Altered thalamocortical rhythmicity in Cav2.3-deficient mice

Molecular and Cellular Neuroscience 39 (2008) 605–618 Contents lists available at ScienceDirect Molecular and Cellular Neuroscience j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y m c n e Altered thalamocortical rhythmicity in Cav2.3-deficient mice Marco Weiergräber a,c,⁎, Margit Henry a, Matthew S.P. Ho b,c, Henrik Struck a, Jürgen Hescheler a,c, Toni Schneider a,c a b c Institute of Neurophysiology, Faculty of Medicine, University of Cologne, Germany Center for Biochemistry, Faculty of Medicine, University of Cologne, Germany Center for Molecular Medicine Cologne (CMMC), University of Cologne, Germany a r t i c l e i n f o Article history: Received 7 April 2008 Revised 28 July 2008 Accepted 13 August 2008 Available online 4 September 2008 Keywords: Cav2.3 Electroencephalogram Hyperoscillation Thalamocortical circuitry Non-convulsive seizure Absence epilepsy a b s t r a c t Voltage-gated calcium channels (VGCCs) are key regulators of neuronal excitability and important factors in epileptogenesis and neurodegeneration. Recent findings suggest a novel, important proictogenic and proneuroapoptotic role of the Cav2.3 E/R-type VGCCs in convulsive generalized tonic–clonic and hippocampal seizures. Though Cav2.3 is also expressed in key structures of the thalamocortical circuitry, their functional relevance in non-convulsive absence seizure activity remains unknown. To this end, we investigated absence specific spike–wave discharge (SWD) susceptibility in control and Cav2.3-deficient mice by systemic administration of γ-hydroxybutyrolactone (GBL, 70 mg/kg i.p.), followed by electrocorticographic radiotelemetric recordings, behavioral analysis and histomorphological characterization. Based on motoric studies, SWD and power-spectrum density (PSD) analysis, our results demonstrate that Cav2.3−/− mice exhibit increased absence seizure susceptibility and altered absence seizure architecture compared to control animals. This study provides evidence for the first time that Cav2.3 E/R-type Ca2+ channels are important in modulating thalamocortical hyperoscillation exerting anti-epileptogenic effects in non-convulsive absence seizures. © 2008 Elsevier Inc. All rights reserved. Introduction Voltage-gated calcium channels (VGCCs) mediate Ca2+ influx into living cells triggering various cellular processes, such as excitation– contraction coupling (Bers, 2002), excitation–secretion coupling (Yang and Berggren, 2005; Kisilevsky and Zamponi, 2008), neurotransmitter and hormone release (Catterall et al., 2005), and also regulation of gene expression (Bito et al., 1997; Hofmann et al., 1999). In the last decade, VGCCs, in particular Cav2.1 P/Q-type and Cav3.2 T-type channels, were shown to be functionally relevant in the etiopathogenesis of various forms of epilepsies in both humans and animal models exhibiting convulsive and non-convulsive seizure phenotypes (Kullmann, 2002; Turnbull et al., 2005). Recently, another VGCC entity, the Cav2.3 E/R-type channel, was also proven to exert proictogenic and proepileptogenic effects in convulsive seizures. Cellular electrophysiology of hippocampal CA1 neurons revealed that Cav2.3 is capable of triggering plateau potentials with superimposed epileptiform bursting following muscarinergic M1/M3 receptor stimulation (Kuzmiski Abbreviations: GBL, γ-hydroxybutyrolactone; RTN, reticular thalamic nucleus; SWD, spike–wave discharge; VGCC, voltage-gated calcium channel. ⁎ Corresponding author. Institute of Neurophysiology and Center for Molecular Medicine Cologne (CMMC), Robert-Koch-Str. 39, D-50931 Cologne, Germany. Fax: +49 0221 478 6965. E-mail address: akp74@uni-koeln.de (M. Weiergräber). 1044-7431/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.mcn.2008.08.007 et al., 2005; Tai et al., 2006). There are further indications, though controversially discussed, that Cav2.3 is involved in epileptogenic afterdepolarizations (ADH) in hippocampal CA1 neurons as well (Metz et al., 2005; Yue et al., 2005). In addition, some anti-epileptic drugs, e.g. lamotrigine, sipatrigine and topiramate, are efficient Cav2.3 E/Rtype Ca2+ channel blockers, among which topiramate is capable of suppressing Cav2.3 mediated epileptiform bursting in the CA1 hippocampal region at therapeutically relevant plasma concentrations (Kuzmiski et al., 2005). The functional properties of Cav2.3 E/R-type Ca2+ channels in convulsive seizures in vivo have been studied in detail recently by gene inactivation of Cacna1E, expressing Cav2.3 (Pereverzev et al., 2002). Electroencephalographic analysis of Cav2.3-deficient mice revealed no spontaneous epileptiform discharges indicative of convulsive seizure activity. Instead, ablation of Cav2.3 resulted in reduced seizure susceptibility to generalized tonic–clonic seizures provoked by pentylenetetrazol (PTZ) and to hippocampal seizures induced by kainic acid (KA) or N-methyl-D-aspartate (NMDA) (Weiergraber et al., 2006a; Weiergraber et al., 2006b; Weiergraber et al., 2007). Intriguingly, Cav2.3 deficiency provides additional neuroprotective effects by suppressing hyperexcitability and excitotoxicity in the CNS (Weiergraber et al., 2007). Furthermore, mutation analysis of EFHC1, a Cav2.3 interaction partner supports that Cav2.3 may mediate epileptogenesis and neurodegeneration (Suzuki et al., 2004). In addition, the molecular chaperone hsp70 known to be involved in neurodegenera- 606 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 Fig. 1. Immunodetection of GABAergic interneurons in the thalamocortical circuitry. (A–H) Coronal sections (Bregma: − 0.1 to −1.0 mm) from control (left) and Cav2.3−/− mice (right) were stained for parvalbumin (1:4000) using indirect immunofluorescent labeling revealing typical staining of the reticular thalamic nucleus (A–D). No obvious structural differences in RTN histomorphology between both genotypes were observed (arrows indicate round “R” neurons; arrowheads point at large fusiform “F” cells; small fusiform cells “f” are rarely present in the RTN). Other than the thalamic region, parvalbumin-positive cells were also detected in the neocortex (E–H). Similar to the RTN, immunofluorescent labeling of cortical GABAergic neurons does not suggest structural alterations in both regions expressing the Cav2.3 E/R-type VGCC. Bar (A, B, E, F): 200 μm; bar (C, D, G, H): 50 μm. tion functionally interacts with the II–III loop of the Cav2.3 E/R-type VGCC and contributes to the protein kinase C mediated effects on this channel (Krieger et al., 2006). Finally, divalent trace metals target Cav2.3 Ca2+ channels (Mathie et al., 2006; Sun et al., 2007) capable of modulating hippocampal seizure susceptibility and neurotoxicity (Dominguez et al., 2003; Takeda et al., 2003a; Takeda et al., 2003b; Takeda et al., 2005). Although the functional involvement of Cav2.3 E/R-type Ca2+ channels in convulsive seizures has been elicited lately, their functional implications in generalized non-convulsive seizures of the absence type are still largely unknown to date. Absence seizures are characterized behaviorally by a paroxysmal loss of consciousness that is normally accompanied with bilateral synchronous spike–wave discharge (SWD) activity, of which the frequency is species specific (Manning et al., 2003). The pathophysiological substrate of absence seizure activity is aberrant hyperoscillation in the underlying thalamocortical–corticothalamic network. M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 Within this circuitry, glutamatergic thalamic relay cells innervate cortical pyramidal neurons which finally resynapse onto the ventrobasal thalamic region. Both structures also exhibit glutamatergic input onto reticular thalamic nucleus (RTN) neurons with the latter providing GABAergic projections not