Download The role of nicotinic acetylcholine receptors and GABAergic

The role of nicotinic acetylcholine receptors
in
Autosomal Dominant Nocturnal Frontal Lobe
Epilepsy
A critical review
Lili Guo
29-06-2009
Table of Contents
ABSTRACT ........................................................................................................................................... 3
INTRODUCTION ................................................................................................................................. 3
Chapter 1: CLINICAL FEATURES................................................................................................... 5
1.1 Symptoms ................................................................................................................................................ 5
1.2 Neuroimaging and EEG ....................................................................................................................... 5
1.3 Seizure origin ......................................................................................................................................... 6
1.4 Comorbidity ............................................................................................................................................ 6
1.5 Diagnosis .................................................................................................................................................. 6
1.6 Treatment ................................................................................................................................................ 6
1.6.1 Carbamazepine (CBZ)...................................................................................................................................7
1.6.2 Nicotine ..............................................................................................................................................................7
1.6.3 Cannabinoids ...................................................................................................................................................8
Chapter 2: GENETICS ....................................................................................................................... 8
2.1 ADNFLE loci ............................................................................................................................................. 8
2.2 nAChRs in epilepsy syndromes ........................................................................................................ 9
2.3 Genetic heterogeneity ......................................................................................................................... 9
Chapter 3: nAChRs ......................................................................................................................... 10
3.1 Structure ............................................................................................................................................... 10
3.2 Localization and function................................................................................................................ 12
3.3 Acetylcholine innervation .............................................................................................................. 12
3.4 nAChR mutations ............................................................................................................................... 13
3.5 Implications for dopamine system .............................................................................................. 14
Chapter 4: ANIMAL MODELS ...................................................................................................... 16
4.1 α4 L9’S and L9’A knock-in mice .................................................................................................... 17
4.2 α4 S248F mice ..................................................................................................................................... 17
4.3 α4 776ins3 and α4 S248F mice ..................................................................................................... 18
4.4 2 V287L transgenic mice ............................................................................................................... 18
4.5 Rat S248L model................................................................................................................................. 19
Chapter 5: CURRENT HYPOTHESES ......................................................................................... 20
5.1 Hypersensitivity ................................................................................................................................. 20
5.2 GABA ....................................................................................................................................................... 21
5.3 Stage II non-REM sleep ..................................................................................................................... 23
5.4 Frontal origin ...................................................................................................................................... 24
5.5 Developmental defects .................................................................................................................... 24
Chapter 6: DISCUSSION & CONCLUSION ................................................................................. 25
REFERENCES .................................................................................................................................... 27
2
ABSTRACT
Autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE), a rare familial form of idiopathic
partial epilepsy, is characterized by short nocturnal seizure episodes, occurring in stage II of the nonREM sleep. The motor attacks during the seizures and the high heritability of this disorder makes
ADNFLE an intriguing topic to study. The seizures are believed to mainly originate in the frontal lobe.
The onset of the disorder is often during early childhood or adolescence. However, due to the
nocturnal onset of the symptoms and diagnostic difficulties, ADNFLE may be diagnosed many years
after the actual disease onset. As the first epileptic disorder to have an identified genetic cause,
ADNFLE is linked to genetic defects in the α4 and β2 subunit of nicotinic acetylcholine receptors
(nAChR) in over hundred affected families. The mutations identified to date show a common
property of hypersensitivity to ligand binding. More insights into the disease mechanism have been
recently provided by animal models. In the developed mouse and rat models harboring
hypersensitive nAChR mutations, important findings have been reported, including evidence for
excessive glutamate transmission, developmental defects, and the paradoxal role for increased
GABAergic transmission. The surprising role for GABA is probably an additional contributor in a
subset of the patients.
The detailed contribution of these finding to ADNFLE pathogenesis is still poorly understood. In
addition to the further research to find the consequences of developmental defects and cause of
excessive glutamate transmission, more studies should be focused on the role of nicotine exposure
during development and low doses of nicotine as a potent therapy. Nicotine has already been shown
to be effective in a few patients, and more data on the therapeutic effects in relation to specific
mutations are needed to establish nicotine as a new treatment and successor of carbamazepine.
3
INTRODUCTION
Epilepsy is a common neurological disorder, affecting approximately 0.5-1% of the world population
(Hirtz et al., 2007). This disorder is characterized by the occurrence of seizures, which can manifest in
different symptoms, such as wild involuntary movements or absences. In general, epilepsy syndromes
are classified into two groups based on the origin of seizures: partial (or focal) and generalized
epilepsy. Ictal activity, i.e. the aberrant hyperactivity of neuronal networks during epileptic seizure,
spreads over the entire brain in generalized epilepsy. In partial or focal epilepsy, this aberrant activity
is limited to certain brain areas. The cause of epilepsy syndromes is referred to as symptomatic and
idiopathic. For a symptomatic epileptic disorder, the underlying cause is a brain abnormality that has
been identified. Known causes include brain damage by trauma, strokes, infections, sclerosis, and
developmental defects. In idiopathic epilepsy, there seems to be no apparent cause of the seizures. No
apparent structural brain abnormalities or other functional dysfunction is found to underlie the seizure
phenotype (Kandel, 2004). One of the idiopathic epilepsy syndromes is autosomal dominant nocturnal
frontal lobe epilepsy (ADNFLE).
ADNFLE is a rare idiopathic epilepsy, which has been diagnosed in over a hundred families
worldwide. This form of familial partial epilepsy had not been described until 1994, and is
characterized by brief motor seizures during sleep. The exclusive occurrence of attacks during sleep
suggests a role for sleep-related activity and circuits in the disease mechanism of ADNFLE. However,
the link between sleep and seizure generation is still poorly studied and understood. Sleep is composed
of cyclic patterns of two phases, referred to as REM (Rapid Eye Movement) and non-REM sleep. The
nocturnal seizures of ADNFLE patients occur mostly in stage II of non-REM sleep. When falling
asleep, transition into the non-REM sleep is the first event to occur. The transition is accompanied by
low neuronal activity and lower EEG frequency. Body temperature and metabolism decrease to their
lowest levels. However, some muscle tone is retained. Non-REM sleep is divided into four stages. The
first stage is the initial transition into the sleeping state, with a duration of only a few minutes. Stage II
non-REM sleep is characterized by 12-14 Hz sleep spindles and K-complexes, which have higher
amplitudes in their EEG patterns. Stages III and IV show delta waves with low frequency (0.5-2 Hz)
and high amplitude, and are therefore also called slow-wave sleep. A phase of non-REM sleep is
followed by REM sleep, which is characterized by rapid eye movements and the absence of muscle
tone. In contrast to the low neuronal activity during non-REM sleep, EEG recordings of REM sleep
show increased neuronal activity and higher frequency EEG patterns. Body temperature and
metabolism rise, but the muscle activity is completely inhibited. Progression of sleep through a cycle
of non-REM (including all four stages) and REM sleep occupies approximately 90-110 minutes, and
this cycle is repeated throughout the sleeping period.
A mutation in the CHRNA4 gene in an ADNFLE family was the first gene identified to underlie
idiopathic epilepsy. To date, several mutations have been found in the CHRNA4 and CHRNB2 gene,
coding for the α4 and β2 subunits of the nicotinic acetylcholine receptors (nAChR), respectively. α4β2
nACh receptors are the major neuronal types of nAChR. The known mutations share one common
feature, they all increase the sensitivity of nAChR for the ligands acetylcholine and nicotine. However,
the contribution of these mutations to the pathogenesis of nocturnal motor seizures remains to be
elucidated (Steinlein, 2007). Identification of the link between mutated nAChR and epileptogenesis is
difficult since ADNFLE is not longer believed to be a Mendelian trait. Overall, the pattern of
inheritance in the affected families suggests an autosomal dominant transmission with reduced
penetrance. Some families show complete penetrance of the trait (100%), whereas other families show
reduced penetrance (29-87%). However, more recent data point to a greater heterogeneity in the
genetics of ADNFLE, involving possibly more than one affected gene (Combi et al., 2004).
In the current review, the clinical features of ADNFLE will be summarized. Available data on the
genetics, nAChR mutations and animal models will be discussed and evaluated, leading to interesting
hypotheses on the possible mechanism of this epileptic disorder. In addition, data is provided to
support the application of nicotine and a new, effective treatment. The dual role of GABA in
developmental and mature central nervous system will be explored, in an effort to understand and
emphasize the complex role of GABA in the excitatory and inhibitory circuits and balance. To
conclude, directions for research in the near future will be pointed out, which can be used for the
deeper understanding of disease mechanism and the establishment of new, reliable therapy.
Chapter 1: CLINICAL FEATURES
The onset of ADNFLE is usually during childhood, before the age of 14 in most patients. However,
older patients have also been reported, up to the age of 52 (Diaz-Otero et al., 2008). The severity and
frequency of seizures appear to decline during later stages of life. The incidence of this disorder is
unknown, partly because of the diagnostic difficulties and unfamiliarity for ADNFLE. To date, over
hundred families have been reported worldwide (Combi et al., 2004). The clinical manifestations in
ADNFLE patients are usually heterogeneous. Many intra- and interfamilial differences have been
observed in the symptom severity, frequency and drug resistance. The most frequently described
symptoms and clinical findings are discussed below.
1.1 Symptoms
The nocturnal seizures of ADNFLE patients are characterized by various types of complex motor
attacks, mainly occurring during stage II of non-REM sleep. These motor symptoms have been
divided into three clinically defined groups: paroxysmal arousals (PA), nocturnal paroxysmal dystonia
(NPD), and episodic nocturnal wandering (ENW). Paroxysmal arousals are characterized by short
episodes (<20 seconds), in which sudden opening of the eyes, head movements, and elevation of the
trunk is observed. In some cases, these movements are accompanied by fearful face expression or
vocalization. Nocturnal paroxysmal dystonia involves longer episodes (20 seconds to 2 minutes), in
which the patients show dystonic posturing or rapid and complex movements of the whole body, and
vocalization. Episodic nocturnal wanderings refer to episodes of 1 to 3 minutes, in which
sleepwalking, screaming and bizarre movements can be observed (Combi et al., 2004). All three
groups of motor symptoms can manifest in the same patient, and the frequency and severity of
symptoms show high variability between patients.
1.2 Neuroimaging and EEG
MRI scans are usually normal in ADNFLE patients, without gross malformations or abnormalities.
Only a few patients from different families show atrophy or, very rarely, cysts in brain areas which do
not seem to be related to the clinical symptoms (Callenbach et al., 2003; Ryvlin, 2006). PET studies
have revealed decreased nAChR density and hypometabolism in the right dorsolateral prefrontal
cortex in five patients, possibly caused by cell loss which is induced by seizure activity (Picard et al.,
2006). In another patient, left mesial frontal hypometabolism was detected (Ryvlin, 2006).
EEG recordings show no interictal abnormalities in over 50% of the ADNFLE patients. Unfortunately,
movement artifacts due to the seizure onset disturb most ictal EEG recordings. Artifact-free recordings
have revealed the frontal origin in a proportion of the patients, and also established the main seizure
occurrence during stage II of non-REM sleep (Marini & Guerrini, 2007). The main EEG
manifestations of stage II non-REM sleep are K-complexes and sleep spindles. Sleep spindles are
oscillations of 11-15 Hz synchronizations, generated by neurons in the reticular nucleus of the
thalamus. Intracerebral EEG recordings have revealed a longer duration of sleep spindles just before
seizure onset in two patients, suggesting a role for abnormal sleep spindle activity in seizure onset
(Picard et al., 2007). K-complexes are slow <1 Hz oscillations that have a cortical origin (Parrino et
al., 2006), consisting of a positive peak between 350-550 ms and a negative peak at 900 ms (Cash et
al., 2009). K-complexes occur randomly during stage II of non-REM sleep, and can also be initiated
5
by sounds. They are proposed to suppress cortical activity and arousal, in order to maintain the
sleeping state of the brain so it does not respond to harmless stimuli (Cash et al., 2009). It has been
shown that the majority of seizures in one ADNFLE patient with a CHRNA4 mutation were
associated with K-complexes in EEG recordings. In this patient, seizure onset mainly occurred on the
descending slope of the K-complex, suggesting a role for K-complexes in the initiation of seizures (El
Helou et al., 2008).
1.3 Seizure origin
Initially, the ADNFLE seizures were believed to have a frontal origin because of their phenotypic
similarities with the observed attacks in orbitofrontal and mesial frontal epilepsy. This assumption has
been partly confirmed by EEG data. Furthermore, PET studies have shown modified metabolism in
the frontal regions. However, exclusively frontal abnormalities are not always observed in ADNFLE
patients, as shown by the inconclusive EEG data. Aberrant EEG waves in the frontal region have only
been recorded in 50% of the patients with abnormal ictal or interictal EEG (Combi et al., 2004).
Therefore, it is possible that ADNFLE seizure activity also involves other brain areas (Ryvlin et al.,
2006). In support of this hypothesis, ictal-onset EEG recordings in a few studied patients suggest an
insular or temporal lobe origin of the nocturnal hypermotor seizures (Picard et al., 2000; Ryvlin et al.,
2006; Ryvlin, 2006) instead of a exclusively frontal origin.
1.4 Comorbidity
Most patients have a normal IQ and do not suffer from comorbid disorders. However, in some
ADNFLE subjects, personality and behavioral disturbances, psychiatric and cognitive defects,
memory deficits, and mental retardation have been reported (Derry et al., 2008; Picard et al., 2000).
Remarkably, schizophrenia and other psychiatric symptoms were associated with the α4 776ins3
mutation in one ADNFLE family (Magnusson et al., 2003). Furthermore, the β2 I312M mutation has
been linked to memory impairments in two families (D. Bertrand et al., 2005; Cho et al., 2008). These
findings suggest that ADNFLE mutations in nAChR subunits may also affect other neuronal circuits
and processes, leading to comorbid psychological and cognitive disorders. It further indicates that the
nAChR subunits are pivotal for the normal functioning of the central nervous system.
