Introduction

Rasmussen encephalitis (RE) is a rare childhood neurological disease characterized by intractable seizures, unilateral neurological deficits, and hemispheric atrophy1. Complications during the perinatal period and facial autoimmune diseases may serve as predisposing factors for RE2. The incidence rate of RE is estimated to be 2.4 cases per 10 million people under 18 years old in Germany3. The pathological characteristics of RE include T-lymphocyte infiltrates, perivascular lymphocytic cuffing, neuronal loss, microglial nodules and astrogliosis in the affected hemisphere4. Currently, the etiology and pathogenesis of RE remain unclear. Immunotherapy can only temporarily delay disease progression. The only effective treatment is functional hemispherectomy, which, although capable of controlling the frequency of seizures, leads to permanent motor and neurological deficits in patients5. Therefore, it is crucial to investigate the pathogenic mechanisms of RE and identify novel therapeutic targets for drug treatment.

The mechanistic target of rapamycin is a highly conserved pathway6. mTOR is composed of two protein complexes, known as mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2). mTORC1 was initially identified as a protein complex consisting of mTOR, regulatory associated protein of mTOR (Raptor), mammalian lethal with Sec13 protein 8 (mLST8), DEP domain containing mTOR interacting protein (DEPTOR), and the 40 kDa proline-rich Akt substrate (PRAS40)7. The clinical drug rapamycin is a direct inhibitor of mTORC1 (with no effect on mTORC2) and is used to treat various diseases, including cancer, neurodegenerative disorders, and immune-related conditions8. mTORC1 activity is primarily activated by two growth signaling pathways: phosphatidylinositol 3-kinase (PI3K)-Akt and Ras-ERK9. The downstream process activates ribosomal protein S6 kinases (S6K1/2) and inhibits the binding of eukaryotic initiation factor 4E (eIF4E) to protein 1 (4E-BP1) to regulate protein synthesis9. In addition, mTORC1 regulates cellular autophagy by activating Stat3 and inhibiting ULK110.

Although overactivation of mTORC1 is generally believed to be a primary cause of neurological symptoms, emerging evidence suggests that dysregulation of mTORC2 is also a major complex leading to neurological disturbances. mTORC2 comprises mTOR, mLST8, DEPTOR, rapamycin-insensitive companion of mTOR (Rictor), and mSIN1 (mammalian stress-activated protein kinase-interacting protein 1)7. mTORC2 is a major upstream regulator of AKT. It directly phosphorylates and activates AKT, specifically at the serine 473 residue. This phosphorylation event is crucial for the translocation of AKT to the plasma membrane, where it exerts its effects by phosphorylating downstream targets involved in various cellular processes, including cell growth, survival, metabolism, and protein synthesis11. Therefore, the dysregulation of mTORC2 and its impact on AKT activation play a significant role in neurobiology, potentially contributing to neurological symptoms and disorders. Further understanding of the mechanisms underlying mTORC2 dysregulation may provide insights into developing novel therapeutic strategies for neurological conditions.

Abnormal regulation of the mTORC1 signaling pathway has been confirmed to promote the occurrence of epilepsy, as observed in cortical developmental disorders such as focal cortical dysplasia (FCD), tuberous sclerosis complex (TSC), and hemimegalencephaly (HME)12. Studies have demonstrated activation of the mTOR signaling pathway in RE through immunohistochemical analysis13. Furthermore, abnormal mutations in the mTOR gene have also been identified in human tissue samples from individuals with RE14. However, to date, there has been no research assessing the specific activation status of the mTOR pathway in RE tissue or the expression status of mTOR-associated pathways in various types of neuronal cells. Therefore, our study further explores the activation status of mTOR-related signaling pathways in RE, control (nonepileptic tissue), and FCD tissues.

Materials and methods

Patients and diagnosis

Surgical specimens were obtained from RE, control (no epilepsy) and FCD patients. The diagnosis of RE was based on rigorous criteria established by Bien et al.15 in 2005, which involve careful consideration of the patient’s clinical features, magnetic resonance imaging (MRI) findings, and neuropathological assessments. The RE epilepsy study population (Suppl. Table S1) consisted of 9 patients (4 females and 5 males) who had undergone surgery for functional hemispherectomies or anatomical hemispherectomies at the Sanbo Brain Hospital, Capital Medical University, China, between July 2019 and September 2022. Three tumor patients without epilepsy and six FCD patients were included in the study as normal and positive controls, respectively. Presurgical evaluation was performed according to previous procedures at Sanbo Brain Hospital16.

