This scientific commentary refers to ‘Motor co-activation in siblings of patients with juvenile myoclonic epilepsy: an imaging endophenotype?’, by Wandschneider et al. (doi: 10.1093/brain/awu175).

In this issue of Brain, Wandschneider and colleagues (2014) report that relatives of patients with juvenile myoclonic epilepsy (JME) show changes in functional MRI, indicating altered motor system activation and functional connectivity, similar to those seen in patients themselves (Vollmar et al., 2011). This study reflects the maturation of the field of imaging, particularly the use of advanced functional MRI analysis. Why is this of interest? We see two major potential contributions of these data; first, the concept of an imaging endophenotype and its specificity and, second, further understanding of the fundamental disease mechanism.

Complex disorders and imaging endophenotypes

Juvenile myoclonic epilepsy is an important epilepsy syndrome because it is common and easily treatable, yet often misdiagnosed. It belongs to the family of generalized epilepsies, formally designated as ‘idiopathic’ or ‘primary’, but now referred to as genetic generalized epilepsies because of the strong evidence of genetic aetiology (Berg et al., 2010), although the specific genetic determinants remain largely elusive. A few very rare families segregate an autosomal dominant gene, the best documented cases being mutations in the GABRA1 gene, which encodes a GABA receptor subunit. Clinical genetic evidence derived from family studies and twin analyses strongly supports the contention that juvenile myoclonic epilepsy is a complex disorder with multiple genetic components (see Helbig et al., 2008 and references therein). Deep sequencing has not yet identified variants that recur with sufficient frequency to be detected with the sample sizes studied (Heinzen et al., 2012) and it does seem that juvenile myoclonic epilepsy is genetically very heterogeneous.

Identifying genes in complex disorders is a major current challenge across all diseases and the concept of an endophenotype, first introduced in the psychiatric genetic literature, is extremely enticing. Here a subclinical phenomenon is sought that segregates in a simpler and hopefully Mendelian fashion within families, even though the disorder itself is polygenic. Thus the endophenotype is used as a clue to identify the underlying genes. Although this concept is not new, and putative endophenotypes have been looked at in many studies, success in identifying relevant disease genes has been limited.

The study reported here attempts to bring the sophistication of endophenotyping to a new level, using cutting-edge MRI techniques. The authors present persuasive evidence that changes similar to those in patients can be seen in unaffected relatives at the group level. However, a general problem with the practical application of an endophenotye is that, for genetic studies, categorization of individuals into two groups—those ‘with’ or ‘without’ the disease/endophenotype—is necessary. Where one is dealing with ‘softer’ markers, be they psychological traits, EEG changes, or functional MRI findings, strict categorical separation of individuals is difficult. Indeed, in this study no differences were shown between treated patients and unaffected first-degree relatives; this is not the way an endophenotype should behave; rather, an ‘intermediate phenotype’ should be identifiable. Of course, this problem might be clarified by studying untreated cases or much larger sample sizes with more sophisticated statistical analyses.

Specificity

Although this study adds to knowledge about the aberrant motor network in patients with juvenile myoclonic epilepsy and their family members, an unanswered question is whether these changes would be found in other forms of genetic generalized epilepsy, such as childhood absence epilepsy and juvenile absence epilepsy. Other electrophysiological measures of the motor system such as motor cortex excitability measured by transcranial magnetic stimulation show effects in all forms of genetic generalized epilepsy, but the effects are greater in juvenile myoclonic epilepsy; changes may also be seen in unaffected relatives (Badawy et al., 2013). Indeed, clinical genetic studies suggest that while juvenile myoclonic epilepsy is related to other generalized epilepsies, there are genetic differences; the relatives in multiplex families with juvenile myoclonic epilepsy probands show varying phenotypes, although the proportion with juvenile myoclonic epilepsy is enriched (Winawer et al., 2005). Therefore, before reaching conclusions on the specificity of these findings to juvenile myoclonic epilepsy, it would be useful to explore this functional MRI paradigm in individuals with non-juvenile myoclonic epilepsy generalized genetic epilepsy. We suspect that abnormalities would be found in all, suggesting that this is a measure of the broad genetic generalized epilepsy trait, rather than being specific to juvenile myoclonic epilepsy.

Insights into mechanism

The key observation here—that of abnormal connections between large-scale intrinsic brain networks—provides a mechanism of atypical network behaviour fundamental to generalized epilepsy. Which networks are hyperconnected may go a long way to explaining the phenotypic variation of the genetic generalized epilepsies. This study demonstrates abnormal co-activation (rather than mutual inhibition) between major networks and suggests that this is an underlying precursor mechanism of juvenile myoclonic epilepsy present in unaffected first-degree relatives as well as patients themselves.

This atypical connectivity between large-scale networks has been demonstrated by functional MRI in patients with juvenile myoclonic epilepsy (Vollmar et al., 2011), childhood absence epilepsy (Masterton et al., 2012), as well as the more severe generalized epilepsy of Lennox-Gastaut syndrome (Pillay et al., 2013), suggesting that abnormal network connections are fundamental to epileptogenesis.

Genetic generalized epilepsies are therefore likely to be a manifestation of primarily abnormal interaction or simultaneous engagement of large-scale networks in this synergistic rather than mutually inhibiting manner. This atypical large network behaviour may also underlie the subtle cognitive deficits seen in these cohorts and in relatives without epilepsy.