only onto the RTN neurons themselves, i.e. lateral inhibition, but also onto thalamic relay cells (Danober et al., 1998; Khosravani and Zamponi, 2006). Both relay neurons of the ventrobasal thalamus and RTN neurons have the intriguing capability to shift between two functional modes, the tonic and burst firing mode, which strongly regulates transmission of external information to the cortex (Blumenfeld and McCormick, 2000). Oscillatory behavior within the thalamocortical circuitry is substantially driven by the RTN which serves as a key modulator of information transfer between thalamus and cortex. It is noteworthy that various extrathalamocortical structures capable of modulating thalamocortical rhythmicity and absence SWD generation project to that circuitry (Danober et al., 1998; Lakaye et al., 2002). Most studies on absence epileptogenesis in humans and animal models in the past predominantly focused on T- and P/Q-type VGCCs. The Cav3.1 VGCC knock-out mouse model lacks rebound burst firing in thalamocortical relay neurons, thus displaying resistance to absence seizures (Kim et al., 2001) and altered sleep architecture (Lee et al., 2004). In addition, Genetic Absence Epilepsy Rats from Strasbourg (GAERS) exhibit increased T-type Ca2+-current in RTN neurons (Tsakiridou et al., 1995) and also alterations in Cav3.1 and Cav3.2 VGCC expression in the adult ventroposterior thalamic nuclei and juvenile RTN neurons, respectively (Talley et al., 2000). Finally, the gene encoding Cav3.2 T-type Ca2+ channel (CACNA1H) was identified as a susceptibility locus of absence epilepsy in humans with gain-of-function mutations triggering thalamocortical hyperoscillation and absence epilepsy (Khosravani et al., 2004; Khosravani et al., 2005; Shin, 2006; Arias-Olguin et al., 2008). Cav2.1−/− mice and various Cav2.1 mouse mutants, e.g. tottering, tottering leaner and rocker, are also prone to absence epilepsy, the mechanism of which still remains poorly understood (Jun et al., 1999; Kullmann, 2002). However, within the thalamocortical circuitry, T- and P/Q-type VGCCs are not the only players. GABAergic interneurons of the RTN and cortex as well as extrathalamocortical structures were clearly shown to express Cav2.3 E/R-type Ca2+ channels (De Borman et al., 1999; Talley et al., 2000; van de Bovenkamp-Janssen et al., 2004; Weiergraber et al., 2006b). Interestingly, de Borman et al. (1999) and Lakaye et al. (2002) detected a significant reduction of Cav2.3 transcript levels in both cerebellum and medulla of GAERS, pointing to a functional involvement of Cav2.3 VGCCs in absence epileptogenesis. Furthermore, the development of SWDs in Wistar Albino Glaxo (WAG/Rij) rats, another model of absence epilepsy, is accompanied with an increased Cav2.1 expression in the reticular thalamic nucleus but reduced Cav2.3 expression in the RTN at the age of absence seizure onset (van de Bovenkamp-Janssen et al., 2004). These observations suggest a functional role of Cav2.3 in the etiopathogenesis of absence epilepsy. In the present study we provide evidence that activation of Cav2.3 can exert anti-absence effects, probably by facilitating the tonic mode of action, and thus being involved in the regulation of thalamocortical rhythmicity and hyperoscillation. These findings provide novel perspectives for Cav2.3 as a potent target in the treatment of absence epilepsy and sleep related disorders. 607 morphology in Cav2.3-deficient mouse brains suggesting that ablation of Cav2.3 VGCCs does not disrupt thalamocortical formation in the murine brain (Weiergraber et al., 2006a). Since immunohistochemistry and in situ hybridization has revealed that Cav2.3 E/R-type VGCCs are expressed in GABAergic interneurons within the RTN and the cortex (Talley et al., 2000; Weiergraber et al., 2006a), we looked into any possible subtle alteration on the cytoarchitectural level in Cav2.3−/− mouse brains compared to controls. One of the molecular markers for GABAergic interneurons is parvalbumin, a calcium-binding protein. Immunofluorescent staining on brain sections of both Cav2.3+/+ and Cav2.3−/− mice displayed a similar staining pattern, with GABAergic RTN interneurons exhibiting intensively positive immunoreactivity (Figs. 1 A–D). The RTN generally appears as a thin layer interposed between the internal capsule and the external medullary lamina, crossed over by bundles of the thalamocortical–corticothalamic fibers, which becomes particularly obvious by using Kluver–Barrera staining (see Supplementary data). The presence of these bundles results in fragmentation of the nucleus into numerous mediolaterally oriented strips of cell bodies. The RTN can be generally divided into three parts, the dorsal (see also Figs. 1A, B), the ventral and the lateral part, exhibiting together a clear shell-like morphology. Parvalbumin staining of the RTN from Cav2.3+/+ (Figs. 1A, C) and Cav2.3−/− mice (Figs. 1B, D) clearly exhibited that different reticular thalamic cell types can be distinguished (Spreafico et al., 1991; Battaglia et al., 1994; Nagaeva and Akhmadeev, 2006): 1. round neurons (“R”) that make up about 37% of RTN neurons predominantly localized in the rostral pole of the RTN harboring 4–8 principal dendrites; 2. large fusiform neurons (F) contributing to about 49% to the RTN cell mass being equally represented throughout the nucleus with 2–4 dendrites from two poles) and finally small fusiform neurons (f) contributing app. 14% of the RTN cell mass mainly distributed in the medial and lateral border of the RTN with two polar dendrites (Figs. 1C, D). The parvalbumin antibody provided specific and intense staining not only of the soma but also the axonal and dendritic processes (Figs. 1C, D). Similarly, intense parvalbumin immunoreactivity in GABAergic interneurons was observed in the neocortex of Cav2.3+/+ and Cav2.3−/− mice (Figs.1E, F). These neurons are also known to express Cav2.3 forming part of the thalamocortical circuitry. In summary, detailed investigation of parvalbumin stained sections from both controls and Cav2.3−/− brains did not show alterations in cell-type specific composition of the RTN or structural connectivity between the GABAergic interneurons. To directly prove that Cav2.3 is coexpressed with parvalbumin in GABAergic interneurons within the RTN and the cortex, double immunofluorescent labeling of the brain section from control and Cav2.3−/− mice was performed. However, two different antibodies against Cav2.3 (see Experimental methods) exhibited non-specific staining throughout the whole mouse brain sections of both genotypes. In stark contrast, the anti-Cav2.3 195A antibody (Grabsch et al., 1999; Weiergraber et al., 2000) exhibited a distinct staining pattern in the RTN on rat brain sections (Figs. 2AII,V–CII,V). These Cav2.3 positively stained cells are in fact GABAergic as they were coimmunostained for parvalbumin (Fig. 2AIII,VI). In contrast, there were low numbers of calretinin positive neurons in the rat RTN (Fig. 2BI,IV) that were only partially Cav2.3 positive (Fig. 2BIII,VI). No immunoreactivity of calbindin was seen in the rat RTN (Fig. 2CI,IV). Cav2.3-deficient mice do not exhibit spontaneous SWDs indicative of absence seizure activity Results Histomorphological characterization of Cav2.3 positive neurons in the reticular thalamic nucleus and neocortex from control and Cav2.3-deficient mice Previously, serial sections of whole brains stained by both Nissland Klüver–Barrera revealed no apparent alteration in gross histo- We previously reported that both video analysis and radiotelemetric electrocorticographic and deep, intracerebral electroencephalographic recordings from Cav 2.3 −/− mice did not exhibit spontaneous epileptiform behavior or ictal discharges indicative of convulsive seizure activity (Weiergraber et al., 2006a; Weiergraber