Brain samples from all 9 RE patients were acquired during surgery. We also obtained six focal cortical dysplasia (FCD) tissue samples as positive controls and three nonepilepsy tissue samples as normal controls. After excision, part of the tissue was immediately frozen in liquid nitrogen and stored at −80 °C until use, and the other part was embedded in paraffin.

All patients were prospectively enrolled in the study, and informed consent was obtained from the parents or legal guardians of all participants. This study was approved by the Capital Medical University Sanbo Brain Hospital Ethics Committee, with ethics approval reference [SBNK-YJ-2020-015-01]. All methods were carried out in accordance with relevant guidelines and regulations. All research was performed in accordance with the Declaration of Helsinki.

Tissue preparation

RE specimens were divided into representative tissue blocks, fixed in 4% formalin, embedded in paraffin and used for immunohistochemistry or immunofluorescence. In addition, frozen specimens were prepared from tissue adjacent to each paraffin-embedded tissue for Western blot analysis. In the FCD group, the lesion site was sampled, while in the nonepileptic group, normal tissue in the distant area of the tumor was sampled.

Immunohistochemistry

The tissues used for immunodetection were cut into 4 μm sections and subjected to immunolabeling with the following antibodies: phospho-S6 (Ser240/244), phospho-Akt (Ser473), PI3K, phospho-p44/42MAPK (Erk1/2), phospho-ULK1 (Ser757), beclin-1, and phospho-Stat3 (Tyr705) (Suppl. Table S2). Then, the sections were deparaffinized by incubation in xylene, followed by rehydration through graded water solutions. Antigen retrieval was performed by heating the sections in sodium citrate (pH 6.0) in a microwave at medium-high heat. The sections were then preincubated with 10% goat serum for 30 min and incubated overnight at 4 °C with primary diluted antibodies. Then, the sections were incubated with a rabbit/mouse two-step detection kit (Zhongshan Golden Bridge Biotechnology, Beijing, China) at room temperature (RT) for 2 h, followed by 1 min of 3,3’-diaminobenzidine tetrahydrochloride hydrate (DAB) exposure. The nuclei were counterstained with hematoxylin, and the sections were then dehydrated and mounted with permanent medium.

For the double-labeling studies, after incubation with primary antibodies, the sections were incubated for 2 h at room temperature with Alexa Fluor™ 488, Alexa Fluor™ 594 (anti-rabbit IgG or anti-mouse IgG; 1:200; Thermo Fisher Scientific, USA) and DAPI (Zhongshan Golden Bridge Biotechnology, Beijing, China) for nuclear staining. Images were visualized with a Leica TCS SP8 confocal microscope and analyzed using DP2-BSW (Olympus) software.

Western blotting

Frozen cortical tissue (150–200 mg) was separated from the underlying white matter and then homogenized in 10 ml of RIPA lysis buffer containing 200 µl of a protease and phosphatase inhibitor cocktail and 200 µl of 0.05 M EDTA, pH 8 (Beyotime Biotechnology, Jiangsu, China). To fully lyse the cell membrane, ultrasonic vibration treatment was applied. After 10 min on ice, the supernatants were centrifuged at 12,000 RPM at 4 °C for 30 min. Protein content was measured by the bicinchoninic acid method, and samples were diluted to equal amounts of protein in each sample in lysis buffer and 5× loading buffer (Beyotime Biotechnology, Jiangsu, China). For electrophoresis, 30 µg of protein from each sample was separated by sodium dodecyl sulfate polyacrylamide gel electrophoretic (SDS‒PAGE) analysis. The samples were run on 5–10% precast gels at 80–120 V for 1.5 h. Then, the proteins were transferred to polyvinylidene difluoride membranes (0.2 μm, Millipore) at 100 V for 1 h using a semidry electroblotting system (Bio-Rad, Transblot SD, Hercules, CA, USA). The membrane was blocked with 5–8% nonfat dry milk for 1 h and then incubated overnight at 4 °C with the primary antibodies listed in Suppl. Table S2. Corresponding horseradish peroxidase-conjugated secondary antibodies (1:4000; Applygen Technologies Inc., Beijing, China) were applied for 1 h at RT. After washing, immunoreactivity was visualized by chemiluminescence using a SuperEnhanced Chemiluminescence Detection Kit (Applygen Technologies Inc., Beijing, China). After the membranes were exposed to Kodak X-ray film, the protein bands were visualized by a Luminescent Image Analyzer (LAS-3000; Fujifilm, Tokyo, Japan). β-Actin (1:2000, monoclonal mouse antibody, Applygen Technologies Inc., Beijing, China) was used to control for differences in the protein loading.

