High cortical spreading depression susceptibility and migraine-associated symptoms in Ca(v)2.1 S218L mice.
ABSTRACT The CACNA1A gene encodes the pore-forming subunit of neuronal Ca(V)2.1 Ca2+ channels. In patients, the S218L CACNA1A mutation causes a dramatic hemiplegic migraine syndrome that is associated with ataxia, seizures, and severe, sometimes fatal, brain edema often triggered by only a mild head trauma.
We introduced the S218L mutation into the mouse Cacna1a gene and studied the mechanisms for the S218L syndrome by analyzing the phenotypic, molecular, and electrophysiological consequences.
Cacna1a(S218L) mice faithfully mimic the associated clinical features of the human S218L syndrome. S218L neurons exhibit a gene dosage-dependent negative shift in voltage dependence of Ca(V)2.1 channel activation, resulting in enhanced neurotransmitter release at the neuromuscular junction. Cacna1a(S218L) mice also display an exquisite sensitivity to cortical spreading depression (CSD), with a vastly reduced triggering threshold, an increased propagation velocity, and frequently multiple CSD events after a single stimulus. In contrast, mice bearing the R192Q CACNA1A mutation, which in humans causes a milder form of hemiplegic migraine, typically exhibit only a single CSD event after one triggering stimulus.
The particularly low CSD threshold and the strong tendency to respond with multiple CSD events make the S218L cortex highly vulnerable to weak stimuli and may provide a mechanistic basis for the dramatic phenotype seen in S218L mice and patients. Thus, the S218L mouse model may prove a valuable tool to further elucidate mechanisms underlying migraine, seizures, ataxia, and trauma-triggered cerebral edema.
- [Show abstract] [Hide abstract]
ABSTRACT: Familial hemiplegic migraine (FHM) is an autosomal dominantly inherited subtype of migraine with aura, characterized by transient neurological signs and symptoms. Typical hemiplegic migraine attacks start in the first or second decade of life. Some patients with FHM suffer from daily recurrent attacks since childhood. Results from extensive studies of cellular and animal models have indicated that gene mutations in FHM increase neuronal excitability and reduce the threshold for spreading depression (SD). SD is a transient wave of profound neuronal and glial depolarization that slowly propagates throughout the brain tissue and is characterized by a high amplitude negative DC shift. After induction of SD, S218L mutant mice exhibited neurological signs highly reminiscent of clinical attacks in FHM type 1 patients carrying this mutation. FHM1 with ataxia is attributable to specific mutations that differ from mutations that cause pure FHM1 and have peculiar consequences on cerebellar Cav2.1 currents that lead to profound Purkinje cell dysfunction and neuronal loss with atrophy. SD in juvenile rats produced neuronal injury and death. Hormonal factors involved in FHM affect SD initiation and propagation. The data identify SD as a possible target of treatment of FHM. In addition, FHM is a useful model to explore the mechanisms of more common types of migraine.Iranian Journal of Child Neurology 03/2014; 8(3):6-11.This article is viewable in ResearchGate's enriched formatRG Format enables you to read in context with side-by-side figures, citations, and feedback from experts in your field.
- [Show abstract] [Hide abstract]
ABSTRACT: The human CACNA1A gene encodes the pore-forming α1 subunit of CaV2.1 (P/Q-type) calcium channels and is the locus for several neurological disorders, including episodic ataxia type 2 (EA2), spinocerebellar ataxia type 6 (SCA6) and Familial Hemiplegic Migraine type 1 (FHM1). Several spontaneous mouse Cacna1a mutant strains exist, among them Rolling Nagoya (tg (rol)), carrying the R1262G point mutation in the mouse Cacna1a gene. tg (rol) mice display a phenotype of severe gait ataxia and motor dysfunction of the hind limbs. At the functional level, the R1262G mutation results in a positive shift of the activation voltage of the CaV2.1 channel and reduced current density. γ-Aminobutyric acid type A (GABAA) receptor subunit expression depends critically on neuronal calcium influx, and GABAA receptor dysfunction has previously been described for the cerebellum of tg (rol) and other ataxic Cacna1a mutant mice. Given the expression pattern of CaV2.1, it was hypothesized that calcium dysregulation in tg (rol) might affect GABAA receptor expression in the forebrain. Herein, functional GABAA receptors in the forebrain of tg (rol) mice were quantified and pharmacologically dissociated using [(3)H] radioligand binding. No gross changes to functional GABAA receptors were identified. Future cell type-specific analyses are required to identify possible cortical contributions to the psychomotor phenotype of tg (rol) mice.Biology and medicine (Aligarh). 01/2014; 6(1).
Article: Update on Animal Models of Migraine.[Show abstract] [Hide abstract]
ABSTRACT: Migraine is a severe and debilitating disorder of the brain that involves a constellation of neurological symptoms alongside head pain. Its pathophysiology is only beginning to be understood, and is thought to involve activation and sensitization of trigeminovascular nociceptive pathways that innervate the cranial vasculature, and activation of brain stem nuclei. Much of our understanding of migraine pathophysiology stems from research conducted in animal models over the last 30 years, and the development of unique assays in animals that try to model specific aspects of migraine pathophysiology related to particular symptoms. This review will highlight some of the latest findings from these established animal models, as well as discuss the latest in the development of novel approaches in animals to study migraine.Current Pain and Headache Reports 11/2014; 18(11):462. · 2.26 Impact Factor
High Cortical Spreading Depression
Susceptibility and Migraine-
Associated Symptoms in
CaV2.1 S218L Mice
Arn M. J. M. van den Maagdenberg, PhD,1,2Tommaso Pizzorusso, PhD,3
Simon Kaja, PhD,2,4Nicole Terpolilli, MD,5Maryna Shapovalova, PhD,6,7
Freek E. Hoebeek, PhD,8Curtis F. Barrett, PhD,1,2Lisa Gherardini, PhD,3
Rob C. G. van de Ven, PhD,1Boyan Todorov, MS, PharmD,1
Ludo A. M. Broos, BSc,1Angelita Tottene, PhD,6,7Zhenyu Gao, MSc,8
Mariann Fodor, PhD,9‡ Chris I. De Zeeuw, MD, PhD,8,10
Rune R. Frants, PhD,1Nikolaus Plesnila, PhD,5Jaap J. Plomp, PhD,2,4
Daniela Pietrobon, PhD,6,7and Michel D. Ferrari, MD, PhD2
Objective: The CACNA1A gene encodes the pore-forming subunit of neuronal CaV2.1 Ca2?channels. In patients,
the S218L CACNA1A mutation causes a dramatic hemiplegic migraine syndrome that is associated with ataxia,
seizures, and severe, sometimes fatal, brain edema often triggered by only a mild head trauma.
Methods: We introduced the S218L mutation into the mouse Cacna1a gene and studied the mechanisms for the
S218L syndrome by analyzing the phenotypic, molecular, and electrophysiological consequences.
Cacna1aS218Lmice faithfully mimic the associated clinical features of the human S218L syndrome. S218L
neurons exhibit a gene dosage–dependent negative shift in voltage dependence of CaV2.1 channel activation,
resulting in enhanced neurotransmitter release at the neuromuscular junction. Cacna1aS218Lmice also display an
exquisite sensitivity to cortical spreading depression (CSD), with a vastly reduced triggering threshold, an increased
propagation velocity, and frequently multiple CSD events after a single stimulus. In contrast, mice bearing the
R192Q CACNA1A mutation, which in humans causes a milder form of hemiplegic migraine, typically exhibit only
a single CSD event after one triggering stimulus.
Interpretation: The particularly low CSD threshold and the strong tendency to respond with multiple CSD events
make the S218L cortex highly vulnerable to weak stimuli and may provide a mechanistic basis for the dramatic
phenotype seen in S218L mice and patients. Thus, the S218L mouse model may prove a valuable tool to further
elucidate mechanisms underlying migraine, seizures, ataxia, and trauma-triggered cerebral edema.
ANN NEUROL 2010;67:85–98
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.21815
Received Oct 22, 2008, and in revised form Jul 8, 2009. Accepted for publication Jul 28, 2009.
Address correspondence to Dr van den Maagdenberg, Department of Human Genetics, Leiden University Medical Centre, Einthovenweg 20, PO
Box 9600, Leiden, The Netherlands. E-mail: firstname.lastname@example.org For CNS electrophysiology, Dr Pietrobon, Department of Biomedical Sciences,
University of Padova, Viale G. Colombo 3, 35121 Padova, Italy. E-mail: email@example.com.
