If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Physiology, Universidade Federal de Sao Paulo, Sao Paulo, BrazilCenter for Research, Education and Innovation, Instituto Jô Clemente, Sao Paulo, Brazil
PTZ kindling was used to explore disease-modifying effects of ANT-DBS in rats.
•
ANT-DBS reduced seizure susceptibility and epileptogenesis.
•
ANT-DBS reduced hippocampal expression of ADK.
•
The clinical utility of ANT-DBS goes beyond seizure control in epilepsy.
Abstract
Background
Deep brain stimulation (DBS) of the anterior nucleus of the thalamus (ANT) is an emerging therapy to provide seizure control in patients with refractory epilepsy, although its therapeutic mechanisms remain elusive.
Objective
We tested the hypothesis that ANT-DBS might interfere with the kindling process using three experimental groups: PTZ, DBS-ON and DBS-OFF.
Methods
79 male rats were used in two experiments and exposed to chemical kindling with pentylenetetrazole (PTZ, 30 mg/kg i.p.), delivered three times a week for a total of 18 kindling days (KD). These animals were divided into two sets of three groups: PTZ (n = 26), DBS-ON (n = 28) and DBS-OFF (n = 25). ANT-DBS (130 Hz, 90 μs, and 200 μA) was paired with PTZ injections, while DBS-OFF group, although implanted remained unstimulated. After KD 18, the first set of PTZ-treated animals and an additional group of 11 naïve rats were euthanized for brain extraction to study adenosine kinase (ADK) expression. To observe possible long-lasting effects of ANT stimulation, the second set of animals underwent a 1-week treatment and stimulation-free period after KD 18 before a final PTZ challenge.
Results
ANT-DBS markedly attenuated kindling progression in the DBS-ON group, which developed seizure scores of 2.4 on KD 13, whereas equivalent seizure scores were reached in the DBS-OFF and PTZ groups as early as KD5 and KD6, respectively. The incidence of animals with generalized seizures following 3 consecutive PTZ injections was 94%, 74% and 21% in PTZ, DBS-OFF and DBS-ON groups, respectively. Seizure scores triggered by a PTZ challenge one week after cessation of stimulation revealed lasting suppression of seizure scores in the DBS-ON group (2.7 ± 0.2) compared to scores of 4.5 ± 0.1 for the PTZ group and 4.3 ± 0.1 for the DBS-OFF group (P = 0.0001). While ANT-DBS protected hippocampal cells, the expression of ADK was decreased in the DBS-ON group compared to both PTZ (P < 0.01) and naïve animals (P < 0.01).
Conclusions
Our study demonstrates that ANT-DBS interferes with the kindling process and reduced seizure activity was maintained after a stimulation free period of one week. Our findings suggest that ANT-DBS might have additional therapeutic benefits to attenuate seizure progression in epilepsy.
Progression of temporal lobe epilepsy in the rat is associated with immunocytochemical changes in inhibitory interneurons in specific regions of the hippocampal formation.
]. Approximately 79 million people worldwide have epilepsy and over 30% of those do not respond to conventional antiseizure drugs and are not candidates for surgical resective therapy. It is therefore imperative to develop alternative treatments leading to seizure remission. One such alternative treatment is deep brain stimulation (DBS) [
DBS is based on the delivery of an electrical stimulus to the brain parenchyma via implanted electrodes and has been increasingly used for the clinical treatment of epilepsy [
]. The (ANT-DBS) SANTE clinical trial and long-term follow up studies demonstrated two main results: (1) ANT-DBS decreased spontaneous recurrent seizures (SRS) acutely over at least three months (anti-ictogenic effect) [
]; (2) After a five-year follow up period, ANT-DBS was associated with further progressive decreases in spontaneous recurrent seizures (SRS), as well as reductions in seizure severity, which were accompanied by both cognitive and clinically significant improvements in quality of life measures [
], its underlying mechanisms remain understudied. Previously, we demonstrated that ANT-DBS performed in the pilocarpine model of temporal lobe epilepsy in rats at parameters that approximate those used in clinical practice [
], we hypothesized that ANT-DBS attenuates kindling through an adenosine-mediated mechanism. We tested this hypothesis using a rodent model of PTZ kindling.
