Ketogenic diet can inhibit epileptic seizures by inhibiting RGMa methylation to promote GluA1 palmitoylation in the epileptic rat model

Abstract

Background

This study investigated the relationship between the ketogenic diet, RGMa methylation, and GluA1 palmitoylation in a rat model of epilepsy, aiming to identify a novel mechanism for treating refractory epilepsy with the ketogenic diet.

Methods

Epileptic rats were treated with a ketogenic diet. Hippocampal tissue was analyzed to assess the expression of DNMT1 and DNMT3a proteins, RGMa DNA methylation and protein expression, FAK and zDHHC3 phosphorylation and expression, and GluA1 palmitoylation and expression. The therapeutic effects of the ketogenic diet on seizure frequency were statistically analyzed, along with the correlation between each parameter and treatment outcomes.

Results

The ketogenic diet treatment reduced seizure frequency in the epileptic rats. It inhibited RGMa DNA methylation in hippocampal tissue while increasing RGMa expression. Additionally, the diet enhanced GluA1 palmitoylation and reduced its expression on the neuronal membrane. These changes were associated with alterations in FAK/Src and zDHHC3 phosphorylation.

Conclusion

The ketogenic diet inhibits RGMa DNA methylation, upregulates RGMa expression, and promotes GluA1 palmitoylation, leading to a reduction in seizure frequency and alleviation of refractory epilepsy in the rat model. These effects may be mediated through the RGMa-FAK/Src-zDHHC3 signaling axis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12986-026-01090-8.

Introduction

Epilepsy is a chronic neurological disorder characterized by abnormal neuronal discharges in the brain, leading to recurrent seizures, loss of consciousness, and other related symptoms. Its prevalence ranges from approximately 3‰ to 9‰ [1]. Although pharmacological treatment remains the primary strategy, some patients develop refractory epilepsy due to poor response to medication. These individuals experience persistent seizures, which elevate the risk of complications and sudden death.

The ketogenic diet (KD) is a high-fat, moderate-protein, and extremely low-carbohydrate dietary regimen that mimics the physiological state of fasting. This dietary approach prompts the body to utilize fat as an energy source instead of carbohydrates, resulting in the production of ketone bodies, which serve as an alternative energy supply. The broad efficacy of the ketogenic diet in neurological disorders has been well-documented [2]. It reduces inflammation, decreases reactive oxygen species, restores neuronal myelin sheaths, promotes mitochondrial formation and regeneration, and enhances neuronal metabolism [3]. The ketogenic diet al.so stabilizes neurotransmitter regulation and neuronal membrane potential, reducing excessive neuronal excitability and abnormal discharge activity [4]. Thus, the ketogenic diet presents a promising alternative treatment for refractory epilepsy.

The ketogenic diet can influence DNA methylation patterns in the body [5]. DNA methylation, a chemical modification of DNA mediated by DNA methyltransferases (DNMT1, DNMT3a), involves the covalent attachment of a methyl group to cytosine-guanine dinucleotide (CpG) sites [6]. This process alters gene expression without changing the underlying DNA sequence [7].

The repulsive guidance molecule-a (RGMa) is a glycoprotein anchored to the membrane by glycosylphosphatidylinositol (GPI). It plays a critical role in neurobiology and various physiological processes and is expressed in both membrane-bound and soluble forms throughout the central nervous system [8, 9]. During neural development, RGMa is involved in the guidance of synapses, neuronal differentiation, and survival, ensuring proper axonal growth and the accurate formation of neural networks [10]. Multiple studies have suggested that RGMa may contribute to the pathogenesis of several central nervous system disorders, including epilepsy [8]. RGMa levels are significantly reduced in both patients with epilepsy and animal models of epilepsy [11, 12]. Overexpression of RGMa can inhibit mossy fiber sprouting and prevent the occurrence of seizures [13]. Additionally, silencing the upstream regulatory factor miR-20a-5p of RGMa can impair neuronal branching and axonal growth via the RGMa-RhoA signaling pathway. Our earlier studies indicated that RGMa expression is downregulated in animal models of status epilepticus, and its upregulation prolongs seizure latency, reduces epilepsy susceptibility, and effectively prevents seizure onset [14].

Alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) is an ionotropic glutamate receptor that mediates rapid excitatory synaptic transmission in the mammalian brain. It is widely distributed in the central nervous system and plays a pivotal role in neuronal signal transmission, being essential for advanced neural functions such as learning and memory [15, 16]. Increasing evidence suggests that the pathophysiology of epilepsy is associated with alterations in AMPAR trafficking, synaptic expression, and signal transduction [17]. S-palmitoylation is a reversible post-translational lipid modification that influences protein homeostasis. This modification involves the covalent attachment of palmitic acid to specific cysteine residues via a thioester bond. This process is mediated by palmitoyltransferases (PATs) from the DHHC family, which are key regulators of palmitoylation [18]. Palmitoylation is essential for controlling protein stability, subcellular localization, membrane trafficking, interactions with effector proteins, enzyme activity, neuronal development, and synaptic plasticity, among other cellular functions [19, 20].

The therapeutic effects of the ketogenic diet, a high-fat regimen, in epilepsy may be mediated through the regulation of RGMa methylation and AMPAR palmitoylation. This hypothesis was tested using the pilocarpine-induced status epilepticus (SE) rat model.

Materials and methods

Animals and grouping

Seven-week-old male Wistar rats (250–275 g) were housed in an SPF-level animal facility (temperature 22–24 °C, relative humidity 50 ± 10%, 12-hour light/dark cycle, with free access to food and water). Thirty-six rats were randomly assigned to one of four groups: the control group, the SE group, the SE + ketogenic diet (SE + KD) group, and the SE + ketogenic diet – control diet (SE + KD - CD) group, with nine rats in each group. After completing seizure recordings, hippocampal tissues were collected from all groups, and proteins and DNA were extracted for subsequent methylation, Western blot (WB), and palmitoylation analysis. All animal procedures were approved by the Animal Ethics Committee of Shengjing Hospital, China Medical University (2024PS967K), and conducted in compliance with the “Guide for the Care and Use of Laboratory Animals” (1996) published by the National Institutes of Health, USA.

Establishment of rat SE model

The SE model was induced by pilocarpine hydrochloride administration [21]. Initially, 1 mg/kg of atropine was injected subcutaneously to reduce peripheral cholinergic responses to pilocarpine. After 30 min, pilocarpine (280 mg/kg) was injected intraperitoneally, and the rats were monitored for behavioral changes. If no SE occurred within 60 min of the first pilocarpine injection, a second dose at half the initial amount was administered. Rats that failed to develop SE after the second dose were excluded from the study. The severity of seizures was assessed using the Racine scale, with seizures graded as IV or higher included in subsequent analyses [22]. If SE persisted for more than 2 h, diazepam (10 mg/kg, subcutaneously) was administered to terminate the seizures, reducing mortality risk. In the control group, normal saline replaced pilocarpine, and atropine and diazepam treatments were administered in the same manner.

Ketogenic diet therapy

Rats that successfully underwent model establishment were randomly divided into three groups, with nine rats in each group. The rats in the SE group were continuously fed a standard diet, consisting of 13.5% fat, 28.5% protein, and 58.0% carbohydrates. In the SD + KD group, the ketogenic diet was introduced starting at the 5th week post-model establishment and continued until the 16th week. The ketogenic diet was composed of 93.4% fat, 4.7% protein, and 1.8% carbohydrates. In the SD + KD-CD group, rats were fed the ketogenic diet starting from the 5th week and continued until the 10th week. In the 11th week, a single intraperitoneal injection of glucose (30% w/v; 2 g/kg) was administered to reverse the diet, and the rats were switched to a standard diet, which they maintained until the 16th week. The control group was continuously fed the standard diet. Throughout the experiment, the mental state, feeding behavior, appearance, weight, and other survival conditions of the rats were monitored and recorded weekly. From weeks 8 to 10 and again from weeks 14 to 16, the frequency of epileptic seizures was recorded (video monitoring for all rats, 8 h per day, 7 days per week). At the end of week 16, rats from each group were sacrificed, and hippocampal tissues were collected for subsequent analysis (Fig. 1).

