Abstract
Ishiura H, Doi K, Mitsui J, Yoshimura J, Matsukawa MK, Fujiyama A, Toyoshima Y, Kakita A, Takahashi H, Suzuki Y, Sugano S, Qu W, Ichikawa K, Yurino H, Higasa K, Shibata S, Mitsue A, Tanaka M, Ichikawa Y, Takahashi Y, Date H, Matsukawa T, Kanda J, Nakamoto FK, Higashihara M, Abe K, Koike R, Sasagawa M, Kuroha Y, Hasegawa N, Kanesawa N, Kondo T, Hitomi T, Tada M, Takano H, Saito Y, Sanpei K, Onodera O, Nishizawa M, Nakamura M, Yasuda T, Sakiyama Y, Otsuka M, Ueki A, Kaida KI, Shimizu J1, Hanajima R, Hayashi T, Terao Y, Inomata-Terada S, Hamada M, Shirota Y, Kubota A, Ugawa Y, Koh K, Takiyama Y, Ohsawa-Yoshida N, Ishiura S, Yamasaki R, Tamaoka A, Akiyama H, Otsuki T, Sano A, Ikeda A, Goto J, Morishita S, Tsuji S. Nat Genet. 2018;50:581–590.
Epilepsy is a common neurological disorder, and mutations in genes encoding ion channels or neurotransmitter receptors are frequent causes of monogenic forms of epilepsy. Here we show that abnormal expansions of TTTCA and TTTTA repeats in intron 4 of SAMD12 cause benign adult familial myoclonic epilepsy (BAFME). Single-molecule, real-time sequencing of BAC clones and nanopore sequencing of genomic DNA identified two repeat configurations in SAMD12. Intriguingly, in two families with a clinical diagnosis of BAFME in which no repeat expansions in SAMD12 were observed, we identified similar expansions of TTTCA and TTTTA repeats in introns of TNRC6A and RAPGEF2, indicating that expansions of the same repeat motifs are involved in the pathogenesis of BAFME regardless of the genes in which the expanded repeats are located. This discovery that expansions of noncoding repeats lead to neuronal dysfunction responsible for myoclonic tremor and epilepsy extends the understanding of diseases with such repeat expansion.
Commentary
The impact of molecular genetics on our understanding of the cause of epilepsies over the last 23 years has been quite incredible. In 2018, approximately 50% of the patients with the most devastating group of epilepsies, the developmental and epileptic encephalopathies, have been solved with the identification of over 200 causative genes (1). Identified mutations (now known as pathogenic variants) are mostly located in the exonic protein–encoding regions of genes. When considering the epilepsies overall, however, known pathogenic variants have been identified in fewer than 10% of individuals. More recently, there have been large-scale efforts placed on gene discovery for the commoner epilepsies with the assistance of international consortia to gather sufficient numbers for genome-wide association studies and sequencing approaches involving tens of thousands of patients. The focal epilepsies, which make up roughly 60% of all epilepsies, have had approximately two dozen genes identified, while the generalized epilepsies, which account for 25% of the epilepsies, have had fewer than a dozen genes found.
One of the reasons for the lack of success in the genetic generalized epilepsies (GGE; also known as idiopathic generalized epilepsies) has been the presumed complex inheritance of this common group of disorders, where multiple genes play a role, potentially with a contribution from environmental factors. The classical GGE comprise the following four syndromes: childhood absence epilepsy, juvenile absence epilepsy, juvenile myoclonic epilepsy, and generalized tonic-clonic seizures (GTCS) alone. In the classical GGE, large families with many affected members are extremely rare with most affected individuals having no affected relatives or only a few. In contrast, gene discovery is far more straightforward in epilepsies following monogenic inheritance and most of the epilepsy genes identified to date have followed dominant, recessive, X-linked, or mitochondrial inheritance.
There is a rare autosomal dominant GGE, which has been the source of some frustration for molecular scientists trying to discover the causative gene, despite very large families with 50 or more affected individuals reported (2). This disease goes by many names including Familial Adult Myoclonic Epilepsy (FAME), Benign Adult Familial Myoclonic Epilepsy (BAFME), Familial Cortical Myoclonic Tremor with Epilepsy, Autosomal Dominant Cortical tremor, Myoclonus and Epilepsy, cortical tremor, Familial Cortical Myoclonic Tremor, Familial Cortical Tremor with Epilepsy, Familial Essential Myoclonus and Epilepsy, Familial benign Myoclonus Epilepsy of Adult onset, and Heredofamilial Tremor and Epilepsy (3).