1.5 Diagnosis
The correct diagnosis of ADNFLE is often difficult due to the lack of reliable EEG recordings, and the
phenotypic similarities of the attacks to parasomnias such as night terror. Therefore, many patients
have been misdiagnosed, or only diagnosed as ADNFLE years after the onset of symptoms. Currently,
patient history, EEG recordings, video observations, and polygraphic recordings such as muscle
activity and heart rate, are often used together to diagnose ADNFLE (Combi et al., 2004). This
combination of diagnostic methods and the increasing knowledge about this rare epilepsy disorder
leads to a more reliable and earlier ADNFLE diagnosis, which is needed to start the appropriate antiepileptic treatment. To date, ADNFLE is diagnosed mainly on the basis of patient history and clinical
observations. When the patient and witnesses report nocturnal motor seizures, the attacks are recorded
on video for deeper analysis. Based on the patient report of the symptoms and video recordings, other
disorders such as night terror and sleepwalking are excluded based on characteristics such as
frequency and onset timepoint of the attacks during sleep. Furthermore, a family history of nocturnal
seizures is needed for ADNFLE to be diagnosed. Abnormal ictal EEG recordings can further verify
the diagnosis but is not always applicable, since this is not present in half of the diagnosed patients
(Combi et al., 2004).
1.6 Treatment
Approximately two-third of the ADNFLE patients are responsive to treatment with anti-epileptic drugs
(AED), which suppress both seizure frequency and severity. Some patients even become seizure-free
by medication. The most effective and commonly prescribed AED in the treatment of ADNFLE
patients is carbamazepine (CBZ). Additional AED’s such as acetazolamide (ACZ) are ocasionally
used to obtain better seizure control (Varadkar et al., 2003). Valproate (VPA), which blocks the
6
transamination step of GABA breakdown and thereby increases GABA concentration, is ineffective in
ADNFLE patients (Picard et al., 1999). With a better understanding of the disease pathogenesis, other
new potential treatments may be proposed. Here, the effects of CBZ and the potential therapeutic
effects of nicotine and cannabinoids are discussed.
1.6.1 Carbamazepine (CBZ)
CBZ is a widely used anti-epileptic drug, which generally acts as a sodium channel blocker, thereby
suppressing the rate of action potentials. In addiction, CBZ has also been shown to act on calcium
channels, NMDA, GABA, and nACh receptors. It is able to inhibit nACh receptor function by acting
as a noncompetitive blocker. Therefore, it is therefore possible that direct actions of CBZ on nAChRs
could contribute to the high efficacy of this drug in the treatment of ADNFLE. In a dose-response
inhibition experiment, α4β2 nAChR with the 776ins3 or S248F mutation showed 3-fold increased
sensitivity to inhibition by CBZ (Picard et al., 1999). In another study using a molecular model of
wild-type and mutant α4β2 nAChR, the affinity of CBZ for the mutant receptor was estimated to be at
least 12-fold increased compared to the wild-type receptor. The increased blocking efficiency for the
S248F mutant by CBZ is caused by an increase in possible binding sites for CBZ in the mutant ion
channel pore. In this molecular structural model, the authors proposed that the higher affinity for CBZ
is probably due to aromatic-aromatic interactions between CBZ and the mutant Phe248 nucleotide
(Ortells & Barrantes, 2002). The higher receptor inhibition may contribute to the effectiveness of CBZ
in some patients carrying hypersensitive nAChRs. However, as the discussed molecular model is not
perfect, CBZ may not show this effect in human nAChR mutations. Furthermore, an increased
sensitivity for CBZ has only been found for two nAChR mutations, which is consistent with the
observation that not all mutations are associated with good CBZ response in patients. Taken together,
the effects of CBZ in most patients is most likely due to sodium channel inhibition, whereas in some
patients with specific mutations, a higher sensitivity to CBZ may add to the general therapeutic
effects.
1.6.2 Nicotine
Evidence has emerged that low doses of nicotine can act as a new treatment for AED-resistant
patients. In one patient, dermal nicotine patches greatly reduced the frequency of seizures, indicating
the possible beneficial effect of nicotine (Willoughby et al., 2003). In 22 other patients from two
families carrying 776ins3 or S248F mutations, seizure frequency was significantly associated with
smoking habits (Brodtkorb & Picard, 2006). A majority of tobacco-consuming mutation carriers (10
out of 14 patients) were seizure-free, whereas 7 non-consumers had frequent, more severe seizures.
The tobacco-consumers which where not seizure-free, consumed the least amounts of tobacco, and
their seizure phenotype was less severe and frequent than the non-consumers. Three patients reported
fluctuations in seizure frequency, which was associated with changes in tobacco consumption. In these
three patients, significant decrease in seizure frequency was reported during periods of increased
tobacco consumption, whereas the seizures reappeared after they stopped smoking. These data suggest
that nicotine could be an environmental factor, which protects ADNFLE patients from severe
symptoms (Brodtkorb & Picard, 2006). This therapeutic effect of nicotine can be explained by an
enhanced desensitization of hypersensitive nAChR, caused by chronic nicotine exposure. It has been
shown that very low doses of nicotine (10nM) effectively desensitize α4β2 nAChR. For these
desensitization experiments, α4 and β2 receptor subunits were transfected into HEK cells. Whole-cell
electrophysiological recordings were conducted when applying nicotine or acetylcholine, at a voltage
clamp of -100 mV. Prolonged application of nicotine (100s) reduced the recorded peak response (in
pA) in HEK cells with 70%, which is more efficient than ACh-induced desensitization. Furthermore,
the recovery time after nicotine-induced desensitization is longer than for ACh (Paradiso & Steinbach,
2003). Thus, there is evidence that nicotine is a potent agent for the desensitization of α4β2 nAChR,
which counteracts hypersensitivity of mutant receptors. After one cigarette, the nicotine level in the
central nervous system of consumers can reach 100nM, which is enough to induce desensitization
(Paradiso & Steinbach, 2003).
7
1.6.3 Cannabinoids
If ADNFLE would indeed involve hyper-inhibition, cannabinoids could be considered as therapeutic
drugs. In the cortex, endo-cannabinoids are released by pyramidal neurons, and inhibit GABAergic
transmission of interneurons. As discussed later, nAChR receptors are expressed on cortical
interneurons which co-express the neuropeptides vasoactive intestinal peptide (VIP) and
cholecystokinin (CCK). CB1 receptors are expressed on GABAergic interneurons which co-express
CCK, as well as on the majority of VIP-expressing interneurons. Thus, most nAChR ligand responsive
GABAergic interneurons express CB1 receptors and may respond to the effects of cannabinoids (Hill
et al., 2007). The presynaptic CB1 receptors regulate GABA release by interneurons. If GABAergic
inhibition is indeed a significant contributor to the pathogenesis of ADNFLE, reduction of GABA
release by stimulation of modulatory CB1 receptors may restore the misbalance between inhibition
and excitation in ADNFLE patients (Mann & Mody, 2008). Evidence for this suggestion is provided
by observations in two mouse models for ADNFLE created in 2006. In these mice carrying known
human mutations in the α4 nAChR subunit, blocking the GABAergic inhibition has been shown to
reduce the seizure phenotype (Klaassen et al., 2006). Therefore, cannabinoids could be tested for their
protective effects on ADNFLE seizures in animal models. However, more detailed characterization
and validation of the role of GABAergic inhibition is needed to establish the therapeutic effects of
cannabinoids.
Chapter 2: GENETICS
In ADNFLE patients, the first genes causing an epilepsy syndrome were identified. Based on the high
prevalence of ADNFLE patients in affected families and the autosomal dominant pattern of
transmission, ADNFLE is classified as a familial epilepsy syndrome. Initially, it was believed that
ADNFLE was caused by single gene defects in the identified susceptibility loci. However,
characterization of more families and patients has shown that ADNFLE might not be a homogeneous
Mendelian trait, but rather involve more underlying (genetic) causes. Patients which show the
ADNFLE symptoms but do not have a familial history of the disorder are referred to as NFLE
(nocturnal frontal lobe epilepsy). There are no differences in the clinical features of the familial and
sporadic cases of NFLE (Combi et al., 2004).
2.1 ADNFLE loci
Three loci associated with ADNFLE have been mapped by linkage studies. In the locus 20q13.2, the
first mutation related to epilepsy was identified in the CHRNA4 gene, coding for the nAChR α4
subunit (Phillips et al., 1995; Steinlein et al., 1995). A second locus was mapped on chromosome
15q24, which contains a cluster of genes coding for the α3, α5, and β4 subunits of nAChR. However,
no mutations have yet been identified in this locus (Phillips et al., 1998). In the third locus, 1p21,
another mutation was found in the CHRNB2 gene, which codes for the β2 subunit of nAChR (Phillips
et al., 2001). After the discovery of the associated genes, several mutations in CHRNA4 and CHRNB2
have been identified (Andermann et al., 2005; Hirose et al., 2005; Kaneko et al., 2002). An overview
of the identified mutations linked to ADNFLE to date and their major effects is provided in Table 1.
Details on the nAChR mutations and Table 1 are discussed in Chapter 3.
8
Recently, a mutation in the gene coding for the α2 nAChR subunit was found in a family with
complex symptoms. However, this finding has not been replicated, since no mutations in this gene
have been identified in other genetic studies on ADNFLE (Gu et al., 2007). In a large study on 47
families, the only identified variation in the CHRNA2 gene is a silent polymorphism I185I, which is
not located in the pore-forming domains of the nAChR (Combi et al., 2008a). These results suggest
that the CHRNA2 mutation is a rare, minor contributor to ADNFLE in very few patients within one
family to date.
Outside the nAChR family, the only gene to be associated with ADNFLE to date is the corticotrophinreleasing hormone (CRH) gene on chromosome 8q. In five ADNFLE patients, two nucleotide
variations in the promoter region have been identified (Combi et al., 2005). The g.-1470C>A variation
was most frequently observed, in 4 out of 44 ADNFLE subjects. The neuropeptide CRH is released by
the hypothalamus, and induces the production of corticotrophin (or adrenocorticotropic hormone,
ACTH) by the pituitary. This results in the production of the stress hormone cortisol. In addition to the
main role of CRH in the modulation of stress and stress-related responses, CRH also exerts other
neuromodulatory effects. The additional effects of CRH are reflected in the expression pattern of the
CRH gene and CRH receptor in the hippocampus, neocortex, and amygdala. Also, a role of CRH in
epilepsy is proposed since it has been shown that CRH induces seizures in young rats (Hollrigel et al.,
1998). Furthermore, massive infantile spams can be treated by ACTH, which decreases CRH levels by
a negative feedback mechanism (Hollrigel et al., 1998). High levels of CRH could contribute in
seizure generation by increasing the excitability of hippocampal neurons, thereby facilitating and
enhancing action potentials and seizure onset (Hollrigel et al., 1998). In line with the proposed role of
high CRH levels in excitability and seizures, the most frequently occurring CRH promoter variation
increases CRH levels and may therefore contribute to ADNFLE pathogenesis by increasing neuronal
excitability (Combi et al., 2008b). However, it is not likely that CRH is a major contributor to
ADNFLE since only very few patients harbor CRH promoter variations. Furthermore, the g.-1470C>A
variation also occurs in healthy controls with a frequency of 2,8% (Combi et al., 2005). It is not known
why the presence of CRH promoter variations in healthy subjects do not cause any phenotypic effects.
2.2 nAChRs in epilepsy syndromes
The role of nAChRs in the disease mechanism of ADNFLE is further underlined by evidence for their
contribution in other epilepsy syndromes. A single nucleotide polymorphism (SNP) in CHRNA4 has
been linked to febrile seizures and idiopathic generalized epilepsy (IGE). Febrile seizures or
convulsions occur in 2-5% of infants during high fever. Children affected by febrile seizures show a
higher risk for the development of epilepsy disorders later in life. IGE refers to types of epilepsy
without apparent damage and abnormalities in the central nervous system, and includes several
common epilepsy syndromes such as childhood absence epilepsy (CAE) and juvenile myoclonic
epilepsy (JME). In children with febrile seizures and IGE, the frequency of the T allele of SNP
1044396 in CHRNA4 is significantly higher than in control children. This silent S543S SNP does not
result in a nucleotide mutation in the α4 receptor subunit, but could rather serve as a genetic marker or
susceptibility factor (Chou et al., 2003; Lee et al., 2007). Association of epilepsy syndromes and the
gene for β2 subunit has not been observed (Lee et al., 2007; Peng et al., 2004). Interestingly, the locus
15q14, containing the gene for the α7 nAChR subunit, has been linked to JME and BECTS (benign
epilepsy of childhood with centrotemporal spikes) patients. α7 is another major neuronal nAChR
subunit, further suggesting the role for nAChRs in epilepsy syndromes (Elmslie et al., 1997; Neubauer
et al., 1998).
2.3 Genetic heterogeneity
As mentioned above, the initial view of ADNFLE as a simple,
homogenous Mendelian trait involving one genetic defect has been
increasingly questioned in the recent years. A majority of the
9
patients (85-90%) does not harbor mutations in the identified loci
and nAChR subunits, suggesting the role of other loci and genes in
the pathogenesis of ADNFLE (Combi et al., 2004). In a genetic screen
on 14 patients from three Italian families, involvement of all known
nAChR subunits has been excluded, which emphasizes the genetic
heterogeneity in ADNFLE (De Marco et al., 2007). Furthermore, two
new putative loci (3p22-p24 and 8q11.2-q21.1) have been identified
in an ADNFLE family (Combi et al., 2005). Together, these findings
suggest that ADNFLE may be a more complex disorder with
oligogenic transmission. It is important to realize the more complex
character of this disorder, and to continue the search for other
genetic contributors in patients without nAChR mutations. It is very
likely that in the majority of the patients, other genes than nAChR
subunits or CRH are underlying the pathogenesis. However, the
identification of possible minor contributors is made difficult by the
availability of only small patient samples, which are insufficient to
show less significant putative loci in a genome-wide scan. Again, the
importance of accurate diagnosis is highlighted, which is needed to
include more patients in genetic studies and to further establish the
genetic contributors of ADNFLE. Chapter 3: nAChRs
The genes CHRNA4 and CHRNB2, in which mutations were identified to cause ADNFLE, code for
subunits of nicotinic acetylcholine receptors (nAChRs). nAChRs are a family of ligand-gated ion
channels, which mediate the effect of the endogenous ligand acetylcholine in cholinergic systems.
