FULL-LENGTH ORIGINAL RESEARCH
Altered thalamic GABAA-receptor subunit expression in the stargazer mouse model of absence epilepsy Steve Seo and Beulah Leitch Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
SUMMARY
Steve Seo is a Ph.D. student in Dr. Beulah Leitch’s lab at the Department of Anatomy, University of Otago, New Zealand.
Purpose: Absence seizures, also known as petit mal seizures, arise from disruptions within the cortico-thalamocortical network. Interconnected circuits within the thalamus consisting of inhibitory neurons of the reticular thalamic nucleus (RTN) and excitatory relay neurons of the ventral posterior (VP) complex, generate normal intrathalamic oscillatory activity. The degree of synchrony in this network determines whether normal (spindle) or pathologic (spike wave) oscillations occur; however, the cellular and molecular mechanisms underlying absence seizures are complex and multifactorial and currently are not fully understood. Recent experimental evidence from rodent models suggests that regional alterations in c-aminobutyric acid (GABA)ergic inhibition may underlie hypersynchronous oscillations featured in absence seizures. The aim of the current study was to investigate whether region-specific differences in GABAA receptor (GABAAR) subunit expression occur in the VP and RTN thalamic regions in the stargazer mouse model of absence epilepsy where the primary deficit is in a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) expression. Methods: Immunofluorescence confocal microscopy and semiquantitative Western blot analysis were used to investigate region-specific changes in GABAAR subunits in the thalamus of the stargazer mouse model of absence epilepsy to determine whether changes in GABAergic inhibition could contribute to the mechanisms underlying seizures in this model of absence epilepsy. Key Findings: Immunofluorescence confocal microscopy revealed that GABAAR a1 and b2 subunits are predominantly expressed in the VP, whereas a3 and b3 subunits are localized primarily in the RTN. Semiquantitative Western blot analysis of VP and RTN samples from epileptic stargazers and their nonepileptic littermates showed that GABAAR a1 and b2 subunit expression levels in the VP were significantly increased (a1: 33%, b2: 96%) in epileptic stargazers, whereas a3 and b3 subunits in the RTN were unchanged in the epileptic mice compared to nonepileptic control littermates. Significance: These findings suggest that region-specific differences in GABAAR subunits in the thalamus of epileptic mice, specifically up-regulation of GABAARs in the thalamic relay neurons of the VP, may contribute to generation of hypersynchronous thalamocortical activity in absence seizures. Understanding region-specific differences in GABAAR subunit expression could help elucidate some of the cellular and molecular mechanisms underlying absence seizures and thereby identify targets by which drugs can modulate the frequency and severity of epileptic seizures. Ultimately, this information could be crucial for the development of more specific and effective therapeutic drugs for treatment of this form of epilepsy. KEY WORDS: GABAA receptors, Thalamocortical network, Absence epilepsy, Stargazer mouse, Immunofluorescence, Western blot.
Accepted October 29, 2013; Early View publication January 13, 2014. Department of Anatomy, Brain Health Research Centre, Otago School of Medical Sciences, University of Otago, Dunedin, New Zealand Address correspondence to Beulah Leitch, Department of Anatomy, Brain Health Research Centre, Otago School of Medical Sciences, University of Otago, PO Box 913, Dunedin, New Zealand. E-mail: [email protected] Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
224
225 Altered GABAARs in Absence Epilepsy Absence epilepsy, which occurs mostly in children, is a type of nonconvulsive generalized epileptic seizure characterized by sudden, momentary impairment of consciousness. Childhood absence epilepsy is the most common pediatric epilepsy, responsible for >10% of all childhood-onset epilepsies.1 Absence seizures, previously termed petit mal seizures, can occur up to several hundred times a day, and have been associated with cognitive and learning difficulties,2 as well as with increased risk of generalized tonic–clonic seizures as patients grow older. The three most common antiepileptic drugs (AEDs) for the treatment of childhood absence epilepsy are ethosuximide, lamotrigine, and valproic acid (sodium valproate).3 However, in some cases, these AEDs are associated with a lack of seizure control; treatment failure occurs in approximately half of newly diagnosed patients.4 In addition, these drugs are frequently accompanied by adverse side effects, ranging from mild to intolerable (general, digestive, cognitive, psychological, and behavioral).4 On electroencephalography, absence seizures appear as 2–4 Hz spike-wave discharges (SWDs), and these are known to arise from disturbances within the thalamocortical system.5–7 Although the underlying molecular and cellular mechanisms have not been fully elucidated, several c-aminobutyric acid A receptor (GABAAR) subunit mutations have been linked with human cases of absence seizures,8,9 and evidence from animal models of absence seizures suggest that altered GABAergic inhibition may contribute to generation of absence seizures.10–14 For example, an increase in either phasic or tonic GABAergic inhibition within the thalamic ventral posterior complex (VP) has been implicated in a number of rodent models of absence epilepsy.11,14 Conversely in mouse knockout models of absence seizures, such as the GABAA b3 subunit knockout mouse, SWDs are linked to loss of GABAAR-mediated inhibition in the reticular thalamic nucleus (RTN).10 However, in the stargazer mouse model of absence epilepsy, which presents with a recessive mutation on chromosome 15, the primary deficit is severely reduced normal expression of the stargazin protein, also known as transmembrane AMPA receptor regulatory protein gamma 2 (TARP-c2).15–17 Stargazin plays a role in membrane trafficking and synaptic insertion of a-amino-3-hydroxy-5methyl-4-isoxazolepropionic acid receptors (AMPARs) at excitatory synapses,18 as well as modulation of the biophysical properties of AMPARs.19 Functional studies on this model have shown that in the thalamus, there is a significant reduction of AMPAR-mediated miniature excitatory postsynaptic current (mEPSC) in the RTN neurons,20 and recent work has revealed that this change could be attributed to selective loss of AMPAR subunits at corticothalamic synapses in the RTN region (CT-RTN synapses).7,21 The underlying mechanism in which loss of AMPAR-mediated excitation at CT-RTN synapses is linked to generation of
SWDs in absence seizures remains controversial, and may involve disinhibition of excitatory relay neurons of the VP following impaired AMPAR-mediated signaling in the RTN. Alternatively, compensatory increase in synaptic N-methyl-D-aspartate receptors (NMDARs) may occur at CT-RTN synapses where AMPARs are lost,22 thereby leading to increased feed-forward inhibition rather than disinhibition. Furthermore, the stargazin mutation is also known to affect GABAAR subunit expression in the hippocampus23 and the cerebellum.24 Therefore investigating region-specific differences in GABAAR subunits in the thalamus of the stargazer mouse may reveal some of the common mechanisms underlying absence seizures and lead to identification of new therapeutic targets for treatment of absence epilepsy. Hence the aim of current study was to investigate whether GABAAR expression is altered in the RTN and VP thalamus of epileptic stargazer mutant mice and thus could contribute to generation of absence seizures.