et al., 2006b). Furthermore, a detailed qualitative and quantitative analysis of long-term ECoG recordings from various cortical regions, 608 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 609 Fig. 2. Immunolocalization of Cav2.3 E/R-type calcium channels in the rat RTN and coexpression with calcium-binding proteins. As Cav2.3 antibodies failed to exhibit specific staining on murine brain sections, we performed immunofluorescent double staining for the calcium-binding proteins parvalbumin (A), calretinin (B) and calbindin (C) together with Cav2.3 Ca2+ channels on rat brain sections using the anti-Cav2.3 195A antibody. Parvalbumin staining (AI,IV) displays the characteristic shell-shaped structure of the RTN localized at the lateral edge of the thalamic region. The number of calretinin positive RTN neurons is considerably less (BI,IV) and there was hardly any immunopositive reaction for calbindin (CI,IV). Immunolocalization of Cav2.3 E/R-type Ca2+ channels (AII,V–CII,V) revealed almost complete staining of the RTN with double staining exhibiting strong colocalization of Cav2.3 Ca2+ channels with parvalbumin (AIII, VI) but to a much lesser extent with calretinin (BIII,VI). Bar (AI–III–CI–III): 200 μm; bar (AIV–VI–CIV–VI): 75 μm. such as the motor cortex (M1/M2) or the somatosensory cortex (Fig. 3Ac1,c2, Fig. 4Ac1,c2) clearly demonstrates that Cav2.3-deficient mice do not spontaneously exhibit SWDs indicative of non-convulsive absence seizure activity. Increased SWD activity in Cav2.3-deficient mice following γ-hydroxybutyrolactone (GBL) administration To investigate the functional relevance of the Cav2.3 E/R-type Ca2+ channel in SWD generation in vivo, we systemically administered γhydroxybutyrolactone (GBL), a prodrug of γ-hydroxybutyric acid (GHB), into control mice (27.86 ± 1.78 g, 16.36 ± 1.24 weeks, n = 8, all ♂) and Cav2.3-deficient animals (32.16 ± 0.99 g, 15.09 ± 1.05 weeks, n = 8, all ♂). Both genotypes were given at least 7 days after implantation to allow for full recovery (Cav2.3+/+: 13.75 ± 1.33 days; Cav2.3−/−: 12.00 ± 0.82 days). GBL is a moderate GABAB-receptor agonist but exerts also strong agonistic effects on the newly characterized GBLreceptors (Andriamampandry et al., 2007) probably via G-protein coupled pathways (Snead, 2000). It is known that GABAB-receptor agonists exacerbate absence seizures, whereas GABAB-receptor antagonists suppress them (Snead, 1992; Hosford et al., 1992; Smith and Fisher, 1996). Thus, GBL injection results in highly organized bilaterally synchronous SWD activity typically associated with behavioral phenomena, such as facial myoclonus, vibrissal twitching and most importantly, motoric arrest (Snead et al., 2000). In controls, administration of GBL at 70 mg/kg results in typical paroxysmal SWD activity (Figs. 3A, B3) predominantly in the frequency range of 2–5 Hz. As depicted in Fig. 4A, Cav2.3−/− mice are capable of exhibiting SWD activity at earlier stage following GBL injection compared to control animals (see also Fig. 5A). Statistical analysis revealed that ictal SWD latency was significantly shorter in Cav2.3−/− mice compared to controls (Cav2.3−/−: 149.8 ± 20.4 s, n = 8 versus Cav2.3+/+: 723.5 ± 136.3 s, n = 8, p b 0.001; Fig. 5A). Furthermore, the total number of SWD episodes within the entire one-hour observation period was increased in Cav2.3−/− mice (22.13 ± 1.01, n = 8) versus controls (7.00 ± 0.46, n = 8, p b 0.001, Fig. 5B) and the same trend held for the total duration of SWD epochs (Cav2.3+/+: 1179.4 ± 122.3 s; n = 8 versus Cav2.3−/−: 2264.4 ± 188.8 s; n = 8; p b 0.001, Fig. 5D). On the contrary, the average duration of each SWD events was reduced in Cav2.3-deficient animals (105.1 ± 12.2 s; n = 8) compared to controls (173.7 ± 22.9 s; n = 8 p = 0.019, Fig. 5C). Intriguingly, these results demonstrate that systemic ablation of the Cav2.3 E/R-type Ca2+ channel results in increased absence seizure susceptibility and altered absence seizure architecture unravelled by the modified electroencephalographic SWD parameters. γ-hydroxybutyrolactone (GBL) induced motoric arrest in Cav2.3−/− mice and controls Gamma-hydroxybutyrolactone, which is metabolized to GHB, is known to cause a typical biphasic motoric phenotype characterized by initial cessation of activity accompanied with dominant absence seizure activity which is then followed by irregular and increasing locomotion again. We therefore investigated the motoric activity 610 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 Fig. 3. Spike–wave discharges induced by γ-hydroxybutyrolactone (GBL) in control and Cav2.3−/− mice (Fig. 4). (A) Representative 1 min radiotelemetric ECoG recordings before (c1, c2) and after the administration of GBL (70 mg/kg i.p.) at the time points indicated. Horizontal bars (labeled 1–3) represent EEG segments that are displayed with an expanded time scale in B1–3. In Cav2.3+/+ (B3) and Cav2.3−/− (Fig. 4B1–3) typical bilaterally synchronous SWD activity becomes apparent. (C) Power-spectrum density analysis of a control segment (Ac2) and EEG epoch 5 min after GBL injection exhibits typical theta- and delta-wave activity due to SWDs in Cav2.3−/− mice (Fig. 4C). Note that latency till first occurrence of absence seizure activity is reduced in Cav2.3-deficient animals (Fig. 4A3', see also Fig. 5A). (activity index) of controls and Cav2.3-deficient mice on a horizontal plane using the implantable radiotelemetry. Although both genotypes exhibit the characteristic biphasic locomotion profile, there are marked differences in their initial hypoactive segment (Fig. 6A). Whereas control animals displayed only a moderate reduction in motoric activity, administration of GBL in Cav2.3−/− mice results in a total cessation of activity for at least 20–25 min following injection (Fig. 6A). Detailed analysis demonstrated that the mean activity index is reduced in Cav2.3−/− mice compared to controls approaching a level of significance (2.46 ± 0.66 rel. units, n = 8 versus 6.90 ± 2.08 rel. units, M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 611 Fig. 4. Spike–wave discharges induced by γ-hydroxybutyrolactone (GBL) in control (Fig. 3) and Cav2.3−/− mice. (A) Representative 1 min radiotelemetric ECoG recordings before (c1, c2) and after the administration of GBL (70 mg/kg i.p.) at the time points indicated. Horizontal bars (labeled 1–3) represent EEG segments that are displayed with an expanded time scale in B1–3. In Cav2.3+/+ (Fig. 3B3) and Cav2.3−/− (B1–3) typical bilaterally synchronous SWD activity becomes apparent. (C) Power-spectrum density analysis of a control segment (Ac2) and EEG epoch 5 min after GBL injection exhibits typical theta- and delta-wave activity due to SWDs in Cav2.3−/− mice (C). Note that latency till first occurrence of absence seizure activity is reduced in Cav2.3-deficient animals (A3', see also Fig. 5A). n = 8, p = 0.061, Fig. 6C). Specifically, the time of total inactivation is significantly increased in Cav2.3-deficient animals (46.13 ± 2.81 min, n = 8 versus 25.00 ± 2.43 min, n = 8, p b 0.001, Fig. 6B). In addition, GBL- induced hypolocomotion was typically accompanied by transient hypothermia which was also measured by radiotelemetry (data not shown). These results clearly illustrate that the increased SWD activity 612 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 Fig. 5. Quantitative SWD analysis in controls and Cav2.3−/− mice following GBL administration. Latency till first occurrence of SWD is shortened in Cav2.3−/− mice (A). Furthermore, the total number of SWD episodes as well as the total SWD duration is increased in Cav2.3-deficient mice compared to controls (B, D). However, the average SWD duration in transgenic mice was reduced (C), indicating increased SWD susceptibility and altered SWD architecture in Cav2.3−/− mice. observed