2. Materials and methods
All animal experiments comply with the ARRIVE guidelines and were carried out in accordance with the EU Directive 2010/63/EU for animal experiments. Eight-week-old male Wistar rats (n = 90; 225–280g) were purchased from the Centre for the Development of Experimental Models in Medicine and Biology (Centro de Desenvolvimento de Modelos Experimentais para Medicina e Biologia – CEDEME; Sao Paulo, SP) and the procedures were approved by the local ethical committee (CEUA: 7365030919/2019). The animals were maintained in the institutional animal facility under controlled conditions (12 h light/dark cycle, at 21 ± 1 C with 50% relative humidity) with chow pellets and water provided ad libitum. Animals were allowed to acclimate to the animal facility for one week before starting the surgical procedures. The experimental design is represented in Fig. 1.
Experiment 1: Interference of ANT-DBS in kindling progression and histopathology (PTZ: n = 12; DBS-OFF: n = 10; DBS-ON: n = 14; naïve: n = 11). Experiment 2: Persistence of kindling suppression effect during PTZ challenge (PTZ: n = 14; DBS-OFF: n = 15, DBS-ON: n = 14). One week before starting the PTZ kindling, 4 animals per group underwent surgery to implant EEG electrodes, in addition to a pair of stimulating electrodes implanted into the anterior nucleus of the thalamus (ANT) in DBS-ON and DBS-OFF groups. Over six weeks all animals received PTZ-injections (30 mg/kg, i.p.) 3 times a week (Mon-Wed-Fri), delivered over a total of 18 PTZ kindling days (KD). Animals from Experiment 1 were sacrificed after the last PTZ injection, while animals from Experiment 2 were allowed to rest for a week before a PTZ challenge.
To evaluate whether ANT-DBS disrupts the kindling process in adult animals (Experiment 1) we implanted bilateral electrodes into the ANT and compared the seizure scores reached after each PTZ injection (30 mg/kg, i.p.) during stimulation (DBS-ON group, n = 14) or in its absence DBS-OFF rats (n = 10). Age-matched animals without electrode implants who received PTZ injections (PTZ group, n = 12) served as non-implanted controls. All animals were sacrificed on KD 18 for histopathological analysis (Experiment 1, Fig. 1). To evaluate whether ANT-DBS exerts lasting therapeutic benefits (Experiment 2), similar sets of animals (PTZ: n = 14, DBS-ON: n = 14, DBS-OFF: n = 15) underwent the protocol described above, with the addition of one week without any intervention. At the end of this ‘wash out’ period all animals received a final PTZ challenge (30 mg/kg, i.p.) and the seizure score was determined. Assessment of the percent of animals that displayed generalized seizures (seizure scores ≥3) during kindling days 16, 17 and 18 included animals from both experiments (PTZ: n = 26, DBS-ON: n = 25, DBS-OFF: n = 28).
2.1 Surgical procedures and ANT stimulation parameters
Under deep ketamine/xylazine (100/7.5 mg/kg i.p.) anesthesia, DBS-ON and DBS-OFF groups received insulated stainless steel electrodes (cathodes; 250 μm diameter; 0.5 mm exposed length) implanted bilaterally into the ANT (AP: −1.5, ML: ± 1.5, DV: 5.4) [
]. A screw electrode implanted over the right somatosensory cortex was used as the anode. Four rats per group had also implanted unilateral hippocampal electrode (AP: +3 mm; ML: ±4 mm; DV: −2.5 mm ventral to the bregma) for electroencephalographic (EEG) recordings during PTZ treatments. Stimulation settings were in the range of those used in our previous studies [
]: 130 Hz, 90 μs, and 200 μA (St Jude MTS, Plano, TX). ANT-DBS was paired with each PTZ-kindling injection, started 2 h prior to each PTZ injection and continued for 30 min after the injection. This stimulation schedule matched a previous study conducted by our group using pilocarpine to induce status epilepticus (SE) [
]. At the end of experiments, correct electrode placement was validated in all DBS animals (Fig. 5A shows the electrode localization in a representative animal).