These receptors are widely expressed in the central nervous system, where they are involved in the
regulation of neuronal processes, e.g. neurotransmitter release and neuronal excitability. As these
processes are involved in many circuits and pathways of neuronal functioning, the cholinergic system
and nAChRs are pivotal for the regulation of neurotransmission in the central nervous system. The
significant role of nAChR has been underlined by the identification of several mutations in the α4 and
β2 nAChR subunits in patients suffering from ADNFLE. However, the contribution of nAChR
mutations to epilepsy pathogenesis remains to be fully elucidated. To explore the nAChRs and the
identified mutations, the normal and mutant receptor properties are discussed below, as well as the
possible consequences of the identified ADNFLE mutations.
3.1 Structure
nACh receptors are formed by five subunits, each containing four transmembrane regions (TM 1-4).
To date, nine α subunits (α2-α10) and three β subunits (β2-β4) have been identified (di Corcia et al.,
2005). nAChR subunits contain an extracellular domain, three transmembrane regions, an intracellular
loop, which is followed by the fourth transmembrane region at the C-terminus (Figure 1a). Both
homomeric and heteromeric receptors are formed by five subunits, creating the ion channel pore in the
center of the receptor composition (Figure 1b). Whereas homomeric receptors are composed by five
identical α subunits, heteromeric receptors usually contain two α and three β subunits. The inner wall
of the ionic pore is formed by the second TM segment (M2) of each subunit (Miyazawa et al., 2003).
10
Homomeric receptors contain five ligand-binding sites, whereas heteromeric receptors have two
binding sites for acetylcholine. In heteromeric receptors, the ligand-binding site is located at the
interface between an α and a β subunit (Figure 1c). Ligand binding induces conformational changes in
the α subunits, resulting in a rotation and the opening of the ion channel. In the open state, the ion
channel of nAChR is permeable to sodium, potassium, and calcium (Gotti & Clementi, 2004).
Figure 1. Schematic overview of nAChR organization (from Gotti & Clementi, 2004).
a. A single subunit contains four transmembrane regions, of which the second region lines the inner
wall of the ion channel.
b. The nAChR is assembled from five subunits, forming an ion channel pore in the center.
c. Schematic view of the homomeric α7 and heteromeric α4β2 receptor subunit composition, with the
ligand binding sites (small circles).
The heteromeric nAChRs can be formed by the subunits α2-6 and β2-4. The β subunits are believed to
regulate the binding property and sensitivity of the nAChR. In general, most receptors have an
α(2)β(3) composition, containing two α and three β subunits. When α and β subunits are equally (1:1)
injected in chick oocytes, the nAChRs in the cell membrane adapt the α(2)β(3) composition. In in vitro
cell lines, the nAChR subunit composition can shift from an α(3)β(2) to an α(2)β(3) structure. The
α(2)β(3) receptors show a higher sensitivity and decreased desensitization. However, it is not known if
this transition and receptor plasticity also occurs in vivo (Gotti & Clementi, 2004).
The accessory α5 and β3 subunits may also contribute to plasticity in receptor properties. These
subunits can only assemble in nAChR as a fifth subunit, together with two other α and β subunits. One
example is the incorporation of the α5 subunit in the α4β2α5 receptor, which is also expressed in the
central nervous system (Kuryatov et al., 2008). Incorporation of the α5 subunit to the α4β2 nAChR
results in increased activation sensitivity and calcium permeability. Therefore, the α5 and β3 subunits
may act as additional subunits to regulate and fine-tune receptor properties and responses, in reaction
to physiological changes. However, only a small proportion of the neuronal nAChRs contain the α5
subunit, as a result of limited expression of this subunit (Gotti & Clementi, 2004; Kuryatov et al.,
2008, Figure 2). It is not known whether the subunit composition is altered in ADNFLE.
11
Figure 2. Subunit composition of nAChRs expressed in the rodent cortex and hippocampus (from
Gotti & Clementi, 2004). In the cortex, α4β2 nAChRs are predominantly present, whereas only a small
amount of the receptors also incorporate the α5 subunit.
3.2 Localization and function
In the central nervous system, the homomeric α7 and heteromeric α4β2 nAChR are the most abundant
receptor types. They are mainly expressed in the hippocampus, thalamus, cortex and the midbrain
(Raggenbass & Bertrand, 2002). The majority of the nAChRs are believed to be located at the
presynaptic compartment, regulating the release of acetylcholine, glutamate, GABA, dopamine and
noradrenalin (NA). Therefore, the presynaptic expression of nAChR on glutamatergic and GABAergic
neurons help to modulate the excitatory and inhibitory networks in the cortex. As mutated nAChR
subunits do not function properly, they are no longer able to modulate and maintain the balance
between excitation and inhibition. A shift of the balance to GABAergic inhibition can explain the
finding that excessive release of GABA is causing the seizures in two mouse models of ADNFLE
(Klaassen et al., 2006). However, the effect of the mutant nAChR on neuronal networks is not fully
understood, because of their expression and modulation on both glutamatergic and GABAergic
neurons.
In addition to their modulatory role on neurotransmitter release, presynaptic nAChRs are also involved
in calcium homeostasis. When activated, nAChRs are permeable to calcium ions and induce the
activation of voltage operated calcium channels (VOCCs), resulting in calcium influx in the
presynaptic compartment. Intracellular calcium is involved in a variety of cellular processes. In
signaling pathways, calcium acts as an important second messenger and interacts with several
calcium-binding proteins or sensors. In the presynaptic terminal, calcium influx leads to fusion of
synaptic vesicles with the presynaptic membrane and release of neurotransmitter. Conversely, the
sensitivity of the nACh receptors is also modulated by extracellular calcium levels (Gotti & Clementi,
2004). When extracellular calcium levels decrease, the sensitivity of nAChR to their ligands also
declines, and vice versa.
In addition to the presynaptic effects, postsynaptic receptors have also been shown to modulate cell
responses, generally fast excitatory neurotransmission (Gotti & Clementi, 2004). In the cortex, only a
subset of the interneurons is activated by administration of nAChR agonists. These interneurons show
co-expression of vasoactive intestinal peptide (VIP) and cholecystokinin (CCK). The direct excitation
of these interneurons by nAChR agonist is mediated by postsynaptic nAChRs containing α4, β2 and
possibly α5 subunits (Porter et al., 1999).
3.3 Acetylcholine innervation
In the cholinergic system, acetylcholine is produced and released by cholinergic neurons. The function
and effects of the cholinergic system is mediated by muscarinic and nicotinic acetylcholine receptors.
The most important system containing the largest group of cholinergic neurons is the magnocellular
basal complex, which provides the main input to the cortex and hippocampus. The second large
system, the peduncolopontine-lateraldorsal tegmental complex (PPN/LDTN), projects to the thalamus
12
and midbrain dopamine neurons. The striatum is also highly innervated by cholinergic neurons (Gotti
& Clementi, 2004). Other smaller cholinergic systems are located in the brain stem, habenula and the
autonomic nervous system.
3.4 nAChR mutations
To date, several mutations in the α4 and β2 subunits have been identified in a small number of affected
families and patients. In all carriers, the mutation is heterozygous. Most mutations affect the TM2
inner-pore segment of the nAChR, indicating that these mutations change the receptor and ion channel
property. Functional experiments have been performed on mutant receptors expressed in Xenopus
oocytes. Initially, homozygous mutants were used, and mutations in the α4 subunit resulted in loss-offunction of the receptor, whereas β2 mutations resulted in a gain-of-function. More recently, studies
were performed using heterozygous mutations, which are more similar to the gene mutations of human
patients. From these studies, the only commonly found consequence of all nAChR mutations is an
increased sensitivity to the ligand (Marini & Guerrini, 2007). To illustrate, Steinlein et al. (1997)
found an EC50 value of 3.6 µM ACh for wild-type α4β2 subunits expressed in Xenopus oocytes,
whereas the sensitivity of α4(776ins3)β2mutant showed an approximately 10-fold increase, with an
EC50 value of 0.28 µM ACh (Steinlein et al., 2007). Gain-of-function in receptor mutants affecting the
TM2 segment can be a direct result of altered ion channel properties. Since nearly all mutations are
located in the TM2 region, changes in the ion channel pore are probably the direct cause for the
increased sensitivity in most mutant receptors. Indirect effects on the receptor properties can cause the
hypersensitivity of mutations outside the TM2 segment. It has been shown that mutations in the
transmembrane TM3 region increase acetylcholine sensitivity by modulating the movements of the
TM2 segment, thereby increasing the probability of channel opening (Hoda et al., 2008).
In addition to the gain-of-function effect, many mutant nAChRs show a reduction of calcium
dependence. In wild-type nAChRs, higher extracellular calcium levels are shown to potentiate the
receptor response, a process called receptor potentiation. Conversely, low levels of extracellular
calcium reduce the nAChR sensitivity. Receptor potentiation by calcium is largely abolished in mutant
α4β2 nAChR, leading to a state of calcium independence (Rodrigues-Pinguet et al., 2005). During
excessive firing, extracellular calcium levels decrease as a result of calcium influx into the presynapse.
However, when mutant nAChR subunits are expressed, the decreased extracellular calcium levels will
no longer inhibit the function of nAChR, resulting in a continuous gain of function and
neurotransmitter release, which is independent of regulation by calcium levels (Mody, 2003).
After longer stimulation and opening of the ion channel, the receptor response to ligands decreases,
which is called desensitization. The transition to the desensitized state can last for several minutes
(Gotti & Clementi, 2004). Higher receptor desensitization is observed in a subset of the mutant
nAChRs (Rodrigues-Pinguet et al., 2003; Rodrigues-Pinguet et al., 2005), resulting in longer periods
of low receptor response. Interestingly, receptor desensitization has only been observed in mutations
in the TM2 domain of the receptor to date (Hoda et al., 2008). This pattern may suggest that TM3 does
not play an important role in the regulation of desensitization properties. Therefore, desensitization is
not significantly affected by mutations in TM3. Another observed property of a subset of mutant
receptors is a decrease in calcium permeability (Bertrand et al., 1998; Lipovsek et al., 2008), which
results in lower calcium influx during channel opening. Other receptor-linked properties such as
subunit composition, internalization and protein synthesis have not been studied. In normal conditions,
the expression of nAChR on the cell surface is overall stable, and little internalization and insertion is
observed (Kumari et al., 2008). It is not known if the process of internalization and protein synthesis is
affected in cells harboring mutant nAChR.
A summary on the identified mutations and their main effects is listed in table 1. Table 1 illustrates the
properties of the mutations, emphasizing both the similarity and differences between them. The table
contains all mutations linked to ADNFLE to date, including the most affected genes CHRNA4 and
CHRNB2, and the minor contributors CHRNA2 and CRH. Listed are the frequency of the mutation,
location of the mutated amino acid in the protein structure, the effectiveness of CBZ treatment, and the
13
known consequences of the mutation based mainly on electrophysiological studies. As mentioned
above, a few conclusions can be drawn from the table. First, the mutation site is mostly in the poreforming M2 segment, affecting the ion channel properties. Second, the response to CBZ treatment is
heterogeneous between families with different mutations. For some mutations, the patients do not
respond or respond normally to CBZ. Two mutations (S248F and 776ins3) seem to show better results
on CBZ treatment, which is consistent with the observed higher affinity of these mutations for CBZ
(Picard et al., 1999). In contrast, patients affected with the I312M mutation in CHRNB2 react
differently to CBZ. Thus, there is no highly consistent pattern or association between a genetic
mutation and responsiveness to CBZ, partly caused by the limited number of families and subjects.
Third, the effect of the mutations in nAChR subunits vary in terms of their calcium permeability,
calcium dependence, desensitization properties, and cognitive effects. The only common effect is a
gain-of-function upon ligand binding. Therefore, the general mechanism of mutant nAChR function in
ADNFLE is most likely related to the gain-of-function of the cholinergic system, although the
identified mutations show different properties and thus variable effects. The less common effect may
either contribute in the increase of nAChR hypersensitivity, or show puzzling counteracting effects.
The higher calcium independence in a subset of the mutant receptors adds to the receptor
hypersensitivity, resulting in a greater gain-of-function and supporting the general mutant nAChR
mechanism. In contrast, the observed increase in receptor desensitization and decrease of calcium
permeability are not in line with the gain-of-function-effect. Longer duration of the desensitized state
and lower calcium influx are expected to decrease receptor function and therefore counteract the
hypersensitivity of mutant nAChR. It is possible that the effects of counteracting properties are less
pronounced than the large increase in receptor sensitivity, and become overruled so that the net effect
of mutant properties remains a gain-of-function. Taken together, hypersensitive nAChR should be
studied for their role in ADNFLE pathogenesis, although the property of specific mutations differs.
Therefore, experiments on mutant nAChR should include several mutants to account for the mutationspecific outcome.