in Cav2.3−/− mice coincides with enhanced motoric arrest typical of GBL-induced absence seizure activity. Power-spectrum density analysis of GBL-induced absence seizure activity in Cav2.3−/− mice and controls In addition to the analysis of SWD activity by visual inspection of the EEG and quantification of indirect behavioral phenomena, e.g. motoric activity, we used a third independent mathematical approach based on power-spectrum density (PSD) analysis to further validate increased bilaterally synchronous SWD activity in Cav2.3deficient mice. In the primary analysis, we performed both continuous and discontinuous PSD analysis of 30 s ECoG segments before and after GBL administration to ascertain that GBL-induced SWD activity is embodied in the theta- and delta-wave frequency range (not shown, but see also Fig. 4C, Figs. 7A, B). Based on these findings, we performed a three-dimensional PSD plot for the total one-hour observation period following GBL injection for the different EEG frequency ranges, δ, θ, α, β and γ. Most importantly, the theta and particularly the delta frequency range known to represent SWDs typical of absence seizure activity are strongly increased in Cav2.3−/− mice compared to controls, particularly within the first 25 min after GBL administration (Figs. 7A, B). During this early period, the enhanced ictal theta- and delta-wave activity in Cav2.3−/− mice was again accompanied by complete cessation of locomotive activity in these animals, whereas there was only moderate theta and delta PSD increase associated with minor hypolocomotion in the control group (Figs. 6A, B). These results were further supported by the analysis of PSD peak latencies of the individual frequency bands (Fig. 7C). In agreement M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 613 Fig. 6. Hypolocomotion in control and Cav2.3−/− mice following GBL administration. (A) Continuous plot of the activity index (representing movement on the horizontal plane) for one-minute episodes averaged for control and Cav2.3−/− mice. Cav2.3−/− mice display a complete cessation of activity within the first 20–25 min after GBL administration, which is only moderate in control mice. Whereas the mean activity does not differ significantly between both genotypes (C), the time of complete inactivity is markedly increased in Cav2.3 deficient animals (B). with three-dimensional PSD analysis (Figs. 7A, B), the delta EEG frequency band reaches its PSD peak much earlier in Cav2.3−/− mice compared to control animals (delta, 0–4 Hz: 10.50 ± 0.82 min in Cav2.3−/− mice, n = 8 versus 16.25 ± 1.94 min in Cav2.3+/+, n = 8, p = 0.016) and the theta band (4–8 Hz) approaches a level of significance (13.00 ± 4.28 min in Cav2.3−/− mice, n = 8 versus 24.75 ± 4.47 min in Cav2.3+/+, n = 8, p = 0.078, Fig. 7C). For the other frequency ranges, alpha (8–12 Hz), beta (12–32 Hz) and gamma (32 – 50 Hz), there was no difference in PSD activity between both genotypes (Figs. 7A, B). In addition, the peak latency for alpha and beta PSD activity remained unchanged. The significant reduction in gamma PSD peak latency in control animals compared to Cav2.3−/− mice is 614 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 based on fluctuating low level PSD values (see also Figs. 7A, B) of which the biological relevance remains to be determined (18.25 ± 8.02 min in Cav2.3+/+, n = 8 versus 51.25 ± 3.18 min in Cav2.3−/−, n = 8, p = 0.002, Fig. 7C). Frequency distribution of GBL-induced thalamocortical hyperoscillation in control and Cav2.3-deficient mice In order to investigate the frequency distribution of thalamocortical hyperoscillation, we performed a continuous PSD analysis of 5 s SWD epochs ranging from 0 to 12 Hz in both controls and Cav2.3−/− mice. Figs. 8A and B illustrates the spectral distribution with PSD normalized to peak values. Both plots display that GBL-induced SWDs characteristic of absence seizure activity predominate in the theta and particularly the delta frequency range. Furthermore, the SWD peak frequency is slightly reduced in Cav2.3−/− mice compared to controls (2.79 ± 0.23 Hz in Cav2.3−/−, n = 8 versus 3.34 ± 0.24 Hz in Cav2.3+/+, n = 8), however, not reaching the level of significance (p = 0.128, Fig. 8C). Previous studies in Cav3.1−/− and control mice have revealed similar frequencies of 3–4 Hz of paroxysmal SWDs in epidural EEG recordings (Kim et al., 2001). Discussion The thalamocortical circuitry is part of a complex neural system that is involved in the control of different stages of vigilance. Within this circuitry thalamic relay cells project to cortical pyramidal neurons that finally reproject to the ventrobasal thalamus. An important structural component in this circuitry is the RTN formation, which is composed of GABAergic interneurons receiving not only collateral glutamatergic projections from cortical pyramidal and thalamic relay cells but also GABAergic projections from other RTN neurons (Manning et al., 2003; Khosravani and Zamponi, 2006). Other than GABA, somatostatin, acetylcholine, CCK and serotonin are also found in RTN terminals (Nagaeva and Akhmadeev, 2006). An intriguing feature of these neurons is that they can exhibit two different modes of action characteristic of different stages of vigilance. At high vigilance, deeper brain structures, e.g. the reticular formation exerts excitatory input on thalamic relay cells and inhibitory effects on RTN neurons. The cells slightly depolarize resulting in the so-called tonic mode of action. Under these conditions, spiking frequency and pattern codes for the information perceived in the periphery which is finally processed to the cortex via ventrobasal thalamic relay cells (Llinas and Steriade, 2006). The ECoG correlate of this tonic behavior is a typical low-amplitude, high-frequency pattern as observed in the beta- and gamma-wave band. With decreasing activity from deeper activating brain structures the thalamic neurons re- and hyperpolarize, passing the intermediate state finally exhibiting the burst mode of action. At hyperpolarizing membrane potentials, low-voltage-activated T-type Ca2+ channels deinactivate, a process known as repriming which allows these channels to be activated upon small depolarization steps due to hyperpolarization and cyclic-nucleotide gated channels (HCN2, HCN4). This results in the generation of low-threshold calcium spikes (LTCSs) superimposed with sodium bursting (Llinas and Steriade, 2006). This rebound burst firing mode in ventrobasal thalamic cells and RTN neurons is a major cellular electrophysiological phenomenon typical of low stages of vigilance, e.g. during slow-wave sleep (Shin, 2006). Subsequent oscillatory thalamocortical activity triggers characteristic low-frequency, high-amplitude theta- and delta-wave EEG Fig. 7. Power-spectrum density analysis of GBL-induced SWD activity in control and Cav2.3-deficient mice. Three-dimensional plot of averaged absolute PSD (mV2/Hz) from controls (A) and Cav2.3−/− mice (B). PSD was calculated for 2 min ECoG epochs. Both genotypes predominately differ in theta- and delta-wave activity representing ictal SWDs. The steep increase in theta and particularly the delta PSD within the first 20 min after GBL administration in Cav2.3−/− mice coincides with enhanced hypolocomotion up to complete motoric cessation in this genotype (Fig. 6A). (C) Power-spectrum density peak latency of different EEG frequency bands following GBL administration. The mean latencies till the occurrence of peak PSD for the individual frequency ranges were calculated by analyzing 2 min ECoG epochs. The ictal delta power-spectrum density peak exhibits reduced latency in Cav2.3−/− mice compared to controls as depicted in (A) and (B). M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 Fig. 8. Frequency distribution