SEARCH

SEARCH BY CITATION

Keywords:

  • Inflammation;
  • Blood–brain barrier;
  • Gliosis;
  • Temporal lobe epilepsy;
  • Status epilepticus

Summary

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Purpose:  Previous studies have shown that inhibition of the mammalian target of rapamycin (mTOR) pathway with rapamycin prevents epileptogenesis after pharmacologically induced status epilepticus (SE) in rat models of temporal lobe epilepsy. Because rapamycin is also known for its immunosuppressant properties we hypothesized that one of the mechanisms by which it exerts this effect could be via suppression of brain inflammation, a process that has been suggested to play a major role in the development and progression of epilepsy.

Methods:  Rats were treated with rapamycin or vehicle once daily for 7 days (6 mg/kg/day, i.p.) starting 4 h after the induction of SE, which was evoked by electrical stimulation of the angular bundle. Hereafter rapamycin was administered every other day until rats were sacrificed, 6 weeks after SE. Video-electroencephalography was used to monitor the occurrence of seizures. Neuronal death, synaptic reorganization, and microglia and astrocyte activation were assessed by immunohistologic staining. Fluorescein was administered to quantify blood–brain barrier leakage.

Key Findings:  Rapamycin treatment did not alter SE severity and duration compared to vehicle treatment rats. Rapamycin-treated rats developed hardly (n = 9) or no (n = 3) seizures during the 6-week treatment, whereas vehicle-treated rats showed a progressive increase of seizures starting 1 week after SE (mean 8 ± 2 seizures per day during the sixth week). Cell loss and sprouting that normally occur after SE were prominent but on average significantly less in rapamycin-treated rats versus vehicle-treated rats. Nevertheless, various inflammation markers (CD11b/c and CD68) were dramatically upregulated and not significantly different between post-SE groups. Of interest, blood–brain barrier leakage was barely detected in the rapamycin-treated group, whereas it was prominent in the vehicle-treated group.

Significance:  mTOR inhibition led to strong reduction of seizure development despite the presence of microglia activation, suggesting that effects of rapamycin on seizure development are not due to a control of inflammation. Whether the effects on blood–brain barrier leakage in rapamycin-treated rats are a consequence of seizure suppressing properties of the drug, or contribute to a real antiepileptogenic effect still needs to be determined.

Temporal lobe epilepsy (TLE) is one of the most common forms of epilepsy in adult patients. It is generally believed that TLE occurs after an initial insult that starts the process of epileptogenesis and leads to TLE, which in 30–40% of the patients cannot be adequately treated with antiepileptic drugs. Therefore, a major objective is to develop a treatment that could prevent epilepsy in patients with high risk of developing epilepsy (Baulac & Pitkanen, 2009; Kelley et al., 2009). However, although some antiepileptogenic or epilepsy-modifying strategies in experimental animal models are promising, an antiepileptogenic drug for patients does not yet exist (Dichter, 2009; Loscher & Brandt, 2010). Recently, rapamycin (RAP), the inhibitor of mammalian target of rapamycin (mTOR) pathway, appears to be a drug that can significantly reduce, and in some instances even prevent the development of epilepsy in animal models (Ljungberg et al., 2006; Zeng et al., 2008, 2009a; Wong, 2010). In the case of TLE, treatment with RAP has been shown to reverse the neuropathology such as mossy fiber sprouting, neuronal cell loss, and neurogenesis, processes that are believed to contribute to epileptogenesis and epilepsy (Buckmaster et al., 2009; Wong, 2010). Although the obtained neuroprotective effect suggests that brain inflammation might also be reduced by RAP, this has not been investigated until now. Recently, it has been reported that RAP reduces brain inflammation after brain injury such as traumatic brain injury and focal ischemia in rat (Erlich et al., 2007; Chauhan et al., 2011). Because accumulating evidence suggests that brain inflammation is an important mechanism that promotes epileptogenesis and seizure development (for review see Vezzani et al., 2011), we hypothesized that the effect of RAP on seizure development might be achieved by an antiinflammatory action. To test this we determined the extent of status epilepticus (SE)–induced brain inflammation after chronic RAP treatment in the electrical stimulated post-SE rat. Activation of astroglia and microglial/macrophages were considered as markers of inflammation. First we confirmed the reported effects of RAP on seizure development. Next we studied the effects of RAP on several neuropathologic characteristics of TLE such as cell loss mossy fiber sprouting, gliosis, and blood–brain barrier leakage. Finally, we correlated these potential proepileptogenic mechanisms with the development of post-SE epilepsy and seizures after RAP treatment.