3.5 Implications for dopamine system
α4-containing nAChRs are expressed very early in development on dopamine neurons in the
nigrostriatal circuit, and enhance dopamine release in the striatum and prefrontal cortex (Azam, Chen,
& Leslie, 2007). In the adult brain, presynaptic nAChRs are involved in motor systems, by modulation
of dopamine release in the midbrain dopaminergic networks. The role of nAChRs on dopamine release
and the motor attacks observed in the patients has led to the study of the midbrain dopamine system in
ADNFLE. A PET study revealed reduced dopamine D1 receptor density in the right putamen in 12
ADNFLE patients carrying the S248F mutation (Fedi et al., 2008). The reduced receptor density in the
putamen, which is a part of the striatum, could be caused by receptor downregulation in response to
increased extracellular dopamine levels. Another explanation could be neuronal loss in the striatum
due to expression of hypersensitive α4 nAChRs. The latter possibility is supported by findings in
knock-in mice, which express the mutated, hypersensitive α4 nAChR subunit (Labarca et al., 2001).
These mice show a significant loss of dopaminergic neurons in the substantia nigra. The authors
propose the continuous activation of the dopaminergic neurons by acetylcholine as the underlying
cause of toxicity and cell death (Labarca et al., 2001). Therefore, it is more likely that the decreased
nAChR expression in the dopamine system is a consequence of hypersensitive nAChRs, rather than a
major contributor to ADNFLE pathogenesis. However, the link between the dopamine abnormalities
and characteristics of motor seizure phenotype has not yet been studied. Thus, it is possible that gainof-function of the cholinergic system causes aberrant midbrain dopamine circuits, which in turn
contribute to the motor manifestations in ADNFLE seizures.
Table 1. Identified mutations in CHRNA4, CHRNB2, other loci, and their major effects.
CBZ; Carbamazepine. AED: anti-epileptic drugs.
CHRNA4
Families Location
Effect
CBZ response
Reference
14
T265I
1
M2
Increased sensitivity
Low penetrance
Interictal abnormality
Increased sensitivity
Ca2+ permeability ↓
Ca2+ dependence ↓
Ca2+ potentiation↓
Desensitization ↑
Increased affinity
Ca2+ permeability ↓
Ca2+ dependence ↓
Ca2+ potentiation↓
Desensitization ↑
Psychiatric illness
Increased sensitivity
Ion permeability ↓
Ca2+ dependence ↓
Ca2+ potentiation↓
Desensitization ↑
Low intellect
No increased
sensitivity
(Leniger et al.,
2003)
S248F(Mice: 4
S252F)
M2
Increased
sensitivity
(Steinlein et al.,
1995)
776ins3
2
(Mice: +L264)
M2
Increased
sensitivity
(Magnusson et
al., 2003;
Steinlein et al.,
1997)
S284L
(Mouse:
S252L, rat:
S256L)
4
M2
L290V
1
M2
Resistant to 10
AED’s
(El Helou et al.,
2008)
R308H
1
2nd
intracellular
loop between
TM3 and
TM4
No good CBZ
response
(Chen et al.,
2009)
No increased
(Hirose et al.,
sensitivity/Ineffecti 1999; Rozycka et
ve
al., 2003)
CHRNB2
Families
Location Effect
CBZ
V287M
2
M2
Effective
V287L
1
M2
I312M
2
M3
L301V
3
M3
V308A
3
M3
CHRNA2
I279N
CRH
1
1
M1
Increased sensitivity
Ca2+ dependence ↓
Ca2+ potentiation↓
Desensitization ↑
Increased sensitivity
Ca2+ dependence ↓
Ca2+ potentiation↓
Desensitization ↑
Increased sensitivity
Opening probability ↑
Memory deficits
Increased sensitivity
Opening probability ↑
Increased sensitivity
Opening probability ↑
Increased sensitivity
Altered levels geneexpression
(Diaz-Otero et al.,
2008; Phillips et
al., 2001)
(De Fusco et al.,
2000)
Ineffective/Effectiv (Bertrand et al.,
e in second family 2005)
(Hoda et al.,
2008)
(Hoda et al.,
2008)
(Aridon et al.,
2006)
(Combi et al.,
2005)
15
Chapter 4: ANIMAL MODELS
To study the phenotypic and molecular effects of the nAChR gain-of-function mutations at a model
system level, several animal models have been created and studied. Previous to the description of
ADNFLE, have already been created to study nAChR function. α4 subunit knockout mice do not show
spontaneous seizures or EEG abnormalities, indicating that the ADNFLE phenotype is not caused by a
loss of receptor function (McColl et al., 2003). Animal models created specifically for ADNFLE
express hypersensitive α4β2 nAChR for the study of mutant receptor function in the whole organism.
In the recent years, more animal models harboring identified human ADNFLE mutations have been
created, which show high face and construct validity. An summary and evaluation of available animal
models is provided in this chapter.
16
4.1 α4 L9’S and L9’A knock-in mice
In a heterozygous L9’S knock-in mouse model, a mutation in the α4 TM2 region results in
hypersensitivity to nicotinic agonists. Homozygous mutant animals were not viable. To generate the
mutant mouse model, embryonic stem cells (ES cells) harboring the 129/SvJ α4 genomic clone with
the mutation were injected into C57BL/6J blastocysts. The L9’S mutation is not identified in
ADNFLE patients, rather, only the effects are similar to human mutations (Fonck et al., 2003). The
resulting transgenic mice do not exhibit spontaneous seizures. In these mice, nicotine induced seizures
at significantly lower concentrations (eight-fold) than in wild-type mice. The nicotine-induced seizures
were completely blocked by the noncompetitive α4β2 nAChR antagonist mecamylamine, a finding
which is in line with the proposed role of hypersensitive nAChRs in epilepsy pathogenesis. The
GABAA receptor antagonist bicuculline caused seizures in both mutant and WT mice at similar
concentrations (Fonck et al., 2003), indicating an absence of major differences in the GABAergic
system between mutant and WT animals. The compounds tacrine and galanthamine, which are
cholinesterase inhibitors and therefore increase ACh concentration in the synapse, were also able to
induce seizures at lower concentrations in the mutant mice compared to WT mice. This finding further
indicates a gain-of-function of the cholinergic neurotransmission. However, a drawback of this model
is the absence of phenotypic hallmarks of human patients such as spontaneous seizures and sleep
disturbances. Therefore, this model is not a valid model for ADNFLE and has only been used to study
the consequences of nAChR gain-of-function, rather than the detailed disease mechanism of
ADNFLE.
Another knock-in mouse line harboring an α4 mutation in the TM2 region is the L9’A mice, in which
the leucine at position 9 is replaced by alanine. Thus, the α4 nAChR gene in these mice differs from
the WT and L9’S mutant mice only by one residue at position 9. Another difference is in the
generation of the L9’A mutant mice. The L9’S mutation in the 129/SvJ genomic clone in the previous
animal model was replaced by the L9’A mutation. The genomic clone was then electroporated into
CJ7 ES cells and placed into C57BL/6 blastocysts. The L9’A mutation has also not been found in
human patients so far. The α4β2 receptors in the L9’A mice show a 30-fold increase in ligand
sensitivity. These mice were shown to be 15-fold more sensitive to nicotine-induced seizures (Fonck
et al., 2005), which are characterized by fast repetitive movements. The seizures in mutant mice were
partial, and no clear EEG ictal activity was recorded. Seizures in homozygous and heterozygous
mutant mice, but not in WT mice, were prevented by pre-treating the mice with very low nicotine
doses (0.1 mg/kg), suggesting higher desensitization in the mutant. EEG recordings showed increased
awakenings in non-REM sleep of L9’A mice, indicating sleep disturbances. These genetic and
phenotypic hallmarks suggest that this model is more similar to the human disease manifestations.
However, no spontaneous seizures occurred in L9’A mutant mice (Fonck et al., 2005). The authors
propose that the lack of spontaneous seizures in mutant mice could be caused by better compensation
strategies in mice than in human patients. Mutant mice show lower expression of α4β2 nAChR, which
might be sufficient to reduce the effect of hypersensitive receptors and prevent spontaneous seizures.
However, the cause of the decreased nAChR levels is not known, and the possible compensation
mechanism does not seem to prevent ADNFLE symptoms in human subjects. Taken together, this
model lacks both face and construct validity, and is therefore not a validated model to study the
disease mechanisms of ADNFLE. Similar to the above described L9’S model, the L’9A model has
only been used to study nAChR hypersensitivity.
4.2 α4 S248F mice
A stronger phenotypic ADNFLE mouse model contains the S248F α4 nAChR mutation, which has
also been found in human patients (Teper et al., 2007). The mutant mice were generated by replacing
the mutation in the 129/SvJ genetic clone used in L9’S animals with the S248F mutation, and
electroporation into 129/Sv ES cells. The ES cells were placed into C57BL/6 blastocytes. The chimere
progeny was then crossed with CD1 animals. The resulting animals which were used for experiments
contained a genetic background of 50% CD1 and 50% 129Sv. Similar to previous described models,
no spontaneous seizures occurred in the S248F mutant mice. The mutant mice showed hypersensitivity
to nicotine-induced seizures, which were suppressed by pretreatment with low doses of nicotine. The
17
anti-epileptic drug carbamazepine also partially suppressed the seizures. Remarkably, the seizure
phenotype in these mice was more similar to the nocturnal motor attacks in human patients. The
authors referred to this specific phenotype as dystonic arousal complex (DAC). DAC is characterized
by aroused, exploration-like head movements, and movements of the whole body. This is followed by
dystonia in the forelimbs, with a stretched trunk. The mice also display a Straub tail, which is a nearly
horizontal tail, bent nearly 180o from natural positions. DAC was not associated with abnormal EEG
recording (Teper et al., 2007). Taken together, this model may be interesting to study the motor attacks
in ADNFLE. However, significant variations were observed in the animals in terms of gender and
genetic background. The use of animals with different genetic backgrounds emphasizes the effect of
distinct genotypes on the phenotype of the animals. The functional effect of a nAChR mutation also
depends on the genetic make-up of a mouse line, resulting in different findings in models using
different mouse strains. Therefore, the genetic differences of the mouse lines should be taken into
account when interpreting the animal model data.
4.3 α4 776ins3 and α4 S248F mice
The first mouse models which exhibit spontaneous seizures are the mice carrying the 776ins3 or
S248F mutation. The latter is the same mutation as in the above-described model. These mice were
created by electroporation of the target genes into 129/SvJ embryonic stem cells, which were placed in
129/SvJ females. The progeny has been backcrossed with 129/SvJ mice and C57BL/6J mice and the
resulting strains were used in experiments. Both heterozygous 776in3 and S248F mutant mice show
frequent spontaneous seizures, abnormal EEG activity, and increased sensitivity to nicotine. Upon
nicotine stimulation, a significant increase (>20 fold) of postsynaptic inhibitory currents was observed
in the pyramidal cells in layers II and III of the cortex. Administration of a GABAA receptor
antagonist, picrotoxin, was able to inhibit EEG abnormalities and seizure occurrence (Klaassen et al.,
2006). This finding suggests that the activation of GABAA receptors is sufficient to trigger seizures in
the current mouse models, which is paradoxal with respect to the inhibitory effects of GABA. To
explain the surprising hyper-inhibition in the mouse models, the authors propose that the gain-offunction of the nAChRs increases GABAergic transmission in the cortex, and subsequently the
pyramidal neurons in layer II and III will show synchronized firing after high GABAergic inhibition
(Figure 3). These interesting findings suggest that excessive GABAergic transmission, as a result of
nAChR hypersensitivity, may underlie the disease mechanism of ADNFLE. However, a major role for
excessive GABAergic transmission has not been established be assumed, since it is not yet known if
GABA plays a primary or accessory role in the underlying mechanisms of ADNFLE.
4.4 2 V287L transgenic mice
Very recently, the only mouse model carrying an identified 2 nAChR subunit mutation, V287L, has
been generated in a FVB background (Manfredi et al., 2009). This model is interesting not only for it’s
similarities to human ADNFLE symptoms, but also for its contribution to study the developmental
effect of the mutation. 24-hour EEG recordings revealed interictal EEG abnormalities in mutant mice,
characterized by frequent, high amplitude spikes. Moreover, the mutant mice suffered from
spontaneous seizures. The EEG abnormality and seizure phenotype was correlated to the expression
level of mutant nAChR, indicating a gene-dosage effect. The mice can be manipulated to show
conditional expression of the mutant receptor, by a tetracycline-controlled expression system. Using
this conditional expression system, the effect of mutant receptors can be studied in temporal windows.
The authors silenced the mutant receptor expression during developmental stages E1-P15, a period in
which nAChRs are proposed to play a critical role in the establishment of neuronal circuits. The
consequence of mutant receptor silencing is that the animals expressed only the wild-type nAChR
during stages E1-P15. The resulting animals did not develop any seizure phenotypes or have any EEG
abnormalities, in contrast to animals which expressed the mutant receptor continuously during
development. This finding indicates that mutant function of nAChR during development is underlying
the disease onset postnatally (Manfredi et al., 2009). To further illustrate this hypothesis, it has been
found that when the mutation is silenced after seizure onset, the seizure phenotype is not affected by
suppression of the mutant. Thus, the removal of mutant receptors in adult mice did not prevent the
18
seizures. This animal model strongly emphasizes the importance of nAChR function during embryonic
development, and points out that developmental defects determine the disease outcome later in life.
Figure 3. Model of cortical synchronization after GABA inhibition (from Klaassen et al., 2006).
The GABAergic interneuron (centered cell) projects to pyramidal neurons in layers II and III of the
cortex. In this model, activation mutant presynaptic nAChR results in a greater GABA release. This
strong GABAergic inhibition will reset all asynchronous firing, and produces synchronous firing of
the pyramidal neurons immediately after the inhibition. In the presence of picrotoxin, a GABA
receptor blocker, GABA inhibition is diminished, resulting in non-pathogenic asynchronous firing.