Statistical analysis

Statistical analyses were performed using the SPSS software package, version 21. Group means were compared by unpaired t tests or Welch’s t tests, while Pearson correlation analysis was used to determine potential associations between variables. The data are expressed as the mean ± standard error of the mean (SEM). Differences were considered statistically significant at p ≤ 0.05.

Results

Clinical and pathological characteristics of the patients

In the RE group (Suppl. Table S3), the mean age at surgery was 7.5 years (range 3.9–13.6 years), the mean age at first seizure was 5.4 years (range 2.7–10.6 years), and the mean duration of epilepsy between the first seizure and surgery was 0.5 years (range 0.2–5.1 years). Epilepsia partialis continua (EPC) was present in 3 of 9 patients. Except for one patient who received only one antiepileptic drug (AED) before surgery, the other patients received at least 3 AEDs. All patients underwent functional or anatomic hemispherectomy, and 7 patients (77.8%) were diagnosed with left hemisphere RE. Four patients (44%) had risk factors for epilepsy, including head trauma, jaundice after birth, premature birth, and hand-foot-mouth disease.

The mean age at surgery in the FCD group was 7.3 years (range 1.2–15.0 years), which was not significantly different from that in the RE group (p = 0.927). The mean age at first seizure onset was 4.7 years (range 0.1–11.0 years), and the mean duration of epilepsy between the first seizure onset and surgery was 2.5 years (range 0.8–6.7 years). There were 2 patients with FCDIIb, 3 patients with FCDIIa, and 1 patient with FCDIb. One patient had a history of febrile convulsion. The mean age of the patients in the seizure-free group was 45 years (range 37–55 years), and the patients were pathologically diagnosed with glioblastoma, anaplastic oligodendroglioma, or cavernous hemangioma.

Expression of key molecules of the mTOR pathway in RE

First, we used Western blotting to analyze the extent of mTORC1 activation by determining the phosphorylation of p-S6 in the control, RE and FCD groups. We found the ratios of p-S6/S6 was significantly higher in the RE group compared to that in the control group (Fig. 1A-B, Suppl. Table S4). The paraffin-embedded sections were used to localize the expression of p-S6 in the control, RE and FCD groups. p-S6 immunostaining of the cortex and white matter revealed weak immunoreactivity in the nonepilepsy group (Fig. 2A), while the RE and FCD groups all showed strong positive immunoreactivity (Fig. 2B-C, G-I). Both dysmorphic neurons and cytomegalic cells displayed mTOR hyperactivation, as shown by p-S6 immunostaining. Interestingly, there were more p-S6 immunoreactive regions in the white matter of the RE tissues than in that of the FCD tissues, especially in the inflamed blood vessels and cell clusters (Fig. 2M-O). In addition, many of these cells were colabelled with not only neurons but also microglia and astrocytes (Fig. 1D-L).

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

In comparison to the control group, patients with RE exhibit activation of both the mTORC1 and mTORC2 pathways. A-C. Analysis of the results of cortical tissue in patients. Showing (A) representative immunoblot images of phospho-S6(Ser240/244), phospho-AKT(Ser473), and the quantification of the ratio of target protein bands to internal controls, expressed as a percentage of the mean value of the control. For (B) p-S6(Ser240/244)/S6(RE, n = 8 vs. Control, n = 3/FCD, n = 6), (C) p-AKT(Ser473)/Actin ((RE, n = 8 vs. Control, n = 3/FCD, n = 6). *: p ≤ 0.05; ns: No significance, using unpaired t-tests. Error bars represent mean ± standard error of the mean (S. E. M). D-L. The representative image of the lesion in RE patients was double-labeled with p-S6(Ser240/244) (green) and NeuN, Iba1, GFAP (red) to demonstrate strong co-labeling with atypical neurons (F), perivascular microglia (I), and astrocytes (L). The nuclear DAPI staining is shown in blue. Scalebars represent 50 μm.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Immunohistochemical staining of representative phospho-S6 (Ser240/244) and phospho-AKT (Ser473) proteins in RE, control and FCD tissues. A-F. The representative tissue staining of phospho-S6 (Ser240/244) and phospho-AKT (Ser473) in the cortex. In the control group, there was a low expression of mTORC1 and mTORC2(A, D). The expression of P-S6 and P-AKT is primarily observed in the cytoplasm of pyramidal neurons, atypical neurons, and astrocyte-like neurons in the cortex of the RE. G-L. Exemplary images of white matter. The mTORC1 and mTORC2 exhibit robust activation in the cytoplasm of astrocytes (H) and ectopic neurons (k) in the RE tissues. M-R. The mTORC1 and mTORC2 pathway demonstrates strong activation in the (M, Amplification at a high multiple of N to M; P, Q is amplified at a high multiple of P) perivascular, characterized by abnormal expansion and (O, R) cellular nodules in the white matter of the RE region. All inserted sections are magnified local images. Scale bars represent 50 μm, except for A, D, M, P, scale bars represent 100 μm.