Current address for Dr Kaja: Department of Ophthalmology, University of Missouri–Kansas City, School of Medicine, Kansas City, MO.
From the Departments of1Human Genetics and2Neurology, Leiden University Medical Centre, Leiden, The Netherlands;3Department of Psychology,
University of Firenze and CNR Institute of Neuroscience, Pisa, Italy;4Department of Molecular Cell Biology–Section Neurophysiology, Leiden University
Medical Centre, Leiden, The Netherlands;5Department of Neurosurgery and Institute for Surgical Research, University of Munich Medical Centre–
Grosshadern, Munich, Germany;6Department of Biomedical Sciences, University of Padova;7CNR Institute of Neuroscience, Padova, Italy;8Department
of Neuroscience, Erasmus Medical Centre, Rotterdam;9Department of Embryology and Anatomy, Leiden University Medical Centre, Leiden; and
10Netherlands Institute for Neuroscience, Royal Dutch Academy for Sciences (KNAW), Amsterdam, The Netherlands.
‡Dr Mariann Fodor is deceased.
A.M.J.M.v.d.M., T.P., S.K., and N.T. contributed equally to this article.
Additional Supporting Information may be found in the online version of this article.
© 2010 American Neurological Association 85
associated with transient hemiparesis and other neurolog-
ical aura symptoms.1We consider FHM a valid mono-
genic model for studying pathogenetic mechanisms that
are involved in the genetically more complex common
forms of migraine. The arguments here for this are mainly
clinical and include: (1) the headache and aura features of
FHM and common migraine attacks are, apart from the
hemiparesis, identical; (2) two-thirds of FHM patients
have, in addition to attacks of FHM, attacks of common
nonhemiplegic migraine; (3) two thirds of their first-
degree relatives have attacks of common nonhemiplegic
migraine; (4) both FHM and common migraine attacks
can be precipitated by similar trigger factors and can be
prevented with the same prophylactic agents; and (5) their
headache phases can be aborted with the same acute treat-
FHM has been linked to three genes, all of which
encode proteins that drive or regulate neuronal excitability
(see Van den Maagdenberg and colleagues6and Pietro-
bon7for review). FHM1 is caused by mutations in the
CACNA1A gene6–10that encodes the ?1A-subunit of neu-
ronal CaV2.1 (P/Q-type) voltage gated calcium channels,
which couple presynaptic Ca2?entry to synaptic neuro-
transmitter release.11Functional studies in heterologous
expression systems support a net gain-of-function effect
of FHM1 mutations12–16(but see also Cao and col-
leagues,17,18and Barrett and colleagues19). In an analysis
of the single-channel properties of eight different FHM1
mutations, open probability (and therefore integrated
Ca2?current) was increased over a broad voltage range,
as a consequence of a shift to lower voltages for channel
activation.12–14Consistent with these findings, cerebellar
granule and cortical pyramidal neurons from knock-in
mice carrying the FHM1 R192Q mutation exhibited sig-
nificantly increased P/Q-type current density at negative
voltages.20,21These mice also showed enhanced excitatory
transmission at cortical synapses caused by increased ac-
tion potential–evoked Ca2?influx and increased proba-
bility of glutamate release,21and as a consequence, a de-
depression (CSD).20,21CSD is a slowly propagating wave
of cortical neuronal and glial depolarization that underlies
the migraine aura22and may possibly activate migraine
Although the CACNA1AR192Q, and most other
FHM1 mutations, cause only mild forms of FHM, which
are not associated with additional major neurological fea-
tures,7,8,24the CACNA1AS218LFHM1 mutation causes a
particularly dramatic clinical syndrome. This “S218L syn-
amilial hemiplegic migraine (FHM) is an autosomal
dominant subtype of migraine in which attacks are
drome” consists of, in addition to attacks of hemiplegic
migraine, slowly progressive cerebellar ataxia, epileptic sei-
zures, and severe, sometimes fatal, cerebral edema which
can be triggered by only a trivial head trauma.25–28
To understand how a single amino acid substitution
can account for such a wide range of neurological fea-
tures, we generated transgenic knock-in mice bearing the
S218L missense mutation in the Cacna1a gene, and ex-
amined the clinical and functional consequences of the
mutation. This provides a unique opportunity to study
shared pathogenetic mechanisms, not only for FHM and
common migraine, but also for cerebellar ataxia, epilepsy,
and mild head trauma–triggered cerebral edema, each of
which has also been linked with common types of mi-
We found that Cacna1aS218Lmice indeed display
the main associated clinical features of the human “S218L
syndrome.” In addition, these mice exhibit a number of
important neurobiological changes. These changes in-
clude: (1) highly increased neuronal Ca2?influx through
CaV2.1 channels; (2) strongly increased spontaneous re-
lease of neurotransmitter at the neuromuscular junction
(NMJ); and (3) a dramatically reduced triggering thresh-
old for CSD, with an increased propagation velocity and
an increased probability of multiple CSD events on a sin-
gle stimulation. We believe that the unique vulnerability
to CSD may explain many of the dramatic features of the
S218L syndrome. Moreover, the Cacna1aS218Lmouse
may prove a valuable tool to study shared pathogenetic
mechanisms of brain disorders that may model or are fre-
quently co-occurring with common types of migraine.
Materials and Methods
Generation of Transgenic Cacna1aS218LMice
The generation of Cacna1aR192Qmice was described previous-
ly.19The Cacna1a gene was modified to generate Cacna1aS218L
mice using a gene-targeting approach such that exon 5 con-
tained the human FHM1 S218L mutation. In the targeting vec-
tor, the original TCA triplet codon 218 was changed into TTA
by site-directed mutagenesis, creating the S218L mutation. Up-
stream of exon 5, a PGK-driven neomycin cassette flanked by
LoxP sites was introduced. E14 strain embryonic stem cells were
electroporated, and positive clones were screened for homologous
recombination by Southern blot analysis using the external probes
indicated in Figure 1A. The presence of the S218L mutation was
confirmed by polymerase chain reaction (PCR) amplification of
exon 5 using primers 271 (5?-CTCCATGGGAGGCACTTG-3?)
and 272 (5?-ACCTGTCCCCTCTTCAAAGC-3?), and subse-
quent digestion with restriction enzyme VspI, as well as direct
sequencing of the PCR products. Correctly targeted embryonic
stem cells were injected into C57Bl/6J blastocysts to create chi-
meric animals. Offspring that were heterozygous for the
S218L?Neo allele were crossed with EIIA-Cre deleter mice36to
ANNALS of Neurology
86Volume 67, No. 1
excise the neomycin cassette. Cacna1aS218Lmice in which the
neomycin cassette was successfully deleted were backcrossed to
C57Bl/6Jico mice. Cacna1aS218L/WT, Cacna1aS218L/S218L, and
wild-type (WT) littermates of the third generation were used for
all analyses (approximately 87.5% C57Bl/6Jico and approxi-
mately 12.5% 129Ola background). Genotyping of mice for all
experiments was performed by PCR analysis as described earlier.
All animal experiments were performed in accordance with the
guidelines of the respective institutions and national legislation.
The studies were performed using a similar number of male and
female mice, and summary data show results from both sexes.
For all experiments, the investigator was blinded to the geno-
For RNA extraction, mice were killed by cervical dislocation,
and forebrain and cerebellum were dissected in ice-cold
phosphate-buffered saline (pH 7.4) and subsequently snap fro-
zen in liquid nitrogen. For Northern blot and reverse transcrip-
tase PCR experiments, RNA was isolated as described previous-
ly.19For Northern blot analysis, 10?g cerebellar RNA was
separated on a 1% agarose gel and subsequently transferred to a
Hybond-N? membrane (Amersham Biosciences, Buckingham-
shire, United Kingdom).32P-labeled PCR products of Cacna1a
or Cyclophilin complementary DNA were used as probes using
standard hybridization and washing conditions.