4.5 Rat S248L model
Two transgenic rat strains harboring a S248L α4 mutation were generated. The S248L mutation has
also been identified in human patients (Zhu et al., 2008). The transgenic S248L nAChR genomic clone
was placed in a Sprague-Dawley genetic background by microinjection. The two obtained strains
showed a phenotype similar to human ADNFLE, of which one was used for experiments. The
transgenic rat phenotype included spontaneous seizures with a frontal origin during slow-wave sleep,
seizure manifestations very similar to the three classes of motor attacks in human patients, absence of
structural or anatomical abnormalities in the central nervous system, and the early onset during life
(seizure onset before 6 weeks postnatally). Furthermore, the disorder in rats is inherited in an
autosomal dominant pattern. These observations indicate that the transgenic rat strains are good
models for ADNFLE, by displaying the most similarities to human ADNFLE of all animal models to
date. Interestingly, a decrease in both synaptic and extrasynaptic GABAergic transmission has been
found in this model. The authors suggest that the attenuated GABAergic transmission could be the
result of rapid nAChR desensitization of the S248L mutant, whereas glutamate transmission is
unchanged. The unchanged glutamate level is unexpected, since desensitization of the nAChRs would
also decrease glutamate transmission. To account for this discrepancy between GABA and glutamate,
the authors discovered abnormalities in glutamate transmission. The basal levels of glutamate release
were higher in the transgenic rats compared to WT rats, both during wakefulness and sleep.
Furthermore, glutamate release in the frontal cortex decreases during slow-wave sleep in WT but not
in transgenic animals. Together, the increased glutamate transmission is sufficient to compensate for
the effects of mutant nAChR function, whereas GABAergic transmission is affected by rapid receptor
19
desensitization. The increased glutamate transmission results in a misbalance in favor of glutamatergic
excitation, and subsequent seizure generation. However, the cause for the aberrant, increased
glutamate transmission is not known. Taken together, the rat ADNFLE model might be the best model
to date, displaying most face and construct validity.
Taken together, the major drawback in mice is the great variability in disease phenotype due to
differences in the genetic background of the animals. Therefore, the results in mice are
inconsistent to date and their reliability should be questioned. The transgenic rat model appears to
be the best ADNFLE animal model to date, which covers all major hallmarks of human ADNFLE
symptoms. However, only two rat strains harboring one mutation have been generated. More
human mutations need to be replicated in rats to further verify the obtained results in the
transgenic rat model. Furthermore, the effect of different genetic background on the seizure
phenotype should also be examined in rats, by generation of rat models using other strain. With
more rat strains and other nAChR mutants, the validity and reliability of the rat as animal model
will be further established. Interestingly, in the rat model it has been found that GABAergic
transmission is attenuated, in contrast to the finding of Klaassen et al. (2006) which has been
discussed above. It is remarkable that both models show high validity, and nevertheless point in
opposite directions. The discrepancy is probably caused by the differences between animal species
and the different mutations. The elevated glutamatergic transmission found in the S248L rats
points out that the results by Klaassen et al. (2006) should be evaluated with care. Two mouse
models are not sufficient to prove aberrant GABA release in ADNFLE, when more conflicting data
are available. More valid animal models with thorough characterization are needed, to further
establish the roles of GABA and glutamate in ADNFLE seizures.Chapter
5: CURRENT
HYPOTHESES
Although gene mutations have been identified in ADNFLE patients and several animal models have
been developed, the underlying mechanism of the pathogenesis is still not fully understood. However,
the genetic findings, in combination with in vitro electrophysiological studies and data from animal
models, lead to the identification of contributors in the disease mechanism. This chapter discusses the
hallmarks which are currently believed to be involved in ADNFLE, and their functional role as far as
it is understood.
5.1 Hypersensitivity of nAChR
With respect to the modulatory function of nAChRs on neurotransmitter release, their role in an
imbalance between excitatory and inhibitory systems can be postulated in several pathways. Since
epilepsy commonly is believed to involve excessive excitation by glutamate, an imbalance in favor of
excitation is sufficient to increase seizure susceptibility. In support of this hypothesis, abnormal high
levels of glutamate transmission have been found in the rat model of ADNFLE (Zhu et al., 2008).
The role of mutant nAChRs in excitation-inhibition imbalance is supported by their predominant
presynaptic expression and modulation of neurotransmitter release. Hypersensitive nAChR could
results in an imbalance of cortical excitability by aberrant modulation. This imbalance in cholinergic
systems, especially the thalamocortical network, may facilitate excitatory input to the cortex and
induce seizure activity (Andermann et al., 2005). In other words, the hypersensitive mutant nAChRs
on thalamic neurons could induce small depolarization of thalamic cells, raising the neuronal
excitability, which may result in spontaneous, synchronized oscillations (Raggenbass & Bertrand,
2002). In addition, it is also possible that mutant nAChRs cause an excitation-inhibition imbalance by
the modulation of other neurotransmitters and pathways. Thus, activation of the hypersensitive nAChR
could result in an increase in the net excitatory input to the cortex, resulting in higher glutamate levels,
and seizures.
20
5.2 Role of GABA
Paradoxally, excessive GABAergic neurotransmission has been found to mediate the seizure
occurrence in two mouse models of ADNFLE (Klaassen et al., 2006), suggesting GABA as a
contributor in ADNFLE. The role of GABAergic systems in epileptic activity and nicotine-induced
seizures has been proposed previously (Dobelis et al., 2003). α4 containing nAChRs are mainly
expressed on presynaptic terminals of GABAergic interneurons in the cortex and hippocampus, where
the release of GABA by activated interneurons provides inhibitory input to the cortical layers. This
GABA release could act to inhibit other inhibitory circuits, resulting in disinhibition and net excitation
of cortical neurons (Ji & Dani, 2000). Another role for GABA in excitatory networks is the
synchronization of glutamatergic neurotransmission (Fonck et al., 2003). Several lines of evidence
support the theory of synchronized excitatory transmission after activation of GABAergic networks.
First, it has been shown that endogenous acetylcholine can activate α4β2 nAChR in the cortex, and
thereby induce synchronized inhibitory interneuron activity in rat prefrontal cortex (Bandyopadhyay et
al., 2006). Whole cell current-clamp recordings were obtained in layer I cortical neurons. The effect of
endogenous ACh was observed by neuronal stimulation after adding an ACh esterase inhibitor to
block ACh breakdown. In the presence of increased endogenous ACh, the recorded GABA activity in
layer I cortex neurons showed higher amplitude and duration. The enhancement of GABA release is
mediated by α4β2 nAChR subunits, since the addition of the α4β2 receptor antagonist abolished the
effect induced by ACh. Based on these results, the authors propose that endogenous ACh is able to
enhance the activity of GABAergic networks (Bandyopadhyay et al., 2006). These inhibitory networks
can generate excitatory oscillations, by synchronization of glutamate transmission. Synchronization
occurs when a large number of glutamatergic neurons fire synchronously after an initial broad
GABAergic inhibition. The resulting excitatory output is stronger than unsynchronized firing.
Therefore, increased activity in GABA networks can enhance neuronal excitation by producing
simultaneous glutamate activity (Mann & Mody, 2008). Second, activation of GABAA receptors and
synchronized GABAergic transmission underlies ictal events in human focal cortical dysplasia (FCD)
slices (D'Antuono et al., 2004). In FCD slices, application of 4-aminopyridine (4-AP) induced
spontaneous epilepsy-like ictal activity, which was blocked by GABA antagonist bicuculline.
Extracellular K+ levels were measured by double-barrelled ion-selective microelectrodes, which
recorded an increase in extracellular [K+] from baseline concentrations of 3.2 mM to 4.8 mM during
ictal activity. Ion-selective microelectrodes can measure the concentration of specific ions both in the
intra- and extracellular space. The ion-specific electrode records the total amount of mV generated by
the selected ion, and the total amount of mV generated in the complete intra- or extracellular
compartment. The specific ion activity and concentration is then obtained by subtracting the signal
measured by the reference electrode (Deveau et al., 2005). By measurement of potassium levels, the
authors proposed that synchronized activation of GABAA receptors causes interictal discharges,
resulting in epileptic activity and increased extracellular [K+] (D'Antuono et al., 2004).
In addition to the indirect excitatory effects of GABA via disinhibition and glutamate synchronization,
recent evidence has shown a direct excitatory role for GABA. It has been shown that cortical
interneurons have a more positive EGABA (Marty & Llano, 2005). Theoretically, a more positive EGABA
compared to resting membrane potential is sufficient to create depolarization of the postsynaptic
neuron. A higher EGABA could be the result of repetitive firing and activation of GABAA receptors,
accompanied by increased intracellular Cl- ([Cl- ]in) and extracellular K+ levels. This mechanism leads
to a short-term increase in [Cl- ] and EGABA. Long-term modification in [Cl- ]in and EGABA are believed
to require increased postsynaptic Ca2+ levels and changes in Cl- transporter property or expression
(Marty & Llano, 2005). Support for the role of EGABA and [Cl- ] is provided by two studies. Recently,
the GABA released by axo-axonic interneurons has shown to be excitatory for the postsynaptic
pyramidal neuron, due to local changes in the reversal potential for GABA (EGABA) (Szabadics et al.,
2006). GABAergic axo-axonic interneurons project solely to the initial segment of axons of cortical
pyramidal cells, which is the initiation site of axon potentials. In the axon initial segment of targeted
pyramidal cells, EGABA showed an increase to -50 mV compared to -73 mV in inhibited postsynaptic
compartments. The excitatory effect of the GABAergic axo-axonic interneurons was mediated by
postsynaptic GABAA receptors and increased intracellular [Cl-]. The elevated GABA reversal potential
21
and [Cl-] at the axon initial segment were correlated with the expression of potassium chloride
cotranspoter 2 (KCC2). KCC2 is involved in chloride efflux, maintaining a low intracellular chloride
level. Density of KCC2 expression was found to be decreased at the axon initial segment, in contrast
to the high expression on somatic and dendritic membranes. The low density of KCC2 leads to more
retained chloride in the cytoplasm, increasing EGABA and [Cl-], which ultimately results in the
excitatory effect of GABA release on the axon initial segment (Szabadics et al., 2006). In another
study, depolarizing GABA transmission and significantly increased EGABA (-55 mV) have been found
in a subset of subicular neurons. Interestingly, ictal activity in these tissues from temporal lobe
epilepsy (TLE) patients was blocked by GABAA receptor antagonists (Cohen et al., 2002). Taken
together, a direct excitatory effect of GABAergic interneurons may contribute to the net excitation of
pyramidal cells under certain conditions.
The findings of Szabadics et al. (2006) highlight the fact that also in the adult central nervous system,
GABA is not solely an inhibitory neurotransmitter. Rather, the properties and function of GABA are
more diverse and change during development. It is known that during early development, GABA is an
excitatory neurotransmitter, and becomes inhibitory in later stages of development. In immature
neurons, the excitatory effect of GABA is caused by an increased intracellular [Cl-], which results in a
resting membrane potential lower than ECl. During later stages in development, intracellular [Cl-]
decrease, and EGABA drops with approximately 20 mV (Figure 4). These changes in reversal potential
are caused by differential expression of two chloride transporters, Na+–K+–2Cl–co- transporter 1
(NKCC1) en KCC2. Whereas KCC2 lowers intracellular chloride levels, NKCC1 accumulates
chloride in the cytoplasm (Stein & Nicoll, 2003). In immature neurons, higher expression of NKCC1
results in higher [Cl- ]in and EGABA. In later stages of development and postnatally, a rapid increase of
KCC2 expression changes the reversal potential of neurons to a lower level, and inverting the effect of
GABA to inhibition (Owens & Kriegstein, 2002; Figure 5). However, as discussed above, a subset of
synapses has retained the excitatory effect of GABA. If we keep in mind that excitation can involve
both glutamate and GABA, neuronal excitability and the excitation-inhibition balance appears to be
more complicated. Here, I will not discuss all consequences and implications in detail. The most
important feature to be noted here, is the potential excitatory effect of GABA and a more complex
interplay between GABA and glutamate, which could be involved not only in ADNFLE, but also in
epilepsy syndromes in general.
Figure 4. EGABA and intracellular [Cl-] decrease during development. From Owens & Kriegstein (2002).
A. During embryonal development, EGABA is higher than spike threshold, resulting in depolarization of the
neuronal membrane. During later developmental stages and postnatally, E GABA decreases and becomes lower
than the threshold potential, resulting in hyperpolarization and inhibition.
B. In the same temporal window and pattern as EGABA decrease, [Cl- ]in also drop during development.
22
Figure 5. Differential expression of NKCC1 and KCC2 mediate the decrease of intracellular chloride levels in
mature neurons. From Stein & Nicoll (2003).
In immature neurons (left), higher NKCC1 expression drives the predominant chloride influx and high [Cl- ]in.
Activation of GABAA receptors results in chloride efflux and depolarization. In mature neurons (right), higher
KCC2 expression lowers [Cl- ]in. As a result, GABA neurotransmission causes influx of chloride and
hyperpolarization.
5.3 Stage II non-REM sleep
During non-REM sleep, the physiological acetylcholine levels in the brain are the lowest. As discussed
above, nAChRs desensitize after prolonged ligand binding. Low levels of ACh during non-REM sleep
results in less receptor desensitization, and the receptors regain their sensitivity. During non-REM
sleep, the nAChR receptors would be in their most sensitive state due to the reduced sensitization.
Acetylcholine is released during the transition from non-REM to REM sleep, as well as in REM sleep,
and during wakefulness. The response of hypersensitive nAChR during non-REM sleep to
acetylcholine could activate pathways to trigger seizure onset. During wakefulness, receptors
desensitize due to higher levels of acetylcholine, and physiological acetylcholine will no longer trigger
seizures (Willoughby et al., 2003).