Materials and Methods

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Experimental animals

Adult male Sprague-Dawley rats (Harlan Netherlands, Horst, The Netherlands) weighing 300–500 g were used in this study, which was approved by the university animal welfare committee. The rats were housed individually in a controlled environment (21 ± 1°C; humidity 60%; lights on 08:00 a.m. to 8:00 p.m.; food and water available ad libitum).

Status epilepticus induction

SE was induced via electrical stimulation of the angular bundle. For more details see Data S1. Behavior was observed during electrical stimulation and several hours thereafter. SE severity was classified as follows: score 1, mild SE: Racine’s scale I and II seizures and exploratory behavior, which is different from natural exploratory behavior, since the rats do not walk at a continuous pace; score 2, moderate SE: head bobbing, periods of immobility and occasional generalized seizures (stage IV or V); and score 3, severe SE: continuous behavioral stage IV–V seizure activity. Immediately after termination of the stimulation, periodic epileptiform discharges (PEDs) occurred at a frequency of 1–2 Hz and lasted several hours (SE). During this period rats had frequent seizures as observed by both their behavior and electroencephalography (EEG). The end of SE could be clearly defined by the disappearance of 1–2 Hz PEDs.

Rapamycin treatment

A stock solution of RAP was made in 100% ethanol (150 mg/ml) and stored at −20°C until use. RAP was diluted prior to use until 4% ethanol with vehicle (VEH) solution (5% Tween 80 + 5% polyethylene glycol 400 [PEG] 400). RAP was given intraperitoneally (6 mg/kg/day under isoflurane anesthesia) starting 4 h after status epilepticus (SE), once daily for 7 days (n = 12). This specific time point was chosen, rather than shortly after SE, since we wanted to investigate the effects on spontaneous seizures, which requires at least 4 h of SE in this animal model (Gorter et al., 2001). Hereafter RAP was given every other day until rats were sacrificed. In addition, VEH SE rats (n = 8) and RAP-treated controls (n = 8) were included. The dose of RAP was based on previous experiments in epileptic rats in which it has been shown to reduce seizure activity (Zeng et al., 2009b).

Immunocytochemistry

The brains were cut on a sliding microtome and 40-μm horizontal sections were collected in 0.1 m phosphate buffer for immunocytochemistry. Horizontal sections between 4,100 and 4,280 μm below cortex surface (dorsal level), 5,100–5,600 μm below cortex surface (mid level), and 7,600–8,100 μm below cortex surface (ventral level) of the contralateral brain part (according to Paxinos & Watson, 1998) of control and post-SE rats were stained with different immunocytochemical markers. Sections were washed in 0.05 m phosphate buffered saline (PBS), pH 7.4 and incubated for 30 min in 0.3% hydrogen peroxide in PBS to inactivate endogenous peroxidase. Sections were then washed (2 × 10 min) in 0.05 m PBS, followed by washing (1 × 60 min) in PBS + 0.5% Triton X-100 + 0.4% bovine serum albumin (BSA). Sections were incubated with mouse-anti neuron specific nuclear protein (NeuN; 1:1,000, MAB377; Millipore, Amsterdam, The Netherlands), mouse-anti Vimentin (VIM, V9; 1:100, M0725; Dako, Glostrup, Denmark), mouse anti-rat CD11b/c (OX-42; 1:100; PharMingen, San Diego, CA, U.S.A.), mouse-anti CD68 (ED1; 1:100, MAB 1435; Millipore), monoclonal rabbit anti-S6 (1:200, 5G10; Cell Signaling Technology, Danvers, MA, U.S.A.), or monoclonal rabbit anti-phospho-S6 (1:200, 91B2; Cell Signaling Technology) in PBS+0.1% Triton X-100 + 0.4% BSA at 4°C. Twenty-four hours after the incubation with the primary antibody, the sections were washed in PBS (3 × 10 min) and then incubated for 1.5 h in biotinylated sheep anti-mouse or anti-rabbit Ig (GE Healthcare, Diegem, Belgium), diluted 1:200 in PBS + 0.1% Triton X-100. This was followed by incubation for 60 min in AB-mix (Vectastain ABC kit, Peroxidase Standard pk-4000; Vector Laboratories, Burlingame, CA, U.S.A.). After washing (3 × 10 min) in 0.05 m Tris–HCl, pH 7.9, the sections were stained with 3,3′-diaminobenzidin tetrahydrochloride (50 mg DAB; Sigma-Aldrich, Zwijndrecht, The Netherlands) and 5 μl 30% hydrogen peroxide in a 10 ml solution of Tris–HCl. The staining reaction was followed under the microscope and stopped by washing the sections in Tris–HCl. After mounting on superfrost plus slides, the sections were air dried, dehydrated in alcohol and xylene, and coverslipped with Entellan (Merck, Darmstadt, Germany). The quantitative analysis is described in Data S1.

