IL-1β is induced in reactive astrocytes in the somatosensory cortex of rats with genetic absence epilepsy at the onset of spike-and-wave discharges, and contributes to their occurrence
Demet Akin a, Teresa Ravizza b, Mattia Maroso b, Nihan Carcak c, Tugba Eryigit d, Ilaria Vanzulli b, Rezzan Gülhan Aker d, Annamaria Vezzani b,⁎, Filiz Yılmaz Onat d,⁎⁎
a b s t r a c t
Interleukin (IL)-1β plays a crucial role in the mechanisms of limbic seizures in rodent models of temporal lobe epilepsy. We addressed whether activation of the IL-1β signaling occurs in rats with genetic absence epilepsy (GAERS) during the development of spike-and-wave discharges (SWDs). Moreover, we studied whether inhibition of IL-1β biosynthesis in GAERS could affect SWD activity.
IL-1β expression and glia activation were studied by immunocytochemistry in the forebrain of GAERS at postnatal days (PN)14, PN20, and PN90 and in age-matched non-epileptic control Wistar rats. In PN14 GAERS, when no SWDs have developed yet, IL-1β immunostaining was undetectable, and astrocytes and microglia showed a resting phenotype similar to control Wistar rats. In 3 out of 9 PN20 GAERS, IL-1β was observed in activated astrocytes of the somatosensory cortex; the cytokine expression was associated with the occurrence of immature-type of SWDs. In all adult PN90 GAERS, when mature SWDs are established, IL-1β was observed in reactive astrocytes of the somatosensory cortex but not in adjacent cortical areas or in extra-cortical regions. An age-dependent c-fos activation was found in the somatosensory cortex of GAERS with maximal levels reached in PN90 rats; c-fos was also induced in some thalamic nuclei in PN20 and PN90 GAERS.
Inhibition of IL-1β biosynthesis in PN90 GAERS by 4-day systemic administration of a specific ICE/Caspase-1 blocker, significantly reduced both SWD number and duration.
These results show that IL-1β is induced in reactive astrocytes of the somatosensory cortex of GAERS at the onset of SWDs. IL-1β has pro-ictogenic properties in this model, and thus it may play a contributing role in the mechanisms underlying the occurrence of absence seizures.
Keywords: Inflammation Absence seizures Interleukins SWD
Introduction
Typical absence epilepsy, a prototype of generalized idiopathic epilepsies, is characterized by several daily episodes of absence seizures which consist of a brief interruption of behavioral activity and a simultaneous electroencephalographic (EEG) expression of bilateral, synchronous, and symmetrical spike-and-wave discharges (SWDs). Genetic Absence Epilepsy Rats from Strasbourg (GAERS), a well defined animal model of typical absence epilepsy in humans, display non- convulsive seizures and SWDs within the cortico-thalamo-cortical circuit involving the somatosensory cortex, the ventrobasal thalamus and the reticular thalamic nucleus (see for review Danober et al., 1998). SWDs in GAERS appear to have an initiation site within the somatosensory cortex, namely in the “the upper lip and nose area of the somatosensory cortex” (S1Ulp) (Meeren et al., 2002; Paxinos and Watson, 2005; Polack et al., 2009). In GAERS, immature-type of SWDs are observed around postnatal day (PN)30 (Carcak et al., 2008; Vergnes et al., 1986) and subsequently the number, duration and frequency of SWDs increase with age, reaching a mature pattern in 3–4 month-old rats (Carcak et al., 2008; Marescaux et al., 1992; Vergnes et al., 1986).
Glial cells have been suggested to play a role in the mechanisms underlying SWDs in GAERS: in particular, it has been shown that the glutamine supply to GABAergic neurons in the cortex and subcortical areas of PN30 GAERS is lower than in control rats (Melo et al., 2006); in adult GAERS, the production of glutamate from astrocytic glutamine was enhanced indicating increased astrocytic metabolism (Melo et al., 2007). Moreover, Dutuit et al. (2000) have described astrocytic activation, as shown by increased levels of glial fibrillary acidic protein (GFAP) in cortex and thalamus of PN30 GAERS before the occurrence of absence seizures, as well as in adult epileptic GAERS as compared to non-epileptic rats.
Activated astrocytes and microglia are major sources of inflam- matory molecules in the brain during epileptic activity induced in experimental models of limbic seizures and temporal lobe epilepsy (TLE) (reviewed in Vezzani et al., 2011). In particular, the prototypical pro-inflammatory pathway activated by interleukin-1(IL)β via its functional receptor IL-1R1, is upregulated both in experimental models as well as in human epileptogenic tissue in TLE and in epilepsies associated with malformations of cortical development (Ravizza et al., 2006a, 2008a). Moreover, the activation of IL-1β signaling in adult rodent forebrain exacerbates limbic seizures (Balosso et al., 2008; Vezzani et al., 2000, 2008) and contributes to precipitation of febrile-like convulsions in immature rodents (Dubé et al., 2005; Heida and Pittman, 2005). Whether IL-1β system is activated and has a proconvulsant role in absence epilepsy is still unknown.
In the present study, we show that IL-1β is induced in activated astrocytes specifically in the somatosensory cortex of adult GAERS but not in non-epileptic control Wistar rats. Importantly, the changes in IL-1β anticipate the age-related onset of mature SWDs, thus suggesting a possible contributing role of this cytokine in SWD generation. We further demonstrate that pharmacological treatment with VX-765, which selectively blocks IL-1β biosynthesis (Randle et al., 2001; Ravizza et al., 2006b; Stack et al., 2005), decreases the frequency of SWDs in adult GAERS, highlighting a contributing role of this cytokine to SWD activity.