The transition from non-REM to REM sleep involves thalamocortical interruption of the sleep
spindles. Cholinergic input to the reticular nucleus of the thalamus is believed to mediate the
thalamocortical oscillations in sleep spindles. Disturbances in the thalamocortical system or its input
may alter the spindle synchronizations (Combi et al., 2004). In support of this theory, sleep spindles
right preceding seizure onset were found to have a longer duration in a few ADNFLE patients (Picard
et al., 2007). The increased duration of sleep spindles could reflect sustained cholinergic input to the
reticular nucleus of the thalamus, which results in prolonged sleep spindle oscillations and cortical
synchronization. The hypothesis of higher cholinergic input to the thalamocortical circuit is in line
with a gain-of-function effect of the cholinergic system. Maintained synchronization in the
thalamocortical circuit caused by increased cholinergic activity may induce prolonged sleep spindles.
Evidence for the role of synchronization is provided by an EEG study, which showed increased
cortical synchronization in the 8-12 Hz frequency band during the recorded seizures in ADNFLE
patients (Ferri et al., 2004). This finding indicates an increase in sleep spindle synchronizations, which
reflects aberrant activity in the thalamocortical circuit and a failure to terminate sleep spindle
synchronization. However, the relation between spindle synchronization and the onset of nocturnal
motor seizures is not yet understood. Increased synchronization could contribute to the seizure onset,
although it is also possible that the observed abnormality is a side effect of hypersensitive nAChRs
and the developmental changes they induced.
23
5.4 Frontal origin
The main seizure onset in the frontal lobe is a puzzling manifestation. This clinical feature of
ADNFLE has not been a prominent subject of research since the seizure origin is still not established.
As mentioned above, abnormal frontal activity has only been observed in 50% of the patients with
abnormal ictal EEG recordings (Combi et al., 2004). Ictal EEG manifestations and frontal
abnormalities have not been correlated to specific mutations to date. One explanation to account for
the frontal origin relies on the properties of thalamocortical network. As mentioned above, the
peduncolopontine-lateraldorsal tegmental complex (PPN/LDTN) provides the main cholinergic input
to the thalamus. A PET study revealed an increase of α4β2 nAChR binding in the interpeduncular
nucleus (IPN) in the midbrain. The IPN projects to the laterodorsal tegmental nucleus (LDT), which is
a part of the PPN/LDTN. The LDT neurons release acetylcholine in the mediodorsal nucleus of the
thalamus (Brandel et al., 1991), which results in arousal and an activated state of the thalamocortical
system (Picard et al., 2006). Interestingly, the mediodorsal nucleus of the thalamus projects to the
orbital and dorsolateral prefrontal cortex, the cingulate cortex, and the insula (Picard et al., 2007).
Thus, the increased α4β2 nAChR density in the IPN could specifically induce a higher excitability of
the frontal and insular regions. Increased density of α4β2 subunits may be a developmental
abnormality in the cholinergic system, caused by the hyperactive mutant receptors during early stages
of development. In turn, the abnormal α4β2 expression in the IPN could contribute in seizure
generation following the pathway described above (Picard et al., 2006). Another hypothesis for the
frontal origin is the direct induction of glutamate release in the prefrontal cortex by ACh binding in the
thalamocortical system. Acetylcholine release in thalamocortical terminals has been shown to increase
glutamate transmission in the prefrontal cortex (Lambe et al., 2003), which results in direct excitation
of the frontal lobe. Taken together, these findings may partly account for the frontal and insular origin
of the seizures. However, the neuronal and molecular properties of the frontal lobe, such as nAChR
subunit expression, should be further studied to determine the cause of the frontal seizure origin in
ADNFLE patients.
5.5 Developmental defects
A role for developmental defects has been suggested by the early disease onset in ADNFLE patients,
and the tendency for seizure remission later in adulthood. The pivotal role of mutated nAChR in
developmental stages has been recently illustrated in an animal model (Manfredi et al., 2009).
Inhibition of mutant β2 nAChR expression during E1-P15 rescues the disease phenotype in transgenic
mice, protecting them completely from seizures. In contrast, mutant mice which express
hypersensitive nAChR throughout development, suffer from spontaneous seizures. Suppression of
mutant nAChR expression in adult mice did also not protect the mice from seizures, indicating that the
expression of mutant nAChR during development is the cause for disease onset early in life (Manfredi
et al., 2009). This interesting finding is supported by the expression pattern and proposed roles of
nAChR subunits during neuronal development. It is known that nAChR are expressed early in the
development of the nervous system. They are involved in neuronal morphogenesis, migration, circuit
formation and synapse formation. During the prenatal development of the central nervous system,
expression of nAChR is the highest between 3-6 months after fertilization. It is possible that during
this period of circuit formation and high nAChR expression, the mutated nAChR cause defects in
cholinergic systems, leading to aberrant neuronal wiring and higher cortical excitability (Combi et al.,
2004). Interestingly, nicotine is known to contribute to the upregulation of α4β2 nAChR at
concentrations above 500 nM (Kuryatov et al., 2005). Chronic exposure to nicotine during embryonic
stages caused by nicotine consumption by the mother may increase the number of nAChRs, resulting
in an even higher expression of hypersensitive receptors, further affecting neuronal circuits and
excitability. Unfortunately, the tobacco consumption of the mothers in ADNFLE families, or mothers
of sporadic patients, have not yet been studied and associated with disease outcome.
24
Chapter 6: DISCUSSION & CONCLUSION
The modulatory function of nAChRs has led to the suggestion of an imbalance of excitatory and
inhibitory signals into the cortex. Recently, the role for excessive GABAergic transmission has been a
surprising finding, which leads to new possible hypotheses on ADNFLE. The emerging evidence that
GABA may not only act in inhibitory pathways, but can rather also be excitatory or synchronize
glutamate networks, has major implication for our understanding of the excitation-inhibition balance.
It could also help to explain why antiepileptic drugs which act on GABA neurotransmission (by
increasing GABA release or inhibiting breakdown) have no effects on epilepsy patients. However, the
role of GABAergic inhibition or excitation in ADNFLE should be not overvalued, since only very few
observations suggest a major role for GABA. To date, two ADNFLE mouse models and a few
electrophysiological studies on epileptic tissue have proposed the involvement of GABA networks in
epilepsy. Generally, it would be more likely that an increased neuronal excitability due to
hypersensitive nAChRs is the major underlying mechanism. Higher glutamate levels and neuronal
excitation are generally believed to underlie epilepsy disorders, and it has not been proved that
ADNFLE is an exception. The modified GABA pathways could also be a secondary effect, caused by
the seizures or intrinsic adaptation mechanisms. Nonetheless, the role of GABA in neuronal
excitability should be further explored to find more support for its contribution in ADNFLE, or to gain
more valuable information on the excitatory-inhibitory balance in the brain. For example, the EGABA
and [Cl- ]in could be studied in post-mortem or surgically removed ADNFLE brain tissue, if possible,
to study the hypothesis of a elevated EGABA and excitatory role of GABA. Another potential source for
support of GABAergic transmission in ADNFLE is the detailed characterization of neuronal activity
in new, valid animal models. In conclusion, a predominant role of GABAergic transmission is unlikely
in ADNFLE pathogenesis, but could rather serve as a minor contributor in a subset of cases, or an
additional mechanism to increased glutamate transmission by synchronization.
Several possible contributors and pathways have been discussed in the previous chapter. It is very
likely that a subset of the proposed neuronal circuits and pathways contribute together to disease onset.
In general, neuronal excitability is enhanced by the gain-of-function of the cholinergic system during
development, leading to a more seizure-susceptible state of the brain. Seizure onset could be induced
by increased cholinergic input to the thalamocortical circuit, caused by hypersensitive α4β2 nAChR.
The aberrant cholinergic input results in excessive release of glutamate in the frontal cortex. In
parallel, GABAergic inhibitory interneurons producing synchronized depolarization may add to the
existing imbalance in excitatory and inhibitory signals. It should be noted that the proposed disease
mechanisms are only broad outline, the details and links between different contributors to ADNFLE
pathogenesis remains to be identified. In this chapter I will propose a direction for further research
which appears to be most promising: the role of nicotine as a environmental factor.
The relationship between prenatal nicotine exposure and ADNFLE disease outcome has not yet been
studied. An interesting and feasible study to gain more insight into the developmental aspect of
ADNFLE involves the examination of nicotine consumption by the mothers in ADNFLE families
during pregnancy. Hypothetically, nicotine in the blood of the mother could reach the fetus and exert
effects on the developing central nervous system. It has been shown that nicotine acts as a chaperone
to upregulate nAChR expression (Kuryatov et al., 2005). Upregulated nAChR levels may be another
pathway for gain-of-function of the cholinergic system. The effects on the immature neuronal
networks could lead to neuronal malformations, such as improper circuit wiring, impaired
synaptogenesis, and higher neuronal excitability. All consequences of higher (mutant) nAChR levels
can contribute to the onset of symptoms during childhood. More expression of nAChR would also
suggest a gain-of-function of the cholinergic system, further establishing the role of hyperactivity of
nAChR in ADNFLE. It would be especially interesting to study the nicotine consumption by the
mothers of patients who do not carry mutations in nAChR subunits. Hypothetically, higher expression
levels of nAChR subunits during development will result in gain-of-function of the cholinergic system
in subject even without genetic defects affecting the cholinergic system. Therefore, nicotine exposure
25
during development may contribute to ADNFLE symptoms in subjects without identified mutations.
Furthermore, if the mothers indeed show high nicotine consumption, which leads to potentially higher
levels of (mutant) nAChR, a possible explanation for the observed decrease in symptoms during adult
life could be proposed. Overall, the nAChR levels decrease during life, which would compensate for
the higher receptor levels, or the hypersensitive receptors (Gotti & Clementi, 2004). This natural
compensation may underlie the remission of ADNFLE over time. The decrease of nAChR is not
linear, nor identical for the different subunits and brain areas, which may cause the variability in
symptom remission and severity.
A drawback of the proposed study is the large amount of tobacco-consuming mothers, which is much
higher than the number of ADNFLE patients. It means that only a very small percentage of the fetuses
exposed to nicotine develop ADNFLE. Several factors may explain why nicotine exposure during
development does not affect more fetuses. First, the dose and time window of nicotine consumption
may be important for the effects on neural development. Not sufficient nicotine levels outside the
crucial developmental stage for neuronal connections will leave the architecture and synapse
formation in the neuronal circuits unaffected. Second, healthy embryos without a genetic defect or
predisposition may not be severely affected by nicotine, and hence do not have apparent neuronal
defects or a phenotype. It is very likely that in ADNFLE patients, nicotine exposure multiplies the
effects of hypersensitive nAChR, or other unidentified mutations, and thereby act as a contributor to
disease onset. Thus, patients have a genetic background which makes them more susceptible to
malformation of neuronal networks as a result of nicotine exposure. Therefore, the study on the
relation between nicotine exposure and ADNFLE disease outcome should start from the ADNFLE
patients and their mothers, instead of looking for nicotine-consuming mothers and the effects on the
embryo.
In conclusion, the study of ADNFLE has made major progress during the recent years. The identified
mutations in nAChR and the subsequent hypersensitivity of the cholinergic system have been
breakthroughs in this field of research. However, more studies are still needed to determine the other
genetic defects which are underlying this disorder. The detailed disease mechanisms and changes in
excitation-inhibition are still not known. Especially the developmental brain malformations caused by
mutant nAChR need to be identified, since they are the first neuronal changes in ADNFLE patients.
Therefore, studies into the pathogenesis should focus more on the more detailed profiling of
developmental changes. However, probably the most promising study is the determination of the link
developmental nicotine exposure and the disease outcome in ADNFLE patients. If a link indeed exists,
these data will fully establish the gain-of-function of the cholinergic system as a main developmental
cause for ADNFLE. In addition, aberrant cholinergic transmission in patients without a genetic cause
can be explained, as well as the changes in seizure severity during life. Therefore, the greatest priority
of research should be focused at the role of nicotine in the upregulation of nAChR during
development. Furthermore, the effect of low doses of nicotine as a therapy should be studied in more
patients to study the effectiveness of nicotine and varieties in response to treatment between different
patients and families. Low doses of nicotine (10nM) is enough to induce desensitization of nAChR
without upregulation of the numbers of nAChR on the cell membrane, for which a higher
concentration is needed (500nM). Even more interesting is to study the link between the responses to
nicotine treatment and the underlying mutation (if identified), and if the mutation determine therapy
outcome. Different responsiveness to nicotine treatment is an indicator for the differential effect of the
mutations. Taken together, these studies have high priority since they are easy, feasible, and may
provide the patients with a more effective treatment, which is in the end the highest interest for
ADNFLE patients.
26
REFERENCES
Andermann, F., Kobayashi, E., & Andermann, E. (2005). Genetic focal epilepsies: State of the art and
paths to the future. Epilepsia, 46 Suppl 10, 61-67. doi:10.1111/j.1528-1167.2005.00361.x
Aridon, P., Marini, C., Di Resta, C., Brilli, E., De Fusco, M., Politi, F., et al. (2006). Increased
sensitivity of the neuronal nicotinic receptor alpha 2 subunit causes familial epilepsy with
nocturnal wandering and ictal fear. American Journal of Human Genetics, 79(2), 342-350.
doi:10.1086/506459
Azam, L., Chen, Y., & Leslie, F. M. (2007). Developmental regulation of nicotinic acetylcholine
receptors within midbrain dopamine neurons. Neuroscience, 144(4), 1347-1360.
doi:10.1016/j.neuroscience.2006.11.011
Bandyopadhyay, S., Sutor, B., & Hablitz, J. J. (2006). Endogenous acetylcholine enhances
synchronized interneuron activity in rat neocortex. Journal of Neurophysiology, 95(3), 19081916. doi:10.1152/jn.00881.2005
Bertrand, D., Elmslie, F., Hughes, E., Trounce, J., Sander, T., Bertrand, S., et al. (2005). The
CHRNB2 mutation I312M is associated with epilepsy and distinct memory deficits.