Experimental Procedures Animals Breeding stocks of stargazer mice were obtained from the Jackson Laboratory (Bar Harbor, ME, U.S.A.). Male epileptic stargazer mice (stg/stg) and nonepileptic control littermates (+/+ and +/stg)15 were raised in University of Otago Hercus Taieri Resource Unit. Only male littermate siblings aged 8–12 weeks were used for experiments. Animals were provided ad libitum access to food and water and maintained at 12 h light/dark cycle (lights on at 7 a.m.). Every effort was made to minimize the number of animals used and The University of Otago Committee on Ethics in the Care and Use of Laboratory Animals approved all animal protocols used in the study. Immunofluorescence confocal microscopy Immunohistochemical procedures were carried out based on well-established protocols previously used in our laboratory.25 After being anesthetized with sodium pentobarbital solution (Nembutal 60 mg/kg, i.p.), mice were perfused transcardially with 5% heparin in phosphate buffered saline (PBS) pH 7.4 for 30 s, then 4% paraformaldehyde (PFA) in 0.1 M Sorensen’s phosphate buffer (PB) pH 7.4 for 10 min. After extraction, brains were cryoprotected in increasing concentration of sucrose solutions (10%, 20%, and 30%). Brains were coronally sectioned into 30-lm–thick sections on a cryostat (Leica CM1950; Leica Microsystems, Wetzlar, Germany) before being treated with 1% sodium borohydride (NaBH4) for 30 min. Sections were submerged in blocking solution (4% normal goat serum, 0.1% bovine serum albumin, 0.3% Triton X-100 in PBS) for 2 h before being incubated in primary antibodies for 48 h. Optimal antibody concentrations for each antibody were determined by using a range of antibody dilutions (serial double-dilution method). Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
226 S. Seo and B. Leitch Primary antibodies used were: rabbit polyclonal GABAAR a1 antibody (1:500, aga-001; Alomone, Jerusalem, Israel); rabbit polyclonal GABAAR a3 antibody (1:200, aga-003; Alomone); rabbit polyclonal GABAAR b2 antibody (1:500, ab8340; Abcam, Cambridge, United Kingdom); rabbit polyclonal GABAAR b3 antibody (1:500, ab4046; Abcam); and mouse monoclonal GAD67 antibody (1:500, mab5406; Chemicon, Temecula, CA, U.S.A.). Sections were doublelabeled with antibodies for GABAAR subunits and GAD67 antibody to clearly delineate the RTN region from the VP region within the thalamus.20,26 Sections were then labeled with the secondary antibodies, Alexa 488 goat anti-rabbit antibody (1:1,000; Invitrogen, Carlsbad, CA, U.S.A.) and Alexa 568 goat anti-mouse antibody (1:1,000; Invitrogen), before being mounted on glass slides and covered with 1,4diazabicyclo[2.2.2]octane (DABCO)-glycerol. Images were taken using LSM 710 confocal microscope (Carl Zeis, Thornwood, NY, U.S.A.). Preadsorption controls were carried out where antigens, against which the antibody was raised, were available (manufacturer supplied control antigens for a1 and a3) in order to test the specificity of the primary antibodies for their respective antigens. Antigen was added at 10 times excess concentration of antibodies and left at room temperature for 4 h before incubation. Omission controls were also carried out where some sections were processed without addition of primary antibodies, to test for the specificity of the secondary antibodies for their respective primary antibodies. Western blotting Tissue samples for the Western blotting procedures were extracted and processed according to the methods described in Barad et al.21 In brief, mice were sacrificed using cervical dislocation method, and brains were extracted and snap frozen on dry ice. Brains were coronally sectioned (300 lm thickness) in a cryostat (SLEE Medical Systems, Mainz, Germany) at 10°C and thaw-mounted onto glass slides. Tissues from RTN and VP regions were identified in reference to Paxinos & Franklin27 and micropunched with a 24 gauge blunted needle on chilled aluminum block under light microscope. Samples were kept in lysis buffer containing phenylmethyl sulfonyl fluoride (PMSF) and protease inhibitor mix (Sigma-Aldrich, St Louis, Missouri, U.S.A.), and homogenized by 5 min of sonication, 2 min of heating on a heat block (100°C), and centrifugation for 5 min at 20,000 g. The supernatant was transferred to microtubes and stored at 80°C until use. Protein concentration was measured using detergent-compatible protein assay (DC protein assay, 500-0116; Bio-Rad Laboratories, Hemel Hempstead, United Kingdom). Ten micrograms to 20 lg of proteins was loaded onto each polyacrylamide gel for electrophoresis (Bio-Rad Laboratories), before being transferred onto nitrocellulose membranes. After drying overnight, membranes were blocked in Odyssey blocking buffer (LICOR Biosciences, Lincoln, NE, U.S.A.) for an hour before Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
being incubated in primary antibodies overnight at 4°C. GABAAR subunit antibodies for Western blotting were used at dilutions of 1:500 (a1), 1:500 (a3), 1:1,000 (b2), and 1:500 (b3). Mouse monoclonal b-actin antibody was used as a loading control (1:1,000, ab8226; Abcam). Membranes were probed with goat anti-mouse IRDye800 (1:10,000, 926-32210; LI-COR Biosciences) and goat anti-rabbit IRDye680 (1:10,000, 926-32221; LI-COR Biosciences) for an hour at room temperature before being scanned on the Odyssey Infrared Imaging System (LI-COR Biosciences). Statistical analysis of band intensity Western blot membrane band intensities were analyzed using the Odyssey v3.0 program (LI-COR Biosciences). Integrated intensity was normalized with regard to respective b-actin control bands in each individual sample lane. The mean of all nonepileptic values was calculated on individual blots, and the normalized intensity values of each nonepileptic and epileptic sample were expressed as a ratio relative to that mean. These ratios are presented as mean standard error of the mean (SEM), and unpaired, two-tailed Mann-Whitney U-test was used to determine statistical significance. Validating the purity of micropunched samples To verify the purity of micropunched samples used in Western blotting, two markers of thalamic subregions were used. Within the thalamus, AMPAR GluA4 subunit is predominantly expressed in the RTN region,21,28 whereas vesicular glutamate transporter vGlut1 is a marker of the VP region.22 Samples were probed with antibodies against GluA4 (1:1,000, ab1508; Millipore, Watford, United Kingdom) and vGlut1 (1:1,000, 135311; Synaptic Systems, Goettingen, Germany) in order to validate the purity of samples used in the current study. Figure preparation Graph figures and final images were prepared using Adobe Photoshop CS5.1 (Adobe Systems, San Jose, CA, U.S.A.) and optimized for brightness, contrast, and resolution.
Results Localization of GABAAR subunits in the thalamic subregions The distribution of GABAAR a1, a3, b2, and b3 subunits in the thalamic RTN and VP regions was examined using immunofluorescence confocal microscopy. Thalamic sections were double-labeled with antibody for GAD67, which converts glutamate to GABA. GAD67 serves as a marker of these thalamic subregions, as it is expressed more prominently in the RTN than in the VP; the RTN contains mainly GABAergic neurons.20,26 In agreement with previous studies on rat tissue sections,29 GAD67 was more strongly expressed in the RTN region than the VP region in the mouse tissue sec-
227 Altered GABAARs in Absence Epilepsy A
B
C
D
E
Figure 1. (A–D) Immunofluorescence confocal microscopy images showing localization of GABAAR subunits and GAD67 in the mouse thalamus. (E) Elimination of virtually all immunostaining in the control sections demonstrates the specificity of primary and secondary antibodies used in the experiments. Epilepsia ILAE
tions analyzed in this study; therefore, it served as a useful marker for clearly identifying the RTN subthalamic region and distinguishing it from VP region in double-labeled tissue samples (Fig. 1). GABAAR a1 and b2 subunits were found
almost exclusively in the VP, with virtually no a1 (Fig. 1A) and very little b2 (Fig. 1B) labeling in the RTN region. Conversely, GABAAR a3 and b3 subunits were expressed mainly in the RTN region, with only diffuse immunoreactivEpilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
228 S. Seo and B. Leitch ity in the VP (Fig. 1C,D). These observations are in accordance with previously published studies on the regional distribution of GABAAR subunits in rat brain.30 Preadsorption of GABAAR a1 and a3 antibodies with their respective specific antigens obtained from the manufacturers, eliminated virtually all immunostaining in control sections (Fig. 1E). Control peptides were unavailable for GABAAR b2 and b3 antibodies from the manufacturers. Omission of primary antibodies also produced no immunolabeling in sections incubated with secondary antibody only (Fig. 1E). These results indicate the specificity of the primary and secondary antibodies used in the current experiment. Selective differences in GABAAR subunits in the VP of epileptic mice Tissue expression levels of GABAAR subunits within the VP (a1, b2) and RTN (a3, b3) regions were compared A
C
B
D
E
between epileptic (E) and nonepileptic (NE) control littermates using semiquantitative Western blotting (Fig. 2). Bands representing each of the GABAAR subunits appeared at their expected molecular weights (~50 kDa for a1, 50– 53 kDa for b2, ~60 kDa for a3, 51–56 kDa for b3; Fig. 2A–D), when compared to the reference protein ladder. The specificity of the primary antibodies used in the Western blotting procedures was tested by preadsorbing the antibodies with an excess of their antigen peptides. The specificity of the secondary antibodies was also tested by omitting the primary antibodies. The corresponding bands were eliminated in all control blots, confirming the specificity of primary and secondary antibodies used. Semiquantitative analyses of the intensities of fluorescent band signals after normalization against b-actin loading controls showed that both a1 and b2 GABAAR subunits were significantly increased in the VP region of epileptic mice compared to the control littermates (Fig. 3). Epileptic stargazer mice showed 33% increase in GABAAR a1 subunit expression compared to controls (NE 1.00 0.06, E 1.33 0.12, n = 8, p < 0.05) and 96% increase in GABAAR b2 subunit expression (NE 1.00 0.06, E 1.96 0.15, n = 8, p < 0.01) (Fig. 3A, B). In contrast, there was no statistically significant difference between expression levels of the a3 and b3 subunits in the RTN of epileptic and nonepileptic mice (Fig. 3C, D). Summary of the relative difference in GABAAR subunit expressions in thalamic regions is presented in Fig. 2E as a change of percentage bar graph. The purity of RTN and VP tissue punch samples was tested by probing the samples with GluA4 and vGlut1 antibodies, which are predominantly expressed in RTN21,28 and VP,22 respectively. GluA4 was shown to be predominantly expressed in the RTN tissue samples (Fig. S1A), whereas vGlut1 was expressed mainly in the VP samples (Fig. S1B). These results are consistent with previous reports that describe localization of GluA4 and vGlut1 in tissue sections of the thalamus, and thus verified the purity of micropunched samples used in the Western blotting procedures.
Discussions Figure 2. Representative Western blot images showing expression of GABAAR subunits (A) a1 (B) b2 (C) a3 (D) b3 in the thalamic subregions in nonepileptic and epileptic mice. All bands were present at expected molecular weights. (E) Summary of differences in GABAAR-subunit expression in the thalamus, expressed as a percentage of nonepileptic controls. The a1 and b2 subunits were increased by 33% and 96%, respectively, in the VP region of epileptic mice. GABAAR a3 and b3 subunits in the RTN region showed no significant difference between epileptic stargazer mice and nonepileptic control mice. Epilepsia ILAE Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
The data presented in the current study demonstrate, for the first time, region- and subunit-specific differences in the expression of GABAAR subunits in the stargazer mouse model of absence epilepsy. The GABAAR subunits a1 and b2, which are preferentially localized in the VP region, are significantly increased in the epileptic stargazer mice (a1: 33%, b2: 96%), whereas the GABAAR subunits a3 and b3, which are mainly present in the RTN region, show no significant difference in expression levels. These results suggest that VP region-specific increases in GABAARs may possibly contribute to generation of hypersynchronous thalamic oscillatory activity in absence seizures in stargazers.