FAME is a relatively mild GGE with two prominent symptoms: cortical myoclonic tremor and GTCS. Onset is in adolescence or early adult life and families differ considerably in the proportion of individuals who have GTCS, ranging from 15% to 100%. GTCS typically occur rarely. Tremor can be moderate to severe and may worsen with age, affecting an individual's daily functioning, such as carrying a glass of water or writing. It may involve the arms, legs, head, and voice. In some families, focal temporal lobe seizures also occur.
FAME was initially described in Japanese families and many were mapped to chromosome 8q24 (4, 5). The second major locus has been the Italian locus (6) at chromosome 2p11.2-q11.2, to which most families of European ancestry map. More recently there have been two further loci reported, chromosome 5p15.31-p15.1 and 3q26.32-3q28 (7), with gene mutations identified in CTNND2 (8) in some, but not all, families linked to the 5p region. With the lack of progress in finding the causative gene in most FAME families, a large international group has merged data to narrow the region at the chromosome 2 locus (9). Substantial combined efforts have involved combing up and down the mapped regions to find the pathogenic variant, with the major focus on exonic regions, without success.
Enter the paper by Ishiura and colleagues. Their findings are transformative, providing our first major insights into the hidden gems in the intronic regions of genes. Using whole genome sequence analysis, they show that repeat expansions in the introns cause FAME and that the nature of these expansions is critical, irrespective of the genes in which they are found. Repeat expansions are not novel; triplet repeat expansions are well known as the molecular mechanism underlying Fragile X syndrome, myotonic dystrophy, and Huntington's disease. What is novel is that FAME is caused by expansion of pentanucleotide repeats in the intron and that specific repeats, TTTCA and TTTTA, are the pathophysiological key to FAME.
A large Japanese consortium identified these pentanucleotide expansions in 85 individuals from 49 families. The expansions were both located in noncoding intron 4 of SAMD12, which encodes sterile alpha-motif domain-containing 12. The expansions of TTTCA and TTTTA varied in size from 2.2 to 18.4 kb, corresponding to 440 to 3680 repeat units in affected individuals. Interestingly, TTTCA repeat expansions were not observed in 1000 control individuals; whereas, 59 of 1000 (5.9%) control individuals had TTTTA repeat expansions. These were 0.5 to 1.5 kb (100–300 repeat units) in size in 28 controls and the remaining 31 probably had longer, undetermined (due to technologic problems) expansions. The expansion size varied in different tissues in autopsied cases, showing somatic instability. Fascinating clinic-molecular insights showed that the length of the repeat inversely correlated with age of onset of epilepsy and myoclonic tremor. Four patients from three families had homozygous expansions; their age of onset was younger than in individuals with heterozygous expansions of similar size, they had a more severe phenotype with cognitive decline and cerebral atrophy after 60 years of age. The length of the repeats showed intergenerational instability, consistent with clinical anticipation. Neuropathologic studies in autopsied brains from six patients were compared with five controls. They showed that these expansions lead to pathogenic RNA foci in the brain due to the accumulation of altered repeat motifs, and suggest that this functional impairment leads to cortical hyperexcitability rather than neurodegeneration, in contrast with other diseases caused by noncoding repeat expansions.
Even more intriguingly, the group then identified the same molecular cause in two of three remaining unsolved families in different genes. They found the same pentanucleotide expansions in genes TNRC6A and RAPGEF2 in these families. The authors invoke that it is the pentanucleotide intronic expansions that lead to FAME, rather than the genes in which they reside.
This discovery of intronic molecular causes of human epilepsy finally unlocks the noncoding DNA puzzle. It is likely that similar pentanucleotide repeats account for other FAME loci; time will tell. But what about the many other epilepsies that remain genetically unsolved? Noncoding DNA will hold the answer to a considerable proportion of the remaining 50% of the unsolved developmental and epileptic encephalopathies. Perhaps just as importantly, both the intronic location and the mechanisms caused by abnormal repeat expansions, will be relevant to the milder self-limited epilepsies, such as GGE and focal epilepsies. These fascinating molecular insights highlight critical ways forward in understanding the molecular pathogenesis of the epilepsies.