Fig. 1.

The overall feeding situation of the four groups of Wistar rats

Bisulfite sequencing PCR(BSP)

The methylation status of the RGMa promoter region in rat hippocampal tissue was assessed using bisulfite sequencing (BSP). DNA was extracted using the Cell/Tissue DNA Extraction Kit (BioTeke, China) and treated with the EZ DNA Methylation Kit (ZYMO, USA). PCR amplification was carried out using the 2×Power Taq PCR MasterMix (BioTeke, China). The PCR conditions were as follows: 95 °C for 5 minutes, 98 °C for 10 seconds, 55 °C for 20 seconds, and 72 °C for 30 seconds, for a total of 40 cycles. The forward primer sequence was 5’-GGATTAGTAAGGGTGGGAT-3’ (19 bp), and the reverse primer sequence was 5’-CCCAAACTAACCTACTACCTC-3’ (21 bp). The PCR product was purified using the multi-functional DNA purification and recovery kit (BioTeke, China) to obtain the target gene fragment. The fragment was then ligated into the T vector system (Promega, USA) for cloning. The RGMa promoter region we analyzed contained a total of 32 CpG sites. The sequencing results were analyzed using the quma website, and the methylation level of each sample was expressed as the percentage of methylated CpG sites out of the total CpG sites.

Co-immunoprecipitation

Co-immunoprecipitation was conducted to detect the phosphorylation level of zDHHC3 in rat hippocampal tissue. Hippocampal tissue samples were lysed in a buffer containing phosphatase inhibitors (Wanleibio, China), and the supernatant was collected after grinding. The protein concentration was quantified, and Protein A + G Agarose Gel was incubated with zDHHC3 antibody (diluted 1:300, Santa Cruz, USA) at 4 °C overnight. The protein was then purified and collected. SDS-PAGE was performed, followed by membrane transfer and WB using zDHHC3 antibody and p-Tyr antibody (diluted 1:1000, Merck Sigma, USA). Rabbit anti-mouse IgG-HRP secondary antibody (diluted 1:5000, Wanleibio, China) was used for detection. β-actin (Wanleibio, China) served as the internal control.

Acyl-biotinyl exchange (ABE)

The acyl-biotin exchange (ABE) method was employed to assess the palmitoylation level of the AMPAR subunit (GluA1) in rat hippocampal tissue. Tissue samples were lysed in a buffer containing 1 mM PMSF (phenylmethylsulfonyl fluoride) and 50 mM NEM (N-ethylmaleimide) and incubated on ice for grinding. The supernatant was collected via centrifugation, and the protein concentration was quantified. Protein A + G Agarose Gel was resuspended in TBS, and GluA1 antibody (Proteintech, China) was added for a 1-hour incubation at room temperature. After centrifugation at 4 °C, the supernatant was discarded, and protein extraction solution was added. The samples were washed three times with lysis buffer and divided into two groups: one with 1 M HAM (hydroxylamine, MACKLIN, China) and one without HAM (used as a control). The reaction was conducted at room temperature for 1 h. Samples were washed with lysis buffer, and the supernatant was discarded. Then, 5 µM Biotin-HPDP-containing lysis buffer was added to each sample, and the reaction proceeded at 4 °C for 1 h. After collecting the samples and washing them with lysis buffer containing PMSF three times, sample buffer containing 5 mM DTT was added, and the samples were heated at 100 °C for 10 min. SDS-PAGE electrophoresis was performed, followed by membrane transfer and WB detection using Streptavidin-HRP antibody (Beyotime, China) and GluA1 antibody.