Materials and methods
Experimental animals
Postnatal day (PN) 14, 20 and adult (PN90) male non-epileptic Wistar rats and GAERS were used (n= 5–9 for each experimental group). PN0 was defined as the day of birth. Wistar rats were obtained from Marmara University Experimental Animal Unit, and GAERS were provided from the breeding colony of Marmara University School of Medicine Department of Pharmacology and Clinical Pharmacology. The rats were housed with their dams at a constant temperature (21 ±3 °C) and relative humidity (60%) with a fixed 12 h light–dark cycle and free access to food and water. The pups were housed with their dams until weaning at PN21. Older animals were housed in groups of two per cage. For each experimental group, male pups were taken from three separate litters. Procedures involving animals and their care were approved by the Marmara University Ethical Committee for Experimental Animals (41.2009.MAR).
Immunocytochemistry
The changes in the expression of glia markers and alterations in cell morphology indicative of glia activation, as well as the expression of IL-1β, were investigated at three different stages of SWD development in GAERS, namely at PN14, when no SWDs have yet developed; at PN20, when “mature” SWDs have not developed yet, and at PN90, when “mature” SWDs are established (Carcak et al., 2008; Vergnes et al., 1986). We did not perform routine EEG analysis to detect the occurrence of SWDs in GAERS used for histological evaluations to avoid compromising the quality of the histological preparations in brain tissue surrounding implanted electrodes. However, 4 out of 9 PN20 GAERS were EEG recorded to evaluate whether “immature-type” of SWDs occurred (Fig. 6).
Transcardial perfusion
Rats were deeply anesthetized with ketamine (100 mg/kg intra- peritoneally, i.p.) and xylazine (10 mg/kg i.p.), and perfused through the ascending aorta with 50 mM cold phosphate-buffered saline (PBS, pH 7.4) followed by chilled 4% PAF in PBS. The brains were post-fixed at 4 °C for 90 min, and then transferred to 20% (PN90 rat) or 30% (PN14 and PN20 rats) sucrose in PBS at 4 °C for 24 h. Then, the brains were immersed in −40 °C (PN14 and PN20 rats) or −50 °C (PN90 rats) isopentane for 3 min and stored at −80 °C until assayed. Serial coronal sections (40 μm) were cut from 0 to −6.6 mm from bregma according to Paxinos and Watson(2005), and collected in 100 mM PBS. We selected 7 different brain levels for the subsequent immunocy- tochemical analysis: bregma 0, −1.4, −2.5, −3.6, −4.8, −5.7 and −6.6. We prepared 7 series of 8 sections each: the .first section was stained for IL-1β, the 2nd for GFAP, the 3rd for OX-42, the 4th for c-fos, while the other sections were used for a double-immunostaining analysis (see below).
IL-1β
IL-1β immunostaining was carried out as previously described (Ravizza et al., 2008a). Briefly, slices were incubated at 4 °C for 10 min in 70% methanol and 2% H2O2 in Tris–HCl-buffered saline (TBS), followed by 30 min incubation in 10% fetal bovine serum (FBS) in 1% Triton X-100 in TBS. The, slices were incubated overnight at 4 °C in the same medium with the primary antibody against IL-1β (1:200, Santa Cruz Bio., CA, USA). Immunoreactivity was tested by the avidin– biotin–peroxidase technique (Vector Labs, Burlingame, CA, USA) using 3′,3′-diaminobenzidine (DAB; Sigma-Aldrich, Munich, Germany) as chromogen, and the signal was amplified by nickel ammonium.
Glial markers
Slices were incubated at 4 °C for 30 min in 0.4% Triton X-100 in PBS followed by a 15 min incubation in 3% FBS in 0.1% Triton X-100 in PBS. They were subsequently incubated overnight at 4 °C in 3% FBS in 0.1% Triton X-100 in PBS, with the following primary antibodies: mouse anti-CD11b (complement receptor type 3, OX-42, 1:100, Serotec Ltd, Oxford, UK), a marker of microglia/macrophages, or with mouse anti-glial fibrillary acidic protein (GFAP, 1:2500, Chemicon Int. Inc., Temecula, USA), a selective marker of astrocytes. Immunoreactivity was tested by avidin–biotin–peroxidase technique (Vectastain ABC kit, Vector Labs, USA), using 3′,3′-diaminobenzidine (DAB; Sigma-Aldrich) as chromogen.
c-fos
This analysis was done since c-fos is a well established surrogate marker of neuronal network activation (Morgan et al., 1987). Slices were incubated at room temperature for 30 min in 0.3% H2O2 in PBS followed by a 2 h incubation in 3% FBS in 0.25% Triton X-100 in PBS. They were subsequently incubated for 48 h at 4 °C with the primary antibody against c-fos (1:10,000, Oncogene, Cambridge, MA, USA) in 3% FBS in 0.25% Triton X-100 in PBS. Immunoreactivity was tested using signal amplification as described above. No immunostaining was observed when the slices were incubated with the primary antibodies pre-absorbed with the corresponding peptides (Marcon et al., 2009; Ravizza et al., 2008a), or without the primary antibodies.
Double-immunostaining
Co-localization studies were carried out to identify the cell types expressing IL-1β. Two slices were used from each brain for each cell type marker. After incubation with the primary antibody against IL- 1β, slices were incubated in biotinylated secondary anti-goat antibody (1:200, Vector Labs), then in streptavidin-HRP and the signal was revealed with tyramide conjugated to Fluorescein using a TSA amplification kit (NEN Life Science Products, Boston, MA, USA). Sections were subsequently incubated with mouse anti-OX-42 (1:100, Serotec Ltd), or mouse anti-GFAP (1:2500, Chemicon). Fluorescence was detected using anti-mouse secondary antibodies conjugated with Alexa546 (Molecular Probes, The Netherlands). Slide-mounted sections were examined with an Olympus Fluorview laser scanning confocal microscope (microscope BX61 and confocal system FV500) using dual excitation of 488 nm (Laser Ar) and 546 nm (Laser He–Ne green) for Fluorescein and Alexa546, respectively. The emission of fluorescent probes was collected on separate detectors. To eliminate the possibility of bleed-through between channels, the sections were scanned in a sequential mode.