Blood–brain barrier permeability

To determine blood–brain barrier permeability, fluorescein (FSC; 100 mg/kg, i.v.; Merck) was injected via the tail vein under isoflurane anesthesia (4 vol%), 6 weeks after SE induction as described previously (van Vliet et al., 2007). EEG recordings were discontinued during anesthesia, which never lasted longer than several minutes. Rats were disconnected from the EEG recording set-up 3.5 h after tracer injection and deeply anesthetized with pentobarbital (Nembutal, intraperitoneally, 60 mg/kg). The animals were perfused through the ascending aorta with 300 ml of 0.37% Na2S solution and 300 ml of 4% paraformaldehyde in 0.1 m phosphate buffer, pH 7.4. The brains were postfixed in situ overnight at 4°C, dissected, and cryoprotected in 30% phosphate-buffered sucrose solution, pH 7.4. After overnight incubation at 4°C, the brains were frozen in isopentane (−30°C) and stored at −80°C until sectioning. The quantitative analysis is described in Data S1.

Statistical comparisons

Statistical comparisons were performed using Mann-Whitney U test. The number of seizures in the last week of each rat was related to the number of NeuN-positive hilar cells, the blood–brain barrier permeability index (see preceding text), and the number of CD11b/c positive cells, and the Spearman rank correlation was performed. p < 0.05 was assumed to indicate a significant difference.

Supplementary methods

See Data S1 for detailed description of electrode implantation and SE induction, video-EEG monitoring, Western blot, plasma levels of rapamycin, C-reactive protein, interleukin [IL]-1β and interleukin-6, quantitative analysis, and Timm staining.

Results

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Status epilepticus and development of spontaneous seizures

At 4 h after the start of the SE, rats were divided into two groups based on their SE behavioral severity scores and injected with either RAP or VEH under light anesthesia. After the injection, SE behavior was strongly diminished in both groups. However, the PEDs continued and were sometimes interrupted by individual EEG seizures that were accompanied by behavioral stage IV/V seizures. The end of SE was defined by the disappearance of 1–2 Hz PEDs. Stimulation intensity and behavioral parameters during stimulation were not significantly different between groups and are all indicated in Table S1. The average SE duration (PED duration) was also not significantly different between groups (Table S1).

Eventually 3 of 12 RAP-treated rats did not develop epilepsy during the 6 weeks of EEG recording, whereas all VEH-treated rats (n = 9) developed spontaneous seizures. The other RAP-treated rats exhibited significant fewer seizures during epileptogenesis, compared to VEH-treated rats (Fig. 1A), which was already evident at the second week after SE induction. A progressive increase in seizure frequency was observed in the VEH-treated group throughout the 6 weeks of EEG monitoring, whereas the number of spontaneous seizures did not increase in the RAP-treated group within this time frame (Fig. 1A). RAP-treated rats had on average 0.5 ± 0.2 seizures/day during the week before they were sacrificed, which was significantly less than VEH-treated rats (8.3 ± 2.4 seizures/day).

image

Figure 1.   Effects of rapamycin treatment on spontaneous seizure development. (A) The average number of spontaneous seizures per day versus time after SE (during 6 weeks). Epileptogenesis was significantly different (p < 0.05) between VEH- and RAP-treated epileptic rats from 2 weeks after SE induction until the end of the experiment at 6 weeks after SE. (B) The seizure duration was significantly shorter in RAP-treated rats compared to VEH-treated rats 5 and 6 weeks after SE (p < 0.05).

Download figure to PowerPoint

Seizure duration was not different throughout epileptogenesis between groups until 4 weeks after SE (Fig. 1B). Hereafter, seizure duration was shorter in RAP-treated rats (47 ± 6 s) compared to VEH-treated rats (55 ± 2 s).

