet on the contralateral hemisphere (Fig. 3A,B), suggestive of a cortical origin of the myoclonus in all patients tested (III-20, IV-10, IV-21). The time course of topographic scalp voltage distribution showed that a central positivity was evident, peaking between 40 and 20 msec before the myoclonus, but the Global Field Power maximum was reached between −10 and −5 msec, corresponding to a centroparietal negativity with a frontopolar positivity (Fig. 3C,D). This configuration suggests that the cortical generator is likely equivalent to a tangentially oriented dipole with the center of mass located in the sensorimotor region, thus implying an involved cortex lying within a vertically oriented sulcus.
Moreover, high-amplitude somatosensory evoked potentials (g-SEPs) were recorded in two of three patients tested, and long latency EMG responses (C-reflex) were recognized in the abductor pollicis brevis (APB) muscle after stimulation of the median nerve in all the three patients tested. TMS was performed in only one patient (IV-21), showing a mild reduction of rest motor threshold.
Interictal EEG showed generalized epileptiform abnormalities (diffuse spike/polyspike-wave discharges) in 4 of 11 affected members tested (25% of the total). The EEG was normal in three cases, whereas it showed frequent runs of diffuse slow activity with intermixed epileptiform activity in four patients (Fig. 4). A photoparoxysmal response (grade 1–4 according Waltz classification) was detected in five patients (31%).
Initial linkage analysis yielded significant LOD scores (>3) only for the FCMTE2 locus (Table S1). Analysis of the extended pedigree confirmed linkage to this locus with maximum location scores >6.0 for six consecutive microsatellite markers from D2S2181 to D2S2264 (Table S2). Haplotyping highlighted that these markers segregated consistently with the affection status in the whole family, and formed a disease-linked haplotype shared by all affected family members, but none of their unaffected relatives. Of the individuals with uncertain phenotype, only three (III-29, IV-6, IV-22) carried this haplotype. The identified critical haplotype narrows the minimal FCMTE2 interval to 10.4 Mb (13.5 cM), excluding 2.9 Mb containing 47 RefSeq genes (http://www.ncbi.nlm.nih.gov/RefSeq) with respect to the previously published candidate region (Crompton et al., 2012). Comparison of the disease-linked haplotype with the genotypes obtained from affected members of the families reported by Madia et al. was consistent with an extended common haplotype ranging from marker D2S2333 to marker D2S160, thus confirming evidence of a founder effect in the Naples area (Fig. 5).
We describe a large new family originating from Southern Italy. The genetic study established linkage to the FCMTE2 locus.
To date, nine families have been linked to FCMTE2 locus (Guerrini et al., 2001; Labauge et al., 2002; De Falco et al., 2003; Striano et al., 2004, 2005; Madia et al., 2008; Suppa et al., 2009; Crompton et al., 2012) and most of them are from Italy. The phenotype of these families is heterogeneous, encompassing families with a severe form (Guerrini et al., 2001) and families showing a milder phenotype with a lower rate of GTCS (Crompton et al., 2012). However, the clinical features of the pedigrees from the province of Naples, sharing the same founder haplotype on chromosome 2, are homogeneous. A recent follow-up study of three of these pedigrees with “classical” FCMTE disclosed a gradual progression of myoclonus severity and mild cognitive impairment with age, together with a slowing of background activity on EEG (Coppola et al., 2011).
Our family originates from the province of Caserta, near Naples. This suggested a common genetic background with the other five families sharing the same FCMTE2 locus. The hypothesis was also supported by the clinical overlap in terms of GTCS frequency (50% vs. 54.8%), age at cortical myoclonus onset (range 19–40 vs. 11–40 years), progression of symptoms, and comorbidities.
As in our family, four patients aged >51 years complained of a slow worsening of the CMT. Moreover, one of the most prominent clinical features in this family was the high recurrence of psychiatric disorders (37.5%), namely anxiety. A high rate (42.8%) of psychiatric manifestations, even if milder, was also disclosed in the other three families from Southern Italy (Coppola et al., 2011), suggesting that this could be a clinical hallmark of FCMTE families from this area. Schizophrenia has also been reported in the large Chinese pedigree unlinked to FCMTE1 and FCMTE2, but only in two unaffected members (Deng et al., 2005).