229 Altered GABAARs in Absence Epilepsy A
C
B
D
Figure 3. Relative expression levels of GABAAR subunits in the RTN and VP of nonepileptic (+/+, +/stg) and epileptic (stg/stg) mice, represented as bar graphs. There are statistically significant increases in normalized relative intensity reading for the GABAAR (A) a1 subunit (NE 1.00 0.06, E 1.33 0.12, n = 8, p < 0.05) and the b2 subunit (NE 1.00 0.06, E 1.96 0.15, n = 8, p < 0.01) in the VP region. There was no statistically significant difference in the a3 and b3 subunit levels between nonepileptic and epileptic mice in the RTN region (*p < 0.05, **p < 0.01). Epilepsia ILAE
Absence seizures are known to arise from disruptions within the cortico-thalamocortical network, but the complex cellular and molecular mechanisms underlying pathophysiology of absence seizures are not well understood and appear to be multifactorial.31 Normal spindle-like oscillations in the thalamocortical network are maintained by the sequential firing of inhibitory GABAergic neurons of the RTN and excitatory glutamatergic thalamocortical (TC) relay neurons in VP region, which are reciprocally connected. The RTN inhibitory neurons receive excitatory inputs from collateral projections of corticothalamic (CT) and TC relay neurons, and project feed-forward inhibition to TC relay neurons in the VP. The TC neurons then become hyperpolarized and fire postinhibitory rebound action potentials as a result of de-inactivation of T-type (lowthreshold) calcium channels.32 This rebound potential reexcites RTN neurons, thereby initiating intrathalamic oscillations.31,33,34 Cortical activation at CT-RTN synapses is more powerful than cortical activation at CT-VP synapses, due to the fact that AMPAR GluA4 subunits are expressed far more abundantly at CT-RTN synapses than at CT-TC synapses.35 Because of this intrinsic connectivity between the cortex and the thalamus, any change in balance between excitation and inhibition within the thalamus could potentially swing normal spindle oscillations into pathologic hypersynchronous SWDs.
In the stargazer mouse model of absence epilepsy, the primary deficit is in AMPAR expression and function, as demonstrated by electrophysiological functional studies20 and immunogold ultrastructural studies.21 These studies demonstrated a selective loss in AMPAR expression and function at CT-RTN synapses but not at CT-VP synapses. The AMPAR subunit GluA4, which is the prominent subunit in the RTN, was significantly reduced in the RTN but not the VP in stargazers. A link between absence epileptic seizures and regional loss of GluA4-AMPAR-mediated excitation in the RTN was also demonstrated by studies on the Gria4 KO model, which exhibits spontaneous SWDs as a result of loss of GluA4.7 These researchers demonstrated that loss of GluA4 caused selective weakening of synaptic strength at CT-RTN synapses, but not at CT-TC relay neuron synapses in the VP or at TC relay-RTN synapses. Collectively, these studies show that loss of AMPARs specifically at CT-RTN synapses is linked to the absence epilepsy phenotype in these mutant and KO mice. However, whether these changes in AMPAR expression and function at CT-RTN synapses are directly causative of hypersynchronous SWDs in absence epilepsy is still unknown. One hypothesis that has been put forward is that the loss of AMPARs at CTRTN synapses and the subsequent reduction in RTN synaptic strength could impair feed-forward inhibition, effectively causing disinhibition of relay neurons and thereby Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
230 S. Seo and B. Leitch resulting in an imbalance of excitation and inhibition within the thalamocortical network and thus leading to hypersynchronous oscillations. However, another hypothesis that has been proposed by Lacey et al.,22 suggests increased excitation of these inhibitory neurons may underlie the hypersynchronous SWDs in this model. These researchers reported that although the amplitude of AMPAR-mediated currents was reduced in stargazers, there was a corresponding increase in the amplitude of NMDARmediated currents at CT-RTN synapses. This indicates that a compensatory increase in synaptic NMDARs may occur where AMPARs are lost, leading to an increase in RTN inhibitory neuron excitability and thus increased feed-forward inhibition.22 However, there is also evidence in the literature that the stargazin mutation may affect GABAAR subunit expression. Changes in GABAAR subunit expression have been reported in hippocampal re-entrant axon collaterals in the dentate gyrus23 and in cerebellar granule cells24 of stargazers. Evidence from other animal models also supports the contention that SWDs underlying absence epilepsy are linked to changes in GABAergic inhibition (phasic or tonic) within thalamocortical networks.10,12–14,36 However, the impact and contribution of altered GABAARmediated inhibition in different thalamic subregions to SWD and absence epilepsy is unclear and somewhat controversial. Intra-RTN inhibition (between reciprocally connected inhibitory RTN neurons) is thought to regulate intrathalamic inhibition and prevent hypersynchronous oscillatory activity leading to the generation of seizures. One hypothesis is that recurrent inhibition within the RTN, which occurs through phasic GABAA a3b3c2 receptors,34 becomes impaired in absence seizures. For example, in the genetically epilepsy-prone WAG/Rij rat, a3 subunit expression is almost completely abolished.13 In the b3 knockout mouse, hypersynchronous thalamic oscillations occur and are linked to reduced GABAA-mediated inhibition specifically in the RTN.10 However in the current study on epileptic stargazers, we found no significant difference in tissue levels of GABAAR a3 or b3, subunits using Western blotting analyses (Fig. 2E). This finding is in line with a study by Schofield et al.,37 which showed that a3 knockout mice do not show spontaneous absence seizure activity. In epileptic mice carrying c2 (R43Q) mutation, no difference in inhibitory postsynaptic potentials (IPSPs) in the RTN has been found,38 suggesting that reduced phasic inhibition within RTN is not a requirement for generation of absence seizures. Conversely, in genetic absence epilepsy rat from Strasbourg (GAERS) and dilute brown non-agouti (DBA) /2J mouse models, an increase in phasic inhibition in the RTN region has been reported.12,14,38 Overall, the published data with regard to the involvement of intra-RTN GABAAR-mediated phasic inhibition in seizure generation are confusing and in some cases contradictory. It seems then that any conclusions Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
on the role of intra-RTN inhibition in the generation of absence seizures cannot be drawn at present. In contrast, evidence for increased GABAergic inhibition in the VP thalamocortical relay neurons has been found in several animal models of absence epilepsy. An overall increase in phasic GABAAR-mediated inhibition in VP relay neurons was observed in the c-hydroxybutyrate model of absence epilepsy by Gervasi et al.11 An increase in tonic GABAergic inhibition in the VP relay neurons has also been reported in a number of rodent models of absence epilepsy including the GAERS rat, which could be partially attributed to malfunction of the GABA transporter GAT-1.14 In our study on the stargazer mouse model of absence epilepsy, semiquantitative Western blot analysis revealed significant increases in tissue levels of GABAAR a1 and b2 subunits in the VP region (33% and 96%, respectively; Fig. 2E). These results concur with previous reports on phasic and tonic inhibition in the VP relay neurons,14 which suggest that augmentation of GABAergic signaling in the thalamic relay nuclei may contribute to generation of hypersynchronous thalamocortical activity. It is worth noting that tonic inhibition in VP relay neurons is mediated via GABAARs having a a4b2d subunit composition, whereas phasic inhibition in the VP is mediated by a1b2c2 containing GABAARs.34 Hence, although the increase in the a1 subunit most likely reflects enhanced phasic inhibition through the a1b2c2 GABAARs, the nature of the increase in the b2 subunit expression (tonic, phasic, or both) remains to be elucidated. Given that an increase in both tonic or phasic inhibition may be possible at thalamocortical VP relay neurons, further investigations at the ultrastructural level are required to determine whether these differences in GABAAR subunits expression at the tissue level are also translated into changes in receptor expression at synapses (or at extrasynaptic sites), and whether these changes are associated with an increase in phasic inhibition or both phasic and tonic inhibition. Furthermore, it still remains to be determined exactly how the mutation of the stargazin gene leads to alterations in GABAAR-subunit expression: whether these differences occur preseizure and could thus be causative, or whether they are the result of seizure-induced activity. In summary, results from the current study demonstrate region-specific changes in GABAAR-subunit expression in the thalamus of epileptic mice. Increase in GABAAR-subunit expression in the VP region may represent up-regulation of phasic inhibition, tonic inhibition, or both. This may lead to an imbalance within the intrathalamic circuit and contribute to hypersynchronous thalamocortical activity and absence seizures. Future studies at the ultrastructural level are needed to determine if these global changes reflect region-specific changes in GABAAR expression at synapses. Furthermore, developmental studies will be required to determine whether such changes occur preseizure and
231 Altered GABAARs in Absence Epilepsy have a proabsence effect or are compensatory changes to reduce the propensity of absence seizures. In addition to the possible involvement of AMPARs, NMDARs, and GABAARs, in the generation of absence seizures, potential underlying mechanisms are further complicated by the presence of T-type (low voltage-activated) calcium channels, which upon activation are known to evoke burst firing and give rise to SWDs.39 GABAB receptors in relay neurons have also been implicated in several animal epilepsy models; their activation has been shown to be partially responsible for abnormally large and hypersynchronous rebound bursts and exacerbation of absence seizures.40 The profound complexity of the cellular and molecular mechanisms underlying generation of absence seizures and the range of targets that may be affected may reflect the variable results obtained from currently available AEDs, where even the best monotherapy fails in almost half of all newly diagnosed cases.4 Given the current situation, and the adverse impact that absence seizures have on childhood learning, it is of huge clinical relevance to further investigate the range of mechanisms whereby absence seizures can arise. Elucidating region-specific changes in synaptic GABAAR subunit expression and the synaptic mechanisms that regulate synchronous thalamocortical network activity may be crucial for identification of novel drug targets and the development of more specific and effective therapeutic intervention methods for treatment of specific subtypes of absence epilepsy.