Cell counting
Quantification of c-fos, IL-1β and GFAP positive cells in GAERS brain was performed as previously described (Ravizza et al., 2006a, 2008a), using one representative slice for each rat in each experimental group. c-fos immunoreactive nuclei were quantified in the following brain areas (Paxinos and Watson, 2005): deep (V–VI) and superficial layers of the somatosensory cortex (in the S1ULp region) at bregma 0 (boxed area in Figs. 1, 3, 5); paraventricular, latero-dorsal and reticular thalamic nuclei, ventrobasal thalamic complex (ventro-posterolateral and ventroposteromedial nuclei) at bregma − 2.5. IL-1β and GFAP immunopositive cells were counted at bregma 0 in deep and superficial layers of the somatosensory cortex (in the S1ULp region).
Briefly, 2 representative 20× non-overlapping fields (612.000 μm2 each) of the deep and superficial layers of the somatosensory cortex (in the S1ULp region) were captured using an Olympus BX61 microscope equipped with a motorized platform; the whole thalamic nuclei were quantified (Fig. 2a). c-fos immunoreactive nuclei, and IL- 1β and GFAP (both resting and activated) immunopositive cells within the captured images were marked by the operator and an automated cell count was generated. Astrocyte and microglia cell activation was determined by well established criteria (Kreutzberg, 1996; Niquet et al., 1994; Ravizza et al., 2008a; Shapiro et al., 2008) by analyzing changes in cell morphology and increases in GFAP or OX-42 immunoreactivity, respectively: resting astrocytes show low GFAP immunoreactivity and stellate-shaped cell bodies with thin processes; activated astrocytes are highly immunoreactive to GFAP and acquire an hypertrophic morphology, characterized by a large cell body, long and thick processes. Resting microglia are small cells with numerous, thin and extensive ramifications, associated with low OX-42 signal. Activated microglia are strongly immunoreactive for OX-42, showing large cell bodies and short processes; their phagocytic phenotype is associated with round-shape morphology without evident processes. In the somatosensory cortex (S1ULp area), the data obtained in each of the two fields captured in the deep or superficial layers per slice were added together providing a single value in each layer for each rat. Although this cell counting method has some limitations as compared to design-based stereological analysis (Schmitz and Hof, 2005), the occurrence of any bias in counting cells should similarly affect GAERS at different ages, since these samples underwent the same methodological procedures in parallel.
Stereotaxic surgery for placement of electrodes and basal EEG recording
A group of PN18 GAERS (n= 4 out of 9) were used for detection of “immature-type of SWDs”. These rats were anesthetized with a diluted ketamine and xylazine solution (1:10 of PN90 GAERS solution, injected intraperitoneally, i.p.). Two bipolar recording electrodes (Plastic’s One Inc., Roanoke, VA, USA, MS303/1-A untwisted) were bilaterally implanted onto the frontoparietal cortices for EEG recording, as previously described (Carcak et al., 2008). A group of PN90 GAERS (n= 10) was used specifically for pharmacological experiments. These rats were deeply anesthetized with administra- tion of ketamine (100 mg/kg, i.p.) and xylazine (10 mg/kg, i.p.). Stainless steel screws were used for extradural ground and recording electrodes; they were placed bilaterally in the skull over the frontal and occipital cortices, using a stereotaxic apparatus (Gulhan Aker et al., 2010).The electrodes for both groups were connected by insulated wires to a microconnector for EEG recordings and fixed to the skull with dental acrylic.
EEG analysis of SWDs
Cortical electrical activity was amplified through a BioAmp ML 136 amplifier, with band pass filter settings at 1–40 Hz and recorded with a PowerLab 8S System running the Chart v.7 program (ADI Instruments, Oxfordshire, U.K.). PN18 GAERS were EEG recorded 2 days after surgery. After a Plexiglass cage adaptation period of 60 min, the EEG recording was done over a 3 h period in the morning (9:00 am–12:00 am). An “immature SWD complex” in PN20 GAERS was identified by a sudden onset of a train of low-frequency (4–5 Hz) discharges lasting at least 1 s with an amplitude of at least twice the EEG background activity superimposed on an unsynchronized waking pattern, as described previously (Carcak et al., 2008). The cumulative total duration of immature SWDs was evaluated in PN20 GAERS over a 3 h recording period. The PN90 GAERS were allowed to recover from surgery for 7 days, then they were placed singly in a Plexiglass cage and a 60 min adaptation period was allowed before each session of EEG recording. SWD activity in PN90 GAERS was defined as previously reported (Onat et al., 2007): briefly, a SWD complex was identified by a train of sharp spikes and slow waves (7.5–9 Hz) with a duration of at least 1 s and an amplitude of at least twice the amplitude of the EEG background activity.
Pharmacological experiments using VX-765 in PN90 GAERS
One day before VX-765 injection, a baseline EEG recording was done in each PN90 GAERS in the morning (from 9 am to 12 am) and in the afternoon (from 2 pm to 5 pm) to assess the occurrence of SWDs after the stereotaxic surgery. These data were not used in the EEG analysis reported in Fig. 7; however, these data were used to compare the pre-injection baseline in each rat with the first day of vehicle or VX-765 injection.
On the day of the pharmacological experiment, PN90 GAERS were randomly divided into 2 groups, to be subsequently injected either with vehicle (n = 5) or with VX-765 (n = 5). After 60 min of adaptation, GAERS received vehicle (3 ml) or VX-765 (200 mg/kg dissolved in 0.1% Tween 80 in 0.5% hydroxy-ethyl-cellulose in distilled water; 3 ml/rat, i.p.) twice a day (i.e. at 9:00 am and at 2:00 pm) for 4 consecutive days. The EEG was continuously recorded for 3 h after each injection of VX-765.
We used this dose and treatment schedule since it provides a brain concentrations of VRT-043198, the active metabolite of VX-765, able to block ICE/Caspase-1 activity (70–140 nM, measured 2 h after last administration), and it has anticonvulsant effects on acute seizures induced by kainic acid (Ravizza et al., 2006b) and in rapid kindling (Ravizza et al., 2008b).