Next, to determine whether mTORC2 was activated, we measured p-Akt levels, and the Western blot results revealed significantly greater p-Akt expression levels in RE patients than in control patients, with no statistically significant differences relative to those in FCD patients (Fig. 1A, C; Suppl. Table S4). Immunostaining for p-Akt in RE patients was observed in both the cortex and white matter (Fig. 2D-F, J-L). Furthermore, there was a significant presence of inflammatory blood vessels and cellular nodules in the white matter of RE patients (Fig. 2P-R).

PI3K-AKT and Ras-MAPK pathway activation in RE patients

To determine whether there are potential upstream mechanisms leading to mTOR dysfunction, we assessed the immunoreactivity of PI3K and p-MAPK. In nonepilepsy control specimens, weak PI3K immunoreactivity was detected (Fig. 3A). In contrast, strong immunoreactivity was observed in the cortex of RE patients, especially in cytomegalic cells (Fig. 3B, C).

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

The expression of PI3K and MAPK is excessively upregulated in RE. A-C . The expression of PI3K in the cortical regions of control, RE, and FCD. In the control group, PI3K demonstrated low levels of positivity, while in the (B) RE and (C) FCD cortical regions, giant and dysmorphic neurons showed strong positivity for PI3K in the cytoplasm. D-I . Expression of p-p44/42MAPK (ERK1/2) in the control, RE, FCD cortex, and white matter. In comparison to the control group, the p-p44/42MAPK (ERK1/2) exhibits strong expression in the cortical regions of RE (E) and FCD (F). Within the cortical regions of RE and FCD (as indicated by the enlarged portion delineated by dashed lines in F), there is observable cytoplasmic and nuclear expression of p-p44/42MAPK (ERK1/2) in cells that exhibit morphological characteristics similar to astrocytes. In the white matter of RE (H), a significant number of cells exhibit nuclear positivity for p-p44/42MAPK (ERK1/2), while there is a small number of positively expressing cells in the FCD white matter (I). J-L . Expression of p-p44/42MAPK (ERK1/2) in the white matter vasculature and cellular nodules. In the white matter, both the microvessels and dilated blood vessels exhibit strong positive expression of p-p44/42MAPK (ERK1/2). Additionally, the astrocyte-like nerve cells gather around the expansion of blood vessels and a small number of p-p44/42MAPK (ERK1/2) positive cells infiltrating around the cellular nodules. Scale bars represent 50 μm. M-N . Western blot analysis of RE tissue samples showing (M) representative bands for individual proteins and (N) quantitative assessment of p-p44/42MAPK (ERK1/2)/Actin (RE, n = 9 vs. Control, n = 3/FCD, n = 6). *, **: p ≤ 0.05, p ≤ 0.01, respectively (unpaired t tests comparing RE vs. control and RE vs. FCD). Error bars indicate S.E.M. O-S. Image of representative fluorescence staining in the RE patient, co-expressed with p-p44/42MAPK (ERK1/2) (green) and GFAP, NeuN, Iba1 (red). The blood vessels of the RE, along with the surrounding astrocytes, exhibit strong co-expression with p-p44/42MAPK (ERK1/2) (O-P). The arrows(O) indicate the blood vessels and neurons, respectively. In addition, a large number of astrocytes co-stained and clustered around the inflammatory blood vessels(P). The neurons surrounding the astrocytes in the white matter (Q), as well as the dysmorphic neurons (R) in the cortex, demonstrate significant co-expression with p-p44/42MAPK (ERK1/2), however, no positive expression of p-p44/42MAPK (ERK1/2) was observed in microglial cells and microglial nodules(S). The nuclear DAPI staining is shown in blue. Scale bars represent 50 μm.