Western blot analysis

WB analysis was performed to detect the expression levels of DNMT1, DNMT3a, RGMa, FAK, Src, pFAK (Tyr397), pSrc (Tyr416), and zDHHC3 in rat hippocampal tissue. Neuronal membrane proteins were isolated to assess GluA1 expression on the neuronal membrane. Total protein was extracted from hippocampal tissue using lysis buffer, while neuronal membrane proteins were separated using the cell membrane protein and cytoplasmic protein extraction kit (Beyotime). Protein quantification was carried out using the BCA method. Samples were incubated overnight at 4 °C with the following primary antibodies (diluted 1:500, from abclonal, China): anti-DNMT1, DNMT3a, RGMa, Src, and pSrc (Tyr416); anti-FAK and pFAK (Tyr397); anti-zDHHC3 (diluted 1:300, Santa Cruz, USA); and anti-GluA1 (diluted 1:20000, Proteintech, China). The samples were then incubated with rabbit anti-mouse IgG-HRP or goat anti-rabbit IgG-HRP secondary antibodies (diluted 1:5000, Wanleibio, China) for 1 h at room temperature. Protein bands were visualized using an ECL luminescent reagent, and the film was scanned. The optical density values of the target bands were analyzed using Gel-Pro-Analyzer software (Media Cybernetics, Rockville, MD, USA). β-actin (Wanleibio, China) was used as the internal control. WB analysis all had three biological replicates.

Statistical analysis

Statistical analysis of the results of the three repetitions of the WB test was performed using GraphPad Prism 9.5 software. Data are presented as mean ± standard deviation ( ± s). Inter-group comparisons were made using one-way analysis of variance (ANOVA) with Tukey’s multiple comparison post hoc test, with P < 0.05 considered statistically significant.

Results

Successful establishment of the SE model in rats

After excluding animals that failed to develop grade 4 epileptic seizures or died during the model establishment process, a total of 27 rats with pilocarpine-induced SE were included in the subsequent experiments. Following the injection of physostigmine, the rats exhibited grade 1 epileptic seizures, characterized by rhythmic facial twitching, at 16.3 ± 2.1 min, which progressively worsened. Multiple limb convulsions or tonic contractions reached grade 4 epileptic seizures at 30.3 ± 2.4 min. Rats with grade 4 or higher seizures were administered subcutaneous diazepam to stop the convulsions. The average duration of epileptic seizures in these 27 rats was 2.0 ± 0.6 h.

Ketogenic diet reduced the frequency of epileptic seizures in the model rats, and this effect partially reversed after the diet was stopped

Throughout the experiment, the survival status of the rats was monitored weekly. Compared to the control group, the epileptic seizure rats showed decreased vitality during the first week, with dull and rough fur and reduced food intake. By the sixth week, the rats gradually returned to normal. In the fifth week, all SE rats undergoing ketogenic diet treatment began to lose their hair. In the SD + KD-CD group, hair loss gradually decreased after diet reversal in the eleventh week, and the fur regained its luster. The rats’ weight was recorded weekly, showing no significant statistical difference among the SE group, SE + KD group, and SD + KD-CD group (Fig. 2a).

Fig. 2.

a: Weight changes in each group of rats. b: Frequency of epileptic seizures in each group of rats during weeks 8–10.(SE vs. SE + KD P = 0.0021,SE + KD vs. SE + KD-CD P > 0.9999) c: Frequency of epileptic seizures in each group of rats during weeks 14–16..(SE vs. SE + KD P < 0.001,SE + KD vs. SE + KD-CD P = 0.0239) ***P < 0.001, **P < 0.01, *P < 0.05

From weeks 8 to 10, the rats in the control group did not exhibit any seizures. The average number of seizures in the other three groups was as follows: 37.9 ± 4.5 seizures in the SE group, 19.6 ± 3.8 seizures in the SE + KD group, and 19.6 ± 3.1 seizures in the SD + KD-CD group (Fig. 2b). Between weeks 14 and 16, the SE group had 38.9 ± 3.4 seizures, the SE + KD group had 13.7 ± 2.4 seizures, and the SD + KD-CD group had 22.7 ± 2.6 seizures (Fig. 2c).

Ketogenic diet inhibits the methylation of the RGMa gene in the hippocampal tissue of model rats and upregulates the expression level of RGMa protein

The methylation status of the RGMa promoter region in rat hippocampal tissues was analyzed using BSP. The results showed a significant increase in the proportion of methylated CpGs in the RGMa promoter in the SE group compared to the control group. In the SE + KD group, this proportion decreased, while in the SE + KD-CD group, it was higher than in the SE + KD group (Fig. 3a).