2.2 Pentylenetetrazole (PTZ) kindling and EEG analysis
Chemical kindling was induced in male Wistar rats (n = 79) according to a standard protocol by giving an i.p. injection of a subconvulsant dose of 30 mg/kg pentylenetetrazole (PTZ, Sigma, USA), on three days each week (Monday, Wednesday, Friday) over the course of 6 weeks amounting to a total of 18 injections [
]. Rats were considered to be fully kindled when 3 consecutive generalized seizures were observed. Following each PTZ injection, the animals were observed for 30 min and the seizure scores were recorded according to a modified Racine scale [
] as follows: stage 0, no response; stage 1, grooming and hyperactivity; stage 2, head nodding and tremor; stage 3, bilateral forelimb clonus; stage 4, hindlimb and forelimb clonus with rearing; stage 5, generalized clonic seizures with loss of postural control. The weight of the animals was measured weekly to adjust the PTZ dosage. The kindling protocol ended once the animals had received all 18 PTZ injections.
In our second experiment, the entire kindling protocol was repeated (Experiment 2, Fig. 1); however, after receiving the last kindling stimulation on KD 18, animals were kept undisturbed for 7 days (PTZ treatment ‘wash-out’, no DBS-stimulation) and then subjected to a single PTZ challenge dose (30 mg/kg, i.p.).
EEGs were recorded for 1h with a BNT-36 system (Lynx, Brazil) using EMSA software (EMSA, Brazil), starting 30 min prior to each PTZ injection and continued 30 min afterwards. Signals were amplified, band pass filtered (0.1–30 Hz) and digitalized (200 samples/second). The quantitative EEG seizure analysis included the most common frequency observed during seizure activity and peak frequency (Hz), peak power (μV2) in each frequency peak, as well as the power in each individual frequency band during each seizure. For the spectral analysis, a frequency heatmap was created based on epileptic activity; lighter colors represent an increased power for the designated frequency, while darker colors represent decreased power. For each frequency interval, the EEG power was calculated using a fast Fourier transform (FFT) algorithm. The epochs for the analysis were delineated by the start and finish of each seizure, in order to avoid the inclusion of data from non-seizure baseline activity.
2.3 Histology, immunofluorescence analysis, and Western blot
In experiment 1 forty-seven animals were either perfused (n = 24) or decapitated (n = 23), (Fig. 1). For this, rats were deeply anesthetized with pentobarbital (thionembutal, 50 mg/kg, ip). For histology and immunofluorescence analysis, the animals were perfused through the left ventricle with 300 ml of saline buffered solution (room temperature), followed by 300–400 ml of 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. Thereafter, the brains were removed, kept overnight in 30% sucrose solution, frozen on dry ice and cut on a cryostat (40-μm thick sections). A one-in-three series of consecutive sections was used for double-labeling immunofluorescence (IF). For IF, sections were incubated overnight with primary antibodies: ADK (1:500, polyclonal rabbit antibody, Bethyl Labs) and RNA binding protein HuR (3A2) (1:500, polyclonal mouse antibody, Santa Cruz Technology) and then with Alexa Fluor® 568 and Alexa Fluor® 488 (1:600, anti-rabbit and anti-mouse, Molecular Probes, US), respectively. Three hippocampal coronal sections per subject were analyzed using a fluorescence microscope (Nikon E600FN, Japan) connected to a high-resolution digital camera (Nikon DXM1200) and converted into digital signals transmitted to a computer. Co-localization analysis included the visual inspection of the size and shape of the cells. Nissl stained section were used for validation of the correct electrode placement and cell counting and. Both IF and Nissl stained sections were counted under the same protocol. In brief, the number of cells was counted in the dorsal hippocampal subregions (dentate gyrus, CA1, CA3, and hilus), bilaterally, and in triplicate. This count was conducted by a researcher unaware of the treatment of the experimental groups with the help of an optical microscope (Nikon, Eclipse E600FN) with a checkered grid. The area of interest had two forbidden lines (exclusion) and two acceptable lines (inclusion). The cells that touched the forbidden lines were not counted, but those that touched the acceptable lines and were within the area of interest were counted. Each hippocampal subregion was divided into three fields, each field 5000 μm2. Thus, the outcome was determined by the average of field values (Fig. 5).