Western blot analysis revealed that p-MAPK levels were significantly greater in RE patients than in nonepilepsy controls and FCD patients (Fig. 3M-N, Suppl. Table S4). Immunohistochemistry for p-MAPK showed faint expression in control specimens throughout all cortical layers (Fig. 3D, G, F, I). In RE patients, strong p-MAPK expression was observed in a large number of astrocytes, and moderate p-MAPK expression was detected in some pyramidal neurons (Fig. 3E, H). In addition, astrocytes and dysmorphic neurons displayed nuclear and cytoplasmic staining, with stronger nuclear staining, which was substantially increased in astrocytes. Many of the vessels showed p-MAPK immunopositivity (Fig. 3J-K). No obvious staining was observed in the nodules of microglia (Fig. 3L). Double labeling demonstrated p-MAPK expression in GFAP-positive astrocytes near blood vessels and in neuronal cells (Fig. 3O-R), but not in activated microglia (Fig. 3S).

Additionally, we suspected that the expression of mTOR-related proteins in RE patients would be correlated with clinical variables (Suppl. Table S4). The results demonstrated a strong correlation between the expression levels of p-MAPK and the epilepsy duration (r = 0.85, p < 0.01).

Dysregulation of ULK1-beclin 1 signaling in RE patients

Proteins activated downstream of the mTOR pathway include the well-characterized p-S6 molecule, which is involved in protein synthesis, and the associated molecules ULK1-beclin 1, which regulate autophagy8. To further explore whether there is aberrant regulation of neuronal autophagy in RE, we next assessed the phosphorylation status and subcellular localization of p-ULK1 and beclin-1. The Western blot results indicated that there were differences in the expression levels of p-ULK1 in RE patients compared to the control group, although these differences did not reach statistical significance (Fig. 4A-B, Suppl. Table S4). Double-labelled immunohistochemistry for p-ULK1 and NeuN revealed that p-ULK1 was expressed in the cytoplasm of dysmorphic neurons (Fig. 4C-H). Interestingly, apoptotic neurons showed abnormal aggregation of round granules with strong p-ULK1 staining (Fig. 4E-Insert section in lower left corner).

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Phosphorylation of ULK in RE regions may be associated with altered neuronal autophagy. A-B. Immunoblot analysis and corresponding quantification of RE, control, FCD showing representative image for individual proteins (A) and quantitative analysis of phospho-ULK1(Ser757)/Actin (B) (RE, n = 9 vs. Control, n = 3/FCD, n = 6). ns: Denotes no significance, using unpaired t-tests comparing RE vs. control and RE vs. FCD. Error bars represent S.E.M. C-H. Representative fluorescent images of RE-organized tissue sections doubled labeled with NeuN (red) and p-ULK1 (Ser757) (green) demonstrating robust p-ULK1 (Ser757) within (C-E) dysmorphic neurons and apoptotic neurons. (F-H) Neurons exhibiting positive staining for p-ULK1 (Ser757) are arranged in a circular pattern. Insets depict an enlarged portion within a dashed outline. The nuclear stained with DAPI appear blue. Scale bars represent 50 μm.

Immunohistochemical staining for beclin-1 revealed neuronal engulfment in the white matter of RE patients (Fig. 5A-F). The dual-labeling results indicated strong colocalization of beclin-1 with nodules of small glial cells and neurons in RE patients, while no colocalization was observed in astrocytes (Fig. 5G-O).

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Dysmorphic neurons and microglial nodules in the cortex and white matter of RE exhibit strong positive expression of Beclin-1, while astrocytes show negative expression. A-F. Representative tissue images of Beclin-1 in the control, RE, FCD. The expression of Beclin-1 was observed in the cortical neurons of the control and FCD groups (A-B, E-F), while its expression was less prominent in the white matter. (C-D) demonstrates a high expression of Beclin-1 in the dysmorphic neurons in the cortex and ectopic neurons in the white matter of RE. Neuronophagia are observed in the white matter, with phagocytic cells arranged in a ring-like pattern around neuronal cell bodies or distributed in a vertical arrangement along the axonal regions of the neurons. G-O. Representative fluorescent staining images of Beclin-1 in RE tissue sections. Beclin-1 demonstrates strong co-localization with microglial nodules (G-I) and dysmorphic neurons (J-L), but no co-labeling was observed with astrocytes (M-O). The nuclear stained with DAPI appear blue. Scale bars represent 50 μm.