Seizure behavior was quantified during epileptogenesis using continuous video recordings. Only partial seizures were observed during the first seizures after the latent period, both in RAP and in VEH-treated rats. Hereafter partial and generalized seizures were observed in both groups. The ratio of generalized/partial seizures was significantly greater in RAP-treated rats during weeks 2, 3, and 4 after SE (53–82% generalized seizures/18–46% partial seizures), compared to VEH-treated rats (17–58% generalized seizures/41–82% partial seizures). Hereafter, no differences were evident (p > 0.05) between VEH and RAP-treated rats (VEH 31–34% generalized seizures/65–69% partial seizures versus RAP 26–32% generalized seizures/67–73% partial seizures).

Activation of the mTOR pathway after electrically induced SE

We determined whether the mTOR pathway was activated after electrically induced SE. The ratio of p-S6/S6 was significantly increased in VEH-treated post-SE rats (3 days after SE) compared to controls (Fig. S1A). After 3-day RAP-treatment in post-SE rats the ratio was similar to control values, indicating that RAP treatment could attenuate seizure induced overexpression of p-S6 and that RAP treatment was effective in inhibiting the mTOR pathway in our animal model. To assess the extent of mTOR activation in the chronic phase, we performed immunostaining of S6 and pS6 in sections of 6-week post-SE rats, and unstimulated controls. The effects of chronic RAP treatment were also investigated. In controls, P-S6 immunoreactivity (IR) was detected at a very low level in some neurons (Fig. S1C). The nonphosphorylated S6 protein was present in all neurons and to lesser extent in glial cells throughout the hippocampus. In VEH-treated epileptic rats, a significant increase of P-S6 IR was observed in neurons, whereas S6 IR was slightly, but significantly, increased in glial cells at all hippocampal levels (Fig. S1D,F,G) compared to controls. RAP-treatment effectively inhibited neuronal expression of P-S6 as shown by the strongly decreased staining compared to VEH-treated epileptic rats (Fig. S1E), although it was still somewhat higher than in controls (Fig. S1F).

Rapamycin and inflammatory protein plasma levels

Blood samples for RAP analysis were collected 6 weeks after SE at different intervals after a single RAP injection. RAP plasma levels were 54 ± 3 μg/L (4 h after a single RAP injection), 28 ± 3 μg/L (24 h after the first bolus injection), and 25 ± 2 μg/L (48 h after the first bolus injection) indicating that RAP levels were sufficient.

Because increased levels of cytokines in blood plasma are considered markers of peripheral inflammation, we also measured plasma levels of IL-1β and the acute phase proteins IL-6 and C-reactive protein. However, plasma levels of IL-1β, IL-6, and C-reactive protein were not different between control, VEH-, and RAP-treated rats and were all below detection limit (respectively, 5 pg/ml, 21 pg/ml, and 1 mg/l).

SE-induced hilar cell loss and mossy fiber sprouting

Because hilar cell loss is a common finding in TLE patients and in the TLE rat model, we quantified the number of cells in this region. Hilar cell loss was obvious in VEH-treated epileptic rats (Fig. 2B) in the dorsal, mid, and ventral hippocampus compared to controls (Fig. 2A). Quantification of the number of hilar neurons in VEH-treated epileptic rats revealed a significant decrease compared to controls (Fig. 2G) at all levels. RAP-treated epileptic rats exerted significant cell loss compared to controls (Fig. 2C,G), but had significant less cell loss compared to VEH-treated epileptic rats, indicating that RAP had a neuroprotective effect.

image

Figure 2.   Effects of rapamycin on hilar cell loss and MFS. NeuN staining in the hippocampus of control RAP-treated rats (A), VEH-treated epileptic rats (B), and RAP-treated epileptic rats (C). Quantification of the number of hilar neurons at three different horizontal levels of the hilar region revealed a significant decrease of hilar cells in both VEH-treated and RAP-treated epileptic rats compared to controls (G). However, significant fewer cells were lost in RAP-treated epileptic rats, compared to VEH-treated rats (G). Timm’s staining of the hippocampus in control RAP-treated rats (D), VEH-treated epileptic rats (E), and RAP-treated epileptic rats (F). Semiquantitative score of Timm’s staining in the dentate gyrus inner molecular layer at the dorsal, mid, and ventral levels revealed MFS in VEH- and RAP-treated epileptic rats compared to controls (H). However, less MFS was observed in RAP-treated epileptic rats compared to VEH-treated epileptic rats (H). *Significantly different from controls (p < 0.05); #Significantly different from VEH-treated epileptic rats (p < 0.05).