Statistical analysis of data
Data were expressed as mean±SEM (n = number of individual rats). The Wilcoxon rank sum test was used for c-fos comparison among the experimental groups. The effects of VX-765 on SWDs was analyzed by repeated measures ANOVA followed by Tukey’s test. Differences due to the treatments were considered significant with pb 0.05.
Results
Immunohistochemistry
Our quantitative immunohistochemical analysis was done in the GAERS brain areas implicated in the circuitry underlying SWDs, namely the somatosensory cortex (S1Ulp area; deep and superficial layers), reticular thalamic nucleus and ventrobasal thalamic complex (reviewed in Aker and Onat, 2002; Danober et al., 1998). Additional analysis was done in lateral–dorsal thalamus and periventricular thalamic nucleus where we observed c-fos positive cells with nuclear staining.
c-fos immunoreactivity in the somatosensory cortex
GAERS showed c-fos staining in the somatosensory cortex to a different extent depending on their age (Figs. 1c, f, h and bar gram; n=5–9); c-fos staining was not observed in age-matched Wistar control rats (Figs. 1a, d, g; n=5 each group). In the deep S1ULp layers, 2 out of 5 PN14 (71.5 ±6.5 nuclei; Fig. 1c) and 3 out of 9 PN20 (118 ±4 nuclei; Fig. 1f) GAERS showed c-fos-immunoreactive nuclei restricted to one hemisphere (bregma 0; boxed area in schematic drawing; see bar gram in Fig. 1). In the same PN20 rats, c-fos was also detected in the S1ULp superficial layers (29.0 ±5.0 nuclei), albeit to a lower extent than in deep layers (pb 0.05; bar gram Fig. 1). The remaining 3 PN14 GAERS and 6 PN20 GAERS did not show c-fos induction in the somatosensory cortex (Figs. 1b, e). In all PN90 GAERS, a bilateral c-fos induction was observed both in the S1ULp deep (204 ±13 nuclei; Fig. 1h) and superficial layers (300.8 ±27.2 nuclei; not shown in Fig. 1 but see bar gram), as well as in the remaining rostro-caudal extent of the deep and superficial layers of the somatosensory cortex (not shown). In PN90 GAERS the number of c-fos positive nuclei was higher in S1ULp superficial vs deep layers (pb 0.05, bar grams in Fig. 1).
c-fos immunoreactivity in the thalamic nuclei
c-fos immunoreactivity was found in some thalamic nuclei of PN20 and PN90 GAERS but not in PN14 GAERS, or in age-matched control Wistar rats. In PN20 GAERS showing unilateral c-fos induction in the deep S1ULp region, c-fos was also increased bilaterally in the paraventricular thalamic nucleus (14.0 ± 3.0, n= 3). In all PN90 GAERS (n= 5), bilateral c-fos immunoreactivity was detected in the reticular thalamic nucleus (88.0 ± 11.0, n = 5; Figs. 2b, f), the ventrobasal thalamic complex (59.6 ± 5.2, n= 5; Figs. 2c, f), as well as in the paraventricular (30.8 ± 3.6, n= 5; Figs. 2d, f) and latero- dorsal (13.6 ± 1.3, n= 5; Figs. 2e, f) thalamic nuclei (see also Slaght et al., 2002; Wallengren et al., 2005; Zhang et al., 1996).
IL-1β immunoreactivity in the somatosensory cortex
The pattern of IL-1β induction in PN20 and PN90 GAERS was closely overlapping with that described for c-fos in the same animals. In 3 out of 9 PN20 GAERS, IL-1β-immunoreactive cells with glia morphology were observed unilaterally in the deep (Fig. 3e) and superficial S1ULp layers (not shown, but see bar gram in Fig. 3) in brain sections adjacent to those showing increased c-fos signal (Fig. 1f); IL-1β signal was mostly localized in close apposition with brain microvasculature (Fig. 3e). Co-localization experiments showed that IL-1β was expressed by GFAP-astrocytes (inset in Fig. 3e). No staining was observed in the adjacent cortical regions or in extra- cortical regions (not shown). No IL-1β staining was observed in the remaining 6 PN20 GAERS (Figs. 3d vs c), or in PN14 rats (Figs. 3b vs a). In PN90 GAERS, IL-1β staining was observed bilaterally in deep layers of S1ULp region (bregma 0) in astrocytes (Fig. 3g, inset) but not in microglia (Figs. 3h–j); IL-1β was also enhanced in GFAP-positive astrocytes bilaterally in the superficial layers (not shown, see bar gram in Fig. 3). IL-1β immunopositive cells were detected throughout the rostro-caudal extent of the somatosensory cortex from bregma 0 to −6.6 (not shown).
Quantification of the number of IL-1β positive astrocytes in the S1ULp region (bar grams in Fig. 3) showed a higher number of cells in deep vs superficial layers in PN20 GAERS (pb 0.05) while a higher number of positive cells were found in the superficial layers in PN90 GAERS (pb 0.05). In PN90 GAERS, no IL-1β positive cells were observed in the cortical areas surrounding the borders of the somatosensory cortex (Fig. 4aI), or in piriform (Fig. 4bI), perirhinal (Fig. 4cI), entorhinal (Fig. 4dI) cortices; no IL-1β staining was detected in extra-cortical brain regions (not shown) with the exception of scattered IL-1β-immunoreactive astro- cytes in the septal aspect of the hippocampus in 1 out of 5 rats (sections at bregma −2.5 and −3.6; not shown).
GFAP immunoreactivity in the somatosensory cortex
In control Wistar rats at all ages, GFAP staining of astrocytes was diffusely low both in the deep (Figs. 5a, c, f) and superficial layers (not shown) of S1ULp area; high magnification morphological analysis showed stellated-shaped astrocytes with thin processes (insets in Figs. 5a, c, f) denoting their resting state. As previously described in Sprague–Dawley rats (Rizzi et al., 2003), the density of astrocytes with a resting phenotype increased with the age of Wistar rats (Figs. 5a, c, f).