Neurobiology of Disease, 20(3), 799-804. doi:10.1016/j.nbd.2005.05.013
Bertrand, S., Weiland, S., Berkovic, S. F., Steinlein, O. K., & Bertrand, D. (1998). Properties of
neuronal nicotinic acetylcholine receptor mutants from humans suffering from autosomal
dominant nocturnal frontal lobe epilepsy. British Journal of Pharmacology, 125(4), 751-760.
doi:10.1038/sj.bjp.0702154
Brandel, J. P., Hirsch, E. C., Malessa, S., Duyckaerts, C., Cervera, P., & Agid, Y. (1991). Differential
vulnerability of cholinergic projections to the mediodorsal nucleus of the thalamus in senile
dementia of alzheimer type and progressive supranuclear palsy. Neuroscience, 41(1), 25-31.
Brodtkorb, E., & Picard, F. (2006). Tobacco habits modulate autosomal dominant nocturnal frontal
lobe epilepsy. Epilepsy & Behavior : E&B, 9(3), 515-520. doi:10.1016/j.yebeh.2006.07.008
27
Callenbach, P. M., van den Maagdenberg, A. M., Hottenga, J. J., van den Boogerd, E. H., de Coo, R.
F., Lindhout, D., et al. (2003). Familial partial epilepsy with variable foci in a dutch family:
Clinical characteristics and confirmation of linkage to chromosome 22q. Epilepsia, 44(10), 12981305.
Cash, S. S., Halgren, E., Dehghani, N., Rossetti, A. O., Thesen, T., Wang, C., et al. (2009). The human
K-complex represents an isolated cortical down-state. Science (New York, N.Y.), 324(5930),
1084-1087. doi:10.1126/science.1169626
Chen, Y., Wu, L., Fang, Y., He, Z., Peng, B., Shen, Y., et al. (2009). A novel mutation of the nicotinic
acetylcholine receptor gene CHRNA4 in sporadic nocturnal frontal lobe epilepsy. Epilepsy
Research, 83(2-3), 152-156. doi:10.1016/j.eplepsyres.2008.10.009
Cho, Y. W., Yi, S. D., Lim, J. G., Kim, D. K., & Motamedi, G. K. (2008). Autosomal dominant
nocturnal frontal lobe epilepsy and mild memory impairment associated with CHRNB2 mutation
I312M in the neuronal nicotinic acetylcholine receptor. Epilepsy & Behavior : E&B, 13(2), 361365. doi:10.1016/j.yebeh.2008.04.017
Chou, I. C., Lee, C. C., Huang, C. C., Wu, J. Y., Tsai, J. J., Tsai, C. H., et al. (2003). Association of
the neuronal nicotinic acetylcholine receptor subunit alpha4 polymorphisms with febrile
convulsions. Epilepsia, 44(8), 1089-1093.
Cohen, I., Navarro, V., Clemenceau, S., Baulac, M., & Miles, R. (2002). On the origin of interictal
activity in human temporal lobe epilepsy in vitro. Science (New York, N.Y.), 298(5597), 14181421. doi:10.1126/science.1076510
Combi, R., Dalpra, L., Ferini-Strambi, L., & Tenchini, M. L. (2005). Frontal lobe epilepsy and
mutations of the corticotropin-releasing hormone gene. Annals of Neurology, 58(6), 899-904.
doi:10.1002/ana.20660
Combi, R., Dalpra, L., Malcovati, M., Oldani, A., Tenchini, M. L., & Ferini-Strambi, L. (2004).
Evidence for a fourth locus for autosomal dominant nocturnal frontal lobe epilepsy. Brain
Research Bulletin, 63(5), 353-359. doi:10.1016/j.brainresbull.2003.12.007
28
Combi, R., Dalpra, L., Tenchini, M. L., & Ferini-Strambi, L. (2004). Autosomal dominant nocturnal
frontal lobe epilepsy--a critical overview. Journal of Neurology, 251(8), 923-934.
doi:10.1007/s00415-004-0541-x
Combi, R., Ferini-Strambi, L., & Luisa Tenchini, M. (2008). CHRNA2 mutations are rare in the NFLE
population: Evaluation of a large cohort of italian patients. Sleep Medicine,
doi:10.1016/j.sleep.2007.11.010
Combi, R., Ferini-Strambi, L., Montruccoli, A., Bianchi, V., Malcovati, M., Zucconi, M., et al. (2005).
Two new putative susceptibility loci for ADNFLE. Brain Research Bulletin, 67(4), 257-263.
doi:10.1016/j.brainresbull.2005.06.032
Combi, R., Ferini-Strambi, L., & Tenchini, M. L. (2008). Compound heterozygosity with dominance
in the corticotropin releasing hormone (CRH) promoter in a case of nocturnal frontal lobe
epilepsy. Journal of Sleep Research, 17(3), 361-362. doi:10.1111/j.1365-2869.2008.00674.x
D'Antuono, M., Louvel, J., Kohling, R., Mattia, D., Bernasconi, A., Olivier, A., et al. (2004). GABAA
receptor-dependent synchronization leads to ictogenesis in the human dysplastic cortex. Brain : A
Journal of Neurology, 127(Pt 7), 1626-1640. doi:10.1093/brain/awh181
De Fusco, M., Becchetti, A., Patrignani, A., Annesi, G., Gambardella, A., Quattrone, A., et al. (2000).
The nicotinic receptor beta 2 subunit is mutant in nocturnal frontal lobe epilepsy. Nature
Genetics, 26(3), 275-276. doi:10.1038/81566
De Marco, E. V., Gambardella, A., Annesi, F., Labate, A., Carrideo, S., Forabosco, P., et al. (2007).
Further evidence of genetic heterogeneity in families with autosomal dominant nocturnal frontal
lobe epilepsy. Epilepsy Research, 74(1), 70-73. doi:10.1016/j.eplepsyres.2006.12.006
Derry, C. P., Heron, S. E., Phillips, F., Howell, S., Macmahon, J., Phillips, H. A., et al. (2008). Severe
autosomal dominant nocturnal frontal lobe epilepsy associated with psychiatric disorders and
intellectual disability. Epilepsia, doi:10.1111/j.1528-1167.2008.01652.x
Deveau, J. S., Lindinger, M. I., & Grodzinski, B. (2005). An improved method for constructing and
selectively silanizing double-barreled, neutral liquid-carrier, ion-selective microelectrodes.
Biological Procedures Online, 7, 31-40. doi:10.1251/bpo103
29
di Corcia, G., Blasetti, A., De Simone, M., Verrotti, A., & Chiarelli, F. (2005). Recent advances on
autosomal dominant nocturnal frontal lobe epilepsy: "understanding the nicotinic acetylcholine
receptor (nAChR)". European Journal of Paediatric Neurology : EJPN : Official Journal of the
European Paediatric Neurology Society, 9(2), 59-66. doi:10.1016/j.ejpn.2004.12.006
Diaz-Otero, F., Quesada, M., Morales-Corraliza, J., Martinez-Parra, C., Gomez-Garre, P., & Serratosa,
J. M. (2008). Autosomal dominant nocturnal frontal lobe epilepsy with a mutation in the
CHRNB2 gene. Epilepsia, 49(3), 516-520. doi:10.1111/j.1528-1167.2007.01328.x
Dobelis, P., Hutton, S., Lu, Y., & Collins, A. C. (2003). GABAergic systems modulate nicotinic
receptor-mediated seizures in mice. The Journal of Pharmacology and Experimental
Therapeutics, 306(3), 1159-1166. doi:10.1124/jpet.103.053066
El Helou, J., Navarro, V., Depienne, C., Fedirko, E., Leguern, E., Baulac, M., et al. (2008). Kcomplex-induced seizures in autosomal dominant nocturnal frontal lobe epilepsy. Clinical
Neurophysiology : Official Journal of the International Federation of Clinical Neurophysiology,
doi:10.1016/j.clinph.2008.07.212
Elmslie, F. V., Rees, M., Williamson, M. P., Kerr, M., Kjeldsen, M. J., Pang, K. A., et al. (1997).
Genetic mapping of a major susceptibility locus for juvenile myoclonic epilepsy on chromosome
15q. Human Molecular Genetics, 6(8), 1329-1334.
Fedi, M., Berkovic, S. F., Scheffer, I. E., O'Keefe, G., Marini, C., Mulligan, R., et al. (2008). Reduced
striatal D1 receptor binding in autosomal dominant nocturnal frontal lobe epilepsy. Neurology,
71(11), 795-798. doi:10.1212/01.wnl.0000316192.52731.77
Ferri, R., Stam, C. J., Lanuzza, B., Cosentino, F. I., Elia, M., Musumeci, S. A., et al. (2004). Different
EEG frequency band synchronization during nocturnal frontal lobe seizures. Clinical
Neurophysiology : Official Journal of the International Federation of Clinical Neurophysiology,
115(5), 1202-1211. doi:10.1016/j.clinph.2003.12.014
Fonck, C., Cohen, B. N., Nashmi, R., Whiteaker, P., Wagenaar, D. A., Rodrigues-Pinguet, N., et al.
(2005). Novel seizure phenotype and sleep disruptions in knock-in mice with hypersensitive
alpha 4* nicotinic receptors. The Journal of Neuroscience : The Official Journal of the Society for
Neuroscience, 25(49), 11396-11411. doi:10.1523/JNEUROSCI.3597-05.2005
30
Fonck, C., Nashmi, R., Deshpande, P., Damaj, M. I., Marks, M. J., Riedel, A., et al. (2003). Increased
sensitivity to agonist-induced seizures, straub tail, and hippocampal theta rhythm in knock-in
mice carrying hypersensitive alpha 4 nicotinic receptors. The Journal of Neuroscience : The
Official Journal of the Society for Neuroscience, 23(7), 2582-2590.
Gotti, C., & Clementi, F. (2004). Neuronal nicotinic receptors: From structure to pathology. Progress
in Neurobiology, 74(6), 363-396. doi:10.1016/j.pneurobio.2004.09.006
Gu, W., Bertrand, D., & Steinlein, O. K. (2007). A major role of the nicotinic acetylcholine receptor
gene CHRNA2 in autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE) is unlikely.
Neuroscience Letters, 422(1), 74-76. doi:10.1016/j.neulet.2007.06.006
Hill, E. L., Gallopin, T., Ferezou, I., Cauli, B., Rossier, J., Schweitzer, P., et al. (2007). Functional
CB1 receptors are broadly expressed in neocortical GABAergic and glutamatergic neurons.
Journal of Neurophysiology, 97(4), 2580-2589. doi:10.1152/jn.00603.2006
Hirose, S., Iwata, H., Akiyoshi, H., Kobayashi, K., Ito, M., Wada, K., et al. (1999). A novel mutation
of CHRNA4 responsible for autosomal dominant nocturnal frontal lobe epilepsy. Neurology,
53(8), 1749-1753.
Hirose, S., Mitsudome, A., Okada, M., Kaneko, S., & Epilepsy Genetic Study Group, Japan. (2005).
Genetics of idiopathic epilepsies. Epilepsia, 46 Suppl 1, 38-43. doi:10.1111/j.00139580.2005.461011.x
Hirtz, D., Thurman, D. J., Gwinn-Hardy, K., Mohamed, M., Chaudhuri, A. R., & Zalutsky, R. (2007).
How common are the "common" neurologic disorders? Neurology, 68(5), 326-337.
doi:10.1212/01.wnl.0000252807.38124.a3
Hoda, J. C., Gu, W., Friedli, M., Phillips, H. A., Bertrand, S., Antonarakis, S. E., et al. (2008). Human
nocturnal frontal lobe epilepsy: Pharmocogenomic profiles of pathogenic nicotinic acetylcholine
receptor beta-subunit mutations outside the ion channel pore. Molecular Pharmacology, 74(2),
379-391. doi:10.1124/mol.107.044545
Hollrigel, G. S., Chen, K., Baram, T. Z., & Soltesz, I. (1998). The pro-convulsant actions of
corticotropin-releasing hormone in the hippocampus of infant rats. Neuroscience, 84(1), 71-79.
31
Ji, D., & Dani, J. A. (2000). Inhibition and disinhibition of pyramidal neurons by activation of
nicotinic receptors on hippocampal interneurons. Journal of Neurophysiology, 83(5), 2682-2690.
Kandel E.R., Schwartz J.H., Jessell T.M. (2000). Principles of Neural Science, 4th ed. McGraw-Hill,
New York. ISBN 0-8385-7701-6
Kaneko, S., Okada, M., Iwasa, H., Yamakawa, K., & Hirose, S. (2002). Genetics of epilepsy: Current
status and perspectives. Neuroscience Research, 44(1), 11-30.