The human connectome and functional connectomics

Recent developments in understanding the structure and function of the human brain at the level of large-scale networks suggest that the brain has ‘small world’ characteristics with a balance between highly efficient ‘neighbourhood’ connections and high cost long-range connections between ‘rich club’ nodes (Sporns, 2011). Functionally, there seem to be a limited number of stable coactivating networks that exist even in the resting state (Smith et al., 2013). This provides a framework for understanding the juvenile myoclonic epilepsy data presented here, i.e. abnormal correlated behaviour between two of these networks, presumably arising because of altered behaviour of one of the highly connected ‘rich club’ nodes that link these networks. This may be a fundamental mechanism underlying the generalized epilepsies, and these data suggest that there may be a genetic predisposition for this abnormal mode of function.

Such studies in diseases like epilepsy will benefit from further theoretical developments and the ability to apply the results of large-scale connectomics projects that are currently underway—the European Human Brain Project (https://www.humanbrainproject.eu), the NIH funded Human Connectome Project (http://www.humanconnectomeproject.org) and the BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative (http://www.nih.gov/science/brain).

Idiopathic no more!

Based on cellular physiology in experimental animals, Gloor (1979) showed that while the intrinsic behaviour of neurons in generalized epilepsy is normal, the abnormality is in the way they oscillate and interact. Many years later, and at a larger scale, the data presented in juvenile myoclonic epilepsy suggest that abnormal interactions between networks, in this case frontal and motor networks, are characteristic of generalized epilepsy in humans. Thus, underscoring the new terminology (Berg et al., 2010), the old term ‘idiopathic’ is now doubly inappropriate—we have made great strides in understanding both the genetic underpinnings and the abnormal network behaviour of this important class of human epilepsy.

Funding

Work of both authors is supported by an NHMRC Program Grant (#400121). G.J is also supported by an NHMRC Practitioner Fellowship (#527800) and by a Victorian Government Operational Infrastructure Support Grant.

References

Badawy
RA
Vogrin
SJ
Lai
A
Cook
MJ
Capturing the epileptic trait: cortical excitability measures in patients and their unaffected siblings
Brain
 , 
2013
, vol. 
136
 (pg. 
1177
-
91
)
Berg
AT
Berkovic
SF
Brodie
MJ
Buchhalter
J
Cross
JH
van Emde Boas
W
, et al.  . 
Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009
Epilepsia
 , 
2010
, vol. 
51
 (pg. 
676
-
85
)
Gloor
P
Generalized epilepsy with spike-and-wave discharge: a reinterpretation of its electrographic and clinical manifestations. The 1977 William G. Lennox Lecture, American Epilepsy Society
Epilepsia
 , 
1979
, vol. 
20
 (pg. 
571
-
88
)
Heinzen
EL
Depondt
C
Cavalleri
GL
Ruzzo
EK
Walley
NM
Need
AC
, et al.  . 
Exome sequencing followed by large-scale genotyping fails to identify single rare variants of large effect in idiopathic generalized epilepsy
Am J Hum Genet
 , 
2012
, vol. 
91
 (pg. 
293
-
302
)
Helbig
I
Scheffer
IE
Mulley
JC
Berkovic
SF
Navigating the channels and beyond: unravelling the genetics of the epilepsies
Lancet Neurol
 , 
2008
, vol. 
7
 (pg. 
231
-
45
)
Masterton
RA
Carney
PW
Jackson
GD
Cortical and thalamic resting-state functional connectivity is altered in childhood absence epilepsy
Epilepsy Res
 , 
2012
, vol. 
99
 (pg. 
327
-
34
)
Pillay
N
Archer
JS
Badawy
RA
Flanagan
DF
Berkovic
SF
Jackson
G
Networks underlying paroxysmal fast activity and slow spike and wave in Lennox-Gastaut syndrome
Neurology
 , 
2013
, vol. 
81
 (pg. 
665
-
73
)
Smith
SM
Vidaurre
D
Beckmann
CF
Glasser
MF
Jenkinson
M
Miller
KL
, et al.  . 
Functional connectomics from resting state fMRI
Trends Cogn Sci
 , 
2013
, vol. 
17
 (pg. 
666
-
82
)
Sporns
O
Networks of the brain
 , 
2011
Cambridge, MA
MIT Press
Vollmar
C
O’Muircheartaigh
J
Barker
GJ
Symms
MR
Thompson
P
Kumari
V
, et al.  . 
Motor system hyperconnectivity in juvenile myoclonic epilepsy: a cognitive functional magnetic resonance imaging study
Brain
 , 
2011
, vol. 
134
 (pg. 
1710
-
9
)
Wandschneider
B
Centeno
M
Vollmar
C
Symms
M
Thompson
PJ
Duncan
JS
, et al.  . 
Motor co-activation in siblings of juvenile myoclonic epilepsy patients: an imaging endophenotype?
Brain
 , 
2014
, vol. 
137
 (pg. 
2469
-
79
)
Winawer
MR
Marini
C
Grinton
BE
Rabinowitz
D
Berkovic
SF
Scheffer
IE
, et al.  . 
Familial clustering of seizure types within the idiopathic generalized epilepsies
Neurology
 , 
2005
, vol. 
65
 (pg. 
523
-
8
)