The accelerating Rotarod (UGO Basile S.R.L., Commerio VA,
Italy) test was performed on a 4cm diameter horizontal rotating
rod. The test was performed in a semidark room with a light
source placed at the bottom to discourage the mice from jump-
ing off the Rotarod. Mice (8–10 weeks of age) were tested in
groups of five. After a training period (in which the mice were
FIGURE 1: Generation of Cacna1aS218Lknock-in mice. (A) Genomic structure of the wild-type (WT) Cacna1a allele, targeting
vector and predicted structure after homologous recombination (S218L?Neo allele), and after Cre-mediated deletion of the
neomycin cassette (S218L allele). LoxP sites are indicated by triangles. Black numbered boxes indicate respective exons,
with the S218L mutation in E5. Thick black lines indicate probes for Southern blot analysis. Primers used for genotyping and
confirmation of the S218L mutation are depicted schematically with arrows. Restriction sites: E, Eam1105I; E1, EcoRI; H,
HindIII, K, KpnI; X, XbaI. (B) Southern blot for all genotypes of S218L?Neo and S218L allele carriers showing genomic DNA
digested with KpnI or Eam1105I and tested with the 5? or 3? probe, respectively. (C) Northern blot of adult WT,
Cacna1aS218L/WT, and Cacna1aS218L/S218Lbrains. Total brain RNA was hybridized with Cacna1a and Cyclophilin complementary
DNA probes. Note that Cacna1a message levels are similar among genotypes.
van den Maagdenberg et al: S218L FHM1 Knock-in Mice
January, 2010 87
placed on the Rotarod turning at a low constant speed of 5 rpm
for 5 minutes), the mice were subjected to 6 consecutive days of
trials (1 trial per day). Each trial started with the Rotarod turn-
ing at a constant speed of 5 rpm for 10 seconds, after which the
speed was gradually increased to 45 rpm over the following 5
minutes. The latency to fall (ie, endurance) was recorded, and
the endurance per trial per genotype is presented as mean ?
standard error of the mean.
Electroencephalographic (EEG) electrodes were surgically im-
planted under isoflurane anesthesia in 10- to 14-week-old mice.
Six 1mm diameter stainless-steel screw electrodes were posi-
tioned as follows: one in each frontal bone and one in each
medial bone for recording activity patterns from primary motor
and parietal cortices, respectively. The remaining two electrodes
(placed in the temporal bone) served as reference (medial) and
ground (lateral) electrodes. Isolated copper wire was used to at-
tach the electrodes to a mini-connector that was cemented to
the skull with dental acrylic. Recording began 2 hours after sur-
gery. For recording, animals were placed in an electrically
shielded sound-proof box for continuous EEG recording and si-
multaneous video observation. Electrical signals were amplified,
filtered (CyberAmp, Molecular Devices, Sunnyvale, CA), digi-
tized (1401Plus; Cambridge Electronic Design, Cambridge,
United Kingdom), and stored for off-line analysis (Spike2; Cam-
bridge Electronic Design). Differential EEG signals were sam-
pled at 500Hz and filtered using a bandpass of 0.1 to 30Hz.
The behavior of the mice was continuously monitored with a
video camera and recorded at 25 frames per second using
custom-made software (Labview; National Instruments, Austin,
TX). Recording lasted up to 48 hours, during which food and
water were available ad libitum.
Survival curves were constructed using the method of Kaplan
and Meier.37The survival curves were compared by the log-rank
test. Survival times were calculated from the date of birth to the
date of death. Mice that died of unknown cause (ie, not related
to an experimental intervention) were considered cases. Statisti-
cal analysis was performed using SPSS 10.0 software (SPSS,
In Vivo Model of Cerebral Impact and
Quantification of Brain Edema
Anesthesia was initiated in an isoflurane chamber (4%) and was
maintained thereafter with 1.5% isoflurane in 30% oxygen/
68.5% nitrous oxide supplied by a face mask. Mild impact was
induced essentially as described previously,38but with the im-
portant difference that the intensity of the weight drop was con-
siderably diminished by reducing the height from where the
weight was dropped from 72 to 15cm. Whereas a weight drop
from 72cm causes severe neurological deficits and high mortality
(?30%) in WT mice, the modified low-height mild-impact
protocol did not cause mortality or any obvious neurological
dysfunction in WT mice. In brief, the animal’s head was fixed
in a stereotactic frame and the skull was exposed by a midline
incision. Body temperature was kept constant at 37°C using a
feedback-controlled heating pad connected to a rectal probe
(Heater Control Module; FHC, Bowdoinham, ME). Impact to
the brain was applied by dropping a 72gm weight guided
through a plastic tube onto the skull from a height of 15cm.
Immediately after the impact the skin was closed and animals
were transferred to an incubator heated to 35°C until recovery
of spontaneous motor activity (30 minutes). Brain water content
was determined as described previously.39,40In brief, 24 hours
after impact, mice were killed and the brains were removed. Af-
ter removal of the cerebellum and olfactory bulb, the brains
were weighed to obtain their wet weight (ww), dried at 110°C
for 24 hours, and then reweighed to obtain their dry weight
(dw). Brain water content (expressed as percentage) was calcu-
lated using this formula: ((ww ? dw)/ww) ? 100.
CaV2.1 Current in Cerebellar Granule Cells
Cerebellar granule cells were grown in primary culture from
6-day-old mice as described previously.41Experiments were per-
formed on cells grown from 6 to 7 days in vitro. Whole-cell
patch-clamp recordings were performed at room temperature as
described previously.20,41The external recording solution con-
tained (in mM): BaCl25, tetraethylammonium-Cl 148, Hepes
10, 0.1mg/ml cytochrome c, pH 7.4 with tetraethylammonium-
methanesulfonate 100, MgCl25, Hepes 30, EGTA 10, adeno-
sine triphosphate 4, guanosine triphosphate 0.5 and cyclic
adenosine monophosphate 1, pH 7.4 with CsOH. Currents
were low-pass filtered at 1kHz and digitized at 5kHz. Compen-
sation (typically 70%) for series resistance was generally used.
Current-voltage (I-V) relationships were obtained only from cells
with a voltage error of less than 5mV and without signs of in-
adequate space clamping such as notchlike current discontinui-
ties, slow components in the decay of capacitive currents (in
response to hyperpolarizing pulses), or slow tails not fully inhib-
ited by nimodipine. The average normalized I-V curves were
multiplied by the average maximal current density obtained
from all cells. I-V curves were fitted with the equation IBa? G ?
(V ? Erev)/(1?e^((V0.5? V)/k)) using a nonlinear regression
method based on the Levenberg–Marquardt algorithm. The liq-
uid junction potentials were such that a value of 12mV should
be subtracted from all voltages to obtain the correct values of
membrane potential in whole-cell recordings.41All drugs were
stored as stock solutions at ?20°C: 250?M ?-conotoxin-GVIA
(?-CgTx-GVIA; Bachem, Budendorf, Switzerland) and 250?M
?-conotoxin-MVIIC (?-CTx-MVIIC; Bachem) in distilled wa-
ter, 10mM nimodipine (a gift from Dr Hof, Sandoz, Basel,
Switzerland) in 95% ethanol.
contained (inmM): Cs-
Ex Vivo Neuromuscular Junction
The NMJ is a model synapse in the peripheral nervous system
exclusively relying on presynaptic Cav2.1 channels42and can be
electrophysiologically analyzed with relative ease to investigate
neurotransmission. Mice (approximately 2 months of age) were
ANNALS of Neurology
88Volume 67, No. 1
killed by carbon dioxide inhalation. Phrenic nerve hemidia-
phragms were dissected and mounted at room temperature in
standard Ringer’s medium containing (in mM) NaCl 116, KCl
4.5, CaCl22, MgSO41, NaH2PO41, NaHCO323, glucose
11, pH 7.4. The solution was continuously bubbled with 95%
Intracellular recordings of miniature end-plate potentials
(MEPPs; the small, spontaneous depolarizing events caused by
uniquantal acetylcholine release) and end-plate potentials (EPPs;
the depolarization resulting from nerve action potential–evoked
acetylcholine release) were performed at NMJs at 28°C using
standard microelectrode equipment as described previously.43,44
At least 30 MEPPs and EPPs were recorded at each NMJ, and
typically 7 to 15 NMJs were sampled per experimental condi-
tion per muscle. Muscle action potentials were prevented with
the muscle-specific sodium channel blocker ?-conotoxin-GIIIB
(3?M; Scientific Marketing Associates, Barnet, Herts, United
Kingdom). EPPs were recorded in Ringer’s medium as described
earlier except containing 0.2mM Ca2?. The nerve was electri-
cally stimulated with 100?s supramaximal pulses delivered at
0.3Hz. Amplitudes of EPPs and MEPPs were normalized to
?75mV, assuming 0mVas
acetylcholine-induced current.45The normalized EPP ampli-
tudes were corrected for nonlinear summation46with an ƒ value
of 0.8. The quantal content at each NMJ (ie, the number of
acetylcholine quanta released per nerve impulse) was calculated
by dividing the normalized, corrected mean EPP amplitude by
the normalized mean MEPP amplitude.