of SWD activity in controls and Cav2.3−/− mice. Three representative 30 s SWD segments from each control and Cav2.3-deficient mouse were analyzed by continuous PSD analysis and plotted as mean normalized PSD distribution in a frequency range from 0–12 Hz (A, B). GBL-induced SWD activity is predominant in the upper delta to lower theta wave range (3–6 Hz) as reported previously. Though slightly reduced, thalamocortical peak oscillation frequency during absence seizure activity does not differ significantly from control mice. activity. Enhanced rebound burst firing of thalamic relay neurons and RTN cells has been clearly implicated to play a crucial role in the etiopathogenesis of absence epilepsy (Shin, 2006). Within the last decade VGCCs have gained major relevance in the etiopathogenesis of absence epilepsy due to their unique electrophysiological properties and cellular distribution (Weiergraber et al., 2006b). The Cav2.3 E/R-type VGCC is expressed in GABAergic interneurons in the cortex and also in the RTN where Cav3.2 and Cav3.3 T-type Ca2+ channels can also be found (Fig. 1; Talley et al., 2000; Weiergraber et al., 2006a). In contrast, thalamic relay cells do not express considerable Cav2.3 amounts, whereas Cav3.1 displays high transcript levels (Talley et al., 1999; Talley et al., 2000). In 615 accordance with its selective expression in the thalamic relay nucleus, Kim et al. (2001) and Lee at al. (2004) demonstrated that Cav3.1 null mutants are resistant to absence seizures induced by GBL or baclofen, exhibiting strong reduction of slow-wave sleep and alteration of sleep architecture. In addition, crossbreeding Cav3.1−/− mice with various Cav2.1 mouse mutants (e.g. tottering, tottering leaner, rocker, rolling Nagoya, lethargic) or Cav2.1−/− mice that display spontaneous absence seizure activity resulted in offspring that were either free from absence seizures or displayed significant reduction in SWD activity (Song et al., 2004). These findings demonstrate that the Cav3.1 T-type Ca2+ channel is of functional relevance in thalamocortical rhythmicity and absence epileptogenesis by contributing to low-threshold Ca2+ spikes and rebound bursting in ventrobasal thalamic neurons. Most interestingly, reticular thalamic neurons were also reported to generate spontaneous oscillations, such as rhythmic spike-burst activities (Von Krosigk et al., 1993; Kim et al., 1995) which exert severe inhibitory input to ventrobasal relay neurons serving as a major driving force for rebound burst firing due to hyperpolarization. Thus, SWD activity in absence epilepsy can originate from RTN neurons. Although gene ablation studies on Cav3.2 and Cav3.3 T-type Ca2+ channel involvement in absence epilepsy in mice have not been published yet, preliminary data suggest a significant contribution of Cav3.2 to the generation of SWD induced by baclofen (Shin, 2006). In addition, several Cav3.2 gain-of-function mutations have been reported in humans associated with childhood absence epilepsy (CAE) (Khosravani et al., 2004; Khosravani et al., 2005; Shin, 2006). Therefore, low-voltage activated T-type Ca2+-current enhancement in both thalamic relay cells and RTN neurons is involved in SWD generation and thus absence epileptogenesis. Though expressed in the RTN, the functional implications of Cav2.3 in thalamocortical rhythmicity still remain unknown. In this study we analyzed Cav2.3-deficient mice with respect to their propensity to generate SWDs using a well-established pharmacological GBL model of absence seizure activity. Our results provide the first and conclusive evidence for a critical role of Cav2.3 E/R-type Ca2+ channels in the generation of and prevention from absence seizures and thalamocortical hyperoscillation. We demonstrated that Cav2.3 ablation results in increased absence seizure susceptibility and altered absence seizure architecture. The increased SWDs activity in Cav2.3−/− mice compared to controls was further shown to coincide with enhanced motoric arrest in Cav2.3-deficient animals, which is typical of simple absence epilepsy not only in humans but also in rodents. There is a general agreement that the Cav2.3 non-Ltype Ca2+ channel exhibits mid- to high-voltage activated behavior (Catterall et al., 2005; Kisilevsky and Zamponi, 2008). Thus, activation of Cav2.3 Ca2+ channels would further depolarize RTN neurons facilitating the tonic firing mode and preventing the cells from exhibiting rebound burst firing (Fig. 9A). This antihyperoscillatory and SWD activity suppressing effect of high-voltage activated Ca2+ channel activation has been reported for GAERS, a rat model of absence epilepsy, in which SWDs are effectively diminished by BayK8644 administration, an L-type Ca2+ channel agonist or increased by L-type Ca2+ channel blockers, e.g. dihydropyridines (van Luijtelaar et al., 2000). As normal thalamocortical rhythmicity is based on a functional equilibrium of low- and high-voltage activated Ca2+current in RTN neurons, ablation of Cav2.3 Ca2+ channels is likely to result in a functional overbalance of the low-voltage activated T-type Ca2+ channel fraction and would thus facilitate rebound burst firing as it is unmasked in Cav2.3-deficient mice following GBL administration (Fig. 9B). Furthermore, our findings clearly parallel data obtained by van de Bovenkamp-Janssen et al. (2004) in WAG/Rij rats, illustrating that occurrence of SWDs in this rat model of absence epilepsy is accompanied by a lack of expression of the Cav2.3 E/Rtype VGCC in the RTN at the age of absence seizure onset. In addition, de Borman et al. (1999) and Lakaye et al. (2002) detected a significant reduction of Cav2.3 transcript levels in both cerebellum 616 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 βlh/lh and Cav2.1tg/tg mice thalami (Caddick et al., 1999), or enhanced 4 GABAB-receptor expression (Hosford et al., 1992), could also result in relatively enhanced GABAergic input and thus facilitate burst firing. Some anti-epileptics such as lamotrigine exert inhibitory actions on Cav2.3 VGCCs (Hainsworth et al., 2003) and were proven to effectively suppress SWDs in both GAERS and WAG/Rij rats (van Luijtelaar et al., 2002; Manning et al., 2003) and thalamocortical burst complexes in rat brain slices (Gibbs et al., 2002). However, lamotrigine strongly inhibits T-type Ca2+ channels (Hainsworth et al., 2003) thus its anti-absence effect is likely due to its predominant inhibitory effect on low-voltage-activated T-type Ca2+ channels. In summary, Cav2.3 E/R-type VGCCs exhibit a Janus-faced behavior in epileptogenesis. Whereas previous studies have demonstrated that the Cav2.3 VGCC is proepileptogenic in convulsive seizures (Weiergraber et al., 2006a; Weiergraber et al., 2007), such as generalized tonic–clonic and hippocampal seizures, it clearly exhibits antiepileptogenic capacity in typical non-convulsive absence epilepsy. These findings illustrate that the channel per se is neither pro- nor anti-epileptogenic. Indeed, it is the functional and structural integration within specific neuronal circuitries that endows its pathophysiological implications. Finally, our results shed new light on Cav2.3 as a possible therapeutic target in absence epilepsy treatment and the involvement of this ion channel in the physiology of slow-wave sleep. Experimental methods Study animals Fig. 9. Voltage-gated Ca2+ channels as functional regulators of thalamocortical burst activity. Depending on the depolarizing or hyperpolarizing input, thalamic relay cells and RTN neurons can switch between different modes, the tonic, intermediate and burst mode of action. Activation of high-voltage activated Ca2+ channels is supposed to result in further depolarization of the cell, thus facilitating the tonic mode of action. In contrast, enhanced low-voltage-activated