SEARCH

SEARCH BY CITATION

Keywords:

  • mTOR;
  • Astrogliosis;
  • Hippocampal sclerosis;
  • Mesial temporal lobe epilepsy (MTLE)

Summary

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Mammalian target of rapamycin (mTOR) is a key protein kinase that regulates basic cellular processes, including development and growth. Mutations in mTOR cause tuberous sclerosis complex (TSC), a condition that is characterized by developmental brain malformations (cortical tubers) and epilepsy. Although considerable insight has been gained recently into the pathologic dysfunction of mTOR in tubers in TSC-related epilepsy, data on the mTOR cascade in mesial temporal lobe epilepsy (MTLE) are lacking. Immunohistochemical investigation with confocal microscopy was performed to evaluate mTOR cascade and to correlate its activity with cellular alterations observed in surgically resected samples of human neocortex and hippocampus in MTLE. We compared results in human tissue to findings in the rat pilocarpine model of sclerotic MTLE. In nonsclerotic and control hippocampus, many neurons in the CA1 subfield expressed high levels of phospho-S6 (p-S6), a reliable marker of mTOR activation. In nonsclerotic and control hippocampus, as well as in magnetic resonance imaging (MRI) normal human neocortex, protoplasmic astrocytes did not express p-S6. In contrast, in sclerotic hippocampus, prominent p-S6 immunostaining was observed mainly in astrocytes and microglia located in the areas of neuronal loss and astrogliosis, whereas neurons in preserved areas of CA1 expressed significantly lower levels of p-S6 immunopositivity than neurons in nonsclerotic or control CA1 subfields. In surgically resected neocortex with chronic astroglial scar tissue, only microglia revealed moderate p-S6 immunoreactivity. Different from human sclerotic epileptic hippocampus, astrogliosis in the chronic rat pilocarpine model of epilepsy was not characterized by glial cells with mTOR activation. The mTOR cascade is activated in astroglial cells in sclerotic MTLE, but not in astrocytes in chronic neocortical scarring or in the pilocarpine model of MTLE. These findings suggest that the astroglial “scar” in sclerotic MTLE has active, ongoing cellular changes. Targeting mTOR in MTLE may provide new pathways for the medical therapy of epilepsy.

The mammalian target of rapamycin (mTOR) cascade is a critical pathway that censors many intracellular processes and extracellular signals and converges them into regulation of cell growth and development through a combination of anabolic and catabolic processes (Wullschleger et al., 2006). It is modulated by a variety of stimuli including trophic factors, cytokines, mitogens, hormones, intracellular ATP/ADP ratio, amount of amino acids, and cellular stress (ischemia, heat shock, DNA damage, and viral infections) (Swiech et al., 2008). Phosphorylation of two main direct downstream mTORC1 substrates S6K1 (p70 ribosomal S6 protein kinase 1) and 4EBP1 (eukaryotic initiation factor 4E-binding protein) leads to translational activation and subsequent cell growth. Phosphorylation of ribosomal protein S6 (S6), the major S6K1substrate responsible for ribosome activation, is considered a reliable immunohistochemical marker of mTOR activation, and is used widely in histologic analysis (Kawasome et al., 1998; Feliciano et al., 2011). It is worth noting that there are several phosphorylation sites in S6 that could be phosphorylated by different kinases. Therefore, Ser240 and Ser244 are specific for S6K1, whereas Ser 235 and Ser236 may be phosphorylated by S6K1 and also by p90 ribosomal S6 (p90RS) kinase in the Ras–mitogen-activated protein kinase pathway (Anjum & Blenis, 2008).

Aberrant mTOR activation due to gene mutations underlies severe developmental brain malformations in tuberous sclerosis complex (TSC) that manifests with epilepsy. mTOR-dependent cellular alterations that predispose to seizures have been studied in detail in TSC and supported by experimental models of the disease (Uhlmann et al., 2002; Wenzel et al., 2004; Zeng et al., 2007; Crino, 2009; Wong, 2010). Activation of the mTOR cascade in cytomegalic neurons and astrocytes was also observed and considered as a possible pathogenetic mechanism in cortical dysplasia (Ljungberg et al., 2006).