Cortical Spreading Depression
For CSD experiments, Cacna1aS218L, and for comparison
Cacna1aR192Q, transgenic and WT mice (20–30g) were anes-
thetized with urethane (20% in saline; 6ml/kg intraperitoneally).
The mice were mounted in a stereotaxic apparatus and contin-
uously monitored for adequate level of anesthesia, temperature,
heart rate, and lack of nociceptive reflexes. Three holes were
drilled in the skull over the left hemisphere to record CSD. The
first corresponded to the occipital cortex and was used for access
of the electrical stimulation electrode. The second hole, at the
parietal cortex (1mm lateral and 1mm caudal to Bregma), and
the third hole, at the frontal cortex (1mm lateral and 1mm ros-
tral to Bregma), were used for placement of the recording elec-
trodes. The dura was carefully removed from a small part of the
cortex using a 28-gauge needle from a small part of the cortex,
sufficient to let the recording electrode (tip diameter, 3?m) en-
ter the cortex. The steady-state (direct current) potential was re-
corded with glass micropipettes placed 200?m below the dural
surface. An Ag/AgCl reference electrode was placed subcutane-
ously above the nose. Cortical stimulation was conducted using
a copper bipolar electrode placed on the cortex after removing
the dura. Pulses of increasing intensity (from 10 up to 800?A)
were applied for 100 milliseconds at 3-minute intervals with a
stimulus isolator/constant current unit (WPI, Sarasota, FL) until
a CSD event was observed. The minimal stimulus intensity at
which a CSD event was elicited was taken as the CSD thresh-
old. Once a CSD event was elicited, the recording continued for
1 hour without stimulation, and additional CSD events were
noted. The percentage of mice with multiple CSD events was
determined only from those mice that could be recorded for 1
full hour after the first detected event. In some WT mice, the
CSD induction threshold was reassessed at the end of the
1-hour monitoring of recurrent CSDs and found to be un-
changed from the initial threshold, suggesting that the general
recording conditions remain stable for at least 1 hour. To mea-
sure CSD propagation velocity, we divided the distance between
the two recording electrodes by the time elapsed between the
CSD onset at the first and second recording sites. The condition
of the cortex was monitored using a surgical microscope, and
animals with lesions were excluded.
Statistical differences were analyzed with paired or unpaired Stu-
dent’s t test, analysis of variance with Tukey’s HSD post hoc
test, Fisher’s exact test, repeated-measures analysis of variance, or
Kruskal–Wallis one-way analysis of variance on ranks followed
by Dunn’s comparison procedure as post hoc test, all where ap-
propriate. Calculations were performed with a statistical software
package (SigmaStat 3.0, Erkrath, Germany). For all experiments,
summary data are presented as mean ? standard error of the
mean; p ? 0.05 was considered to be statistically significant.
Generation of the S218L Knock-in Mice
Using a gene-targeting approach with homologous recom-
bination, we introduced the human CACNA1AS218Lmuta-
tion into the orthologous Cacna1a gene (see Figs 1A, B) to
generate Cacna1aS218Lmice. Both RNA (see Fig 1C) and
protein analyses (see Supplementary Fig 1) indicated that
neither the mutation nor the introduction of the LoxP site
appeared to have interfered with any major regulatory se-
quence because no major effect on Cacna1a expression with
respect to quantity or location was observed. However, the
experiments were not designed to demonstrate minor dif-
ferences in Cacna1a expression.
S218L Mice Exhibit Main Features of the
Human S218L Syndrome
Cacna1aS218L/S218Lmice showed a complex neurological
phenotype that is remarkably similar to that of what is
seen in CACNA1AS218Lpatients (Fig 2A). Most of the
time, the mice appeared phenotypically normal, except for
mild cerebellar ataxia. The ataxia was reflected by poor
performance in the Rotarod test (see Fig 2B), accompa-
nied by reduced arborization solely of proximal, primary
dendrites of cerebellar Purkinje neurons (see Fig 2C and
Supplementary Fig 2A). These primary dendrites in
Cacna1aS18L/S218Lhad a decreased length (see Supplemen-
tary Fig 2B). Purkinje cells of 2-month-old S218L mice
showed abnormal distal turns and a “weeping willow” ap-
van den Maagdenberg et al: S218L FHM1 Knock-in Mice
January, 2010 89
ANNALS of Neurology
90 Volume 67, No. 1
pearance (see Fig 2C) similar to what has been reported
for aged (?1 year) Rocker mice, which carry the naturally
occurring Cacna1aT1310Kmissense mutation.47
We observed two types of spontaneous attacks in
mice: (1) attacks of hemiparesis,
which are consistent with the attacks of hemiparesis as
seen in S218L FHM1 patients; and (2) attacks of gener-
alized seizures that were fatal in some cases. The hemipa-
retic attacks consisted of brief periods of apparent tran-
sient unilateral weakness, manifesting as slow, circular
locomotion (because of leg movement only on one side of
the body) and attempts to support themselves against the
cage wall on the side of the hemiparesis. After such epi-
sodes, the mouse would remain immobile for about 20
minutes and then recovered. Some of these attacks were
captured on video (see Supplementary movie for an ex-
ample). In addition to the hemiparetic attacks, we also
observed attacks that started with myoclonic jerks and
subsequently evolved into generalized tonic-clonic (“grand
mal”) seizures. We were able to capture and analyze sev-
eral of such episodes in three mice with EEG recording
coupled to video monitoring (see Fig 2D; see also Sup-
plementary note 1). All three mice died at the end of such
In line with the observation that mice can die after
spontaneous epileptic attacks, the life expectancy of
Cacna1aS218L/S218Lmice, kept in their home cage envi-
ronment (without experimental interventions), was signif-
icantly decreased (see Fig 2E). Postmortem analysis indi-
cated that the likely cause of death was lung edema, a
frequent finding in epilepsy-related deaths, often second-
ary to cardiac arrest (data not shown). Pathological exam-
ination excluded myocardial infarction, or other morpho-
logical abnormalities, in these cases. Additional histology
of homozygous S218L brain material at the light micro-
scope level did not demonstrate obvious structural abnor-
malities such as neurodegeneration (data not shown). The
earlier-described spontaneous episodes of hemiparesis, fa-
tal seizures, and reduced life expectancy all appear to be
unique to Cacna1aS218L/S218Lmice; none was observed in
Cacna1aS218L/WT(see Fig 2) or in mice carrying the
R192Q mutation (Cacna1aR192Q/R192Q), which in pa-
tients is associated with a much milder form of hemiple-
Consistent with observations in S218L patients,
Cacna1aS218L/S218Lmice, but not WT or Cacna1aS218L/WT
mice, exhibited significant brain edema 24 hours after mild
head impact as applied in our modified low-height weight-
drop model (see Fig 2F). The brain water content increased
by only 1.51% in WT mice (from 76.99 ? 0.67 to
78.50 ? 1.14%) and by only 1.49% in Cacna1aS218L/WT
mice (from 77.12 ? 0.94 to 78.61 ? 1.07%). In contrast,
in Cacna1aS218L/S218Lmice, the brain water content in-
creased significantly by 2.84% (from 76.89 ? 0.64 to
79.73 ? 0.93 %; p ? 0.02 vs Cacna1aS218L/WTand WT
mice). Baseline brain water content was similar in all three
genotypes. Twenty percent of the Cacna1aS218L/S218Lmice,
but none of the WT or Cacna1aS218L/WTmice, died after
these mild impact experiments.
Taken together, these results demonstrate that
Cacna1aS218L/S218Lmice faithfully mimic the broad spec-
trum of spontaneous episodic, mild impact-triggered, and
permanent clinical features seen in S218L patients.
Gain-of-Function Effect on Whole-Cell Ca2?