No apparent changes in astrocytes resting morphology were observed in PN14 GAERS (Figs. 5b vs a, and insets). In 3 out of 9 PN20 GAERS, GFAP immunoreactivity was enhanced unilaterally in S1ULp deep (Fig. 5e, and inset) and superficial (not shown, but see Fig. 5i) layers in astrocytes with an activated phenotype. These changes occurred in slices adjacent to those showing increased IL-1β (Fig. 3e) and c-fos (Fig. 1f). In the remaining 6 PN20 GAERS, resting GFAP- immunopositive astrocytes were observed (Fig. 5d and inset); in these animals IL-1β staining was undetectable (Fig. 3d) and c-fos signal was absent (Fig. 1e). In PN90 GAERS, we observed GFAP- positive activated astrocytes bilaterally in the S1ULp deep (Fig. 5g and inset) and superficial layers (not shown, but see Fig. 5i). This staining encompassed the rostro-caudal extent of the somatosensory cortex from bregma 0 to −6.6 (not shown). In 1 out of 5 PN90 GAERS, scattered activated astrocytes were detected in the hippocampi (not shown). The changes in GFAP-positive astrocytes occurred in the same areas of increased IL-1β immunoreactivity (Fig. 3g) and c-fos signal (Fig. 1h) as detected in adjacent slices.
Cell counting in the S1ULp area (Fig. 5i) confirmed the absence of activated astrocytes in PN14 GAERS; in 3 out of 9 PN20 GAERS the activated astrocytes were predominant in deep cortical layers (pb 0.05) whereas in PN90 GAERS these cells were more abundant in the superficial layers (pb 0.05). Additionally, the total number of astrocytes (including resting and activated cells) increased in age-dependent manner in GAERS (Fig. 5h) similarly to control Wistar rats (Figs. 5a, c, f). In PN20 and PN90 GAERS, the pattern of GFAP staining in extra- cortical areas or cortical regions adjacent to the somatosensory cortex was similar to that observed in control Wistar rats (not shown). No evidence of microglia activation was detected in GAERS vs Wistar rats at all ages (see for ref. Figs. 3i, j).
EEG recording of SWDs in P20 GAERS
No SWDs were observed in the basal EEG of 3 out of 4 PN20 GAERS over the 3 h recording session in the morning, while one animal showed “immature” SWD activity with a total cumulative duration of 33 s and a mean duration of 1.3 ± 0.1 s (Fig. 6).
Effect of VX-765, an inhibitor of IL-1β biosynthesis, on SWDs in PN90 GAERS
Fig. 7 reports the daily cumulative frequency and duration of SWDs as assessed by EEG analysis during the morning (9:00 am–12:00 am) and the afternoon (2:00 pm–5:00 pm) recording sessions in PN90 GAERS during treatment with VX-765. On the first day of treatment (see Materials and methods for details), VX-765 did not modify SWD number and duration during the 3 h post-injection recording: SWD values were similar to those measured in GAERS rats treated with vehicle (Figs. 7a, b), and to those obtained in the same rats the day before treatment during time-matched EEG recording sessions (not shown). On the second day of treatment (i.e. after the 3rd drug injection), VX-765 significantly reduced the cumulative duration (Fig. 7a) and number (Fig. 7b) of SWDs by 55% on average as compared to vehicle-injected GAERS. This inhibitory effect on SWDs was maintained for the following 3rd and 4th days of treatment. The lack of effect of VX-765 on the first day of the injection is compatible with previous data showing that repetitive administration is required to achieve therapeutic drug concentrations in brain (Ravizza et al., 2006a, 2008b). Vehicle injection did not change the cumulative duration and number of SWDs in GAERS during the 4 days of injection (Figs. 7a, b). The reduction in SWDs induced by VX-765 was concomitant with a reduction in the behavioral correlates associated with SWD occurrence (i.e. arrest and immobility).
Discussion
The present data show that IL-1β is specifically induced in reactive astrocytes in the somatosensory cortex of adult GAERS when mature SWDs re established (Carcak et al., 2008). This induction was first observed in about 30% GAERS at PN20, thus before the development of mature SWDs (Carcak et al., 2008; Marescaux et al., 1992; Vergnes et al., 1986) while it was absent in PN14 GAERS when SWDs had not yet developed. In PN20 GAERS, IL-1β induction was observed in concomitance with “immature SWDs”, and it was restricted to the S1ULp area of the somatosensory cortex, predominantly in the deep layers, thus in the region involved in SWD generation (Meeren et al., 2002; Polack et al., 2007, 2009). Accordingly, inhibition of neuronal activity in this specific area by topical application of tetrodotoxin or injection of ethosuximide prevented the generation of ictal activities in adult GAERS (Manning et al., 2004; Polack et al., 2009). The pattern of IL-1β activation in 2 out of 5 PN20 GAERS which were not EEG recorded, was the same observed in 1 out of 4 PN20 GAERS with the “immature SWD activity”, thus suggesting that this type of immature activity likely occurs in about 30% of animals at this age. The functional meaning of the unilateral induction of IL-1β in the S1ULp area of PN20 GAERS is presently unclear.
In PN90 GAERS, IL-1β induction was not limited to the S1ULp area but spread to all rostro-caudal extension of the somatosensory cortex including also, and more prominently, the superficial cortical layers where the thalamo-cortical glutamatergic projections terminate (Depaulis and Van Luijtelaar, 2006). This evidence indicates that the cytokine expression in the somatosensory cortex accompanies the age-dependent development of SWD activity (Carcak et al., 2008), suggesting that it may be triggered by the underlying neuronal network activation. We can envisage that once mature SWDs develop, this activity further activates astroglia thus determining the larger and widespread expression of the cytokine observed in the somatosensory cortex in PN90 GAERS. Accordingly, recurrent seizure activity has been shown to promote brain inflammation in rats (for review see Vezzani et al., 2011).