Mesial temporal lobe epilepsy (MTLE) is one of the most prevalent human epilepsy syndromes. Despite a long history of investigations of MTLE, the mTOR cascade in MTLE has been poorly studied. Herein we show that the mTOR cascade in sclerotic MTLE is differentially activated in glial cells in comparison to neurons, and that cellular involvement depends on the hippocampal pathology.

Material and Methods

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Human tissue specimens

Surgical specimens were obtained from patients with medically intractable epilepsy and as part of the resection of brain lesions in patients without epilepsy. All patient protocols were approved by the Columbia University Medical Center Institutional Review Board. We examined 21 nonsclerotic hippocampi (mean age at surgery, 31.4 years, age range 12–56 years) and 34 sclerotic hippocampi (mean age 42 years, age range 17–69 years). We also studied 14 samples of neocortex, all of which were without magnetic resonance imaging (MRI) or neuropathologically identified abnormality. These neocortical samples were acquired during the lateral temporal resection carried out to provide access to the mesial temporal structures as part of MTLE or benign brain lesional surgery. We also studied two samples of neocortex from epilepsy patients in which a chronic glial scar had developed: (1) following previous surgery 6 years earlier and (2) along the wall of a postischemic cyst. In addition, we studied three specimens from patients without seizures (mean age 61 years, age range 52–69 years) who had their hippocampi resected in association with adjacent lesions.

Animals

Male Sprague-Dawley rats (100–200 g) were used in the experiment under the guidelines approved by Columbia University’s Institutional Animal Care and Use Committee. After premedication with scopolamine (5 mg/kg, i.p.) to prevent the effects of peripheral cholinergic stimulation, pilocarpine (330 mg/kg, i.p.) was administered to induce seizures. Seizures were graded on the modified Racine scale (Morton & Leavens, 2000), and only animals with grade 4–5 seizures for 2 h were used in experiments. After 2 h of continuous seizures, ketamine (80 mg/kg, i.p.) was given to stop seizures. Rats were housed in standard cages with free access to food and water on a 12-h light/dark cycle. Animals were sacrificed in 8–12 months after induced seizures by intracardiac perfusion with 4% paraformaldehyde under deep anesthesia.

Histology and immunohistochemistry

Specimens were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 12–16 h (4°C), and 40-μm sections were prepared with a vibratome (Leica VT1000S, Leica Microsystems Inc., Buffalo, IL, U.S.A.) and stored in cryoprotectant solution at −20°C. Standard procedure for Nissl staining with cresyl violet was used for routine analysis of tissue.

Antibodies

Primary antibodies were used against:

  • 1
     markers of astroglial cells—(a) glial fibrillary acidic protein (GFAP): monoclonal (1:1,000, G3893; Sigma-Aldrich, St. Louis, MO, U.S.A.), rabbit polyclonal (1:1,000, Z 0334; Dako, Carpinteria, CA, U.S.A.), and chicken polyclonal (1:500, PCK-591P; Covance, Berkeley, CA, U.S.A.); (b) vimentin: monoclonal (1:500, M 0725; Dako); (c) nestin: rabbit polyclonal (1:500, PRB-570; Covance); (d) glutamine synthetase (GS): monoclonal (1:1,000, MAB302; Chemicon, Temecula, CA, U.S.A.) and rabbit polyclonal (1:200, sc-9067; Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A.); astrocyte specific glutamate transporters: (e) excitatory amino-acid transporter 1 (glutamate aspartate transporter): monoclonal (1: 100, clone 10D4; Novocastra Lab, Newcastle upon Tyne, United Kingdom); (f) glutamate transporter 1 (excitatory amino-acid transporter 2): mouse monoclonal (1:500, 611654; BD Transduction Lab., Franklin Lakes, NJ, U.S.A.);
  • 2
     markers specific for neurons—(a) microtubule-associated protein 2: mouse monoclonal (1:250, M9942; Sigma-Aldrich); (b) panneuronal neurofilament marker SMI 311: mouse monoclonal (1:1,000, SMI 311; Covance); and (c) markers of inhibitory interneurons: glutamate decarboxylase 65 (GAD65): mouse monoclonal (1:200, clone GAD-6, 559931; BD Pharmingen, Sparks, MD, U.S.A.) and parvalbumin: mouse monoclonal (1:1,000, clone PARV-19; Sigma-Aldrich);
  • 3
     markers of microglial cells—(a) cluster of differentiation 68 (CD68): monoclonal (1:300; M 0814; Dako, for human tissue) and CD68: monoclonal (1:100; MCA341GA; AbDSerotec, Raleigh, NC, U.S.A., for rat tissue); (b) LN-3: monoclonal (1:50; 69303; ICN Biomedicals, Inc., Aurora, OH, U.S.A.); and (c) ionized calcium binding adaptor molecule 1: rabbit polyclonal (1:500, CP 290 A, B; BiocareMadical, LLC, Concord, CA, U.S.A.);
  • 4
     marker of mTOR cascade activation—non-phospho-S6 ribosomal protein (S6): rabbit polyclonal (1:300, 2317; Cell Signaling Technology, Inc, Boston, MA, U.S.A.), phospho-S6 ribosomal protein (phosphorylated at Ser235/236) (p-S6(235/236)): rabbit polyclonal (1:100, 4857; Cell Signaling), phospho-S6 ribosomal protein (phosphorylated at Ser240/244) (p-S6(240/244)): rabbit polyclonal (1:100, 5364; Cell Signaling).