Acknowledgments This work was supported by grants from the University of Otago Research Grants (UORG) and Deans Bequest Fund awarded to B.L. The authors thank the staff of the Otago Centre for Confocal Microscopy (OCCM) for excellent technical support, and the University of Otago for Doctoral Scholarship awarded to Steve Seo.
Disclosure The authors have no conflicts of interest to declare. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Jallon P, Loiseau P, Loiseau J. Newly diagnosed unprovoked epileptic seizures: presentation at diagnosis in CAROLE study. Epilepsia 2001;42:464–475. 2. Pavone P, Bianchini R, Trifiletti RR, et al. Neuropsychological assessment in children with absence epilepsy. Neurology 2001;56:1047–1051. 3. Wheless JW, Clarke DF, Carpenter D. Treatment of pediatric epilepsy: expert opinion. J Child Neurol 2005;20(Suppl. 1):S1–S56. 4. Glauser TA, Cnaan A, Shinnar S, et al. Ethosuximide, valproic acid, and lamotrigine in childhood absence epilepsy. N Engl J Med 2010;362:790–799. 5. McCormick DA, Contreras D. On the cellular and network bases of epileptic seizures. Annu Rev Physiol 2001;63:815–846. 6. Crunelli V, Leresche N. Childhood absence epilepsy: genes, channels, neurons and networks. Nat Rev Neurosci 2002;3:371–382.
7. Paz JT, Bryant AS, Peng K, et al. A new mode of corticothalamic transmission revealed in the Gria4( / ) model of absence epilepsy. Nat Neurosci 2011;14:1167–1173. 8. Wallace RH, Marini C, Petrou S, et al. Mutant GABA(A) receptor c2subunit in childhood absence epilepsy and febrile seizures. Nat Genet 2001;28:49–52. 9. Maljevic S, Krampfl K, Cobilanschi J, et al. A mutation in the GABA (A) receptor a1 subunit is associated with absence epilepsy. Ann Neurol 2006;59:983–987. 10. Huntsman MM, Porcello DM, Homanics GE, et al. Reciprocal inhibitory connections and network synchrony in the mammalian thalamus. Science 1999;283:541–543. 11. Gervasi N, Monnier Z, Vincent P, et al. Pathway-specific activation of gamma-hydroxybutyric acid in sensory thalamus and its relevance to absence seizures. J Neurosci 2003;23:11469–11478. 12. Bessa€ıh T, Bourgeais L, Badiu CI, et al. Nucleus-specific abnormalities of GABAergic synaptic transmission in a genetic model of absence seizures. J Neurophysiol 2006;96:3074–3081. 13. Liu X-B, Coble J, van Luijtelaar G, et al. Reticular nucleus-specific changes in a3 subunit protein at GABA synapses in genetically epilepsy-prone rats. Proc Natl Acad Sci USA 2007;104: 12512–12517. 14. Cope DW, Di Giovanni G, Fyson SJ, et al. Enhanced tonic GABAA inhibition in typical absence epilepsy. Nat Med 2009;15:1392–1399. 15. Noebels JL, Qiao X, Bronson RT, et al. Stargazer: a new neurological mutant on chromosome 15 in the mouse with prolonged cortical seizures. Epilepsy Res 1990;7:129–135. 16. Letts VA, Kang MG, Mahaffey CL, et al. Phenotypic heterogeneity in the stargazin allelic series. Mamm Genome 2003;14:506–513. 17. Letts VA. Stargazer—a mouse to seize. Epilepsy Curr 2005;5:161– 165. 18. Chen L, Chetkovich DM, Petralia RS, et al. Stargazin regulates synaptic targeting of AMPA receptors by two distinct mechanisms. Nature 2000;408:936–943. 19. Tomita S, Adesnik H, Sekiguchi M, et al. Stargazin modulates AMPA receptor gating and trafficking by distinct domains. Nature 2005;435:1052–1058. 20. Menuz K, Nicoll RA. Loss of inhibitory neuron AMPA receptors contributes to ataxia and epilepsy in stargazer mice. J Neurosci 2008;28:10599–10603. 21. Barad Z, Shevtsova O, Arbuthnott GW, et al. Selective loss of AMPA receptors at corticothalamic synapses in the epileptic stargazer mouse. Neuroscience 2012;217:19–31. 22. Lacey CJ, Bryant A, Brill J, et al. Enhanced NMDA receptordependent thalamic excitation and network oscillations in stargazer mice. J Neurosci 2012;32:11067–11081. 23. Payne HL, Donoghue PS, Connelly WM, et al. Aberrant GABA(A) receptor expression in the dentate gyrus of the epileptic mutant mouse stargazer. J Neurosci 2006;26:8600–8608. 24. Payne HL, Connelly WM, Ives JH, et al. GABAA alpha6-containing receptors are selectively compromised in cerebellar granule cells of the ataxic mouse, stargazer. J Biol Chem 2007;282:29130–29143. 25. Seo S, Liu P. Leitch B spatial learning-induced accumulation of agmatine and glutamate at hippocampal CA1 synaptic terminals. Neuroscience 2011;192:28–36. 26. Turner CP, DeBenedetto D, Ware E, et al. Postnatal exposure to MK801 induces selective changes in GAD67 or parvalbumin. Exp Brain Res 2010;201:479–488. 27. Paxinos G, Franklin KB. The mouse brain in stereotaxic coordinates. 2nd ed. San Diego, CA: Academic Press; 2001. 28. Mineff EM, Weinberg RJ. Differential synaptic distribution of AMPA receptor subunits in the ventral posterior and reticular thalamic nuclei of the rat. Neuroscience 2000;101:969–982. 29. Esclapez M, Tillakaratne NJ, Tobin AJ, et al. Comparative localization of mRNAs encoding two forms of glutamic acid decarboxylase with nonradioactive in situ hybridization methods. J Comp Neurol 1993;331:339–362. 30. Pirker S, Schwarzer C, Wieselthaler A, et al. GABA(A) receptors: immunocytochemical distribution of 13 subunits in the adult rat brain. Neuroscience 2000;101:815–850. 31. Beenhakker MP, Huguenard JR. Neurons that fire together also conspire together: is normal sleep circuitry hijacked to generate epilepsy? Neuron 2009;62:612–632. Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
232 S. Seo and B. Leitch 32. Kim U, Sanchez-Vives MV, McCormick DA. Functional dynamics of GABAergic inhibition in the thalamus. Science 1997;278:130–134. 33. Steriade M. Sleep, epilepsy and thalamic reticular inhibitory neurons. Trends Neurosci 2005;28:317–324. 34. Huguenard JR, McCormick DA. Thalamic synchrony and dynamic regulation of global forebrain oscillations. Trends Neurosci 2007;30:350–356. 35. Golshani P, Liu X-B, Jones EG. Differences in quantal amplitude reflect GluR4-subunit number at corticothalamic synapses on two populations of thalamic neurons. Proc Natl Acad Sci USA 2001;98:4172–4177. 36. Tan HO, Reid CA, Chiu C, et al. Increased thalamic inhibition in the absence seizure prone DBA/2J mouse. Epilepsia 2008;49:921–925. 37. Schofield CM, Kleiman-Weiner M, Rudolph U, et al. A gain in GABAA receptor synaptic strength in thalamus reduces oscillatory activity and absence seizures. Proc Natl Acad Sci USA 2009;106:7630–7635. 38. Tan HO, Reid CA, Single FN, et al. Reduced cortical inhibition in a mouse model of familial childhood absence epilepsy. Proc Natl Acad Sci USA 2007;104:17536–17541.
Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
39. Tringham E, Powell KL, Cain SM, et al. T-type calcium channel blockers that attenuate thalamic burst firing and suppress absence seizures. Sci Transl Med 2012;4:121ra19. 40. Manning J-PA, Douglas AR, Bowery NG. Pharmacology of absence epilepsy. Trends Pharmacol Sci 2003;24:542–549.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Expression of GluA4 and vGlut1 in the thalamic subregions, verifying the purity of RTN and VP samples used in Western blotting procedures.