T-type Ca2+-currents or reduced high-voltage activated Ca2+ channel activity facilitates the burst mode of action. As Cav2.3 E/R-type channels exhibit mid- to high-voltage activated properties (A), ablation of this channel results in functional overbalance of the T-type Ca2+ channel population which would subsequently favor thalamocortical burst activity in Cav2.3−/− mice (B). As Cav2.3 is also expressed in extrathalamocortical structures, modulatory influence on RTN membrane potential cannot be excluded. and medulla of GAERS, pointing to a functional involvement of Cav2.3 in absence epileptogenesis and further stressing the role of Cav2.3 positive extrathalamocortical structures projecting to the thalamocortical circuitry. It is conceivable that the lack of Cav2.3 in the neocortex may also contribute to the increased absence seizure susceptibility in Cav2.3−/− mice. The neocortex has been well documented to be involved in the generation of SWDs. Interestingly, athalamic animals are also capable of exhibiting SWDs in the cortex following bicuculline administration, indicating an important role of the cortex in the genesis of SWDs (Steriade and Contreras, 1998). This might be of particular relevance as Cav2.3 is also expressed in GABAergic cortical interneurons (Rhee et al., 1999; Timmermann et al., 2002). However, GBL-induced absence seizures are predominantly thalamus-dependent due to GABAB and GBL receptor agonistic effects (Steriade and Contreras, 1998; Seidenbecher et al., 1998; Manning et al., 2004; Shin, 2006). Finally, it cannot be excluded that there may be compensatory changes in other VGCC expression in thalamic relay neurons or RTN cells. However, previous investigation of VGCC transcript levels isolated from total thalamic preparation did not reveal significant differences between controls and Cav2.3-deficient animals (Weiergraber et al., 2006a). Furthermore, other mechanisms, such as reduced excitatory but normal inhibitory synaptic transmission as reported in Cav2.3-deficient mice backcrossed into C57Bl/6 have been generated and described previously (Pereverzev et al., 2002; Weiergraber et al., 2006a). Male Cav2.3-deficient animals and male control mice (with identical genetic C57Bl/6 background) were used in this study. Mice were housed in Makrolon cages type II and maintained at a conventional 12-h light/dark cycle with food and water available ad libitum. All animal experimentation was approved by the local institutional committee on animal care. Antibodies Mouse monoclonal parvalbumin and calbindin antibodies (both 1: 4000) were purchased from Sigma (Germany), the monoclonal murine calretinin antibody (1:15) was obtained from Acris Antibodies (Germany). Two different sources of antibodies against Cav2.3 were used to detect for the channel on both murine and rat brain sections, namely, the polyclonal anti-Cav2.3 195A antibody (1:100) raised in rabbit with the target epitope being localized in pore-loop (IS5-6) of domaine I (Grabsch et al., 1999; Weiergraber et al., 2000), and a commercially available Cav2.3 antibody (1:50) raised in rabbits and directed against the II–III loop of the channel (Alomone Labs, Israel). Immunofluorescence and immunohistochemistry Brains from Cav2.3+/+ and Cav2.3−/− mice were fixed by perfusion with 4% formaldehyde in 0.1 M PB-buffer (pH 7.4) and embedded in paraffin. The paraffin sections (6 μm) were dewaxed, cleared in xylene, rehydrated in decreasing percentage of ethanol and finally transferred into TBS (50 mM Tris–HCl, pH 7.4; 150 mM NaCl). Microwave antigen retrieval was performed in 0.1 M sodium citrate buffer (pH 6.0). Sections were blocked either by non-serum based blocking reagent (DAKO, Germany) or M.O.M blocking reagent (Vector Lab., USA), depending on the secondary antibodies used. For immunofluorescent staining using the mouse parvalbumin, calbindin and calretinin monoclonal antibodies, the M.O.M. blocking reagent was applied on sections for 2 h at RT. The primary mouse monoclonal antibodies against parvalbumin and calbindin (Sigma, Germany) were diluted M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 1:4000, the calretinin antibody 1:15 in TBS solution containing 10% normal goat serum and 0.5% Triton X-100 and was incubated at RT for 1.5 h. A goat-anti-mouse secondary antibody (1:1000) conjugated with Alexa 546 (Invitrogen, Germany) was diluted in the same solution and incubated on sections at RT for 1.5 h. Unbounded antibodies were removed by rinsing sections 3 × 5 min in TBS buffer. The nuclei were stained by 0.1% Hoechst 33258 (1:500, Invitrogen, Germany) for 10 min. Sections were finally mounted with coverslips using DAKO fluorescent mounting medium. For enzyme-linked indirect immunohistochemical labeling, DAKO Envision+ (mouse Ig) kit was used and the staining was performed according to the manufacturer's instruction. The primary parvalbumin antibody was diluted in 3% (w/v) BSA/TBS and the color development of DAB chromogen was controlled by checking its intensity under the light microscope. DAB-stained sections were counter-stained by Haematoxylin (Sigma), dehydrated in ethanol series and cleared in xylene. The sections were mounted with coverslips using DAPX mounting medium (Fluka, Germany). Negative controls were carried out using normal mouse serum. Histology of brain slices from Cav2.3+/+ and Cav2.3−/− mice was further investigated using standard Nissl and Klüver–Barrera staining. For double immunostaining of Cav2.3 and parvalbumin on rat brain, sections were first incubated with anti-parvalbumin antibody according to the above mentioned procedure, and then followed by incubation of anti-Cav 2.3 195A antibody at RT for 1 h. The immunoreactivity was detected by goat-anti mouse Ig conjugated with Alexa 488 (Invitrogen) and goat-anti rabbit Ig conjugated with Alexa 546 (Invitrogen), respectively. Radiotelemetric surface EEG recordings (Electrocorticograms, ECoG) The TA10ETA-F20 transmitter (DSI, St. Paul, MN, USA) was used for electrocorticographic (surface) recordings in Cav2.3+/+ and Cav2.3−/− mice. The radiotelemetry system, implantation procedure, and postoperative treatment including pain management are as previously described (Weiergraber et al., 2005; Weiergraber et al., 2006a; Weiergraber et al., 2007). Epidural leads were positioned at the primary somatosensory cortex (S1), the transition zone of barrel field, dysgranular region and shoulder region at the following stereotaxic coordinates: (+)-lead, Bregma −1 mm, lateral of Bregma 2.5 mm (right hemisphere); (−)-lead, Bregma −1 mm, lateral of Bregma 2.5 mm (left hemisphere) and finally fixed at the neurocranium using dental cement. Animals were given at least 10 days to fully recover before initiating injection experiments (Kramer and Kinter, 2003). Thalamocortical hyperoscillation induced by GBL administration To evaluate the impact of Cav2.3 in SWD generation, γ-hydroxybutyrolactone (GBL, Sigma, Germany), a prodrug of γ-hydroxybutyric acid (GHB) was systemically administered to a radiotransmitter implanted Cav2.3+/+ (27.86 ± 1.78 g, 16.36 ± 1.24 weeks, n = 8, all ♂) and Cav2.3−/− mice (32.16 ± 0.99 g, 15.09 ± 1.05 weeks, n = 8, all ♂) at a dosage of 70 mg/kg. GBL provokes electroencephalographically recordable absence seizures in rodents which are associated with behavioural arrest (Snead et al., 2000). Gamma-hydroxybutyrolactone was freshly dissolved in physiological 0.9% NaCl before injection. Both genotypes received GBL injection at least 10 days after implantation to ensure full recovery from radiotransmitter implantation (Cav2.3+/+: 15.50 ± 0.92 days; Cav2.3−/−: 12.60 ± 1.12 days). Each animal was isolated for N30 min before intraperitoneal (i.p.) administration of the non-convulsant epileptogenic agent. Collecting data for absence seizure susceptibility analysis