Note: only those antibodies that gave reliable and consistent immunostaining were included on the list.

Secondary antibodies include conjugated to fluorophores: anti-mouse Alexa Fluor 488, 594, and 633; anti-chicken Alexa Fluor 594, 633; anti-rabbit Alexa Fluor 594; all from goat or donkey (1:300; Molecular Probes, Eugene, OR, U.S.A.).

For light microscopy, sections were pretreated in 3% H2O2 in methanol (30 min) and after blocking in 10% donkey serum incubated overnight (4°C) with primary antibodies. After washing, sections were incubated with biotinylated secondary antibodies (1:1,000, 1 h, at room temperature [RT]) and then, after washing, with avidin–biotin-peroxidase complex (ABC, Vector Laboratories Inc., Burlingame, CA, U.S.A.) (1 h, RT). Peroxidase activity was visualized with 0.03% DAB (Sigma) with 0.005% H2O2 in 0.05 m Tris buffer. Sections were mounted on glasses, dehydrated in a graded ethanol, and included in DPX.

For double immunofluorescence, after blocking with 10% normal goat (or donkey) serum (30 min, at RT), free-floating sections were incubated in a mixture of primary antibodies raised in different species for overnight (4°C). For visualization, Alexa Fluor-conjugated secondary antibodies were used for 1 h at RT. Fluorescent Nissl reagent (NeuroTrace 640/660 deep-red, 1:150, Molecular Probes) was used (30 min, RT) for visualization of general histologic structure in double immunostaining. Blocking serum, primary and secondary antibodies, and fluorescent Nissl reagent were applied in 0.2% Triton X-100 in PBS. Sections for fluorescent microscopy were mounted on slides in Vectashield (Vector Lab).

To control the specificity of immunostaining, primary antibodies were omitted and substituted with appropriate normal serum.

Slides were viewed using a confocal microscope (Bio-Rad Radiance 2000, Nikon E800, Nikon Instruments Inc., Melville, NY, U.S.A.).

Quantitative immunohistochemical analysis

The numbers of immunolabeled cells were quantified in the images merged from a stack of adjacent six images (1,024 × 1,024 pixel resolution, observed area 606 × 606 μm) captured at a distance of 0.5 μm from each other. Only cells with clearly outlined nuclei were taken into consideration. Three sections were used from one specimen and at least five images were captured from each section.

Statistical analysis

Data are expressed as mean ± standard error of the mean (SEM). Student’s t-test with p < 0.05 was considered significant.

Results

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

The mTOR cascade is differentially activated in nonsclerotic and sclerotic hippocampus in MTLE

In nonsclerotic and control hippocampi, immunostaining with p-S6(235/236) was observed predominantly in neurons in CA1 subfield (Fig. 1A,B). Intensity of immunofluorescence in neurons varied, being especially high in polygonal and oval neurons (Fig. 1D,E). Many of these cells expressed markers of inhibitory neurons: GAD65 (∼65% GAD+ neurons were p-S6+, 166 of 250) and parvalbumin (∼55% paralbumin+ neurons were p-S6+, 140 of 250) (Fig. 1F,G) and were considered as inhibitory interneurons. It should be noted that pyramidal neurons in CA2 and CA3 areas did not show immunoreactivity for p-S6(235/236) (Fig. A1,B); some p-S6(235/236) immunopositive neurons found in these subfields immunostained for GAD65 and parvalbumin (Fig. 1H,I). Granule neurons in the dentate gyrus did not show immunoreactivity for p-S6(235/236). There was little detected p-S6 astroglial expression in either control or nonsclerotic hippocampus.