Influx and Synaptic Transmission
To understand the molecular consequences of the S218L
mutation on CaV2.1 channel functioning, we performed
whole-cell patch-clamp recordings of cerebellar granule
neurons (Fig 3A). Neurons expressing S218L channels
displayed significantly increased whole-cell CaV2.1 cur-
rent density at negative voltages as reflected by a pro-
nounced gene dosage–dependent leftward shift in voltage-
dependent activation (see Fig 3B). In contrast, current
density was not changed at positive voltages, indicating
that the number of functional Cav2.1 channels at the so-
matodendritic plasma membrane is similar across geno-
types (see also Supplementary note 2). Moreover, the en-
hanced Ca2?influx was not accompanied by significant
changes in the density of other high-voltage–activated
CaVchannels, except for a slight decrease in L-type cur-
rents (see Fig 3C).
FIGURE 2: Phenotypic consequences in Cacna1aS218Lmice.
(A) Venn diagram depicting the overlapping and distinct
clinical characteristics of pure familial hemiplegic mi-
graine (FHM), as well as the “S218L syndrome” with se-
vere associated features exhibited in S218L patients. (B)
Rotarod testing for ataxia in 2-month-old mice shows se-
vere impairments in both performance and learning for
Cacna1aS218L/S218Lmice (n ? 10–11). Trials were performed
on consecutive days. (C) Golgi-Cox staining of cerebellar
Purkinje neurons from 2-month-old wild-type (WT) and
Cacna1aS218L/S218Lmice. Arrows indicate decreased branch-
ing and downturned ends in the Cacna1aS218L/S218Lneuron.
Cacna1aS218L/S218Lmouse showing sequential phases (1–6)
of a “grand mal” fatal seizure. Concurrent video monitor-
ing was used to assess the overt clinical phenotype, de-
scribed as follows: IP, interictal period; MJ, myoclonic
jerks; C, clonic seizure, TC, tonic-clonic seizure; D, death.
(E) Kaplan–Meier plot showing significantly decreased sur-
vival of Cacna1aS218L/S218Lmice (p < 0.001 vs WT). (F)
Twenty-four hours after mild head impact (see Materials
and Methods), Cacna1aS218L/S218Lmice had increased cere-
bral water content (n ? 9, 15, and 5 mice for WT,
Cacna1aS218L/WT, and Cacna1aS218L/S218L, respectively; *p <
0.05 vs WT; #p < 0.05 vs S218L/WT).
van den Maagdenberg et al: S218L FHM1 Knock-in Mice
January, 2010 91
The effects in Cacna1aS218L/S218Lneurons were rem-
iniscent of the changes seen in neurons of the clinically
milder R192Q mice,20but were more pronounced. As a
consequence of the even lower activation threshold of
S218L channels,14the gain-of-function of the CaV2.1
currentat low voltages
Cacna1aS218L/S218Lneurons than the gain-of-function we
previously observed in Cacna1aR192Q/R192Qneurons, lead-
ing to Ca2?influx close to the resting potential. While
the CaV2.1 current at ?40mV was approximately 4 times
larger in Cacna1aR192Q/R192Qneurons compared with
WT neurons,20the CaV2.1 current in Cacna1aS218L/S218L
neurons was 6.6 times larger than that in WT neurons
(see Fig 3B).
Consistent with Ca2?influx occurring close to the
resting potential, the S218L mutation produced a much
was more dramatic in
larger increase in the frequency of spontaneous transmit-
ter release events (ie, the MEPPs) at the NMJ when
compared with the R192Q mutation. While the MEPP
frequency in Cacna1aR192Q/R192Qmice was only 2.5
times greater than in WT mice,20the MEPP frequency
Cacna1aS218L/S218Lmice 12.5 times greater in compari-
son with WT mice (Fig 4A).
Remarkably, although a similar increase in CaV2.1
current at ?40mV, relative to their respective WT mice,
was8.5 times andin
FIGURE 3: CaV2.1 (P/Q-type) current density (IBa) is increased
in cerebellar granule neurons from Cacna1aS218Lmice. (A)
Peak amplitude of whole-cell Ba2?current recorded from a
Cacna1aS218L/WTcerebellar granule cell in primary culture dur-
ing successive step depolarizations to ?10mV that were ap-
plied at 10-second intervals from a holding potential of
?80mV (5mM Ba2?as the charge carrier). Nimodipine (5?M),
?-conotoxin GVIA (1?M), and ?-conotoxin MVIIC (3?M) were
applied to the bath as indicated. (left inset) Representative
current traces taken at times a, b, c, d and difference trace
representing the CaV2.1 (P/Q-type) component (c-d). (right
inset) Representative traces at increasing voltage from ?50
to ?10mV, measured at I-V c and I-V d. Scale bars ? 20
milliseconds, 200pA. (B) CaV2.1 current density as a function
of voltage in wild-type (WT; white circles), Cacna1aS218L/WT
(gray circles), and Cacna1aS218L/S218Lmice (black circles). The
CaV2.1 current was isolated pharmacologically as shown in
(A). Peak whole-cell P/Q-type Ba2?currents were divided by
the membrane capacitance, normalized to their maximal
value, and averaged; average normalized I-V curves (n ? 14,
12, and 10 cells for WT, Cacna1aS218L/WT, and Cacna1aS218L/
S218L, respectively) were multiplied by the average maximal
current densities (n ? 35, 44, and 28 cells, respectively). In-
dividual I-V curves were fit with the equation IBa? G ? (V ?
Erev)/(1?e^((V0.5? V)/k)). Average (? standard error of the
mean) V0.5 values were ?15 ? 1 (WT), ?20.1 ? 0.8
(Cacna1aS218L/WT; p < 0.005 vs WT), and ?25 ? 1mV
(Cacna1aS218L/S218L; p < 0.005 vs Cacna1aS218L/WT, p <
0.00005 vs WT). (inset) P/Q-type Ba2?currents elicited at the
voltages indicated (pooled from different cells). (C) Current
densities of L-, N-, and R-type Ca2?channels in WT and
Cacna1aS218Lmice, measured at ?10mV. The different cur-
rent types were isolated pharmacologically as shown in (A).
The current densities for the L-type component were 10.5 ?
0.7 (n ? 24), 12.7 ? 0.7 (n ? 27), and 13.8 ? 1.1 (n ? 29)
pA/pF in Cacna1aS218L/S218L
(gray bars), and WT neurons (white bars), respectively; cur-
rent densities for the N-type were 8.0 ? 0.7 (n ? 27), 8.8 ?
0.8 (n ? 29), and 6.5 ? 0.8 (n ? 29) pA/pF; current densities
for the R-type were 23.0 ? 1.6 (n ? 28), 20.4 ? 1.0 (n ?
43), and 22.9 ? 1.1 (n ? 35) pA/pF. The sum of the three
components (total non-CaV2.1 current density) was similar for
all three genotypes (41.5 ? 1.9, 41.9 ? 1.5, and 43.2 ?
(black bars), Cacna1aS218L/WT
ANNALS of Neurology
92Volume 67, No. 1
Cacna1aS218L/WT(2.7 times larger; see Fig 3B) and
Cacna1aR192Q/R192Qmice (3.8 times larger20), the relative
increase in MEPP frequency was much greater in
Cacna1aS218L/WTmice, consistent with more Ca2?influx
at rest in Cacna1aS218L/WTnerve terminals. However,
quantal content (ie, the number of neurotransmitter
quanta released by an action potential) measured in low
extracellular Ca2?was 2.5 times increased in S218L syn-
apses (compared with WT; see Fig 4B) to a similar extent
as in R192Q synapses (3.4 times compared with WT20).
This observation suggests a ceiling effect for release prob-
ability at the NMJ.
Extreme Susceptibility of S218L Mice for
Cortical Spreading Depression
Because CSD is the likely underlying mechanism for mi-
graine aura,22and has also been implicated in epilepsy
and edema,48increased susceptibility for CSD in S218L
compared with R192Q mice could explain many features
of the dramatic S218L syndrome. To test this hypothesis,
we assessed the triggering threshold for CSD, the CSD
propagation velocity, and the probability of multiple CSD
events on a single stimulus in the two knock-in mouse
Indeed, S218L mice showed both a lower threshold
for CSD induction (Fig 5A; for examples of CSD traces,
see Supplementary Fig 3) and a faster propagation of
CSD waves (see Fig 5B) compared with R192Q mice. In
addition, S218L mice had a gene dosage–dependent in-
creased probability of experiencing successive CSD events
on a single stimulus: the probability was the greatest in
homozygous Cacna1aS218L/S218Lmice, which carry two
copies of the S218L allele, lowest in WT mice, and in
between in heterozygous Cacna1aS218L/WTmice, which
carry only one copy of the S218L allele (see Fig 5C).
