John Libbey Eurotext

Molecular genetics of the idiopathic epilepsies: the next steps Volume 2, numéro 4, Décembre 2000

Progress to date

Currently there are six genes established as causative for idiopathic epilepsy syndromes (table) and all code for ion channel subunits [1, 3-8]. In addition, a number of other syndromes have been mapped, but the specific genetic defects await identification [9]. These discoveries have shown that the idiopathic epilepsies are, at least in part, a family of channelopathies. This fundamental observation is driving the search for other channel genes as well as deepening our understanding of the pathophysiology of the idiopathic epilepsies. Secondly, as can be seen from the table, genetic heterogeneity is the rule - that is, clinically identical syndromes may be caused by abnormalities in different (albeit related) genes.

It has also become clear that a single genetic defect may contribute to a wide variety of phenotypes. Such phenotypic variability is best illustrated in the syndrome of generalized epilepsy with febrile seizures plus, where sodium channel mutations in the one family may be associated with simple febrile seizures, febrile seizures going on to later afebrile attacks (febrile seizures plus) or more severe phenotypes. The explanation of this phenotypic variability is not known but is probably due to modifying genes, with environmental factors possibly also playing a role [10, 11].

The road ahead

Less satisfying has been the realisation that the discoveries over the last six years have been largely confined to uncommon epilepsies that show single gene inheritance. Success has come from mapping large pedigrees and then proceeding to gene identification by positional cloning, or the positional candidate approach. Although there are undoubtedly more monogenetic epilepsies to discover, most of the genetic contributions to the epilepsies are likely to be complex, involving multiple, possibly interacting, genes, with or without environmental factors [9].

Mapping and identifying genes in epilepsies with complex inheritance is much more challenging. Large families with the common syndromes are very rare, if they exist at all, and other strategies must be used such as sib-pair analysis, association studies, etc. To date, the results of these studies have not been definitive. Whilst a number of linkages have been reported, most have not been consistently replicated and no genes have been firmly associated with the idiopathic epilepsies [9]. There have been a few reports of mutations in certain channel genes in occasional cases, but the pathogenic relevance of these finding is unclear [12, 13]. The rapid progress of the human genome project, the ability to do much more rapid mutation screening, and improved bioinformatic methodology, will help to solve the problems encountered in studies of common diseases with complex inheritance, like the majority of the idiopathic epilepsies.

The second issue which researchers must address in the next few years is the "big picture" of the genetics of epilepsies in the community. This will require integration of epidemiological techniques in the loop of clinical phenotyping and molecular genetics (figure). For example, epidemiological studies are needed to assess the prevalence and phenotypic spectrum of newly identified mutations. Currently, little is known about the proportion of patients with particular epilepsy syndromes who have mutations in the epilepsy genes identified so far (table). Epidemiological studies designed to test genetic hypotheses are challenging because of the large sample sizes they require. Molecular analyses can be very expensive, and the accuracy and detail of clinical diagnoses are generally not comparable to those in specifically targeted families that are intensively investigated for genetic analysis. Nonetheless, major insights into the inheritance of the epilepsies have come from genetic epidemiological studies. These include the importance of genetic factors in partial as well as generalized epilepsies, the genetic relationships of partial and generalized epilepsies and appreciation of the interaction between genetic and acquired factors. Genetic epidemiological data provide estimates of the risks to family members, and have demonstrated the puzzling "maternal effect", involving higher risks of epilepsy in offspring of affected mothers than in offspring of affected fathers (for review see [14]).

It is with this background that the study of Doose et al. (this issue) on a cohort of children with benign neonatal epilepsies should be viewed. Neonatal epilepsies may have a genetic contribution. An autosomal dominant form, benign familial neonatal convulsions (BFNC) was the first idiopathic epilepsy to be mapped [15], and subsequently two potassium channel genes (KCNQ2, KCNQ3) were shown to be responsible for the majority of families (table). Although BFNC is said to be rare, its incidence in the population is not known, and it is probably under recognized. However, additional genes are likely to also contribute to neonatal epilepsies. It is not yet clear whether mutations in KCNQ2 and KCNQ3 may contribute to cases where there is no family history, but preliminary evidence suggests that they have a relatively small role. Unfortunately, analysis of these genes was not reported in the study of Doose et al.

The relationship of benign neonatal convulsions to later epilepsies has not been well studied until now. Although cases of neonatal convulsions evolving into benign partial epilepsies have been reported, Doose et al. show that this is a not an infrequent occurrence. In addition, there appears to be a relationship between benign neonatal epilepsies and febrile seizures, although this requires further exploration because of the high population frequency of febrile convulsions. Nonetheless, the data do suggest that the benign neonatal epilepsies may share some genetic factors with the benign partial epilepsies of childhood. Moreover, this study highlights the need for rigorous genetic epidemiological studies of the epilepsies to understand the relationships of various syndromes. Such studies will provide clues for molecular genetic approaches and deepen our understanding of the phenotypic spectrum of particular epilepsies.

Our challenge is to understand the molecular genetics of the common epilepsies. Progress will require close interaction between epileptologists expert in phenotyping, genetic epidemiologists, and molecular geneticists. The road ahead will be bumpy, but the route can now be seen!