Fig. 3.

Ketogenic diet inhibits RGMa methylation in the hippocampal tissue of SE rats and upregulates RGMa expression. a: Percentage of methylation at RGMa CpGs. b-d: Changes in the expression of DNMT1 (b: Ctr vs. SE and SE vs. SE + KD P < 0.001,SE + KD vs. SE + KD-CD P = 0.0029), DNMT3a (c: Ctr vs. SE P < 0.001, SE vs. SE + KD P = 0.0011,SE + KD vs. SE + KD-CD P = 0.0267), and RGMa (d: Ctr vs. SE and SE vs. SE + KD P < 0.001,SE + KD vs. SE + KD-CD P = 0.2892). e: Immunofluorescence staining of the hippocampal tissue (x400). ***P < 0.001, **P < 0.01, *P < 0.05

WB analysis was performed to detect the expression of DNMT1 and DNMT3a in the hippocampal tissues of model rats. The results indicated that the expression levels of both DNMT1 and DNMT3a were elevated in the SE group, decreased in the SE + KD group compared to the SE group, and increased again in the SD + KD-CD group, consistent with the trend observed in the methylation status of the RGMa promoter region (Fig. 3b, c). Additionally, the expression levels of RGMa in the rat hippocampal tissues were assessed. WB results revealed a significant decrease in RGMa expression in the SE group compared to the control group and an increase in expression in the SE + KD group. No significant difference was observed between the SE + KD-CD group and the SE + KD group (Fig. 3d). Immunofluorescence double staining was used to detect the expression of RGMa in the hippocampal tissue. The nuclei of hippocampal neurons were stained green by NeuN, blue by DAPI, and RGMa was stained green. Compared with the control group, the immunofluorescence intensity of RGMa in the epilepsy group significantly decreased. The intensities in the SE + KD group and the SE + KD-CD group were higher than those in the SE group, but there was no significant difference between the two groups (Fig. 3e).

Ketogenic diet upregulates the palmitoylation of the AMPAR GluA1 subunit and downregulates its expression on the neuronal cell membrane

WB analysis was employed to assess the expression levels of the AMPAR GluA1 subunit on the hippocampal neuronal cell membranes across different groups. The results revealed a significant increase in GluA1 expression in the SE group compared to the control group. In contrast, the SE + KD group exhibited lower expression levels than the SE group, while the SE + KD-CD group showed higher expression than the SE + KD group (Fig. 4a).

Fig. 4.

Ketogenic diet downregulates the expression of the AMPAR GluA1 subunit on the neuronal cell membrane and upregulates its palmitoylation. a: Changes in GluA1 expression on the neuronal cell membrane.(Ctr vs. SE P < 0.001,SE vs. SE + KD P = 0.0018,SE + KD vs. SE + KD-CD P = 0.0134) b: Changes in GluA1 palmitoylation. (Ctr vs. SE P < 0.001,SE vs. SE + KD P = 0.001,SE + KD vs. SE + KD-CD P = 0.0017) ***P < 0.001, **P < 0.01, *P < 0.05

The ABE method was used to evaluate the palmitoylation of the GluA1 subunit in rat hippocampal tissue. Compared to the control group, the SE group displayed significantly lower palmitoylation levels of the GluA1 subunit. Palmitoylation in the SE + KD group was higher than in the SE group, whereas the SE + KD-CD group had lower palmitoylation levels than the SE + KD group (Fig. 4b).

Upregulated RGMa inhibits the FAK/Src pathway and increases the expression of zDHHC3

No significant differences were observed in the expression levels of FAK and Src across the groups. Further analysis of FAK phosphorylation at Tyr397 and Src phosphorylation at Tyr416 revealed increased phosphorylation in the SE group relative to the control. In the SE + KD group, phosphorylation levels were lower than in the SE group, whereas the SE + KD-CD group showed higher phosphorylation than the SE + KD group (Fig. 5a, b).