Altered thalamic GABAA-receptor subunit expression in the stargazer mouse model of absence epilepsy Steve Seo and Beulah Leitch Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
SUMMARY
Steve Seo is a Ph.D. student in Dr. Beulah Leitch’s lab at the Department of Anatomy, University of Otago, New Zealand.
Purpose: Absence seizures, also known as petit mal seizures, arise from disruptions within the cortico-thalamocortical network. Interconnected circuits within the thalamus consisting of inhibitory neurons of the reticular thalamic nucleus (RTN) and excitatory relay neurons of the ventral posterior (VP) complex, generate normal intrathalamic oscillatory activity. The degree of synchrony in this network determines whether normal (spindle) or pathologic (spike wave) oscillations occur; however, the cellular and molecular mechanisms underlying absence seizures are complex and multifactorial and currently are not fully understood. Recent experimental evidence from rodent models suggests that regional alterations in c-aminobutyric acid (GABA)ergic inhibition may underlie hypersynchronous oscillations featured in absence seizures. The aim of the current study was to investigate whether region-specific differences in GABAA receptor (GABAAR) subunit expression occur in the VP and RTN thalamic regions in the stargazer mouse model of absence epilepsy where the primary deficit is in a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) expression. Methods: Immunofluorescence confocal microscopy and semiquantitative Western blot analysis were used to investigate region-specific changes in GABAAR subunits in the thalamus of the stargazer mouse model of absence epilepsy to determine whether changes in GABAergic inhibition could contribute to the mechanisms underlying seizures in this model of absence epilepsy. Key Findings: Immunofluorescence confocal microscopy revealed that GABAAR a1 and b2 subunits are predominantly expressed in the VP, whereas a3 and b3 subunits are localized primarily in the RTN. Semiquantitative Western blot analysis of VP and RTN samples from epileptic stargazers and their nonepileptic littermates showed that GABAAR a1 and b2 subunit expression levels in the VP were significantly increased (a1: 33%, b2: 96%) in epileptic stargazers, whereas a3 and b3 subunits in the RTN were unchanged in the epileptic mice compared to nonepileptic control littermates. Significance: These findings suggest that region-specific differences in GABAAR subunits in the thalamus of epileptic mice, specifically up-regulation of GABAARs in the thalamic relay neurons of the VP, may contribute to generation of hypersynchronous thalamocortical activity in absence seizures. Understanding region-specific differences in GABAAR subunit expression could help elucidate some of the cellular and molecular mechanisms underlying absence seizures and thereby identify targets by which drugs can modulate the frequency and severity of epileptic seizures. Ultimately, this information could be crucial for the development of more specific and effective therapeutic drugs for treatment of this form of epilepsy. KEY WORDS: GABAA receptors, Thalamocortical network, Absence epilepsy, Stargazer mouse, Immunofluorescence, Western blot.
Accepted October 29, 2013; Early View publication January 13, 2014. Department of Anatomy, Brain Health Research Centre, Otago School of Medical Sciences, University of Otago, Dunedin, New Zealand Address correspondence to Beulah Leitch, Department of Anatomy, Brain Health Research Centre, Otago School of Medical Sciences, University of Otago, PO Box 913, Dunedin, New Zealand. E-mail: [email protected] Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
224
225 Altered GABAARs in Absence Epilepsy Absence epilepsy, which occurs mostly in children, is a type of nonconvulsive generalized epileptic seizure characterized by sudden, momentary impairment of consciousness. Childhood absence epilepsy is the most common pediatric epilepsy, responsible for >10% of all childhood-onset epilepsies.1 Absence seizures, previously termed petit mal seizures, can occur up to several hundred times a day, and have been associated with cognitive and learning difficulties,2 as well as with increased risk of generalized tonic–clonic seizures as patients grow older. The three most common antiepileptic drugs (AEDs) for the treatment of childhood absence epilepsy are ethosuximide, lamotrigine, and valproic acid (sodium valproate).3 However, in some cases, these AEDs are associated with a lack of seizure control; treatment failure occurs in approximately half of newly diagnosed patients.4 In addition, these drugs are frequently accompanied by adverse side effects, ranging from mild to intolerable (general, digestive, cognitive, psychological, and behavioral).4 On electroencephalography, absence seizures appear as 2–4 Hz spike-wave discharges (SWDs), and these are known to arise from disturbances within the thalamocortical system.5–7 Although the underlying molecular and cellular mechanisms have not been fully elucidated, several c-aminobutyric acid A receptor (GABAAR) subunit mutations have been linked with human cases of absence seizures,8,9 and evidence from animal models of absence seizures suggest that altered GABAergic inhibition may contribute to generation of absence seizures.10–14 For example, an increase in either phasic or tonic GABAergic inhibition within the thalamic ventral posterior complex (VP) has been implicated in a number of rodent models of absence epilepsy.11,14 Conversely in mouse knockout models of absence seizures, such as the GABAA b3 subunit knockout mouse, SWDs are linked to loss of GABAAR-mediated inhibition in the reticular thalamic nucleus (RTN).10 However, in the stargazer mouse model of absence epilepsy, which presents with a recessive mutation on chromosome 15, the primary deficit is severely reduced normal expression of the stargazin protein, also known as transmembrane AMPA receptor regulatory protein gamma 2 (TARP-c2).15–17 Stargazin plays a role in membrane trafficking and synaptic insertion of a-amino-3-hydroxy-5methyl-4-isoxazolepropionic acid receptors (AMPARs) at excitatory synapses,18 as well as modulation of the biophysical properties of AMPARs.19 Functional studies on this model have shown that in the thalamus, there is a significant reduction of AMPAR-mediated miniature excitatory postsynaptic current (mEPSC) in the RTN neurons,20 and recent work has revealed that this change could be attributed to selective loss of AMPAR subunits at corticothalamic synapses in the RTN region (CT-RTN synapses).7,21 The underlying mechanism in which loss of AMPAR-mediated excitation at CT-RTN synapses is linked to generation of
SWDs in absence seizures remains controversial, and may involve disinhibition of excitatory relay neurons of the VP following impaired AMPAR-mediated signaling in the RTN. Alternatively, compensatory increase in synaptic N-methyl-D-aspartate receptors (NMDARs) may occur at CT-RTN synapses where AMPARs are lost,22 thereby leading to increased feed-forward inhibition rather than disinhibition. Furthermore, the stargazin mutation is also known to affect GABAAR subunit expression in the hippocampus23 and the cerebellum.24 Therefore investigating region-specific differences in GABAAR subunits in the thalamus of the stargazer mouse may reveal some of the common mechanisms underlying absence seizures and lead to identification of new therapeutic targets for treatment of absence epilepsy. Hence the aim of current study was to investigate whether GABAAR expression is altered in the RTN and VP thalamus of epileptic stargazer mutant mice and thus could contribute to generation of absence seizures.