The increased c-fos signaling was concomitant with the IL-1β induction in the somatosensory cortex of PN20 and PN90 GAERS, thus supporting that neuronal network activation triggers IL-1β biosyn- thesis in activated astrocytes. However, no IL-1β induction was found in PN14 rats showing a mild c-fos increase in the S1ULp area. Similarly, c-fos, but not IL-1β was induced in thalamic nuclei directly involved in the generation of SWDs (i.e. the reticular thalamic nucleus and the ventrobasal thalamic complex), as well as in other thalamic areas possibly recruited in the oscillatory thalamo-cortical network (Depaulis and Van Luijtelaar, 2006; Meeren et al., 2002). One likely explanation of these findings is that IL-1β induction in astrocytes occurs (or is detectable) only if a threshold of neuronal activation, as assessed by c-fos, is reached. Indeed, no increase in IL-1β was observed in areas where the number of c-fos positive cells ranged between 14 and 88 (i.e. in thalamic nuclei of PN20 and PN90 GAERS, and in somatosensory cortex of PN14 GAERS). On the contrary, IL-1β was overexpressed in somatosensory cortex of PN20 and PN90 GAERS where c-fos positive cells were N 115. C-fos activation is especially sensitive to excitatory glutamatergic neurotransmission (Morgan et al., 1987; Sagar et al., 1988) which may explain the more prominent induction of c-fos in GAERS somatosensory cortex as compared to the thalamic relay nuclei where increased GABAergic activity predomi- nates (Danober et al., 1998; Depaulis and Van Luijtelaar, 2006).
This evidence extends the link between brain inflammation and seizures which was originally described in models of limbic epilepsy (Vezzani et al., 2011). Recurrent limbic seizures in rodents induce IL- 1β in activated glial cells predominantly in the hippocampus and limbic cortex where seizures originate and spread (Balosso et al., 2008; De Simoni et al., 2000; Maroso et al., 2011; Ravizza et al., 2008a; Vezzani et al., 1999, 2000); thus, the commonality with GAERS is the induction of IL-1β in brain areas of seizure origin. One notable difference with models of limbic seizures is the lack of microglia activation and its IL-1β production in GAERS, which may be attributed to the different types of epileptic activity. Moreover, inflammation induced by limbic seizures is developmentally regulated, as in GAERS, and its age of onset appears to depend on the type of proconvulsant challenge (Dubé et al., 2010; Marcon et al., 2009; Rizzi et al., 2003).
Pharmacological studies in models of limbic seizures demonstrat- ed that IL-1β contributes to precipitation and recurrence of seizures caused by chemoconvulsants and electrical stimulation both in adult and immature rodents (Auvin et al., 2010; Balosso et al., 2008; De Simoni et al., 2000; Marchi et al., 2009; Maroso et al., 2011; Ravizza et al., 2006b, 2008b; Vezzani et al., 1999, 2000). IL-1β also contributes to determine the threshold to experimental febrile seizures in immature rats and mice (Dubé et al., 2005; Heida and Pittman, 2005). These findings raised the crucial question of whether IL-1β induction in astrocytes in GAERS is an epiphenomenon of SWD activity or it contributes to this activity. Pharmacological inhibition of IL-1β biosynthesis using a selective ICE/Caspase-1 inhibitor (Randle et al., 2001; Stack et al., 2005), significantly reduced SWD activity in adult GAERS, thus demonstrating that this cytokine is causally involved in this type of epileptic activity. Accordingly, previous evidence showed that injection of lipopolysaccharide, which induces IL-1β and downstream inflammatory mediators in brain, increased the number of SWDs in WAG/Rij rats (Kovacs et al., 2006).
About the mechanisms by which IL-1β could contribute to SWDs, recent evidence showed that astrocytes activated by neuronal activity can release glutamate which in turn triggers neuronal slow inward currents (SICs; see for review Halassa et al., 2007). SICs have a role in neuronal synchronization and can trigger action potentials in neurons; these currents are increased in models of seizures and their pharmacological inhibition significantly attenuates ictal events (Fellin et al., 2006, 2009). SICs are mediated by neuronal NMDA receptors containing NR2B subunit, and IL-1β has been shown to activate neuronal Ca2+ influx via NR2B phosphorylation (Balosso et al., 2008; Viviani et al., 2003). Indeed, this latter mechanism mediates the proconvulsant actions of IL-1β in models of limbic seizures (Balosso et al., 2008) and could be involved also in IL-1β effects in SWDs.
The cytokine effects on SWDs may also involve IL-1β-mediated inhibition of glial glutamate uptake, leading to an increase in glutamate extracellular levels (Hu et al., 2000; Ye and Sontheimer, 1996). A reduction in the cortical expression of astrocytic glutamate transporter GLT1 and GLAST was indeed reported in PN30 GAERS, thus before the onset of mature SWDs (Carcak et al., 2008; Dutuit et al., 2002). Moreover, glutamate uptake was decreased in primary culture of cortical astrocytes obtained from newborn GAERS (PN1) as compared to non-epileptic controls (Dutuit et al., 2002).
In summary, this study reports the first experimental findings showing that mature SWDs in GAERS are associated with, and preceded by, IL-1β induction in activated astrocytes in the somato- sensory cortex. Moreover, a specific anti-inflammatory approach, which blocks IL-1β biosynthesis, demonstrates that this cytokine contributes to SWD activity. This evidence may open the perspective to develop specific anti-inflammatory approaches for managing this non-convulsive form of epilepsy.
References
Aker, R.G., Onat, F.Y., 2002. Cardiovascular regulation through hypothalamic GABA(A) receptors in a genetic absence epilepsy model in rat. Epilepsia 43, 107–114.
Auvin, S., Shin, D., Mazarati, A., Sankar, R., 2010. Inflammation induced by LPS enhances epileptogenesis in immature rat and may be partially reversed by IL1RA. Epilepsia 51 (Suppl 3), 34–38.
Balosso, S., Maroso, M., Sanchez-Alavez, M., Ravizza, T., Frasca, A., Bartfai, T., et al., 2008. A novel non-transcriptional pathway mediates the proconvulsive effects of interleukin-1beta. Brain 131, 3256–3265.