Download figure to PowerPoint

Extensive mossy fiber sprouting was evident in VEH-treated epileptic rats (Fig. 2E,H). This was not observed in controls (Fig. 2D) at the dorsal and mid level of the hippocampus, whereas at the ventral level a low Timm score was observed in these rats. RAP-treated epileptic rats had mossy fiber sprouting at all hippocampal levels, but significantly less compared to VEH-treated epileptic rats (Fig. 2F,H).

CD11b/c and CD68 immunoreactive monocytes

CD11b/c is a marker for microglia cells in the brain. Using Western blotting we showed that CD11b/c significantly increased 3 days after SE in both VEH-treated and RAP-treated rats, compared to controls. No significant differences were detected between VEH- and RAP-treated epileptic rats (Fig. S1B). Immunostaining showed that CD11b/c IR cells were abundantly present throughout the hippocampus of control rats (median number score 2) with moderate IR (score 1), with a typical morphology of resting microglia (Fig. 3A and Table S2). The number and IR of CD11b/c positive cells significantly increased in VEH-treated epileptic rats compared to controls (Fig. 3B and Table S2) at all hippocampal levels. CD11b/c cells that had an increased IR had a typical morphology of activated microglia. CD11b/c positive cells that had a low IR score had morphology of resting microglia cells. A similar increase in the number of CD11b/c and IR was observed in RAP-treated epileptic rats, which were not different from VEH-treated epileptic rats (Fig. 3C and Table S2).

image

Figure 3.   Effects of rapamycin on microglia/macrophage and monocyte activation. CD11b/c IR cells were abundantly present throughout the hippocampus of control rats, with a typical morphology of resting microglia (A). The number and IR of CD11b/c positive cells significantly increased in VEH-treated epileptic rats compared to controls (B). CD11b/c cells that had increased IR had a typical morphology of activated microglia. A similar increase in the number and IR of CD11b/c cells has been observed in RAP-treated epileptic rats, which were not different from VEH-treated epileptic rats (C). CD68 staining in the hippocampus of RAP-treated control rats (D), VEH-treated epileptic rats (E), and RAP-treated epileptic rats (F). No CD68 staining in individual cells was observed in control rats (D). In contrast, CD68 positive cells were abundantly present in VEH-treated epileptic rats (E) and RAP-treated epileptic rats (F). No significant differences were observed in the number of CD68 positive cells between VEH and RAP-treated epileptic rats.

Download figure to PowerPoint

CD68 is a marker for macrophages/monocytes. CD68 staining was absent in control rats (Fig. 3D). In contrast, CD68 positive cells were abundantly present throughout all hippocampal subfields of VEH-treated epileptic rats (Fig. 3E) and RAP-treated epileptic rats (Fig. 3F). No significant differences were observed in the number of CD68 positive cells between VEH- and RAP-treated epileptic rats (Table S3).

Vimentin immunoreactive astrocytes

The intermediate filament protein VIM was used to detect reactive astrocytes. In controls, VIM was expressed only in blood vessels (Fig. 4A). The number and IR of VIM positive cells in the hippocampus increased significantly in VEH-treated epileptic rats compared to controls (Fig. 4B, Table S4A,B) along the septal-temporal axis. These cells had a typical morphology of activated astrocytes. A similar increase in the number of VIM cells and IR was observed in RAP-treated epileptic rats, which were not different from VEH-treated epileptic rats (Fig. 4C, Table S4A,B).

image

Figure 4.   Effects of rapamycin on astrogliosis. The intermediate filament protein VIM was used to detect reactive astrocytes. In controls, VIM was expressed in only some blood vessels (A). The number and IR of VIM positive cells significantly increased in VEH-treated epileptic rats compared to controls (B) at all hippocampal levels. These cells had a typical morphology of activated astrocytes. A similar increase in the number of VIM and IR was observed in RAP-treated epileptic rats, which were not different from VEH-treated epileptic rats (C).