SWD latencies were calculated as the time interval from the moment of GBL injection to the first electroencephalographic 617 observation of SWDs. If an animal did not exhibit SWDs, it was assigned the maximum latency of the total observation period, i.e. 60 min. In addition, the frequencies and total duration of SWD episodes (ictal phases) and interictal epochs during the 1 h observation period were evaluated using radiotelemetric ECoG recordings. SWD episodes with an interictal period N3 s were regarded as separate events. Evaluation of EEG data and statistical analysis To acquire and analyze EEG data, the Dataquest A.R.T. 4.1 software (DSI) was used. In addition to EEG recordings following GBL administration, 24 h control recordings were carried out at day 7 and day 10 post-implantation for each individual animal. EEG activity was sampled at 1000 Hz with no a priori filter cut-off. Absolute powerspectrum density (PSD, mV2/Hz) was calculated from 2 min segments using the periodogram function (FFT based with Hamming windowing method). Frequency ranges were defined as follows: delta, δ (1– 4 Hz), theta, (4–8 Hz), alpha, α (8–12 Hz), beta, β (12–32 Hz), and gamma, γ (32–50 Hz). Additional to this discontinuous PSD evaluation, continuous PSD analysis was also performed to elicit the SWD frequency distribution. All data are calculated and displayed as means ± SEM. Continuous variables were analyzed using the parametric Student's t-test, considering p b0.05 as statistically significant. Acknowledgments We thank Mrs. Renate Clemens and the animal keepers of the central facility and Dr. B. Wagner for their excellent and permanent assistance. The work was financially supported by the Center of Molecular Medicine Cologne (CMMC, to MW, JH, and TS, Faculty of Medicine, University of Cologne). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.mcn.2008.08.007. References Andriamampandry, C., Taleb, O., Kemmel, V., Humbert, J.P., Aunis, D., Maitre, M., 2007. Cloning and functional characterization of a gamma-hydroxybutyrate receptor identified in the human brain. FASEB J. 21, 885–895. Arias-Olguin, I.I., Vitko, I., Fortuna, M., Baumgart, J.P., Sokolova, S., Shumilin, I.A., Van Deusen, A., Soriano-Garcia, M., Gomora, J.C., Perez-Reyes, E., 2008. Characterization of the gating brake in the I–II loop of Cav3.2 T-type Ca2+ channels. J. Biol. Chem. 283, 8136–8144. Battaglia, G., Lizier, C., Colacitti, C., Princivalle, A., Spreafico, R., 1994. A reticuloreticular commissural pathway in the rat thalamus. J. Comp. Neurol. 347, 127–138. Bers, D.M., 2002. Cardiac excitation–contraction coupling. Nature 415, 198–205. Bito, H., Deisseroth, K., Tsien, R.W., 1997. Ca2+-dependent regulation in neuronal gene expression. Curr. Opin. Neurobiol. 7, 419–429. Blumenfeld, H., McCormick, D.A., 2000. Corticothalamic inputs control the pattern of activity generated in thalamocortical networks. J. Neurosci. 20, 5153–5162. Caddick, S.J., Wang, C., Fletcher, C.F., Jenkins, N.A., Copeland, N.G., Hosford, D.A., 1999. Excitatory but not inhibitory synaptic transmission is reduced in lethargic (Cacnb4 (lh)) and tottering (Cacna1atg) mouse thalami. J. Neurophysiol. 81, 2066–2074. Catterall, W.A., Perez-Reyes, E., Snutch, T.P., Striessnig, J., 2005. International Union of Pharmacology. XLVIII. Nomenclature and structure–function relationships of voltage-gated calcium channels. Pharmacol. Rev. 57, 411–425. Danober, L., Deransart, C., Depaulis, A., Vergnes, M., Marescaux, C., 1998. Pathophysiological mechanisms of genetic absence epilepsy in the rat. Prog. Neurobiol. 55, 27–57. De Borman, B., Lakaye, B., Minet, A., Zorzi, W., Vergnes, M., Marescaux, C., Grisar, T., 1999. Expression of mRNA encoding a1E and a1G subunit in the brain of a rat model of absence epilepsy. NeuroReport 10, 569–574. Dominguez, M.I., Blasco-Ibanez, J.M., Crespo, C., Marques-Mari, A.I., Martinez-Guijarro, F.J., 2003. Zinc chelation during non-lesioning overexcitation results in neuronal death in the mouse hippocampus. Neuroscience 116, 791–806. Gibbs III, J.W., Zhang, Y.F., Ahmed, H.S., Coulter, D.A., 2002. Anticonvulsant actions of lamotrigine on spontaneous thalamocortical rhythms. Epilepsia 43, 342–349. Grabsch, H., Pereverzev, A., Weiergräber, M., Schramm, M., Henry, M., Vajna, R., Beattie, R.E., Volsen, S.G., Klöckner, U., Hescheler, J., Schneider, T., 1999. Immunohistochemical detection of a1E voltage-gated Ca2+ channel isoforms in cerebellum, INS-1 618 M. Weiergräber et al. / Molecular and Cellular Neuroscience 39 (2008) 605–618 cells, and neuroendocrine cells of the digestive system. J. Histochem. Cytochem. 47, 981–993. Hainsworth, A.H., McNaughton, N.C., Pereverzev, A., Schneider, T., Randall, A.D., 2003. Actions of sipatrigine, 202W92 and lamotrigine on R-type and T-type Ca2+ channel currents. Eur. J. Pharmacol. 467, 77–80. Hofmann, F., Lacinová, L., Klugbauer, N., 1999. Voltage-dependent calcium channels: from structure to function. Rev. Physiol. Biochem. Pharmacol. 139, 33–87. Hosford, D.A., Clark, S., Cao, Z., Wilson Jr., W.A., Lin, F.H., Morrisett, R.A., Huin, A., 1992. The role of GABAB receptor activation in absence seizures of lethargic (lh/lh) mice. Science 257, 398–401. Jun, K., Piedras-Renteria, E.S., Smith, S.M., Wheeler, D.B., Lee, S.B., Lee, T.G., Chin, H.M., Adams, M.E., Scheller, R.H., Tsien, R.W., Shin, H.S., 1999. Ablation of P/Q-type Ca2+ channel currents, altered synaptic transmission, and progressive ataxia in mice lacking the a1A-subunit. Proc. Natl. Acad. Sci. U. S. A. 96, 15245–15250. Khosravani, H., Zamponi, G.W., 2006. Voltage-gated calcium channels and idiopathic generalized epilepsies. Physiol. Rev. 86, 941–966. Khosravani, H., Altier, C., Simms, B., Hamming, K.S., Snutch, T.P., Mezeyova, J., McRory, J.E., Zamponi, G.W., 2004. Gating effects of mutations in the Cav3.2 T-type calcium channel associated with childhood absence epilepsy. J. Biol. Chem. 279, 9681–9684. Khosravani, H., Bladen, C., Parker, D.B., Snutch, T.P., McRory, J.E., Zamponi, G.W., 2005. Effects of Cav3.2 channel mutations linked to idiopathic generalized epilepsy. Ann. Neurol. 57, 745–749. Kim, D., Song, I., Keum, S., Lee, T., Jeong, M.J., Kim, S.S., McEnery, M.W., Shin, H.S., 2001. Lack of the burst firing of thalamocortical relay neurons and resistance to absence seizures in mice lacking a1G T-type Ca2+ channels. Neuron 31, 35–45. Kim, U., Bal, T., McCormick, D.A., 1995. Spindle waves are propagating synchronized oscillations in the ferret LGNd in vitro. J. Neurophysiol. 74, 1301–1323. Kisilevsky, A.E., Zamponi, G.W., 2008. Presynaptic calcium channels: structure, regulators, and blockers. Handb. Exp. Pharmacol. 45–75. Kramer, K., Kinter, L.B., 2003. Evaluation and applications of radiotelemetry in small laboratory animals. Physiol. Genomics 13, 197–205. Krieger, A., Radhakrishnan, K., Pereverzev, A., Siapich, S.A., Banat, M., Kamp, M.A., Leroy, J., Klockner, U., Hescheler, J., Weiergraber, M., Schneider, T., 2006. The molecular chaperone hsp70 interacts with the cytosolic II–III loop of the Cav2.3 E-type voltage-gated Ca2+ channel. Cell. Physiol. Biochem. 17, 97–110. Kullmann, D.M., 2002. The neuronal channelopathies. Brain 125, 1177–1195. Kuzmiski, J.B., Barr, W., Zamponi, G.W., MacVicar, B.A., 2005. Topiramate inhibits the initiation of plateau potentials in CA1 neurons by depressing R-type calcium channels. Epilepsia 46, 481–489. Lakaye, B., Thomas, E., Minet, A., Grisar, T., 2002. The genetic absence epilepsy rat from Strasbourg (GAERS), a rat model of absence epilepsy: computer modeling and differential gene expression. Epilepsia 43 (Suppl. 5), 123–129. Lee, J., Kim, D., Shin, H.S., 2004. Lack of delta waves and sleep disturbances during nonrapid eye movement sleep in mice lacking alpha1G-subunit of T-type calcium channels. Proc. Natl. Acad. Sci. U. S. A. 101, 18195–18199. Llinas, R.R., Steriade, M., 2006. Bursting of thalamic neurons and states of vigilance. J. Neurophysiol. 