Fig. 5.

Upregulated RGMa inhibits the FAK/Src pathway and increases the expression of zDHHC3. a: Changes in FAK expression (Ctr vs. SE P = 0.7327,SE vs. SE + KD P = 0.5669,SE + KD vs. SE + KD-CD P = 0.6452) and FAK phosphorylation(Ctr vs. SE, SE vs. SE + KD and SE + KD vs. SE + KD-CD P < 0.001). b: Changes in Src expression (Ctr vs. SE P > 0.9999,SE vs. SE + KD P = 0.9943,SE + KD vs. SE + KD-CD P = 0.3802) and Src phosphorylation(Ctr vs. SE P > 0.9999,SE vs. SE + KD P = 0.9943,SE + KD vs. SE + KD-CD P = 0.3802). c: Changes in zDHHC3 expression and zDHHC3 phosphorylation(Ctr vs. SE, SE vs. SE + KD P < 0.001, SE + KD vs. SE + KD-CD P = 0.0031). d: Immunofluorescence staining of the hippocampal tissue (x400).***P < 0.001, **P < 0.01, *P < 0.05, ns P > 0.05

WB analysis also indicated no significant differences in the expression levels of zDHHC3 among the groups. Phosphorylation levels of zDHHC3 were measured using tyrosine phosphatase assays, showing significant increases in the SE and SE + KD-CD groups compared to the control. No significant difference in phosphorylation was observed between the SE + KD group and the control group (Fig. 5c).

Discussion

Epilepsy is a chronic neurological disorder characterized by recurrent and highly synchronized abnormal neuronal discharges, leading to brain dysfunction. Despite the continuous advancement in medicine and the increasing variety of anti-epileptic drugs, some patients continue to experience frequent seizures even with consistent drug treatment, progressing to refractory epilepsy. The causes of refractory epilepsy are multifaceted, contributing to the lack of specific treatment options. The ketogenic diet, a specialized dietary therapy, involves a high-fat, low-carbohydrate, and moderate-protein regimen that induces a state of ketosis in the body. This metabolic shift has been shown to help reduce the frequency of epileptic seizures. Studies indicate that in patients undergoing a standardized ketogenic diet, seizures can be reduced by over 90%, or may even disappear, particularly in pediatric populations [23]. In this study, a rat model of SE induced by pilocarpine was used to investigate the effects of the ketogenic diet on RGMa methylation, the FAK/Src-zDHHC3 signaling axis, and the palmitoylation of the AMPAR GluA1 subunit, aiming to explore the molecular mechanisms by which the ketogenic diet may aid in the treatment of epilepsy.

In this study, the ketogenic diet was found to reduce seizure frequency in the epilepsy rat model, with a slight increase in seizure frequency observed after returning to a normal diet. This further corroborates the anticonvulsant effects of the ketogenic diet, indicating that these effects do not immediately dissipate upon diet reversal. Our results demonstrated a significant increase in the methylation levels of the RGMa promoter region and the expression of DNMTs in the hippocampal tissues of the model rats, alongside a marked decrease in RGMa protein expression. Following ketogenic diet intervention, both the methylation levels of the RGMa promoter and DNMT expression were reduced, while RGMa expression was upregulated. Upon reversion to a standard diet, although methylation levels and DNMT expression slightly increased, RGMa expression showed a downward trend compared to pre-reversal levels, though this change was not statistically significant. These findings suggest that the ketogenic diet may inhibit DNMT activity and reduce RGMa promoter methylation and that this effect may persist for a prolonged period. AMPARs are crucial for mediating rapid excitatory synaptic transmission, and the phosphorylation and palmitoylation of the GluA1 subunit play important regulatory roles in receptor membrane trafficking and synaptic function [24]. Palmitoylation influences GluA1 expression on the cell membrane—higher palmitoylation leads to a decrease in membrane expression, whereas lower palmitoylation results in an increase [25]. In the present study, the expression of GluA1 on the neuronal membranes of SE rats was significantly elevated, while its palmitoylation level was markedly reduced. After ketogenic diet intervention, the palmitoylation of GluA1 was normalized, and its membrane expression decreased. These findings indicate that the ketogenic diet may mitigate excessive neuronal excitability by promoting GluA1 palmitoylation and preventing its over-accumulation on the membrane. Palmitoylation, a reversible post-translational modification, facilitates the dynamic shuttling of proteins between the cell membrane and cytoplasm, playing a pivotal role in neurodegenerative diseases, demyelinating conditions associated with neuroinflammation, and epilepsy. Palmitoylation of proteins is primarily catalyzed by DHHC family proteins (e.g., zDHHC3) [18, 26–28]. Therefore, the ketogenic diet may help restore normal neuronal excitability and reduce epileptic seizures by modulating AMPAR palmitoylation.