There was, however, no difference in average latency until
(1,323 ? 189 seconds) and Cacna1aS218L/S218L(1,108 ?
211 seconds) (p ? 0.35).
It is noteworthy to emphasize that, although CSD
induction threshold and propagation velocity are similar
in Cacna1aS218L/WTand Cacna1aR192Q/R192Qmice (see
Figs 5A, B), multiple CSD events were seen only rarely in
Cacna1aR192Q/R192Qmice (see Fig 5C).
The selectively high incidence of multiple CSDs in
S218L mice makes it unlikely that recurring CSDs would
be merely caused by tissue damage during CSD record-
ings. If tissue damage had been the underlying cause, one
would have expected a more evenly distributed increased
probability of repetitive CSD events over the various ge-
notypes. In contrast, R192Q mice hardly exhibited recur-
ring CSD events. Lastly, great care was taken in monitor-
ing the condition of the cortex. Mice with visible cortical
lesions, as identified using a surgical microscope, were ex-
cluded from the analyses.
Here, we present the generation and behavioral, electro-
physiological, and neurobiological characterization of
S218L knock-in mice bearing the human pathogenic
CACNA1AS218Lmissense mutation in CaV2.1 calcium
channels. In humans, this mutation causes a broad and
FIGURE 4: Synaptic transmission at the neuromuscular junc-
tion (NMJ). Synaptic transmission at the NMJ is increased
in Cacna1aS218Lmice. (A) Summary of miniature end-plate
potential (MEPP) frequency (n ? 6; ‡p < 0.001 vs wild-type
[WT]; #p < 0.01 vs Cacna1aS218L/WT). Exemplar recordings
are shown at the right. Vertical scale bar ? 0.75mV. (B)
Summary of neurotransmitter release in 0.2mM extracellu-
phrenic nerve (n ? 3 mice; †p < 0.01 vs WT). At the right
are exemplar end-plate potentials. Black triangles mark the
van den Maagdenberg et al: S218L FHM1 Knock-in Mice
FIGURE 5: Increased susceptibility to cortical spreading depression (CSD) in Cacna1aS218Land Cacna1aR192Qmice. (A)
Summary of the threshold required to elicit at least one CSD event in wild-type (WT), Cacna1aS218L, and Cacna1aR192Qmice.
Cacna1aS218Lstrain: n ? 30, 39, and 22 mice for WT, Cacna1aS218L/WT, and Cacna1aS218L/S218L, respectively. Cacna1aR192Q
strain: n ? 28 and 31 mice for WT and Cacna1aR192Q, respectively. *p < 0.05 versus WT; †p < 0.05 versus S218L/WT; ‡p <
0.05 versus R192Q/R192Q. (B) Summary of CSD velocity measured using two recording electrodes (see Materials and
Methods). Sample sizes are as in (A). *p < 0.05 versus WT; †p < 0.05 versus S218L/WT; ‡p < 0.05 versus R192Q/R192Q.
(C) Proportion of mice (expressed as a percentage of all mice) with no response, a single CSD, or multiple CSD events
as a function of stimulation intensity. Cacna1aS218Lstrain: n ? 27, 34, and 18 mice for WT, Cacna1aS218L/WT, and
Cacna1aS218L/S218L, respectively. Cacna1aR192Qstrain: n ? 13 and 15 mice for WT and Cacna1aR192Q, respectively.
ANNALS of Neurology
94 Volume 67, No. 1
devastating, sometimes even fatal, clinical syndrome, with
both episodic and progressive features of brain dysfunc-
tion. In this respect, the S218L syndrome is at the very
extreme of the clinical spectrum of migraine. Our findings
can be summarized as follows: First, S218L mice display a
complex phenotype consisting of mild permanent cerebel-
lar ataxia, spontaneous attacks of hemiparesis and/or
(sometimes fatal) seizures, and brain edema after only a
mild head impact. These features faithfully mimic the
clinical spectrum of patients carrying the CACNA1AS218L
Second, most of the functional consequences of the
S218L mutation are qualitatively similar to, but quantita-
tively more pronounced, than those of the R192Q muta-
tion; this correlates well with the milder phenotype that is
seen in patients with the R192Q mutation8and strongly
supports the view that these changes are causally relevant.
Specifically, when compared with Cacna1aR192Q/R192Q
neurons20,21,44(and this study), Cacna1aS218L/S218Lneu-
rons have: (1) a greater increase in CaV2.1-mediated Ca2?
current density on weak depolarization; (2) a larger in-
crease in spontaneous neurotransmitter release at the
NMJ, a model synapse in the peripheral nervous system
that relies exclusively on CaV2.1 channels and can be elec-
trophysiologically analyzed with relative ease; and (3) a
greater susceptibility for CSD, as reflected by a lower trig-
gering threshold, higher propagation velocity, and a
higher probability of successive CSD events on only a sin-
gle stimulus (see also later). Moreover, heterozygous
Cacna1aS218L/WTand homozygous Cacna1aR192Q/R192Q
mice showed quantitatively similar gain-of-function ef-
fects in Ca2?current density, and CSD threshold and
velocity. However, spontaneous neurotransmitter release
at the NMJ was much larger in Cacna1aS218L/WTmice,
suggesting a larger Ca2?influx at rest in nerve terminals.
In contrast with the good correlation between severity
of the phenotype and facilitation of both neuronal Ca2?
current and CSD, the facilitation of evoked release at
the NMJ at low extracellular [Ca2?] was similar in
Cacna1aS218L/S218Land Cacna1aR192Q/R192Qmice. In fact,
at physiological extracellular [Ca2?], evoked release at the
NMJ of both mutant mice20(also data not shown) was not
different from that at WT NMJs. Tottene and colleagues21
reported an enhanced evoked glutamate release at excitatory
cortical pyramidal cell synapses of Cacna1aR192Q/R192Q
mice at physiological [Ca2?], but an unaltered evoked
GABA release from inhibitory fast-spiking interneuron syn-
apses (despite being also controlled by CaV2.1 channels).
Thus, the NMJ may perhaps be a better model for cortical
inhibitory fast-spiking interneuron synapses than cortical
excitatory synapses, whose gain-of-function was shown to
be causally linked to CSD facilitation.21
Third, the S218L FHM1 mutation, but not the
R192Q FHM1 mutation, increased the probability of mul-
tiple CSD events in response to only a single threshold
stimulus. Whereas WT and R192Q mice generally experi-
enced a single CSD event on a single stimulus, the S218L
mice showed a gene dosage–dependent increased probabil-
ity of having successive CSD events. This tendency of
S218L mice for successive CSD events is particularly re-
markable given that, despite similar CSD induction thresh-
old and propagation velocity for Cacna1aR192Q/R192Qand
Cacna1aS218L/WTmice (see Figs 5A, B), multiple CSD
events are seen only rarely in Cacna1aR192Q/R192Qmice (see
Fig 5C). Thus, the cortical susceptibility to repetitive CSD
events is apparently unique to the S218L mutation.
The exact mechanism for recurrent CSD events af-
ter a single stimulus and the exact downstream conse-
quences of multiple CSD events remain to be determined.
The combination of the more negative activation thresh-
old of S218L CaV2.1 channels, coupled with both their
impaired inactivation during long depolarization and faster
recovery from inactivation,14suggests the involvement of a
process that is particularly sensitive to Ca2?influx at rest,
channel inactivation, or both. Perhaps prolonged Ca2?in-
flux during the first CSD event may be involved as well.
Our findings suggest that the high sensitivity of the
S218L brain to even mild stimuli (eg, low-impact head
trauma) may be explained, at least in part, by the unique
combination of a particularly low CSD trigger threshold
and a high propensity for multiple CSD events. Increased
susceptibility to the effects of successive CSD events may
also play a role. By analogy, we propose that similar mech-
anisms underlie the severe and clinically broad phenotype
that is seen in S218L patients, but not in patients carrying
other, milder FHM1 mutations. Thus, our data provide a
mechanistic basis for the overlapping clinical manifestations
of both R192Q and S218L patients, as well as for the ad-
ditional severe clinical features seen only in S218L patients.
There is ample evidence that migraine, mild head
trauma–triggered migraine attacks (such as in footballer’s
migraine), brain edema, epilepsy, and (subclinical) cere-
bellar ataxia can co-occur in various combinations.10,29–
33,48,49 This suggests shared molecular disease pathways.