Increased Stat3 signaling in RE tissue samples

To further investigate the downstream activation of mTOR-associated molecules, we examined Stat3, which is also a direct substrate of mTORC1. Nuclear Stat3 regulates autophagy by modulating the transcription of the autophagy gene beclin-1. Previous studies have indicated that Stat3 signaling is involved in the pathogenesis of various brain disorders, including Alzheimer’s disease, depression, and epilepsy17. Our results demonstrated a significant increase in p-Stat3 levels in the RE tissue samples compared to those in the control tissue samples (Fig. 6A-B, Suppl. Table S4). Immunohistochemical staining revealed abundant nuclear-positive cells in both the cortex and white matter of the RE samples (Fig. 6C-H). Positive cells infiltrated around blood vessels and nodules (Fig. 6I-J).

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

Increased expression of phospho-Stat3 (Tyr705) was observed in the tissue samples of the RE group. A-B. The immunoblot analysis and subsequent quantification of control, RE and FCD samples revealed a representative image of individual proteins (A) and a quantitative assessment of phospho-Stat3 (Tyr705)/Actin (B) (RE, n = 4 vs. Control, n = 3/FCD, n = 3). ns: Indicates no statistical significance, as determined by unpaired t-tests comparing RE vs. control and RE vs. FCD. Error bars represent S.E.M. C-J. Representative immunohistochemical images of p-Stat3 (Tyr705) in the control samples of RE and FCD tissue sections. The control group showed negative expression of p-Stat3 (Tyr705) (C-D). A small amount of cell expression of p-Stat3 (Tyr705) is observed in the cortex and white matter of FCD (G-H). In the cortex of the RE, there are Cytoplasm-positive astrocyte-like cells expressing p-Stat3 (Tyr705) (E). Additionally, there is a significant presence of p-Stat3 (Tyr705) nuclear-positive cells in the white matter (F), as well as infiltration of p-Stat3 (Tyr705) positive cells around blood vessels (I) and nodules (J) from RE samples. Insets illustrate a magnified section enclosed by a dashed border. The nuclei, stained with DAPI, are visible as blue. Scale bars represent 50 μm.

Discussion

In recent years, multiple studies have provided compelling evidence of widespread expression of the mTOR signaling pathway in various epilepsy-associated conditions, such as tuberous sclerosis, focal cortical dysplasia, hemimegalencephaly, and ganglioglioma13,18,19,20,21,22. Overactivation of mTOR has been linked to the development of brain anomalies such as dysplastic neurons, atypical cortical arrangement, and astrogliosis. Previous research focused only on the immunohistochemical assessment of ps6 and lacked a comprehensive quantitative comparison of the distinct expression patterns of the mTORC1 and mTORC2 pathways, especially in RE patients13. Our study makes a major effort to comprehensively investigate the quantitative and observational aspects of the mTOR signaling pathway, including its upstream and downstream components, along with their expression patterns in neural cells within the context of RE. Our findings reveal a notable increase in both mTORC1 and mTORC2 activity in RE-affected tissues compared to their nonepileptic counterparts. Notably, the PI3K/AKT and MAPK/ERK signaling pathways play crucial roles as upstream activators of mTORC1 and mTORC2, suggesting potential targets for therapeutic interventions. Notably, we found a correlation between the duration of epilepsy and the level of p-MAPK, providing further insight into RE progression. Furthermore, our findings demonstrated widespread expression of p-ULK1, p-Stat3, and beclin-1 in RE tissues. However, beclin-1 was not detected in astrocytes, suggesting a potential role for this protein in the pathogenesis of RE.

mTORC1 and mTORC2 mediated epileptogenesis

Our research highlights the role of the mTOR signaling pathway in the epileptogenesis of RE, focusing on both mTORC1 and mTORC2. mTORC1 regulates neuronal processes like cell proliferation, protein synthesis, and cortical development23, with its dysfunction linked to RE pathogenesis. Inflammation and the presence of cytomegalic pyramidal neurons (CPNs) are key features of RE24,25, though the role of CPNs in seizure onset remains unclear. Our findings confirm the presence of giant cell neurons in RE, with mTORC1-related molecules like p-S6, p-AKT, and PI3K showing colocalization with these neurons. We also observed upregulation of mTORC2 signaling, marked by increased p-AKT levels26, supporting its involvement in epileptogenesis. Targeting mTORC2 with antisense oligonucleotides shows therapeutic potential, suggesting it could be a promising target for RE treatment27.