RGMa is involved in the guidance of synapses and neuronal survival and may also regulate the FAK/Src signaling pathway, influencing the onset of epilepsy [13]. The present study revealed a significant increase in the phosphorylation levels of FAK (Tyr397) and Src (Tyr416) in the hippocampal tissues of the SE group rats. However, following ketogenic diet intervention, phosphorylation levels were significantly reduced. FAK and Src are key intracellular non-receptor tyrosine kinases that interact with each other. FAK’s signaling capability relies on the phosphorylation of multiple kinase domains (Tyr397, Tyr567, and Tyr577), and the activation of this phosphorylation cascades to regulate downstream molecules, including the activity of the PAT zDHHC3 [29]. Li et al. demonstrated that in breast cancer, a reduction in RGMa expression leads to increased FAK phosphorylation, which activates the FAK/Src pathway. Src, in turn, further regulates the activity and expression of zDHHC3 through phosphorylation [30]. The findings of this study align with those observations, suggesting that RGMa may exert its effects by inhibiting the FAK/Src pathway, reducing zDHHC3 phosphorylation, and thus modulating the palmitoylation of GluA1.

In summary, this research has identified correlations between RGMa methylation, AMPAR GluA1 subunit palmitoylation, refractory epilepsy, and ketogenic diet therapy. In the hippocampal tissues of epileptic model rats treated with a ketogenic diet, a decrease in DNMT1/DNMT3a expression, a reduction in RGMa DNA methylation, and an increase in RGMa expression were observed. This upregulation of RGMa was associated with enhanced GluA1 palmitoylation and a reduction in epileptic seizures. The modulation of GluA1 palmitoylation by RGMa may be mediated through the FAK/Src-zDHHC3 pathway. However, due to time constraints, this study has certain limitations. While this study examined changes in the phosphorylation levels of FAK/Src and zDHHC3 in conjunction with RGMa, further experimental validation of the role of this signaling axis was not performed. Furthermore, alterations in other signaling pathways were not investigated. These aspects will be explored in future research.

Conclusion

This study demonstrates that in the rat epilepsy model, the ketogenic diet may partially reduce the frequency of epileptic seizures and improve the phenotype of refractory epilepsy by inhibiting RGMa DNA methylation, upregulating RGMa protein expression, and enhancing the palmitoylation of the GluA1 subunit. These findings suggest that the RGMa-FAK/Src-zDHHC3 signaling axis may be involved in the antiepileptic effect of the ketogenic diet, but further research is needed to determine its specific contribution in multiple mechanisms.

Author contributions

T.Y and W.X conceived and designed research; JN.X, L.H and JT.L collected data and conducted research; T.Y and JN.X analyzed and interpreted data; JN.X wrote the initial paper; T.Y revised the paper; T.Y had primary responsibility for the final content. All authors read and approved the final manuscript.

Funding

Liaoning Province Science and Technology Plan Joint Program(2023-MSLH-391). National Natural Science Foundation of China (82470042).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All animal experiments were performed according to the National Institutes of Health guidelines for the Care and Use of Laboratory Animals (NIH publication 80 − 23, revised in 1996) and the necessary approval (2024PS967K) was obtained from the Animal. Ethics Committee of Shengjing Hospital Affiliated with China Medical University.

Competing interests

The authors declare no competing interests.