For instance, both migraine attacks and seizures are asso-
ciated with excessive neuronal excitability in the cortex.31
Whereas CSD is well recognized as the pathophysiological
correlate of the migraine aura,50,51its role in epilepsy is
less clear. Clearly, there is a distinction between spreading
depression and hypersynchronous activity in seizures, but
it appears that spreading depression may gradually evolve
van den Maagdenberg et al: S218L FHM1 Knock-in Mice
January, 2010 95
into “spreading convulsions” when waves occur at high
frequency.52Eikermann-Haerter and coworkers53recently
showed that generalized seizures may occur 45 to 75 min-
utes after a single KCl-induced CSD in homozygous
Cacna1aS218Lmice, but not in Cacna1aR192Qmice, indi-
cating that CSD and epileptic events are separated in
time. Follow-up studies, comparing the exact conditions
under which CSD and epilepsy may occur in Cav2.1 mu-
tant mouse types, are likely to shed further light on the
similarities and dissimilarities of these two apparently op-
posing neuronal mechanisms underlying episodic brain
CSD has also been implicated in edema.48In ro-
dents, waves of spreading depression caused transient severe
neuronal swelling, changes in dendritic structures,54and
prolonged severe vascular leakage with impairment of the
blood–brain barrier.55Opening of the blood–brain barrier,
which possibly is secondary to multiple CSD events, pre-
ceded cortical edema in a patient with a severe attack of
FHM type 2 caused by an ATP1A2 mutation.56We, there-
fore, like to speculate that the unique vulnerability of the
S218L brain to multiple CSD events may underlie the dra-
matic and sometimes fatal clinical features in both S218L
patients and mice. Future experiments are needed to ad-
dress the issue of whether the edema is a direct conse-
quence of trauma, or is due to either trauma-induced CSD
As yet, it is unclear how the S218L mutation may
cause cerebellar ataxia. One can speculate that the ataxia
may be caused by excessive Ca2?influx in cerebellar Pur-
kinje cells, to a differential effect of the mutation on the
afferent inputs of Purkinje cells, or both. These mecha-
nisms are likely to disturb the delicately regulated firing
patterns of the Purkinje cells, as suggested earlier by stud-
ies in naturally occurring Cacna1a mouse mutants.57,58
Despite the advantages that our R192Q and S218L
mice have to study disease mechanisms, it is also clear
that the models needs further validation. For instance, it
needs to be established whether R192Q and S218L mice
can experience migraine attacks and whether transgenic
mice are, indeed, useful models to study migraine. Al-
though the S218L mice appear helpful in investigating
how a single CaV2.1 mutation may lead to a combination
of clinical phenotypes (ie, hemiplegic migraine, seizures,
cerebellar ataxia, and mild head trauma induced edema)
seen in FHM1 S218L patients, it is also clear that the
co-occurrence of phenotypes in these mice makes it much
harder to study the effect of a CaV2.1 mutation on specific
phenotypes. In this respect, it needs to be shown that
these mice will further advance our knowledge on under-
standing disease mechanisms of, for instance, hemiplegic
migraine or cerebellar ataxia. A direct extrapolation of re-
sults from mice to patients should be done with great cau-
tion, as there are obvious differences in the physiology
between humans and experimental animal models. Future
studies need to show the benefits and limitations of these
transgenic mouse models.
In summary, we report here that S218L FHM1 mice
faithfully appear to mimic the salient clinical features of the
broad human S218L hemiplegic migraine syndrome, and
we propose a mechanistic explanation for these severe clin-
ical features. Despite the limitations mentioned earlier, we
believe that the S218L and R192Q knock-in FHM1
mouse models will likely serve as valuable tools to unravel-
ing, at least to some extent, mechanisms for FHM, com-
mon migraine, epilepsy, and cerebral edema.
This work was supported by the Netherlands Organiza-
tion for Scientific Research (NWO; 903-52-291, M.D.F.,
R.R.F.; Vici 918.56.602, M.D.F.); the EU “EURO-
HEAD” (LSHM-CT-2004-504837, D.P., M.D.F) and
the Center of Medical System Biology established by the
Netherlands Genomics Initiative/Netherlands Organisa-
tion for Scientific Research and Community (M.D.F.,
R.R.F., A.M.J.M.v.d.M.); Telethon Italy (GGP05236,
T.P.), Italian Ministry of Education University Research
(Grants PRIN2005 and FIRB2002; D.P.); the Prinses
Beatrix Fonds (MAR01-0105, J.J.P.); the Hersenstichting
Nederland (9F01(2).24, J.J.P.); KNAW van Leersum-
fonds (AFD/ML/2901, J.J.P.); the Dutch ZON-MW Or-
ganization for Medical Sciences (912-07-022, C.I.d.Z.);
ALW Life Sciences, Senter (Neuro-Bsik, BSIK03053,
C.I.d.Z.); Prinses Beatrix Fonds (OP05-16/100506,
C.I.d.Z.); the SENSOPAC program of the European
Community (028056, C.I.d.Z.); and European Molecular
Biology Organization postdoctoral fellowship (ALTF810-
2005, S.K.) and a Michael Smith Foundation for Health
Research trainee award (ST-PDF-00140[05-1]BM, S.K.).
We thank M. L. Maat-Schieman, S. van Duijnen, and S.
Verbeek for help in histological experiments and embry-
onic cell stem injections, and E. Hess for expert opinions
on the evaluation of the behavioral phenotype. In addi-
tion, we thank E. Putignano, U. Nehrdich, and R. van
der Giessen for technical assistance.
1.Headache Classification Subcommittee of the International Head-
ache Society.The international
disorders: 2nd edition. Cephalalgia 2004;24:1–160.
2.Terwindt GM, Ophoff RA, Haan J, et al. Variable clinical expres-
sion of mutations in the P/Q-type calcium channel gene in fa-
milial hemiplegic migraine. Neurology 1998;50:1105–1110.
ANNALS of Neurology
96Volume 67, No. 1
3.Ducros A, Denier C, Joutel A, et al. The clinical spectrum of
familial hemiplegic migraine associated with mutations in a neu-
ronal calcium channel. N Engl J Med 2001;345:17–24.
4.Ferrari MD, Goadsby PJ. Migraine as a cerebral ionopathy with
abnormal central sensory processing. In: Gilman S, Pedley T,
eds. Neurobiology of disease. New York: Elsevier, 2006:333–348.
5. Ferrari MD. Migraine. Lancet 1998;351:1043–1051.
6.Van den Maagdenberg AM, Haan J, Terwindt GM, Ferrari MD.
Migraine: gene mutations and functional consequences. Curr
Opin Neurol 2007;20:299–305.
7.Pietrobon D. Familial hemiplegic migraine. Neurotherapeutics
8.Ophoff RA, Terwindt GM, Vergouwe MN, et al. Familial hemiplegic
migraine and episodic ataxia type-2 are caused by mutations in the
Ca2?channel gene CACNL1A4. Cell 1996;87:543–552.
9.Adams PJ, Snutch TP. Calcium channelopathies: voltage-gated
calcium channels. Subcell Biochem 2007;45:215–251.
10. Stam AH, van den Maagdenberg AM, Haan J, et al. Genetics of
migraine: an update with special attention to genetic comorbid-
ity. Curr Opin Neurol 2008;21:288–293.
11.Catterall WA. Structure and function of neuronal Ca2? channels
and their role in neurotransmitter release. Cell Calcium 1998;24:
12.Hans M, Luvisetto S, Williams ME, et al. Functional conse-
quences of mutations in the human alpha(1A) calcium channel
subunit linked to familial hemiplegic migraine. J Neurosci 1999;
13.Tottene A, Fellin T, Pagnutti S, et al. Familial hemiplegic mi-
graine mutations increase Ca2?influx through single human
CaV2.1 channels and decrease maximal CaV2.1 current density in
neurons. Proc Natl Acad Sci U S A 2002;99:13284–13289.
14. Tottene A, Pivotto F, Fellin T, et al. Specific kinetic alterations of
human CaV2.1 calcium channels produced by mutation S218L
edema and coma after minor head trauma. J Biol Chem 2005;
and delayed cerebral
15.Kraus RL, Sinnegger MJ, Glossmann H, et al. Familial hemiplegic
migraine mutations change alpha(1A) Ca2?channel kinetics.