Experimental Procedures Animals Breeding stocks of stargazer mice were obtained from the Jackson Laboratory (Bar Harbor, ME, U.S.A.). Male epileptic stargazer mice (stg/stg) and nonepileptic control littermates (+/+ and +/stg)15 were raised in University of Otago Hercus Taieri Resource Unit. Only male littermate siblings aged 8–12 weeks were used for experiments. Animals were provided ad libitum access to food and water and maintained at 12 h light/dark cycle (lights on at 7 a.m.). Every effort was made to minimize the number of animals used and The University of Otago Committee on Ethics in the Care and Use of Laboratory Animals approved all animal protocols used in the study. Immunofluorescence confocal microscopy Immunohistochemical procedures were carried out based on well-established protocols previously used in our laboratory.25 After being anesthetized with sodium pentobarbital solution (Nembutal 60 mg/kg, i.p.), mice were perfused transcardially with 5% heparin in phosphate buffered saline (PBS) pH 7.4 for 30 s, then 4% paraformaldehyde (PFA) in 0.1 M Sorensen’s phosphate buffer (PB) pH 7.4 for 10 min. After extraction, brains were cryoprotected in increasing concentration of sucrose solutions (10%, 20%, and 30%). Brains were coronally sectioned into 30-lm–thick sections on a cryostat (Leica CM1950; Leica Microsystems, Wetzlar, Germany) before being treated with 1% sodium borohydride (NaBH4) for 30 min. Sections were submerged in blocking solution (4% normal goat serum, 0.1% bovine serum albumin, 0.3% Triton X-100 in PBS) for 2 h before being incubated in primary antibodies for 48 h. Optimal antibody concentrations for each antibody were determined by using a range of antibody dilutions (serial double-dilution method). Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
226 S. Seo and B. Leitch Primary antibodies used were: rabbit polyclonal GABAAR a1 antibody (1:500, aga-001; Alomone, Jerusalem, Israel); rabbit polyclonal GABAAR a3 antibody (1:200, aga-003; Alomone); rabbit polyclonal GABAAR b2 antibody (1:500, ab8340; Abcam, Cambridge, United Kingdom); rabbit polyclonal GABAAR b3 antibody (1:500, ab4046; Abcam); and mouse monoclonal GAD67 antibody (1:500, mab5406; Chemicon, Temecula, CA, U.S.A.). Sections were doublelabeled with antibodies for GABAAR subunits and GAD67 antibody to clearly delineate the RTN region from the VP region within the thalamus.20,26 Sections were then labeled with the secondary antibodies, Alexa 488 goat anti-rabbit antibody (1:1,000; Invitrogen, Carlsbad, CA, U.S.A.) and Alexa 568 goat anti-mouse antibody (1:1,000; Invitrogen), before being mounted on glass slides and covered with 1,4diazabicyclo[2.2.2]octane (DABCO)-glycerol. Images were taken using LSM 710 confocal microscope (Carl Zeis, Thornwood, NY, U.S.A.). Preadsorption controls were carried out where antigens, against which the antibody was raised, were available (manufacturer supplied control antigens for a1 and a3) in order to test the specificity of the primary antibodies for their respective antigens. Antigen was added at 10 times excess concentration of antibodies and left at room temperature for 4 h before incubation. Omission controls were also carried out where some sections were processed without addition of primary antibodies, to test for the specificity of the secondary antibodies for their respective primary antibodies. Western blotting Tissue samples for the Western blotting procedures were extracted and processed according to the methods described in Barad et al.21 In brief, mice were sacrificed using cervical dislocation method, and brains were extracted and snap frozen on dry ice. Brains were coronally sectioned (300 lm thickness) in a cryostat (SLEE Medical Systems, Mainz, Germany) at 10°C and thaw-mounted onto glass slides. Tissues from RTN and VP regions were identified in reference to Paxinos & Franklin27 and micropunched with a 24 gauge blunted needle on chilled aluminum block under light microscope. Samples were kept in lysis buffer containing phenylmethyl sulfonyl fluoride (PMSF) and protease inhibitor mix (Sigma-Aldrich, St Louis, Missouri, U.S.A.), and homogenized by 5 min of sonication, 2 min of heating on a heat block (100°C), and centrifugation for 5 min at 20,000 g. The supernatant was transferred to microtubes and stored at 80°C until use. Protein concentration was measured using detergent-compatible protein assay (DC protein assay, 500-0116; Bio-Rad Laboratories, Hemel Hempstead, United Kingdom). Ten micrograms to 20 lg of proteins was loaded onto each polyacrylamide gel for electrophoresis (Bio-Rad Laboratories), before being transferred onto nitrocellulose membranes. After drying overnight, membranes were blocked in Odyssey blocking buffer (LICOR Biosciences, Lincoln, NE, U.S.A.) for an hour before Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
being incubated in primary antibodies overnight at 4°C. GABAAR subunit antibodies for Western blotting were used at dilutions of 1:500 (a1), 1:500 (a3), 1:1,000 (b2), and 1:500 (b3). Mouse monoclonal b-actin antibody was used as a loading control (1:1,000, ab8226; Abcam). Membranes were probed with goat anti-mouse IRDye800 (1:10,000, 926-32210; LI-COR Biosciences) and goat anti-rabbit IRDye680 (1:10,000, 926-32221; LI-COR Biosciences) for an hour at room temperature before being scanned on the Odyssey Infrared Imaging System (LI-COR Biosciences). Statistical analysis of band intensity Western blot membrane band intensities were analyzed using the Odyssey v3.0 program (LI-COR Biosciences). Integrated intensity was normalized with regard to respective b-actin control bands in each individual sample lane. The mean of all nonepileptic values was calculated on individual blots, and the normalized intensity values of each nonepileptic and epileptic sample were expressed as a ratio relative to that mean. These ratios are presented as mean standard error of the mean (SEM), and unpaired, two-tailed Mann-Whitney U-test was used to determine statistical significance. Validating the purity of micropunched samples To verify the purity of micropunched samples used in Western blotting, two markers of thalamic subregions were used. Within the thalamus, AMPAR GluA4 subunit is predominantly expressed in the RTN region,21,28 whereas vesicular glutamate transporter vGlut1 is a marker of the VP region.22 Samples were probed with antibodies against GluA4 (1:1,000, ab1508; Millipore, Watford, United Kingdom) and vGlut1 (1:1,000, 135311; Synaptic Systems, Goettingen, Germany) in order to validate the purity of samples used in the current study. Figure preparation Graph figures and final images were prepared using Adobe Photoshop CS5.1 (Adobe Systems, San Jose, CA, U.S.A.) and optimized for brightness, contrast, and resolution.
Results Localization of GABAAR subunits in the thalamic subregions The distribution of GABAAR a1, a3, b2, and b3 subunits in the thalamic RTN and VP regions was examined using immunofluorescence confocal microscopy. Thalamic sections were double-labeled with antibody for GAD67, which converts glutamate to GABA. GAD67 serves as a marker of these thalamic subregions, as it is expressed more prominently in the RTN than in the VP; the RTN contains mainly GABAergic neurons.20,26 In agreement with previous studies on rat tissue sections,29 GAD67 was more strongly expressed in the RTN region than the VP region in the mouse tissue sec-
227 Altered GABAARs in Absence Epilepsy A
B
C
D
E
Figure 1. (A–D) Immunofluorescence confocal microscopy images showing localization of GABAAR subunits and GAD67 in the mouse thalamus. (E) Elimination of virtually all immunostaining in the control sections demonstrates the specificity of primary and secondary antibodies used in the experiments. Epilepsia ILAE
tions analyzed in this study; therefore, it served as a useful marker for clearly identifying the RTN subthalamic region and distinguishing it from VP region in double-labeled tissue samples (Fig. 1). GABAAR a1 and b2 subunits were found
almost exclusively in the VP, with virtually no a1 (Fig. 1A) and very little b2 (Fig. 1B) labeling in the RTN region. Conversely, GABAAR a3 and b3 subunits were expressed mainly in the RTN region, with only diffuse immunoreactivEpilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
228 S. Seo and B. Leitch ity in the VP (Fig. 1C,D). These observations are in accordance with previously published studies on the regional distribution of GABAAR subunits in rat brain.30 Preadsorption of GABAAR a1 and a3 antibodies with their respective specific antigens obtained from the manufacturers, eliminated virtually all immunostaining in control sections (Fig. 1E). Control peptides were unavailable for GABAAR b2 and b3 antibodies from the manufacturers. Omission of primary antibodies also produced no immunolabeling in sections incubated with secondary antibody only (Fig. 1E). These results indicate the specificity of the primary and secondary antibodies used in the current experiment. Selective differences in GABAAR subunits in the VP of epileptic mice Tissue expression levels of GABAAR subunits within the VP (a1, b2) and RTN (a3, b3) regions were compared A
C
B
D
E
between epileptic (E) and nonepileptic (NE) control littermates using semiquantitative Western blotting (Fig. 2). Bands representing each of the GABAAR subunits appeared at their expected molecular weights (~50 kDa for a1, 50– 53 kDa for b2, ~60 kDa for a3, 51–56 kDa for b3; Fig. 2A–D), when compared to the reference protein ladder. The specificity of the primary antibodies used in the Western blotting procedures was tested by preadsorbing the antibodies with an excess of their antigen peptides. The specificity of the secondary antibodies was also tested by omitting the primary antibodies. The corresponding bands were eliminated in all control blots, confirming the specificity of primary and secondary antibodies used. Semiquantitative analyses of the intensities of fluorescent band signals after normalization against b-actin loading controls showed that both a1 and b2 GABAAR subunits were significantly increased in the VP region of epileptic mice compared to the control littermates (Fig. 3). Epileptic stargazer mice showed 33% increase in GABAAR a1 subunit expression compared to controls (NE 1.00 0.06, E 1.33 0.12, n = 8, p < 0.05) and 96% increase in GABAAR b2 subunit expression (NE 1.00 0.06, E 1.96 0.15, n = 8, p < 0.01) (Fig. 3A, B). In contrast, there was no statistically significant difference between expression levels of the a3 and b3 subunits in the RTN of epileptic and nonepileptic mice (Fig. 3C, D). Summary of the relative difference in GABAAR subunit expressions in thalamic regions is presented in Fig. 2E as a change of percentage bar graph. The purity of RTN and VP tissue punch samples was tested by probing the samples with GluA4 and vGlut1 antibodies, which are predominantly expressed in RTN21,28 and VP,22 respectively. GluA4 was shown to be predominantly expressed in the RTN tissue samples (Fig. S1A), whereas vGlut1 was expressed mainly in the VP samples (Fig. S1B). These results are consistent with previous reports that describe localization of GluA4 and vGlut1 in tissue sections of the thalamus, and thus verified the purity of micropunched samples used in the Western blotting procedures.
Discussions Figure 2. Representative Western blot images showing expression of GABAAR subunits (A) a1 (B) b2 (C) a3 (D) b3 in the thalamic subregions in nonepileptic and epileptic mice. All bands were present at expected molecular weights. (E) Summary of differences in GABAAR-subunit expression in the thalamus, expressed as a percentage of nonepileptic controls. The a1 and b2 subunits were increased by 33% and 96%, respectively, in the VP region of epileptic mice. GABAAR a3 and b3 subunits in the RTN region showed no significant difference between epileptic stargazer mice and nonepileptic control mice. Epilepsia ILAE Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
The data presented in the current study demonstrate, for the first time, region- and subunit-specific differences in the expression of GABAAR subunits in the stargazer mouse model of absence epilepsy. The GABAAR subunits a1 and b2, which are preferentially localized in the VP region, are significantly increased in the epileptic stargazer mice (a1: 33%, b2: 96%), whereas the GABAAR subunits a3 and b3, which are mainly present in the RTN region, show no significant difference in expression levels. These results suggest that VP region-specific increases in GABAARs may possibly contribute to generation of hypersynchronous thalamic oscillatory activity in absence seizures in stargazers.