Klaassen, A., Glykys, J., Maguire, J., Labarca, C., Mody, I., & Boulter, J. (2006). Seizures and
enhanced cortical GABAergic inhibition in two mouse models of human autosomal dominant
nocturnal frontal lobe epilepsy. Proceedings of the National Academy of Sciences of the United
States of America, 103(50), 19152-19157. doi:10.1073/pnas.0608215103
Kumari, S., Borroni, V., Chaudhry, A., Chanda, B., Massol, R., Mayor, S., et al. (2008). Nicotinic
acetylcholine receptor is internalized via a rac-dependent, dynamin-independent endocytic
pathway. The Journal of Cell Biology, 181(7), 1179-1193. doi:10.1083/jcb.200709086
Kuryatov, A., Luo, J., Cooper, J., & Lindstrom, J. (2005). Nicotine acts as a pharmacological
chaperone to up-regulate human alpha4beta2 acetylcholine receptors. Molecular Pharmacology,
68(6), 1839-1851. doi:10.1124/mol.105.012419
Kuryatov, A., Onksen, J., & Lindstrom, J. (2008). Roles of accessory subunits in alpha4beta2(*)
nicotinic receptors. Molecular Pharmacology, 74(1), 132-143. doi:10.1124/mol.108.046789
Labarca, C., Schwarz, J., Deshpande, P., Schwarz, S., Nowak, M. W., Fonck, C., et al. (2001). Point
mutant mice with hypersensitive alpha 4 nicotinic receptors show dopaminergic deficits and
increased anxiety. Proceedings of the National Academy of Sciences of the United States of
America, 98(5), 2786-2791. doi:10.1073/pnas.041582598
Lambe, E. K., Picciotto, M. R., & Aghajanian, G. K. (2003). Nicotine induces glutamate release from
thalamocortical terminals in prefrontal cortex. Neuropsychopharmacology : Official Publication
of the American College of Neuropsychopharmacology, 28(2), 216-225.
doi:10.1038/sj.npp.1300032
32
Lee, C. C., Chou, I. C., Tsai, C. H., Wan, L., Shu, Y. A., Tsai, Y., et al. (2007). Association of
idiopathic generalized epilepsy with polymorphisms in the neuronal nicotinic acetylcholine
receptor subunits. Journal of Clinical Laboratory Analysis, 21(2), 67-70. doi:10.1002/jcla.20155
Lena, C., Popa, D., Grailhe, R., Escourrou, P., Changeux, J. P., & Adrien, J. (2004). Beta2-containing
nicotinic receptors contribute to the organization of sleep and regulate putative micro-arousals in
mice. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 24(25),
5711-5718. doi:10.1523/JNEUROSCI.3882-03.2004
Leniger, T., Kananura, C., Hufnagel, A., Bertrand, S., Bertrand, D., & Steinlein, O. K. (2003). A new
Chrna4 mutation with low penetrance in nocturnal frontal lobe epilepsy. Epilepsia, 44(7), 981985.
Lipovsek, M., Plazas, P., Savino, J., Klaassen, A., Boulter, J., Elgoyhen, A. B., et al. (2008).
Properties of mutated murine alpha4beta2 nicotinic receptors linked to partial epilepsy.
Neuroscience Letters, 434(2), 165-169. doi:10.1016/j.neulet.2007.12.061
Magnusson, A., Stordal, E., Brodtkorb, E., & Steinlein, O. (2003). Schizophrenia, psychotic illness
and other psychiatric symptoms in families with autosomal dominant nocturnal frontal lobe
epilepsy caused by different mutations. Psychiatric Genetics, 13(2), 91-95.
doi:10.1097/01.ypg.0000056173.32550.b0
Manfredi, I., Zani, A. D., Rampoldi, L., Pegorini, S., Bernascone, I., Moretti, M., et al. (2009).
Expression of mutant {beta}2 nicotinic receptors during development is crucial for
epileptogenesis. Human Molecular Genetics, doi:10.1093/hmg/ddp004
Mann, E. O., & Mody, I. (2008). The multifaceted role of inhibition in epilepsy: Seizure-genesis
through excessive GABAergic inhibition in autosomal dominant nocturnal frontal lobe epilepsy.
Current Opinion in Neurology, 21(2), 155-160. doi:10.1097/WCO.0b013e3282f52f5f
Marini, C., & Guerrini, R. (2007). The role of the nicotinic acetylcholine receptors in sleep-related
epilepsy. Biochemical Pharmacology, 74(8), 1308-1314. doi:10.1016/j.bcp.2007.06.030
Marty, A., & Llano, I. (2005). Excitatory effects of GABA in established brain networks. Trends in
Neurosciences, 28(6), 284-289. doi:10.1016/j.tins.2005.04.003
33
Mathews, G. C. (2007). Is too much inhibition to blame in autosomal dominant nocturnal frontal lobe
epilepsy? Epilepsy Currents / American Epilepsy Society, 7(4), 114-116. doi:10.1111/j.15357511.2007.00193.x
McColl, C. D., Horne, M. K., Finkelstein, D. I., Wong, J. Y., Berkovic, S. F., & Drago, J. (2003).
Electroencephalographic characterisation of pentylenetetrazole-induced seizures in mice lacking
the alpha 4 subunit of the neuronal nicotinic receptor. Neuropharmacology, 44(2), 234-243.
Miyazawa, A., Fujiyoshi, Y., & Unwin, N. (2003). Structure and gating mechanism of the
acetylcholine receptor pore. Nature, 423(6943), 949-955. doi:10.1038/nature01748
Mody, I. (2003). Calcium and autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE).
Epilepsy Currents / American Epilepsy Society, 3(6), 221-222. doi:10.1046/j.15357597.2003.03603.x
Neubauer, B. A., Fiedler, B., Himmelein, B., Kampfer, F., Lassker, U., Schwabe, G., et al. (1998).
Centrotemporal spikes in families with rolandic epilepsy: Linkage to chromosome 15q14.
Neurology, 51(6), 1608-1612.
Ortells, M. O., & Barrantes, G. E. (2002). Molecular modelling of the interactions of carbamazepine
and a nicotinic receptor involved in the autosomal dominant nocturnal frontal lobe epilepsy.
British Journal of Pharmacology, 136(6), 883-895. doi:10.1038/sj.bjp.0704786
Owens, D. F., & Kriegstein, A. R. (2002). Is there more to GABA than synaptic inhibition? Nature
Reviews.Neuroscience, 3(9), 715-727. doi:10.1038/nrn919
Paradiso, K. G., & Steinbach, J. H. (2003). Nicotine is highly effective at producing desensitization of
rat alpha4beta2 neuronal nicotinic receptors. The Journal of Physiology, 553(Pt 3), 857-871.
doi:10.1113/jphysiol.2003.053447
Parrino, L., Halasz, P., Tassinari, C. A., & Terzano, M. G. (2006). CAP, epilepsy and motor events
during sleep: The unifying role of arousal. Sleep Medicine Reviews, 10(4), 267-285.
doi:10.1016/j.smrv.2005.12.004
Peng, C. T., Chou, I. C., Li, C. I., Hsu, Y. A., Tsai, C. H., & Tsai, F. J. (2004). Association of the
nicotinic receptor beta 2 subunit and febrile seizures. Pediatric Neurology, 30(3), 186-189.
doi:10.1016/j.pediatrneurol.2003.08.001
34
Phillips, H. A., Favre, I., Kirkpatrick, M., Zuberi, S. M., Goudie, D., Heron, S. E., et al. (2001).
CHRNB2 is the second acetylcholine receptor subunit associated with autosomal dominant
nocturnal frontal lobe epilepsy. American Journal of Human Genetics, 68(1), 225-231.
doi:10.1086/316946
Phillips, H. A., Scheffer, I. E., Berkovic, S. F., Hollway, G. E., Sutherland, G. R., & Mulley, J. C.
(1995). Localization of a gene for autosomal dominant nocturnal frontal lobe epilepsy to
chromosome 20q 13.2. Nature Genetics, 10(1), 117-118. doi:10.1038/ng0595-117
Phillips, H. A., Scheffer, I. E., Crossland, K. M., Bhatia, K. P., Fish, D. R., Marsden, C. D., et al.
(1998). Autosomal dominant nocturnal frontal-lobe epilepsy: Genetic heterogeneity and evidence
for a second locus at 15q24. American Journal of Human Genetics, 63(4), 1108-1116.
doi:10.1086/302047
Picard, F., Baulac, S., Kahane, P., Hirsch, E., Sebastianelli, R., Thomas, P., et al. (2000). Dominant
partial epilepsies. A clinical, electrophysiological and genetic study of 19 european families.
Brain : A Journal of Neurology, 123 ( Pt 6)(Pt 6), 1247-1262.
Picard, F., Bertrand, S., Steinlein, O. K., & Bertrand, D. (1999). Mutated nicotinic receptors
responsible for autosomal dominant nocturnal frontal lobe epilepsy are more sensitive to
carbamazepine. Epilepsia, 40(9), 1198-1209.
Picard, F., Bruel, D., Servent, D., Saba, W., Fruchart-Gaillard, C., Schollhorn-Peyronneau, M. A., et
al. (2006). Alteration of the in vivo nicotinic receptor density in ADNFLE patients: A PET study.
Brain : A Journal of Neurology, 129(Pt 8), 2047-2060. doi:10.1093/brain/awl156
Picard, F., Megevand, P., Minotti, L., Kahane, P., Ryvlin, P., Seeck, M., et al. (2007). Intracerebral
recordings of nocturnal hyperkinetic seizures: Demonstration of a longer duration of the preseizure sleep spindle. Clinical Neurophysiology : Official Journal of the International Federation
of Clinical Neurophysiology, 118(4), 928-939. doi:10.1016/j.clinph.2006.12.014
Porter, J. T., Cauli, B., Tsuzuki, K., Lambolez, B., Rossier, J., & Audinat, E. (1999). Selective
excitation of subtypes of neocortical interneurons by nicotinic receptors. The Journal of
Neuroscience : The Official Journal of the Society for Neuroscience, 19(13), 5228-5235.
35
Raggenbass, M., & Bertrand, D. (2002). Nicotinic receptors in circuit excitability and epilepsy.
Journal of Neurobiology, 53(4), 580-589. doi:10.1002/neu.10152
Rodrigues-Pinguet, N., Jia, L., Li, M., Figl, A., Klaassen, A., Truong, A., et al. (2003). Five ADNFLE
mutations reduce the Ca2+ dependence of the mammalian alpha4beta2 acetylcholine response.
The Journal of Physiology, 550(Pt 1), 11-26. doi:10.1113/jphysiol.2002.036681
Rodrigues-Pinguet, N. O., Pinguet, T. J., Figl, A., Lester, H. A., & Cohen, B. N. (2005). Mutations
linked to autosomal dominant nocturnal frontal lobe epilepsy affect allosteric Ca2+ activation of
the alpha 4 beta 2 nicotinic acetylcholine receptor. Molecular Pharmacology, 68(2), 487-501.
doi:10.1124/mol.105.011155
Rozycka, A., Skorupska, E., Kostyrko, A., & Trzeciak, W. H. (2003). Evidence for S284L mutation of
the CHRNA4 in a white family with autosomal dominant nocturnal frontal lobe epilepsy.
Epilepsia, 44(8), 1113-1117.
Ryvlin, P. (2006). Avoid falling into the depths of the insular trap. Epileptic Disorders : International
Epilepsy Journal with Videotape, 8 Suppl 2, S37-56.
Ryvlin, P., Minotti, L., Demarquay, G., Hirsch, E., Arzimanoglou, A., Hoffman, D., et al. (2006).
Nocturnal hypermotor seizures, suggesting frontal lobe epilepsy, can originate in the insula.
Epilepsia, 47(4), 755-765. doi:10.1111/j.1528-1167.2006.00510.x
Ryvlin, P., Rheims, S., & Risse, G. (2006). Nocturnal frontal lobe epilepsy. Epilepsia, 47 Suppl 2, 8386. doi:10.1111/j.1528-1167.2006.00698.x
Stein, V., & Nicoll, R. A. (2003). GABA generates excitement. Neuron, 37(3), 375-378.
Steinlein, O. K. (2007). Genetic disorders caused by mutated acetylcholine receptors. Life Sciences,
80(24-25), 2186-2190. doi:10.1016/j.lfs.2007.03.007
Steinlein, O. K., Magnusson, A., Stoodt, J., Bertrand, S., Weiland, S., Berkovic, S. F., et al. (1997). An
insertion mutation of the CHRNA4 gene in a family with autosomal dominant nocturnal frontal
lobe epilepsy. Human Molecular Genetics, 6(6), 943-947.
Steinlein, O. K., Mulley, J. C., Propping, P., Wallace, R. H., Phillips, H. A., Sutherland, G. R., et al.
(1995). A missense mutation in the neuronal nicotinic acetylcholine receptor alpha 4 subunit is
36
associated with autosomal dominant nocturnal frontal lobe epilepsy. Nature Genetics, 11(2), 201203. doi:10.1038/ng1095-201
Szabadics, J., Varga, C., Molnar, G., Olah, S., Barzo, P., & Tamas, G. (2006). Excitatory effect of
GABAergic axo-axonic cells in cortical microcircuits. Science (New York, N.Y.), 311(5758),
233-235. doi:10.1126/science.1121325
Taly, A., Corringer, P. J., Grutter, T., Prado de Carvalho, L., Karplus, M., & Changeux, J. P. (2006).
Implications of the quaternary twist allosteric model for the physiology and pathology of
nicotinic acetylcholine receptors. Proceedings of the National Academy of Sciences of the United
States of America, 103(45), 16965-16970. doi:10.1073/pnas.0607477103
Tapper, A. R., McKinney, S. L., Nashmi, R., Schwarz, J., Deshpande, P., Labarca, C., et al. (2004).
Nicotine activation of alpha4* receptors: Sufficient for reward, tolerance, and sensitization.
Science (New York, N.Y.), 306(5698), 1029-1032. doi:10.1126/science.1099420
Teper, Y., Whyte, D., Cahir, E., Lester, H. A., Grady, S. R., Marks, M. J., et al. (2007). Nicotineinduced dystonic arousal complex in a mouse line harboring a human autosomal-dominant
nocturnal frontal lobe epilepsy mutation. The Journal of Neuroscience : The Official Journal of
the Society for Neuroscience, 27(38), 10128-10142. doi:10.1523/JNEUROSCI.3042-07.2007
Varadkar, S., Duncan, J. S., & Cross, J. H. (2003). Acetazolamide and autosomal dominant nocturnal
frontal lobe epilepsy. Epilepsia, 44(7), 986-987.
Willoughby, J. O., Pope, K. J., & Eaton, V. (2003). Nicotine as an antiepileptic agent in ADNFLE: An
N-of-one study. Epilepsia, 44(9), 1238-1240.
Zhu, G., Okada, M., Yoshida, S., Ueno, S., Mori, F., Takahara, T., et al. (2008). Rats harboring S284L
Chrna4 mutation show attenuation of synaptic and extrasynaptic GABAergic transmission and
exhibit the nocturnal frontal lobe epilepsy phenotype. The Journal of Neuroscience : The Official
Journal of the Society for Neuroscience, 28(47), 12465-12476. doi:10.1523/JNEUROSCI.296108.2008
37