image

Figure 1.   Differential expression of S6 and p-S6(235/236)/(240/244) in nonsclerotic (A, B) and sclerotic (C) hippocampi. Note: in nonsclerotic hippocampus many neurons in Ammon’s horn are p-S6 immunoreactive (A1, A3); in Ca2-CA3 subfields (arrows) neurons are immunopositive only for p-S6(240/244) (A1, A3, B, B1). (C) In sclerotic hippocampus high level of p-S6 expression is observed in the area of neuronal loss and reactive gliosis (arrows). DI) Interneurons in Ammon’s horn reveal high level of p-S6(235/236)immunoreactivity. (D, E) Oval and polygonal neurons express high level of p-S6 immunoreactivity (arrows) in CA1 (D) and CA3 (E) subfields. (FI) Neurons with high level of p–S6 immunoreactivity (arrows) express markers of inhibitory interneurons (GAD65 in F, H and parvalbumin in G, I); (F, G) – CA1 subfield; (H, I) – CA3 subfield. B and B1– enlarged areas of CA2 and CA3 from A1 and A2, respectively. F′, G′, H′, and I′– cleaved F, G, H, and I images, respectively. Scale bars: (AC): 1.5 mm; (D, E): 300 μm; (FI): 200 μm.

Download figure to PowerPoint

In contrast, in sclerotic hippocampi, predominantly glial mTOR activation was observed. There was essentially no neuronal p-S6 staining in areas of astrogliosis characterized by increased GFAP staining and little GS staining. Only scattered neurons in areas without astrogliosis expressed p-S6 (Fig. 1C1,C2). However, the majority of astrocytes and microglial cells in sclerotic CA1 and CA3 subfields and hilus showed high level of p-S6 immunostaining (Fig. 2A–D). Granule cells in dentate gyrus were p-S6 immunonegative in all studied cases including dispersion (n = 11) and depletion (n = 5) of granule cell layer.

image

Figure 2.   Expression of p-S6(235/235) in the areas of astrogliosis in CA1 subfield in sclerotic hippocampi (AF) and chronic glial scar tissue (G, H). (A) Astrocytes in sclerotic CA1 express high level of GFAP and p-S6 immunoreactivity. (B) The borders of sclerotic CA1 can be clearly outlined (dotted line) by a sharp demarcation between protoplasmic glutamine synthetase+/p-S6− astrocytes and “scar” glutamine synthetase-/p-S6+ astrocytes. In sclerotic areas only astrocytes (C) and microglia (D) are p-S6 immunoreactive whereas NG2-cells (E) and preserved oligodendrocytes (F) are p-S6 immunonegative. Corresponding cells are marked with arrows in cleaved images. (G, H) Glial scar developed after first surgery performed 6 years ago. Note that only microglial cells express p-S6 (H), whereas astrocytes are p-S6 immunonegative (G). B′H′– cleaved BH, respectively. F1 and G1—enlarged areas from F and G images, respectively. Cleaved images are marked with . Scale bars: (A, B): 300 μm; (CF): 45 μm; (F, G): 120 μm; (G1, H1): 55 μm.

Download figure to PowerPoint

Taking into consideration that S6 phosphorylation at Ser235/236 sites might also be performed by p90RS kinase (Anjum & Blenis, 2008), we checked immunoreactivity for p-S6(240/244), phosphorylation known to be performed only by S6K1. Similar dichotomy in neuronal and glial immunostaining was observed in nonsclerotic and sclerotic hippocampi with application of p-S6(240/244) antibody (Fig. 1A3,B1,C2). It is of interest that in contrast to p-S6(235/236), p-S6(240/244) was observed in pyramidal neurons in CA2-CA3 subfields in nonsclerotic hippocampi (Fig. 1B1). In two cases of sclerotic hippocampus with dispersion of granule cell layer in dentate gyrus, granule neurons revealed immunoreactivity for p-S6 (240/244) (not shown).