Of interest, in our family the concomitance of psychiatric symptoms with mild and isolated CMT was one of the major factors responsible for the delay in diagnosis. Given the tremor fluctuations related to situational stress, CMT was frequently underestimated by the patients themselves, and considered a manifestation of the anxiety state rather than an organic disease per se. Moreover, none of the patients were being treated with beta-blockers or primidone, suggesting that the diagnosis of essential tremor was not considered.
The presence of epileptic seizures could also be misleading. If considered in isolation, the epileptic phenotypic trait shows a lower penetrance, and an autosomal dominant pattern of inheritance can be missed. Moreover, VPA is the first choice drug for these patients given the rare GTCS often provoked by photic stimulation. If not carefully investigated, the mild postural tremor can be erroneously attributed to the antiepileptic medication.
Despite the relatively large number of FCMTE families linked to the 2p11.1-q12.2 locus (10 families including the present one), recombinations reducing the critical region were rarely observed. This is attributable to the presence of the centromere within the locus, as previously pointed out by different authors (Madia et al., 2008; Crompton et al., 2012). The chromosomal region identified by the disease-linked haplotype in the present family narrows the FCMTE2 candidate interval to 10.4 Mb containing 59 RefSeq genes. So far efforts to identify FCMTE2 causative mutations by candidate gene analysis have been frustrated (Saint-Martin et al., 2008). Currently, Next Generation Sequencing (NGS) approaches, such as whole exome or targeted genome sequencing, make it easier to explore large genomic regions containing hundreds of genes, but interpretation of NGS results can still be problematic, especially in dominant disorders. On the one hand, exome sequencing typically leads to the identification of a single candidate mutation within a linkage peak (Rabbani et al., 2012), but the intrinsic technological limitations of this technology such as incomplete exonic capture or preferential capture of one of the two alleles can prevent detection of the disease-causing mutation. On the other hand, targeted resequencing of the whole genomic region might disclose additional candidate mutations in noncoding sequences, which are difficult to discriminate from neutral and deleterious variants. Therefore, fine genetic mapping is still crucial to shorten the list of candidate genes that should be scanned for mutations. For this reason, the exclusion of 47 positional candidates by the genetic study of the present family will be helpful in future investigations of the FCMTE2 locus by NGS. Finally, we provided evidence that the present family descends from the same common ancestor of the previously reported families from the province of Naples showing a common founder haplotype (Madia et al., 2008). This also will have implications in future gene-mapping efforts, since it confirms that all the Italian families originating from this area most likely share the same causative mutation.
The present study describes a large, new Italian family with FCMTE linked to chromosome 2.
Comparison of genetic data from the present family and from families reported previously by Madia et al. confirmed the presence of an extended founder haplotype, as supposed from the common geographic origin and the clinical similarities. The presence of psychiatric disorders with mild CMT led to a diagnosis delay in our patients, suggesting that although this entity is rare, it could be underestimated. The further refinement of the critical FCMTE2 interval will direct efforts for gene identification toward a restricted number of genes.
We thank Professor SF Berkovic, Prof. I Scheffer, Dr. Mark Corbett, and Dr. J. Gecz for their valuable collaboration and exchange of views. We also thank Dr. Patrizia Avoni, Dr. Margherita Santucci, and Prof. Antonia Parmeggiani for their contribution to the data review. Among our technicians, particular thanks are due to Dr. Monica Lucchi for performing the SEP and C-reflex studies. We are particularly grateful to Dr. Zara, Dr. Madia, and all their collaborators for providing us the DNA of individuals belonging to the five FCMTE families from the province of Naples studied in Madia et al. (2008). Moreover, thanks to Silverio Perrotta and Saverio Scianguetta for providing DNA of the 50 control individuals from Naples. We acknowledge the Italian Ministry of Health, Young Investigators Award, Project GR-2009-1574072 for financial support.
None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.