ERK/MAPK involvement in the pathogenic mechanism of RE

Furthermore, our investigation revealed that the Ras-MAPK pathway is a key upstream influencer in activating mTOR signaling in RE. P-44/42 MAPK, also known as ERK1/2, is part of the family of extracellular signal-regulated kinases28. ERK1/2 plays a pivotal role in regulating various cellular activities, including oxidative stress, proliferation, differentiation, and apoptosis. Notably, the expression of MAPK was primarily observed in neurons, vascular endothelial cells, and peripheral astrocytes, consistent with earlier research demonstrating the upregulation of the p-MAPK signaling pathway in several epilepsy disorders, such as TSC, TLE, and FCD29,30,31,32. Previous studies by our team have demonstrated that TLR4 is positively expressed in neurons, vascular endothelial cells and surrounding glial cells in RE patients33. Subsequently, activated TLR4 in astrocytes phosphorylates ERK1/2, which is involved in the induction of MMP-934. The use of a toll-like receptor 9 (TLR9) inhibitor has the potential to mitigate the migration of reactive astrocytes and suppress the expression of proinflammatory cytokines and chemokines by inhibiting MAPK35. Furthermore, our research revealed a positive correlation between activated ERK levels and the duration of epileptic seizures. Enhanced ERK signaling is directly associated with increased susceptibility to epilepsy36. Therefore, adjunctive therapy with ERK inhibitors may offer significant clinical benefits in controlling seizures in patients with refractory epilepsy.

Aberrant p-ULK, beclin-1 regulation and upregulated STAT3 as mechanisms exacerbating epilepsy in RE

In addition to mTOR pathway activation, we suspect that impaired autophagy in neuronal cells may worsen RE progression. Our findings show upregulation of phosphorylated ULK and reduced beclin-1 expression in RE tissue, suggesting altered autophagic regulation and impaired astrocytic phagocytic function37. ULK-1 plays a key role in autophagy initiation, and its phosphorylation can either activate or inhibit this process38. Furthermore, we observed STAT3 activation in RE cortical tissues, which may contribute to the disease by inhibiting autophagy, as seen in other studies39. These findings point to autophagy dysfunction and aberrant STAT3 activation as potential mechanisms in RE pathogenesis.

Limitations of the study

Several of the antibodies used in our investigation had limitations in terms of their application range, which prevented further analysis of their specific expression levels within the tissue samples. Additionally, the nature of studying human epileptic tissue presents inherent constraints, and our findings may not necessarily reflect the intrinsic characteristics of the brains of RE patients in other clinical phases. Notably, the majority of the surgical patients included in our study were in the acute clinical phase, and the potential influence of various antiepileptic medications on the results cannot be ruled out. The limited number patients and age-mismatched controls in our study prevented us from conducting further robust statistical analyses. Although a strong correlation may exist between p-MAPK expression and epilepsy duration, it is challenging to attach significant weight to this observation due to the limitations of the study.

Conclusion

These results may have important implications for potential therapeutic interventions. Total mTOR blockade using dual inhibitors targeting mTORC1 and mTORC2 may compromise the neuroprotective mechanisms triggered by mTORC2 activation. However, the chronic use of high-dose rapalogs may also negatively affect mTORC2 function. Therefore, alternative approaches that selectively modulate the downstream activities of mTORC1 and mTORC2, such as p70S6K inhibitors and autophagy inducers, alone or in combination, may be preferable treatment options for epilepsy. Furthermore, negative regulation of selective pathways upstream of mTORC1 and mTORC2, such as ERK inhibitors, may hold promise as potential approaches for the treatment of refractory epilepsy in some patients.

Overall, these findings provide important insights into the dysregulation of the mTOR pathway in RE and suggest potential therapeutic strategies that can be explored for better management of this condition. Future research should focus on further investigating these pathways and developing targeted interventions that can effectively modulate mTOR signaling and improve outcomes for RE patients.