Quantitative evaluation of a number of p-S6 immunopositive neurons in nonsclerotic areas of CA1 in sclerotic hippocampi and in corresponding areas in nonsclerotic hippocampi showed significant differences for both types of phospho-S6 antibodies: for p-S6(235/236): 84.2 ± 1.9% in nonsclerotic versus 28.6 ± 2.4% in sclerotic hippocampus, p < 0.001; for p-S6(240/244): 86.9 ± 1.6% in nonsclerotic versus 50.4 ± 2.2% in sclerotic hippocampus, p < 0.001; data presented as percentage of p-S6 immunopositive neurons to total amount stained with Nissl. No difference in the pattern of glial cells staining for p-S6(235/236) and p-S6(240/244) was observed. It should be noted that NG2-cells (Fig. 2E) and oligodendrocytes (Fig. 2F) did not reveal immunostaining for p-S6 (both types) in sclerotic subfields in hippocampi with neuronal loss.

Expression of p-S6 (both types) in astrocytes and microglia in sclerotic hippocampi was observed in every studied case and did not depend on the age of patients and the duration of seizure syndrome. High level of glial p-S6 expression was found consistently, even in patients >60 years (n = 4), with a long (>30 year) history of epilepsy.

mTOR is not activated in glial cells in “control” temporal neocortex

For comparison with our findings in nonsclerotic and sclerotic hippocampi, we examined “control” neocortical samples obtained from patients with MTLE or benign mesial temporal pathology. In control neocortex, obtained during the surgical approach to either epileptic or nonepileptic mesial temporal pathologies, astrocytes were characteristically GS positive but did not stain for p-S6. These findings were similar to those obtained for control and nonsclerotic MTLE hippocampus.

mTOR is not expressed by astrocytes in neocortical chronic scar tissue

To analyze expression of p-S6 in chronic glial scar tissue, we examined two cases in which neocortical glial scar tissue developed in association with medically refractory epilepsy several years following brain insults. In both of these cases, only a few p-S6 immunoreactive astrocytes were found within the studied scar tissue (Fig. 2G,G1). p-S6 expression in microglia was more prominent, although varied in different areas of the scar (Fig. 2H,H1).

The mTOR cascade is not activated in chronic gliosis in the rat pilocarpine model of epilepsy

The rodent pilocarpine model of MTLE is one of most widely used limbic epilepsy models. Administration of pilocarpine and induced seizures produced severe neuronal loss in the hippocampus (mainly CA1 subfield) and pyriform cortex (II and III layers), with subsequent development of a glial scar in several months after the initial insult (Fig. 3A,E). Scar astrocytes located in the glial scar in 8–12 months after initial induced seizures did not show immunoreactivity for p-S6 (Fig. 3B,B1,F). Microglial cells also did not reveal activation (based on CD68 immunoreactivity) and were p-S6 immunonegative (Fig. 3C,D).

image

Figure 3.   Astrocytes and microglial cells do not show p-S6 immunoreactivity in glial scars 1 year after pilocarpine-induced seizures in rat. (A) One hemisphere with prominent neuronal loss and glial scars in hippocampus and pyriform cortex (arrows). (B, B1) Scar astrocytes do not show p-S6 immunoreactivity in sclerotic hippocampus. Note that only granule neurons and few preserved pyramidal neurons are p-S6 immunopositive. (C, D) Microglial cells are p-S6 immunonegative in the glial scar in sclerotic hippocampus. Note that there are many microglial cells detected with pan-marker of microglia (Iba-1) in scar tissue (D). (E, F) Scar astrocytes do not reveal immunoreactivity for p-S6 in scar tissue in pyriform cortex. (BF) confocal microscopy. B1 is enlarged area from B. B′, B1′, C′, D′, F′ are cleaved B, B1, C, D, F images, respectively. Scale bars: (A): 2.5 mm; (B, C): 150 μm; (B1): 50 μm; (D): 110 μm: (E, F): 400 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

The present study shows that control and nonsclerotic hippocampi are characterized by neuronal activation of mTOR. In contrast, in hippocampal sclerosis in MTLE, mTOR activity is not seen in remaining neurons within areas of damage, whereas astrocytes and microglia in these areas of neuronal loss show marked mTOR activation, as detected by p-S6 staining. Of interest, mTOR activation is not a general feature of chronic astrogliosis, as cases of neocortical chronic gliosis showed astroglia with morphology and GFAP expression typical for gliosis, but without p-S6 staining.