Carcak, N., Aker, R.G., Ozdemir, O., Demiralp, T., Onat, F.Y., 2008. The relationship between age-related development of spike-and-wave discharges and the resistance to amygdaloid kindling in rats with genetic absence epilepsy. Neurobiol. Dis. 32, 355–363.
Danober, L., Deransart, C., Depaulis, A., Vergnes, M., Marescaux, C., 1998. Pathophys- iological mechanisms of genetic absence epilepsy in the rat. Prog. Neurobiol. 55, 27–57.
De Simoni, M.G., Perego, C., Ravizza, T., Moneta, D., Conti, M., Marchesi, F., et al., 2000. Inflammatory cytokines and related genes are induced in the rat hippocampus by limbic status epilepticus. Eur. J. Neurosci. 12, 2623–2633.
Depaulis, A., Van Luijtelaar, E.L., 2006. Genetic model of absence epilepsy in the rat. In: Pitkanen, A., et al. (Ed.), Models of Seizures and Epilepsy. Elsevier Inc, San Diego, CA, pp. 233–248.
Dubé, C., Vezzani, A., Behrens, M., Bartfai, T., Baram, T.Z., 2005. Interleukin-1beta contributes to the generation of experimental febrile seizures. Ann. Neurol. 57, 152–155.
Dubé, C.M., Ravizza, T., Hamamura, M., Zha, Q., Keebaugh, A., Fok, K., et al., 2010. Epileptogenesis provoked by prolonged experimental febrile seizures: mechanisms and biomarkers. J. Neurosci. 30, 7484–7494.
Dutuit, M., Didier-Bazes, M., Vergnes, M., Mutin, M., Conjard, A., Akaok, H., et al., 2000. Specific alteration in the expression of glial fibrillary acidic protein, glutamate dehydrogenase, and glutamine synthetase in rats with genetic absence epilepsy. Glia 32, 15–24.
Dutuit, M., Touret, M., Szymocha, R., Nehlig, A., Belin, M.F., Didier-Bazes, M., 2002. Decreased expression of glutamate transporters in genetic absence epilepsy rats before seizure occurrence. J. Neurochem. 80, 1029–1038.
Fellin, T., Gomez-Gonzalo, M., Gobbo, S., Carmignoto, G., Haydon, P.G., 2006. Astrocytic glutamate is not necessary for the generation of epileptiform neuronal activity in hippocampal slices. J. Neurosci. 26, 9312–9322.
Fellin, T., Halassa, M.M., Terunuma, M., Succol, F., Takano, H., Frank, M., et al., 2009. Endogenous nonneuronal modulators of synaptic transmission control cortical slow oscillations in vivo. Proc. Natl. Acad. Sci. U. S. A. 106, 15037–15042.
Gulhan Aker, R., Tezcan, K., Carcak, N., Sakalli, E., Akin, D., Onat, F.Y., 2010. Localized cortical injections of ethosuximide suppress spike-and-wave activity and reduce the resistance to kindling in genetic absence epilepsy rats (GAERS). Epilepsy Res. 89, 7–16.
Halassa, M.M., Fellin, T., Takano, H., Dong, J.H., Haydon, P.G., 2007. Synaptic Belnacasan islands defined by the territory of a single astrocyte. J. Neurosci. 27, 6473–6477.
Heida, J.G., Pittman, Q.J., 2005. Causal links between brain cytokines and experimental febrile convulsions in the rat. Epilepsia 46, 1906–1913.
Hu, S., Sheng, W.S., Ehrlich, L.C., Peterson, P.K., Chao, C.C., 2000. Cytokine effects on glutamate uptake by human astrocytes. Neuroimmunomodulation 7, 153–159.
Kovacs, Z., Kekesi, K.A., Szilagyi, N., Abraham, I., Szekacs, D., Kiraly, N., et al., 2006. Facilitation of spike-wave discharge activity by lipopolysaccharides in Wistar Albino Glaxo/Rijswijk rats . Neuroscience 140, 731–742.
Kreutzberg, G.W., 1996. Microglia: a sensor for pathological events in the CNS. Trends Neurosci. 19, 312–318.
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.
Marchi, N., Fan, Q., Ghosh, C., Fazio, V., Bertolini, F., Betto, G., et al., 2009. Antagonism of peripheral inflammation reduces the severity of status epilepticus. Neurobiol. Dis. 33, 171–181.
Marcon, J., Gagliardi, B., Balosso, S., Maroso, M., Noé, F., Morin, M., et al., 2009. Age- dependent vascular changes induced by status epilepticus in rat forebrain: implications for epileptogenesis. Neurobiol. Dis. 34, 121–132.
Marescaux, C., Vergnes, M., Depaulis, A., 1992. Genetic absence epilepsy in rats from Strasbourg—a review. J. Neural Transm. Suppl. 35, 37–69.
Maroso, M., Balosso, S., Ravizza, T., Iori, V., Wright, C., French, J., et al., 2011. ICE/Caspase 1 inhibition reduces acute seizures and drug resistant chronic epileptic activity in mice. Neurotherapeutics 8, 304–315.
Meeren, H.K., Pijn, J.P., Van Luijtelaar, E.L., Coenen, A.M., Lopes da Silva, F.H., 2002. Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J. Neurosci. 22, 1480–1495.
Melo, T.M., Sonnewald, U., Touret, M., Nehlig, A., 2006. Cortical glutamate metabolism is enhanced in a genetic model of absence epilepsy. J. Cereb. Blood Flow Metab. 26, 1496–1506.
Melo, T.M., Sonnewald, U., Bastholm, I.A., Nehlig, A., 2007. Astrocytes may play a role in the etiology of absence epilepsy: a comparison between immature GAERS not yet expressing seizures and adults. Neurobiol. Dis. 28, 227–235.
Morgan, J.I., Cohen, D.R., Hempstead, J.L., Curran, T., 1987. Mapping patterns of c-fos expression in the central nervous system after seizure. Science 237, 192–197.
Niquet, J., Ben-Ari, Y., Represa, A., 1994. Glial reaction after seizure induced hippocampal lesion: immunohistochemical characterization of proliferating glial cells. J. Neurocytol. 23, 641–656.