229 Altered GABAARs in Absence Epilepsy A
C
B
D
Figure 3. Relative expression levels of GABAAR subunits in the RTN and VP of nonepileptic (+/+, +/stg) and epileptic (stg/stg) mice, represented as bar graphs. There are statistically significant increases in normalized relative intensity reading for the GABAAR (A) a1 subunit (NE 1.00 0.06, E 1.33 0.12, n = 8, p < 0.05) and the b2 subunit (NE 1.00 0.06, E 1.96 0.15, n = 8, p < 0.01) in the VP region. There was no statistically significant difference in the a3 and b3 subunit levels between nonepileptic and epileptic mice in the RTN region (*p < 0.05, **p < 0.01). Epilepsia ILAE
Absence seizures are known to arise from disruptions within the cortico-thalamocortical network, but the complex cellular and molecular mechanisms underlying pathophysiology of absence seizures are not well understood and appear to be multifactorial.31 Normal spindle-like oscillations in the thalamocortical network are maintained by the sequential firing of inhibitory GABAergic neurons of the RTN and excitatory glutamatergic thalamocortical (TC) relay neurons in VP region, which are reciprocally connected. The RTN inhibitory neurons receive excitatory inputs from collateral projections of corticothalamic (CT) and TC relay neurons, and project feed-forward inhibition to TC relay neurons in the VP. The TC neurons then become hyperpolarized and fire postinhibitory rebound action potentials as a result of de-inactivation of T-type (lowthreshold) calcium channels.32 This rebound potential reexcites RTN neurons, thereby initiating intrathalamic oscillations.31,33,34 Cortical activation at CT-RTN synapses is more powerful than cortical activation at CT-VP synapses, due to the fact that AMPAR GluA4 subunits are expressed far more abundantly at CT-RTN synapses than at CT-TC synapses.35 Because of this intrinsic connectivity between the cortex and the thalamus, any change in balance between excitation and inhibition within the thalamus could potentially swing normal spindle oscillations into pathologic hypersynchronous SWDs.
In the stargazer mouse model of absence epilepsy, the primary deficit is in AMPAR expression and function, as demonstrated by electrophysiological functional studies20 and immunogold ultrastructural studies.21 These studies demonstrated a selective loss in AMPAR expression and function at CT-RTN synapses but not at CT-VP synapses. The AMPAR subunit GluA4, which is the prominent subunit in the RTN, was significantly reduced in the RTN but not the VP in stargazers. A link between absence epileptic seizures and regional loss of GluA4-AMPAR-mediated excitation in the RTN was also demonstrated by studies on the Gria4 KO model, which exhibits spontaneous SWDs as a result of loss of GluA4.7 These researchers demonstrated that loss of GluA4 caused selective weakening of synaptic strength at CT-RTN synapses, but not at CT-TC relay neuron synapses in the VP or at TC relay-RTN synapses. Collectively, these studies show that loss of AMPARs specifically at CT-RTN synapses is linked to the absence epilepsy phenotype in these mutant and KO mice. However, whether these changes in AMPAR expression and function at CT-RTN synapses are directly causative of hypersynchronous SWDs in absence epilepsy is still unknown. One hypothesis that has been put forward is that the loss of AMPARs at CTRTN synapses and the subsequent reduction in RTN synaptic strength could impair feed-forward inhibition, effectively causing disinhibition of relay neurons and thereby Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
230 S. Seo and B. Leitch resulting in an imbalance of excitation and inhibition within the thalamocortical network and thus leading to hypersynchronous oscillations. However, another hypothesis that has been proposed by Lacey et al.,22 suggests increased excitation of these inhibitory neurons may underlie the hypersynchronous SWDs in this model. These researchers reported that although the amplitude of AMPAR-mediated currents was reduced in stargazers, there was a corresponding increase in the amplitude of NMDARmediated currents at CT-RTN synapses. This indicates that a compensatory increase in synaptic NMDARs may occur where AMPARs are lost, leading to an increase in RTN inhibitory neuron excitability and thus increased feed-forward inhibition.22 However, there is also evidence in the literature that the stargazin mutation may affect GABAAR subunit expression. Changes in GABAAR subunit expression have been reported in hippocampal re-entrant axon collaterals in the dentate gyrus23 and in cerebellar granule cells24 of stargazers. Evidence from other animal models also supports the contention that SWDs underlying absence epilepsy are linked to changes in GABAergic inhibition (phasic or tonic) within thalamocortical networks.10,12–14,36 However, the impact and contribution of altered GABAARmediated inhibition in different thalamic subregions to SWD and absence epilepsy is unclear and somewhat controversial. Intra-RTN inhibition (between reciprocally connected inhibitory RTN neurons) is thought to regulate intrathalamic inhibition and prevent hypersynchronous oscillatory activity leading to the generation of seizures. One hypothesis is that recurrent inhibition within the RTN, which occurs through phasic GABAA a3b3c2 receptors,34 becomes impaired in absence seizures. For example, in the genetically epilepsy-prone WAG/Rij rat, a3 subunit expression is almost completely abolished.13 In the b3 knockout mouse, hypersynchronous thalamic oscillations occur and are linked to reduced GABAA-mediated inhibition specifically in the RTN.10 However in the current study on epileptic stargazers, we found no significant difference in tissue levels of GABAAR a3 or b3, subunits using Western blotting analyses (Fig. 2E). This finding is in line with a study by Schofield et al.,37 which showed that a3 knockout mice do not show spontaneous absence seizure activity. In epileptic mice carrying c2 (R43Q) mutation, no difference in inhibitory postsynaptic potentials (IPSPs) in the RTN has been found,38 suggesting that reduced phasic inhibition within RTN is not a requirement for generation of absence seizures. Conversely, in genetic absence epilepsy rat from Strasbourg (GAERS) and dilute brown non-agouti (DBA) /2J mouse models, an increase in phasic inhibition in the RTN region has been reported.12,14,38 Overall, the published data with regard to the involvement of intra-RTN GABAAR-mediated phasic inhibition in seizure generation are confusing and in some cases contradictory. It seems then that any conclusions Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
on the role of intra-RTN inhibition in the generation of absence seizures cannot be drawn at present. In contrast, evidence for increased GABAergic inhibition in the VP thalamocortical relay neurons has been found in several animal models of absence epilepsy. An overall increase in phasic GABAAR-mediated inhibition in VP relay neurons was observed in the c-hydroxybutyrate model of absence epilepsy by Gervasi et al.11 An increase in tonic GABAergic inhibition in the VP relay neurons has also been reported in a number of rodent models of absence epilepsy including the GAERS rat, which could be partially attributed to malfunction of the GABA transporter GAT-1.14 In our study on the stargazer mouse model of absence epilepsy, semiquantitative Western blot analysis revealed significant increases in tissue levels of GABAAR a1 and b2 subunits in the VP region (33% and 96%, respectively; Fig. 2E). These results concur with previous reports on phasic and tonic inhibition in the VP relay neurons,14 which suggest that augmentation of GABAergic signaling in the thalamic relay nuclei may contribute to generation of hypersynchronous thalamocortical activity. It is worth noting that tonic inhibition in VP relay neurons is mediated via GABAARs having a a4b2d subunit composition, whereas phasic inhibition in the VP is mediated by a1b2c2 containing GABAARs.34 Hence, although the increase in the a1 subunit most likely reflects enhanced phasic inhibition through the a1b2c2 GABAARs, the nature of the increase in the b2 subunit expression (tonic, phasic, or both) remains to be elucidated. Given that an increase in both tonic or phasic inhibition may be possible at thalamocortical VP relay neurons, further investigations at the ultrastructural level are required to determine whether these differences in GABAAR subunits expression at the tissue level are also translated into changes in receptor expression at synapses (or at extrasynaptic sites), and whether these changes are associated with an increase in phasic inhibition or both phasic and tonic inhibition. Furthermore, it still remains to be determined exactly how the mutation of the stargazin gene leads to alterations in GABAAR-subunit expression: whether these differences occur preseizure and could thus be causative, or whether they are the result of seizure-induced activity. In summary, results from the current study demonstrate region-specific changes in GABAAR-subunit expression in the thalamus of epileptic mice. Increase in GABAAR-subunit expression in the VP region may represent up-regulation of phasic inhibition, tonic inhibition, or both. This may lead to an imbalance within the intrathalamic circuit and contribute to hypersynchronous thalamocortical activity and absence seizures. Future studies at the ultrastructural level are needed to determine if these global changes reflect region-specific changes in GABAAR expression at synapses. Furthermore, developmental studies will be required to determine whether such changes occur preseizure and
231 Altered GABAARs in Absence Epilepsy have a proabsence effect or are compensatory changes to reduce the propensity of absence seizures. In addition to the possible involvement of AMPARs, NMDARs, and GABAARs, in the generation of absence seizures, potential underlying mechanisms are further complicated by the presence of T-type (low voltage-activated) calcium channels, which upon activation are known to evoke burst firing and give rise to SWDs.39 GABAB receptors in relay neurons have also been implicated in several animal epilepsy models; their activation has been shown to be partially responsible for abnormally large and hypersynchronous rebound bursts and exacerbation of absence seizures.40 The profound complexity of the cellular and molecular mechanisms underlying generation of absence seizures and the range of targets that may be affected may reflect the variable results obtained from currently available AEDs, where even the best monotherapy fails in almost half of all newly diagnosed cases.4 Given the current situation, and the adverse impact that absence seizures have on childhood learning, it is of huge clinical relevance to further investigate the range of mechanisms whereby absence seizures can arise. Elucidating region-specific changes in synaptic GABAAR subunit expression and the synaptic mechanisms that regulate synchronous thalamocortical network activity may be crucial for identification of novel drug targets and the development of more specific and effective therapeutic intervention methods for treatment of specific subtypes of absence epilepsy.