J Biol Chem 1998;273:5586–5590.
16. Kraus RL, Sinnegger MJ, Koschak A, et al. Three new familial
hemiplegic migraine mutants affect P/Q-type Ca(2?) channel ki-
netics. J Biol Chem 2000;275:9239–9243.
17.Cao YQ, Piedras-Renterı ´a ES, Smith GB, et al. Presynaptic Ca2?
channels compete for channel type-preferring slots in altered
neurotransmission arising from Ca2?
18.Cao YQ, Tsien RW. Effects of familial hemiplegic migraine type 1
mutations on neuronal P/Q-type Ca2? channel activity and in-
hibitory synaptic transmission. Proc Natl Acad Sci U S A 2005;
19. Barrett CF, Cao YQ, Tsien RW. Gating deficiency in a familial
hemiplegic migraine type 1 mutant P/Q-type calcium channel.
J Biol Chem 2005;280:24064–24071.
20. Van den Maagdenberg AM, Pietrobon D, Pizzorusso T, et al. A
Cacna1a knock-in migraine mouse model with increased suscepti-
bility to cortical spreading depression. Neuron 2004;41:701–710.
21.Tottene A, Conti R, Fabbro A, et al. Enhanced excitatory trans-
mission at cortical synapses as the basis for facilitated spreading
depression in Ca(v)2.1 knockin migraine mice. Neuron 2009;61:
22. Lauritzen M. Pathophysiology of the migraine aura. The spread-
ing depression theory. Brain 1994;117:199–210.
23. Bolay H, Reuter U, Dunn AK, et al. Intrinsic brain activity triggers
trigeminal meningeal afferents in a migraine model. Nat Med
24.Haan J, Kors EE, Vanmolkot KR, et al. Migraine genetics: an
update. Curr Pain Headache Rep 2005;9:213–220.
25. Fitzsimons RB, Wolfenden WH. Migraine coma. Meningitic migraine
with cerebral oedema associated with a new form of autosomal
dominant cerebellar ataxia. Brain 1985;108:555–577.
26. Kors EE, Terwindt GM, Vermeulen FL, et al. Delayed cerebral
edema and fatal coma after minor head trauma: role of
CACNA1A calcium channel subunit gene and relationship with
familial hemiplegic migraine. Ann Neurol 2001;49:753–760.
27. Curtain RP, Smith RL, Ovcaric M, Griffiths LR. Minor head
trauma-induced sporadic hemiplegic migraine coma. Pediatr
28. Chan YC, Burgunder JM, Wilder-Smith E, et al. Electroencepha-
lographic changes and seizures in familial hemiplegic migraine
patients with the CACNA1A gene S218L mutation. J Clin Neu-
29.Sa ´ndor PS, Mascia A, Seidel L, et al. Subclinical cerebellar
dimensional analysis of reaching movements. Ann Neurol
typesof migraine:a three-
30.Ambrosini A, Sa ´ndor PS, De Pasqua V, et al. Performances in
cerebellar and neuromuscular transmission tests are correlated in
migraine with aura. J Headache Pain 2008;9:29–32.
31. Rogawski MA. Common pathophysiologic mechanisms in mi-
graine and epilepsy. Arch Neurol 2008;65:709–714.
32.Haan J, van den Maagdenberg AM, Brouwer OF, et al. A review
of the genetic relation between migraine and epilepsy. Cepha-
33.Blau JN, Solomon F. Migraine and intracranial swelling: an ex-
periment of nature. Lancet 1985;28:718.
34.Meaney JF, Williams CE, Humphrey PR. Case report: transient
unilateral cerebral oedema in hemiplegic migraine: MR imaging
and angiography. Clin Radiol 1996;51:72–76.
35.Cha YH, Millett D, Kane M, et al. Adult-onset hemiplegic mi-
graine with cortical enhancement and oedema. Cephalalgia
36.Lakso M, Pichel JG, Gorman JR, et al. Efficient in vivo manipu-
lation of mouse genomic sequences at the zygote stage. Proc
Natl Acad Sci U S A 1996;93:5860–5865.
37.Kaplan EL, Meier P. Nonparametric estimation from incomplete
observations. J Am Stat Assoc 1958;53:457–481.
38. Chen Y, Constantini S, Trembovler V, et al. An experimental
model of closed head injury in mice: pathophysiology, histo-
pathology, and cognitive deficits. J Neurotrauma 1996;13:
39.Vakili A, Kataoka H, Plesnila N. Role of arginine vasopressin V1
and V2 receptors for brain damage after transient focal cerebral
ischemia. J Cereb Blood Flow Metab 2005;25:1012–1019.
40.Zweckberger K, Ero ¨s C, Zimmermann R, et al. Effect of early and
delayed decompressive craniectomy on secondary brain damage
after controlled cortical impact in mice. J Neurotrauma 2006;23:
41.Fletcher CF, Tottene A, Lennon VA, et al. Dystonia and cerebel-
lar atrophy in CACNA1A null mice lacking P/Q calcium channel
activity. FASEB J 2001;15:1288–1290.
42.Uchitel OD, Protti DA, Sanchez V, et al. P-type voltage-
dependent calcium channel mediates presynaptic calcium influx
and transmitter release in mammalian synapses. Proc Natl Acad
Sci U S A 1992;89:3330–3333.
43.Kaja S, van de Ven RC, Broos LA, et al. Characterization of
acetylcholine release and the compensatory contribution of non-
van den Maagdenberg et al: S218L FHM1 Knock-in Mice
January, 2010 97
Ca(v)2.1 channels at motor nerve terminals of leaner Ca(v)2.1-
mutant mice. Neuroscience 2007;144:1278–1287.
44.Kaja S, van de Ven RC, Broos LA, et al. Gene dosage-dependent
CACNA1A R192Q knockin mice are non-progressive and do not
lead to morphological changes or muscle weakness. Neuro-
45.Magleby KL, Stevens CF. The effect of voltage on the time
course of end-plate currents. J Physiol 1972;223:151–171.
46.McLachlan EM, Martin AR. Non-linear summation of end-plate
potentials in the frog and mouse. J Physiol 1981;311:307–324.
47.Zwingman TA, Neumann PE, Noebels JL, et al. Rocker is a new
variant of the voltage-dependent calcium channel gene Cacna1a.
J Neurosci 2001;21:1169–1178.
48. Gorji A. Spreading depression: a review of the clinical relevance.
Brain Res Brain Res Rev 2001;38:33–60.
49.Solomon S. John Graham Senior Clinicians Award Lecture. Post-
traumatic migraine. Headache 1998;38:772–778.
50.Hadjikhani N, Sanchez Del Rio M, Wu O, et al. Mechanisms of
migraine aura revealed by functional MRI in human visual cortex.
Proc Natl Acad Sci U S A 2001;98:4687–4692.
51. Bowyer SM, Aurora KS, Moran JE, et al. Magnetoencephalo-
graphic fields from patients with spontaneous and induced mi-
graine aura. Ann Neurol 2001;50:582–587.
52. Whieldon JA, van Harreveld A. Cumulative effects of minimal
cortical stimulation. Electroencephalogr Clin Neurophysiol 1950;
53. Eikermann-Haerter K, Dileko ¨z E, Kudo C, et al. Genetic and hor-
monal factors modulate spreading depression and transient
hemiparesis in mouse models of familial hemiplegic migraine
type 1. J Clin Invest 2009;119:99–109.
54. Takano T, Tian GF, Peng W, et al. Cortical spreading depression
causes and coincides with tissue hypoxia. Nat Neurosci 2007;10:
55.Gursoy-Ozdemir Y, Qiu J, Matsuoka N, et al. Cortical spreading
depression activates and upregulates MMP-9. J Clin Invest 2004;
56. Dreier JP, Jurkat-Rott K, Petzold GC, et al. Opening of the
blood-brain barrier preceding cortical edema in a severe attack
of FHM type II. Neurology 2005;64:2145–2147.
57.Hoebeek FE, Stahl JS, van Alphen AM, et al. Increased noise
level of Purkinje cell activities minimizes impact of their
58.Walter JT, Alvin ˜a K, Womack MD, et al. Decreases in the pre-
cision of Purkinje cell have making cause cerebellar dysfunction
and ataxia. Nat Neurosci 2006;9:389–397.
ANNALS of Neurology
98Volume 67, No. 1