Consistent activation of mTOR in scar astrocytes in sclerotic hippocampi was an unexpected finding. mTOR cascade activation in acutely reactive astrocytes and microglial cells in the first days and weeks following brain insult is a recently described phenomenon in a number of experimental conditions (Codeluppi et al., 2009; Dello Russo et al., 2009; Park et al., 2011). To our knowledge, there are no data on mTOR analysis in glial cells in the long-term period (several months to years) following brain insult.

In the chronic pilocarpine model of MTLE in rats, we did not find p-S6 immunoreactivity in hippocampal or neocortical glial cells. Similarly, in the genetically predisposed astrogliosis seen in cortical tubers in TSC, the majority of astrocytes with immunohistological (GFAP+/GS−) and morphological properties similar to sclerotic hippocampal astrocytes do not express p-S6 (Sosunov et al., 2008). There is thus something unique about chronic human sclerotic hippocampus in MTLE that is in marked contrast with chronic neocortical human brain scarring or the rat pilocarpine model of MTLE. Scar astrocytes in hippocampal sclerosis are usually considered to be functionally limited cells that largely carry out a mechanical role, filling the space following neuronal demise. Based on our results and known data on mTOR activation, we speculate that scar astrocytes in hippocampal sclerosis are active cells that perform some as yet undetermined function related to protein synthesis.

The absence of p-S6 immunostaining in granule neurons in dentate gyrus in the majority of studied sclerotic hippocampi (in 32 of 34 studied cases with hippocampal sclerosis) may be of special interest. In a rat experimental model of limbic epilepsy, the mTOR inhibitor rapamycin blocks mossy fiber sprouting (Buckmaster et al., 2009), indicating that mTOR cascade is responsible for this characteristic epilepsy-related neuronal pathologic change in the hippocampus. Although we did not assess mossy fiber sprouting in studied cases, literature data indicate that axonal sprouting in granule neurons is found in the majority of, if not all, sclerotic hippocampi (Sutula et al., 1989; Houser et al., 1990; Babb et al., 1991). The absence of mTOR activation in granule cells in nearly all of our studied sclerotic cases might indicate that axonal reorganization and growth occurred long before surgery.

Our data demonstrating mTOR cascade activation in Ammon’s horn interneurons in nonsclerotic hippocampus are in line with known data that the axonal arborization of inhibitory interneurons found in epileptic hippocampus (Mathern et al., 1995) is rapamycin-sensitive (Buckmaster & Wen, 2011). We do not have a clear explanation for the different pattern of neuronal expression of S6 phosphorylation at different sites in CA2 and CA3 subfields; this issue needs additional investigation.

This study raises as many questions as it answers. What is the role of mTOR expression in hippocampal and cortical neurons in control conditions? Why is mTOR selectively expressed more by inhibitory neurons in nonsclerotic hippocampus? Why do seizures in the setting of sclerotic MTLE increase mTOR expression in astrocytes while seizures in nonsclerotic MTLE do not? If sclerosis/glial scarring/neuronal death results in mTOR activation in astrocytes, why do we not see this happen in chronic neocortical areas of scarring or in the rat pilocarpine model of TLE?

Analysis of brain pathology removed during epilepsy surgery in humans provides interesting observations of unclear pathophysiologic significance. It is impossible to know whether the observed findings are causative or a result of chronic epilepsy. Although studying rodent models of epilepsy is the preferred way to follow epileptogenesis over time, models do not always recapitulate the human findings, as shown above. We readily admit that we do not know whether ongoing mTOR activation in sclerotic MTLE is contributing directly to ongoing seizures or is an adaption by the human hippocampus to try to limit seizures.

Acknowledgments

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

This work was supported in part by K08-NS048064 (GM) and the Tuberous Sclerosis Alliance. In studying human tissue, there is frequently a cause versus effect debate. It has been our great pleasure to discuss this dualism with Jürgen Wenzel over the years, in relation to many aspects of astrocytic and neuronal (dys)function in the human epileptic brain, as well as in experimental models of epilepsy. Jürgen has always been an amazingly skilled investigator; a critical and demanding thinker, and one of the nicest people on the planet. The epilepsy research world will miss him.

Disclosure

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

None of the authors has any conflicts of interest in relation to this work. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References