Download figure to PowerPoint

Blood–brain barrier permeability

Bloodbrain barrier leakage, as detected by fluorescein (FSC) brain entry, was evident in VEH-treated epileptic rats in hippocampal brain regions including dentate gyrus, CA3, and CA1 (Fig. 5B,D). No leakage was observed in the cerebellum of these rats (which is not involved in temporal lobe epilepsy), indicating region specific bloodbrain barrier damage. No FSC was visible in any brain region of control rats (Fig. 5A). BBB leakage was significantly reduced in all investigated brain regions of RAP-treated rats compared to VEH-treated rats (Fig. 5C,D).

image

Figure 5.   Effects of rapamycin treatment on bloodbrain barrier permeability. Bloodbrain barrier permeability was assessed in various brain regions by fluorescein entry. Fluorescein was not detected in the hippocampus of control rats (A), indicating an intact bloodbrain barrier. Bloodbrain barrier leakage, as detected by fluorescein brain entry (green dots), was evident in VEH-treated epileptic rats (B and D) in all analyzed brain regions, except the cerebellum. In RAP-treated epileptic rats (C) fluorescein could be detected (except the cerebellum), but the fluorescein permeability index was significantly reduced in all analyzed brain regions of RAP-treated rats compared to VEH-treated rats (D). gcl, granule cell layer, ml, molecular layer. *Significantly different from controls (p < 0.05); #Significantly different from VEH-treated epileptic rats (p < 0.05).

Download figure to PowerPoint

Correlations between seizure development, structural changes, and inflammation

To determine whether there is a correlation between seizure progression and any of the measured histologic parameters we constructed three-dimensional graphs with the individual data points of RAP- and VEH-treated post-SE rats and controls.

Figure 6A shows the relationship between seizure progression, hilar cell loss, and mossy fiber sprouting (MFS). The graph shows that SE led to significant cell loss and MFS. However, cell loss and MFS were less in 8 of 12 RAP-treated rats compared to the VEH-treated rats. In four RAP-treated rats the extent of MFS or cell loss was comparable to VEH-treated epileptic rats, indicating that considerable sprouting and cell loss occurred without obvious development of daily seizures. Figure 6B shows the relationship between seizure progression, microglia activation, and blood–brain barrier permeability. There was no correlation between microglia activation and seizure progression. Only very few (or even no) seizures were observed in individual rats that showed considerable hippocampal CD11b/c or CD68 staining. However, a clear positive correlation between blood–brain barrier permeability and seizure progression could be observed (Spearman rank, p < 0.05).

image

Figure 6.   Three-dimensional plots with the individual data points of RAP- and VEH-treated post-SE rats and controls. (A) The relationship between seizure frequency, the number of hilar cells, and MFS. SE leads to significant cell loss and MFS; however, in 8 of 12 RAP-treated rats this was reduced if compared to the VEH-treated rats. In four RAP-treated rats the extent of MFS or cell loss was comparable to VEH-treated epileptic rats, indicating that considerable MFS and cell loss can occur without obvious development of daily seizures. (B) The relationship between seizure frequency, microglia activation (CD11b/c score), and blood–brain barrier permeability (fluorescein permeability index). There is no correlation between microglia activation and seizure progression. No, or only very few, seizures were observed in individuals that showed considerable CD11b/c staining. However, a clear positive correlation between blood–brain barrier permeability and seizure progression can be observed (Spearman rank, p < 0.05).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

The main findings of this study can be summarized as follows: chronic RAP treatment starting 4 h after the start of electrically induced SE, strongly reduced—and in some rats even prevented—the development of epilepsy. RAP treatment also reduced SE-induced neuronal cell loss, mossy fiber sprouting, and bloodbrain barrier leakage. However, RAP did not reduce hippocampal microglia or astrocyte activation.

Reduction of seizure development by rapamycin

We observed that chronic RAP treatment effectively reduced the development of epilepsy. In 3 of 12 RAP-treated rats, epileptic seizures were never detected, a phenomenon that we have never observed in VEH-treated post-SE rats or any other post-SE rat in experiments that we performed in the past. The effect on seizure development is comparable with the observations made by Wong et al. in the kainate TLE rat model (Zeng et al., 2009b). The investigators could prevent the development of epilepsy with RAP pretreatment; RAP posttreatment starting 1 day after SE led to a strong reduction in seizure development. However, using the amygdala stimulation TLE rat model, Sliwa et al. (2012) showed that when treatment was continued for only 2 weeks, no effect on seizure development (monitored for 6 weeks) was detected, suggesting that treatment has to continue for at least >2 weeks.