Onat, F.Y., Aker, R.G., Gurbanova, A.A., Ates, N., van Luijtelaar, G., 2007. The effect of generalized absence seizures on the progression of kindling in the rat. Epilepsia 48 (Suppl 5), 150–156.
Paxinos, G., Watson, C., 2005. The Rat Brain in Stereotaxic Coordinates. Academic Press, New York.
Polack, P.O., Guillemain, I., Hu, E., Deransart, C., Depaulis, A., Charpier, S., 2007. Deep layer somatosensory cortical neurons initiate spike-and-wave discharges in a genetic model of absence seizures. J. Neurosci. 27, 6590–6599.
Polack, P.O., Mahon, S., Chavez, M., Charpier, S., 2009. Inactivation of the somatosensory cortex prevents paroxysmal oscillations in cortical and related thalamic neurons in a genetic model of absence epilepsy. Cereb. Cortex 19, 2078–2091.
Randle, J.C., Harding, M.W., Ku, G., Schonharting, M., Kurrle, R., 2001. ICE/Caspase-1 inhibitors as novel anti-inflammatory drugs. Expert Opin. Investig. Drugs 10, 1207–1209.
Ravizza, T., Boer, K., Redeker, S., Spliet, W.G., van Rijen, P.C., Troost, D., et al., 2006a. The IL-1beta system in epilepsy-associated malformations of cortical development. Neurobiol. Dis. 24, 128–143.
Ravizza, T., Lucas, S.M., Balosso, S., Bernardino, L., Ku, G., Noé, F., et al., 2006b. Inactivation of caspase-1 in rodent brain: a novel anticonvulsive strategy. Epilepsia 47, 1160–1168.
Ravizza, T., Gagliardi, B., Noé, F., Boer, K., Aronica, E., Vezzani, A., 2008a. Innate and adaptive immunity during epileptogenesis and spontaneous seizures: evidence from experi- mental models and human temporal lobe epilepsy. Neurobiol. Dis. 29, 142–160.
Ravizza, T., Noé, F., Zardoni, D., Vaghi, V., Sifringer, M., Vezzani, A., 2008b. Interleukin Converting Enzyme inhibition impairs kindling epileptogenesis in rats by blocking astrocytic IL-1beta production. Neurobiol. Dis. 31, 327–333.
Rizzi, M., Perego, C., Aliprandi, M., Richichi, C., Ravizza, T., Colella, D., et al., 2003. Glia activation and cytokine increase in rat hippocampus by kainic acid-induced status epilepticus during postnatal development. Neurobiol. Dis. 14, 494–503.
Sagar, S.M., Sharp, F.R., Curran, T., 1988. Expression of c-fos protein in brain: metabolic mapping at the cellular level. Science 240, 1328–1331.
Schmitz, C., Hof, P.R., 2005. Design-based stereology in neuroscience. Neuroscience 130, 813–831.
Shapiro, L.A., Wang, L., Ribak, C.E., 2008. Rapid astrocyte and microglial activation following pilocarpine-induced seizures in rats. Epilepsia 49 (Suppl 2), 33–41.
Slaght, S.J., Leresche, N., Deniau, J.M., Crunelli, V., Charpier, S., 2002. Activity of thalamic reticular neurons during spontaneous genetically determined spike and wave discharges. J. Neurosci. 22, 2323–2334.
Stack, J.H., Beaumont, K., Larsen, P.D., Straley, K.S., Henkel, G.W., Randle, J.C., et al., 2005. IL-converting enzyme/caspase-1 inhibitor VX-765 blocks the hypersensitive response to an inflammatory stimulus in monocytes from familial cold autoinflammatory syndrome patients. J. Immunol. 175, 2630–2634.
Vergnes, M., Marescaux, C., Depaulis, A., Micheletti, G., Warter, J.M., 1986. Ontogeny of spontaneous petit mal-like seizures in Wistar rats. Brain Res. 395, 85–87.
Vezzani, A., Conti, M., De Luigi, A., Ravizza, T., Moneta, D., Marchesi, F., et al., 1999. Interleukin-1beta immunoreactivity and microglia are enhanced in the rat hippocampus by focal kainate application: functional evidence for enhancement of electrographic seizures. J. Neurosci. 19, 5054–5065.
Vezzani, A., Moneta, D., Conti, M., Richichi, C., Ravizza, T., De Luigi, A., et al., 2000. Powerful anticonvulsant action of IL-1 receptor antagonist on intracerebral injection and astrocytic overexpression in mice. Proc. Natl. Acad. Sci. U. S. A. 97, 11534–11539.
Vezzani, A., Ravizza, T., Balosso, S., Aronica, E., 2008. Glia as a source of cytokines: implications for neuronal excitability and survival. Epilepsia 49 (Suppl 2), 24–32.
Vezzani, A., French, J., Bartfai, T., Baram, T.Z., 2011. The role of inflammation in epilepsy. Nat. Rev. Neurol. 7, 31–40.
Viviani, B., Bartesaghi, S., Gardoni, F., Vezzani, A., Behrens, M.M., Bartfai, T., et al., 2003. Interleukin-1beta enhances NMDA receptor-mediated intracellular calcium in- crease through activation of the Src family of kinases. J. Neurosci. 23, 8692–8700. Wallengren, C., Li, S., Morris, M.J., Jupp, B., O’Brien, T.J., 2005. Aggravation of absence seizures by carbamazepine in a genetic rat model does not induce neuronal c-Fos activation. Clin. Neuropharmacol. 28, 60–65.
Ye, Z.C., Sontheimer, H., 1996. Cytokine modulation of glial glutamate uptake: a possible involvement of nitric oxide. Neuroreport 7, 2181–2185.
Zhang, X., Le Gal La Salle, G., Ridoux, V., Yu, P.H., Ju, G., 1996. Fos oncoprotein expression in the rat forebrain following muscimol-induced absence seizures. Neurosci. Lett. 210, 169–172.