95, 3297–3308. Manning, J.P., Richards, D.A., Bowery, N.G., 2003. Pharmacology of absence epilepsy. Trends Pharmacol. Sci. 24, 542–549. Manning, J.P., Richards, D.A., Leresche, N., Crunelli, V., Bowery, N.G., 2004. Cortical-area specific block of genetically determined absence seizures by ethosuximide. Neuroscience 123, 5–9. Mathie, A., Sutton, G.L., Clarke, C.E., Veale, E.L., 2006. Zinc and copper: pharmacological probes and endogenous modulators of neuronal excitability. Pharmacol. Ther. 111, 567–583. Metz, A.E., Jarsky, T., Martina, M., Spruston, N., 2005. R-type calcium channels contribute to afterdepolarization and bursting in hippocampal CA1 pyramidal neurons. J. Neurosci. 25, 5763–5773. Nagaeva, D.V., Akhmadeev, A.V., 2006a. Structural organization, neurochemical characteristics, and connections of the reticular nucleus of the thalamus. Neurosci. Behav. Physiol 36, 987–995. Pereverzev, A., Mikhna, M., Vajna, R., Gissel, C., Henry, M., Weiergräber, M., Hescheler, J., Smyth, N., Schneider, T., 2002. Disturbances in glucose-tolerance, insulin-release and stress-induced hyperglycemia upon disruption of the Cav2.3 (a1E) subunit of voltage-gated Ca2+ channels. Mol. Endocrinol. 16, 884–895. Rhee, J.S., Ishibashi, H., Akaike, N., 1999. Calcium channels in the GABAergic presynaptic nerve terminals projecting to Meynert neurons of the rat. J. Neurochem. 72, 800–807. Seidenbecher, T., Staak, R., Pape, H.C., 1998. Relations between cortical and thalamic cellular activities during absence seizures in rats. Eur. J. Neurosci. 10, 1103–1112. Shin, H.S., 2006. T-type Ca2+ channels and absence epilepsy. Cell Calcium 40, 191–196. Smith, K.A., Fisher, R.S., 1996. The selective GABAB antagonist CGP-35348 blocks spike– wave bursts in the cholesterol synthesis rat absence epilepsy model. Brain Res. 729, 147–150. Snead III, O.C., 1992. Evidence for GABAB-mediated mechanisms in experimental generalized absence seizures. Eur. J. Pharmacol. 213, 343–349. Snead III, O.C., 2000. Evidence for a G protein-coupled gamma-hydroxybutyric acid receptor. J. Neurochem. 75, 1986–1996. Snead III, O.C., Banerjee, P.K., Burnham, M., Hampson, D., 2000. Modulation of absence seizures by the GABA(A) receptor: a critical role for metabotropic glutamate receptor 4 (mGluR4). J. Neurosci. 20, 6218–6224. Song, I., Kim, D., Choi, S., Sun, M., Kim, Y., Shin, H.S., 2004. Role of the alpha1G T-type calcium channel in spontaneous absence seizures in mutant mice. J. Neurosci. 24, 5249–5257. Spreafico, R., Battaglia, G., Frassoni, C., 1991. The reticular thalamic nucleus (RTN) of the rat: cytoarchitectural, Golgi, immunocytochemical, and horseradish peroxidase study. J. Comp. Neurol. 304, 478–490. Steriade, M., Contreras, D., 1998. Spike–wave complexes and fast components of cortically generated seizures. I. Role of neocortex and thalamus. J. Neurophysiol. 80, 1439–1455. Sun, H.S., Hui, K., Lee, D.W., Feng, Z.P., 2007. Zn2+ sensitivity of high- and low-voltage activated calcium channels. Biophys. J. 93, 1175–1183. Suzuki, T., Delgado-Escueta, A.V., Aguan, K., Alonso, M.E., Shi, J., Hara, Y., Nishida, M., Numata, T., Medina, M.T., Takeuchi, T., Morita, R., Bai, D., Ganesh, S., Sugimoto, Y., Inazawa, J., Bailey, J.N., Ochoa, A., Jara-Prado, A., Rasmussen, A., Ramos-Peek, J., Cordova, S., Rubio-Donnadieu, F., Inoue, Y., Osawa, M., Kaneko, S., Oguni, H., Mori, Y., Yamakawa, K., 2004. Mutations in EFHC1 cause juvenile myoclonic epilepsy. Nat. Genet. 36, 842–849. Tai, C., Kuzmiski, J.B., MacVicar, B.A., 2006. Muscarinic enhancement of R-type calcium currents in hippocampal CA1 pyramidal neurons. J. Neurosci. 26, 6249–6258. Takeda, A., Hirate, M., Tamano, H., Nisibaba, D., Oku, N., 2003a. Susceptibility to kainateinduced seizures under dietary zinc deficiency. J. Neurochem. 85, 1575–1580. Takeda, A., Minami, A., Seki, Y., Oku, N., 2003b. Inhibitory function of zinc against excitation of hippocampal glutamatergic neurons. Epilepsy Res. 57, 169–174. Takeda, A., Tamano, H., Nagayoshi, A., Yamada, K., Oku, N., 2005. Increase in hippocampal cell death after treatment with kainate in zinc deficiency. Neurochem. Int. 47, 539–544. Talley, E.M., Cribbs, L.L., Lee, J.H., Daud, A., Perez-Reyes, E., Bayliss, D.A., 1999. Differential distribution of three members of a gene family encoding low voltageactivated (T-type) calcium channels. J. Neurosci. 19, 1895–1911. Talley, E.M., Solórzano, G., Depaulis, A., Perez-Reyes, E., Bayliss, D.A., 2000. Low-voltageactivated calcium channel subunit expression in a genetic model of absence epilepsy in the rat. Mol. Brain Res. 75, 159–165. Timmermann, D.B., Westenbroek, R.E., Schousboe, A., Catterall, W.A., 2002. Distribution of high-voltage-activated calcium channels in cultured gamma-aminobutyric acidergic neurons from mouse cerebral cortex. J. Neurosci. Res. 67, 48–61. Tsakiridou, E., Bertollini, L., de Curtis, M., Avanzini, G., Pape, H.C., 1995. Selective increase in T-type calcium conductance of reticular thalamic neurons in a rat model of absence epilepsy. J. Neurosci. 15, 3110–3117. Turnbull, J., Lohi, H., Kearney, J.A., Rouleau, G.A., Delgado-Escueta, A.V., Meisler, M.H., Cossette, P., Minassian, B.A., 2005. Sacred disease secrets revealed: the genetics of human epilepsy. Hum. Mol. Genet. 14, 2491–2500. van de Bovenkamp-Janssen, M., Scheenen, W.J., Kuijpers-Kwant, F.J., Kozicz, T., Veening, J.G., van Luijtelaar, E.L., McEnery, M.W., Roubos, E.W., 2004. Differential expression of high voltage-activated Ca2+ channel types in the rostral reticular thalamic nucleus of the absence epileptic WAG/Rij rat. J. Neurobiol. 58, 467–478. van Luijtelaar, G., Wiaderna, D., Elants, C., Scheenen, W., 2000. Opposite effects of T- and L-type Ca2+ channels blockers in generalized absence epilepsy. Eur. J. Pharmacol. 406, 381–389. van Luijtelaar, E.L., Drinkenburg, W.H., van Rijn, C.M., Coenen, A.M., 2002. Rat models of genetic absence epilepsy: what do EEG spike–wave discharges tell us about drug effects? Methods Find. Exp. Clin. Pharmacol. 24 (Suppl. D), 65–70. Von Krosigk, M., Bal, T., McCormick, D.A., 1993. Cellular mechanisms of a synchronized oscillation in the thalamus. Science 261, 361–364. Weiergraber, M., Pereverzev, A., Vajna, R., Henry, M., Schramm, M., Nastainczyk, W., Grabsch, H., Schneider, T., 2000. Immunodetection of alpha1E voltage-gated Ca2+ channel in chromogranin-positive muscle cells of rat heart, and in distal tubules of human kidney. J. Histochem. Cytochem. 48, 807–819. Weiergraber, M., Henry, M., Hescheler, J., Smyth, N., Schneider, T., 2005. Electrocorticographic and deep intracerebral EEG recording in mice using a telemetry system. Brain Res. Brain Res. Protoc. 14, 154–164. Weiergraber, M., Henry, M., Krieger, A., Kamp, M., Radhakrishnan, K., Hescheler, J., Schneider, T., 2006a. Altered seizure susceptibility in mice lacking the Cav2.3 E-type Ca2+ channel. Epilepsia 47, 839–850. Weiergraber, M., Kamp, M.A., Radhakrishnan, K., Hescheler, J., Schneider, T., 2006b. The Cav2.3 voltage-gated calcium channel in epileptogenesis—shedding new light on an enigmatic channel. Neurosci. Biobehav. Rev. 30, 1122–1144. Weiergraber, M., Henry, M., Radhakrishnan, K., Hescheler, J., Schneider, T., 2007. Hippocampal seizure resistance and reduced neuronal excitotoxicity in mice lacking the Cav2.3 E/R-type voltage-gated calcium channel. J. Neurophysiol. 97, 3660–3669. Yang, S.N., Berggren, P.O., 2005. {beta}-Cell CaV channel regulation in physiology and pathophysiology. Am. J. Physiol. Endocrinol. Metab. 288, E16–E28. Yue, C., Remy, S., Su, H., Beck, H., Yaari, Y., 2005. Proximal persistent Na+ channels drive spike afterdepolarizations and associated bursting in adult CA1 pyramidal cells. J. Neurosci. 25, 9704–9720.