As in previous publications related to RAP treatment in post-SE animals (Zeng et al., 2009b; Sliwa et al., 2012), RAP-treated rats were compared to VEH-treated rats that received ethanol as solvent for RAP. Because the volume is rather small (approximately 0.4 ml of 4% ethanol, i.p.) this did not affect the normal progression of seizure development. Therefore, we can assume that the observed effects on seizure development in the RAP group is produced by RAP itself and not by its solvent.

In our study, EEG monitoring was assessed during RAP treatment so that we cannot exclude that RAP produces the effect on seizure development via an anticonvulsive action. However, we have reasons to assume that RAP is not anticonvulsive. First, the duration and severity of the SE were not affected by RAP in our model, which was similar to the observation previously made in RAP-treated kainate rats (Zeng et al., 2009b). This may be explained by the fact that it is difficult to pharmacologically block a self-sustained SE. Secondly, several electrophysiologic in vitro studies indicate that RAP does not change excitability in rat hippocampal slices, measured either with evoked field potential (Daoud et al., 2007) or sodium and potassium currents and single unit recordings of individual hippocampal neurons (Ruegg et al., 2007). Finally, no effects on seizure frequency could be detected in mice using the same RAP treatment protocol (Buckmaster & Lew, 2011). Nevertheless, more recent studies suggest that we cannot entirely rule out an anticonvulsant effect of RAP, since anticonvulsant effects can be found when the mTOR pathway is activated: Huang et al. (2010) showed a reduction in spontaneous seizures in chronic epileptic kainate-treated rats. Cepeda et al. (unpublished observations, data presented during the 11th Workshop on the Neurobiology of Epilepsy 2011, Grottaferrata, Italy) showed that RAP reduces cortical excitability in cortical slices of phosphatase and tensin homolog (PTEN) knockout mice (in which the mTOR pathway is activated). In the present study we started chronic RAP treatment 4 h after SE and continued RAP treatment until rats were sacrificed 6 weeks later. In future studies we will use other RAP treatment paradigms (e.g., start RAP treatment earlier/later or discontinue RAP treatment before rats are sacrificed) in order to determine whether RAP can have truly antiepileptogenic effects.

Neuronal cell loss and mossy fiber sprouting

Neuronal cell loss and MFS are two biologic processes that are not only a consequence of SE but that also are believed to play a proepileptogenic role. Therefore, neuroprotection as well as suppression of MFS might have an antiepileptogenic function. Although suppression of MFS by RAP appears to be a consistent finding by different research groups, the effect on neuronal cell loss appears to depend on timing of the treatment. RAP treatment that started before SE did produce a significant neuroprotective effect (Zeng et al., 2009b); however, this did not occur in rats that in which treatment started 1 day post-SE in both kainate and pilocarpine models (Zeng et al., 2009b; Buckmaster & Lew, 2011). We started RAP treatment at 4 h after SE and observed less cell death and suppression of MFS compared to VEH-treated post-SE rats. However, significant cell loss and MFS were still observed in some RAP-treated rats that did not exhibit any significant progression of epilepsy. This indicates that RAP probably obtained its inhibitory effect on seizure development by mechanisms other than neuroprotection and suppression of MFS. Of interest, Buckmaster et al. (2009) noted suppression of MFS only as long as RAP was present. When RAP was discontinued for several weeks, a strong expression of MFS was observed. EEG monitoring was not performed in these studies. Future studies in which EEG is continuously monitored after discontinuation of RAP treatment need to be performed to get insight into whether RAP treatment is truly antiepileptogenic or just suppressing seizure activity. In the latter case seizure development will be only delayed, and will be manifest as soon as RAP treatment is discontinued.