Acknowledgments This work was supported by grants from the University of Otago Research Grants (UORG) and Deans Bequest Fund awarded to B.L. The authors thank the staff of the Otago Centre for Confocal Microscopy (OCCM) for excellent technical support, and the University of Otago for Doctoral Scholarship awarded to Steve Seo.
Disclosure The authors have no conflicts of interest to declare. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Jallon P, Loiseau P, Loiseau J. Newly diagnosed unprovoked epileptic seizures: presentation at diagnosis in CAROLE study. Epilepsia 2001;42:464–475. 2. Pavone P, Bianchini R, Trifiletti RR, et al. Neuropsychological assessment in children with absence epilepsy. Neurology 2001;56:1047–1051. 3. Wheless JW, Clarke DF, Carpenter D. Treatment of pediatric epilepsy: expert opinion. J Child Neurol 2005;20(Suppl. 1):S1–S56. 4. Glauser TA, Cnaan A, Shinnar S, et al. Ethosuximide, valproic acid, and lamotrigine in childhood absence epilepsy. N Engl J Med 2010;362:790–799. 5. McCormick DA, Contreras D. On the cellular and network bases of epileptic seizures. Annu Rev Physiol 2001;63:815–846. 6. Crunelli V, Leresche N. Childhood absence epilepsy: genes, channels, neurons and networks. Nat Rev Neurosci 2002;3:371–382.
7. Paz JT, Bryant AS, Peng K, et al. A new mode of corticothalamic transmission revealed in the Gria4( / ) model of absence epilepsy. Nat Neurosci 2011;14:1167–1173. 8. Wallace RH, Marini C, Petrou S, et al. Mutant GABA(A) receptor c2subunit in childhood absence epilepsy and febrile seizures. Nat Genet 2001;28:49–52. 9. Maljevic S, Krampfl K, Cobilanschi J, et al. A mutation in the GABA (A) receptor a1 subunit is associated with absence epilepsy. Ann Neurol 2006;59:983–987. 10. Huntsman MM, Porcello DM, Homanics GE, et al. Reciprocal inhibitory connections and network synchrony in the mammalian thalamus. Science 1999;283:541–543. 11. Gervasi N, Monnier Z, Vincent P, et al. Pathway-specific activation of gamma-hydroxybutyric acid in sensory thalamus and its relevance to absence seizures. J Neurosci 2003;23:11469–11478. 12. Bessa€ıh T, Bourgeais L, Badiu CI, et al. Nucleus-specific abnormalities of GABAergic synaptic transmission in a genetic model of absence seizures. J Neurophysiol 2006;96:3074–3081. 13. Liu X-B, Coble J, van Luijtelaar G, et al. Reticular nucleus-specific changes in a3 subunit protein at GABA synapses in genetically epilepsy-prone rats. Proc Natl Acad Sci USA 2007;104: 12512–12517. 14. Cope DW, Di Giovanni G, Fyson SJ, et al. Enhanced tonic GABAA inhibition in typical absence epilepsy. Nat Med 2009;15:1392–1399. 15. Noebels JL, Qiao X, Bronson RT, et al. Stargazer: a new neurological mutant on chromosome 15 in the mouse with prolonged cortical seizures. Epilepsy Res 1990;7:129–135. 16. Letts VA, Kang MG, Mahaffey CL, et al. Phenotypic heterogeneity in the stargazin allelic series. Mamm Genome 2003;14:506–513. 17. Letts VA. Stargazer—a mouse to seize. Epilepsy Curr 2005;5:161– 165. 18. Chen L, Chetkovich DM, Petralia RS, et al. Stargazin regulates synaptic targeting of AMPA receptors by two distinct mechanisms. Nature 2000;408:936–943. 19. Tomita S, Adesnik H, Sekiguchi M, et al. Stargazin modulates AMPA receptor gating and trafficking by distinct domains. Nature 2005;435:1052–1058. 20. Menuz K, Nicoll RA. Loss of inhibitory neuron AMPA receptors contributes to ataxia and epilepsy in stargazer mice. J Neurosci 2008;28:10599–10603. 21. Barad Z, Shevtsova O, Arbuthnott GW, et al. Selective loss of AMPA receptors at corticothalamic synapses in the epileptic stargazer mouse. Neuroscience 2012;217:19–31. 22. Lacey CJ, Bryant A, Brill J, et al. Enhanced NMDA receptordependent thalamic excitation and network oscillations in stargazer mice. J Neurosci 2012;32:11067–11081. 23. Payne HL, Donoghue PS, Connelly WM, et al. Aberrant GABA(A) receptor expression in the dentate gyrus of the epileptic mutant mouse stargazer. J Neurosci 2006;26:8600–8608. 24. Payne HL, Connelly WM, Ives JH, et al. GABAA alpha6-containing receptors are selectively compromised in cerebellar granule cells of the ataxic mouse, stargazer. J Biol Chem 2007;282:29130–29143. 25. Seo S, Liu P. Leitch B spatial learning-induced accumulation of agmatine and glutamate at hippocampal CA1 synaptic terminals. Neuroscience 2011;192:28–36. 26. Turner CP, DeBenedetto D, Ware E, et al. Postnatal exposure to MK801 induces selective changes in GAD67 or parvalbumin. Exp Brain Res 2010;201:479–488. 27. Paxinos G, Franklin KB. The mouse brain in stereotaxic coordinates. 2nd ed. San Diego, CA: Academic Press; 2001. 28. Mineff EM, Weinberg RJ. Differential synaptic distribution of AMPA receptor subunits in the ventral posterior and reticular thalamic nuclei of the rat. Neuroscience 2000;101:969–982. 29. Esclapez M, Tillakaratne NJ, Tobin AJ, et al. Comparative localization of mRNAs encoding two forms of glutamic acid decarboxylase with nonradioactive in situ hybridization methods. J Comp Neurol 1993;331:339–362. 30. Pirker S, Schwarzer C, Wieselthaler A, et al. GABA(A) receptors: immunocytochemical distribution of 13 subunits in the adult rat brain. Neuroscience 2000;101:815–850. 31. Beenhakker MP, Huguenard JR. Neurons that fire together also conspire together: is normal sleep circuitry hijacked to generate epilepsy? Neuron 2009;62:612–632. Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
232 S. Seo and B. Leitch 32. Kim U, Sanchez-Vives MV, McCormick DA. Functional dynamics of GABAergic inhibition in the thalamus. Science 1997;278:130–134. 33. Steriade M. Sleep, epilepsy and thalamic reticular inhibitory neurons. Trends Neurosci 2005;28:317–324. 34. Huguenard JR, McCormick DA. Thalamic synchrony and dynamic regulation of global forebrain oscillations. Trends Neurosci 2007;30:350–356. 35. Golshani P, Liu X-B, Jones EG. Differences in quantal amplitude reflect GluR4-subunit number at corticothalamic synapses on two populations of thalamic neurons. Proc Natl Acad Sci USA 2001;98:4172–4177. 36. Tan HO, Reid CA, Chiu C, et al. Increased thalamic inhibition in the absence seizure prone DBA/2J mouse. Epilepsia 2008;49:921–925. 37. Schofield CM, Kleiman-Weiner M, Rudolph U, et al. A gain in GABAA receptor synaptic strength in thalamus reduces oscillatory activity and absence seizures. Proc Natl Acad Sci USA 2009;106:7630–7635. 38. Tan HO, Reid CA, Single FN, et al. Reduced cortical inhibition in a mouse model of familial childhood absence epilepsy. Proc Natl Acad Sci USA 2007;104:17536–17541.
Epilepsia, 55(2):224–232, 2014 doi: 10.1111/epi.12500
39. Tringham E, Powell KL, Cain SM, et al. T-type calcium channel blockers that attenuate thalamic burst firing and suppress absence seizures. Sci Transl Med 2012;4:121ra19. 40. Manning J-PA, Douglas AR, Bowery NG. Pharmacology of absence epilepsy. Trends Pharmacol Sci 2003;24:542–549.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Expression of GluA4 and vGlut1 in the thalamic subregions, verifying the purity of RTN and VP samples used in Western blotting procedures.