Inflammation and astrogliosis

Based on recent literature and the immunosuppressant properties of RAP, we expected to find a reduction of microglia activation and inflammation after the treatment. This would have helped to explain RAP’s suppressing effect on epileptogenesis after SE, which, according to the hypothesis, can be modulated by antiinflammatory treatment (Vezzani et al., 2011). Recently, it has been reported that RAP produces a neuroprotective and antiinflammatory effect in the brain in a focal cerebral ischemia rat model (Chauhan et al., 2011); moreover, microglia activation is reduced by RAP in a closed head injury mouse model together with an increased neuroprotection (Erlich et al., 2007). However, contrary to our expectation we did not detect any containment of microglia or monocyte activation by RAP in the brain. Similarly, extensive astrogliosis was still evident after SE in RAP-treated rats. RAP treatment in mouse models for tuberous sclerosis complex reverses the histologic abnormalities, including astrocytosis (Zeng et al., 2008). We did not find any evidence that RAP decreased the number of reactive astrocytes. Vimentin staining was intense in various hippocampal areas. Despite the absence of any suppressing effect of RAP on parenchymal inflammation and gliosis, seizure development was strongly reduced. These data do not support the involvement of brain inflammation or gliosis as epileptogenic mechanism, although we cannot exclude that the drug affects specific inflammatory pathways, or that it targets inflammation at the level of the blood–brain barrier, leading to RAP-induced reduction of blood–brain barrier leakage (see following text).

Bloodbrain barrier leakage

Blood–brain barrier dysfunction is a common feature that is observed after SE (Lorenzo et al., 1972; Zucker et al., 1983; Lassmann et al., 1984; Cornford & Oldendorf, 1986). It may be caused by several mechanisms that include inflammation, metabolic failure, and angiogenesis (Oby & Janigro, 2006; Rigau et al., 2007; Shlosberg et al., 2010). Increased plasma levels of IL-1β and acute phase proteins might also increase blood–brain barrier leakage (Blamire et al., 2000). In our model, blood–brain barrier leakage is still evident at 6 weeks after SE. Previous studies have shown that manipulation and opening of the blood–brain barrier leads to the induction of seizures in pigs and human (Marchi et al., 2007), to an increase in seizure progression in rats (van Vliet et al., 2007), and to epileptiform activity in an in vitro blood–brain barrier model (Seiffert et al., 2004). This makes the blood–brain barrier an interesting target to investigate treatment of epileptogenesis or seizure progression (Fabene et al., 2008; Marchi et al., 2011a,b; Shorvon & Ferlisi, 2011). The decreased blood–brain barrier permeability in the RAP group was not due to decreased plasma levels of the proinflammatory cytokine IL-1β or changes in acute phase proteins such as IL-6 and C-reactive protein, since these did not appear to be changed at 6 weeks after SE. It is tempting to speculate that RAP can mediate an antiepileptogenic effect via restoration and strengthening of blood–brain barrier. We cannot exclude, however, that the reduced FSC permeability is just a consequence of the reduced number of seizures in RAP-treated post-SE rats. However, it has been recently reported that RAP decreases the paracellular permeability of porcine proximal tubular epithelial cells via modulation of the tight junction protein claudin-1 (Martin-Martin et al., 2010), which would support the notion of a restorative function of RAP. Whether this occurs at the level of the blood–brain barrier can be tested in a transcellular electrical resistance model using brain endothelial cells. In future studies we will also determine the time course of blood–brain barrier dysfunction during RAP treatment at different time points during epileptogenesis, to assess whether RAP can have direct effects on the blood–brain barrier.

Conclusion

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

RAP treatment leads to a suppression of seizure development in the rat TLE model. This suppression is not linked to a containment of microglia/macrophage activation after SE. Whether RAP affects specific inflammatory pathways needs to be studied in more detail. Of interest, blood–brain barrier leakage is strongly reduced by RAP treatment, suggesting that the barrier might be an important target. Whether reduction of BBB leakage plays a causative role in the potential antiepileptogenic effect of RAP will be subject of future studies.

Acknowledgments

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

This work was supported by The Netherlands Organisation for Scientific Research (NWO), Veni grant 863.08.017 (EAvV), Nationaal Epilepsie Fonds, grant 11-03 (JAG) and EU FP7 project NeuroGlia, Grant Agreement Number 2021 (EA).

Disclosure

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

None of the authors has any conflict of interest to disclose. The authors have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Data S1. Supplementary methods.

Figure S1. Western blot analysis of phosphorylated S6 (p-S6), S6 protein and CD11b/c in rat hippocampus and the effect of RAP on phosphorylated S6 (p-S6) and S6 protein expression.

Table S1. Electrical stimulation and status epilepticus parameters.

Table S2. (A) CD11b/c cell score. (B) CD11b/c immunoreactivity score.

Table S3. CD68 score.

Table S4. (A) Vimentin cell score. (B) Vimentin immunoreactivity score.

FilenameFormatSizeDescription
EPI_3513_sm_Methods-TableS1-S4.doc144KSupporting info item
EPI_3513_sm_FigS1.tif19024KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.