For immunoblot analysis, animals were decapitated, the hippocampus dissected, weighed and frozen in liquid nitrogen vapor, then stored at −80 °C until use. Samples were homogenized in radioimmunoprecipitation assay buffer (RIPA) containing protease inhibitors (#11697498001, Roche, Switzerland). Protein content was assessed using a Thermo Fisher Bicinchoninic Acid protein assay kit. For electrophoresis, 30 μg of aqueous protein extracts were loaded and separated on 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels and transferred to Hybond®-P polyvinylidene fluoride (PVDF) membranes (Amersham, US). The blots were incubated overnight at 4 °C in Tris buffered saline (TBS: 20 mM Tris, 150 mM NaCl, 0.1% Tween, pH 7.4) containing 3% non-fat dry milk and polyclonal rabbit anti-ADK (Bethyl Labs, 1:4500). The membranes were washed in TBS and incubated in TBS containing 5% non-fat dry milk and goat anti-rabbit secondary antibody (Thermo Fisher Scientific, 1:5000). Immunoreactivity was scanned using a Bio-Rad Touch imager and digitalizer using Bio-Rad Image Lab software. Protein quantification was performed using Image J V. 1.53 (NHI, USA) and presented as the ratio of ADK/alpha-tubulin.
2.4 Statistical analysis
Parametric assumptions from histology, IF and WB results were tested using the Shapiro-Wilk test (for normality) and followed by one-way ANOVA. Results are expressed as mean ± SEM. Repeated measures analysis of variance (RM ANOVA) was used for the analysis of kindling data. Both were followed by Tukey-Kramer post hoc tests. Significance was set at P < 0.05.
3. Results
3.1 Seizure data- experiment 1
With cumulative PTZ injections, seizures became gradually more severe for all groups, but especially for the PTZ and DBS-OFF groups (Fig. 2A). One animal from the PTZ group died from severe seizures during kindling and was excluded from this study. An average seizure stage (SS) of 3 was reached on the 8th KD for both DBS-OFF (SS: 3.2 ± 0.1) and PTZ (SS: 3.1 ± 0.1) animals, while the DBS-ON group exhibited substantially lower seizure scores on the 8th KD (SS: 2.1 ± 0.3). While seizure scores continued to increase towards the end of the PTZ kindling protocol for the PTZ and PTZ-SHAM-DBS groups, these scores remained stable for the DBS-ON group. Post hoc comparisons indicate that DBS-OFF and PTZ groups did not differ in their seizure scores during the study. The DBS-ON group had lower seizure scores than all other groups during all kindling sessions subsequent to KD 6. Significant differences in seizure scores compared to KD1 were only observed at KD 13 for the DBS-ON group (mean SS = 2.4 ± 0.1; P = 0.04; Fig. 2A), which was in marked contrast to increased SS on KD 5 (mean SS = 3.0 ± 0.1; P < 0.001) and on KD 6 (mean SS = 2.9 ± 0.2; P < 0.001) for the DBS-OFF and PTZ groups, respectively.
Fig. 2Suppression of PTZ kindling epileptogenesis by ANT-DBS in adult rats.
A: Intra group difference of mean seizure score (SS) from KD1 was observed on KD 5 for PTZ (P = 0.009) and on KD 6 for DBS-OFF groups (P = 0.01), respectively, whereas the DBS-ON group reached the same seizure score difference on KD 13 (P = 0.04). Differences in seizure score (SS) between DBS-ON and other groups were observed for the first time during KD 6, and continued from KD 8 until the end of the protocol. In B, the mean SS for DBS-ON group was lower than the other two groups. Intra-group difference of SS from KD1 only occurred on KD 14 in DBS-ON group, whereas it occurred on KD 3 in PTZ and on KD 6 in DBS-OFF groups. One week of washout did not reverse the ANT-DBS effects after a PTZ challenge, which resulted in a significant attenuation of the mean seizure score of the DBS-ON group compared to the PTZ and DBS-OFF groups (P < 0.001). C. The percentage of animals that reached seizure scores ≥3 for three consecutive days (KD 16, 17 and 18) in the ANT-DBS group was significantly smaller (21%) than in the DBS-OFF (74%) and PTZ (94%) groups, P < 0.001. ˆ indicates intra-group differences when compared to the mean SS from KD1. ∗ indicates differences between DBS-ON and PTZ. # indicates differences between DBS-ON and DBS-OFF.
To determine lasting effects of ANT-DBS stimulation, animals had a one-week ANT-DBS and PTZ free ‘washout period’ following the 18 PTZ-kindling sessions (Experiment 2), before being challenged with a final PTZ injection, that was not paired with ANT-DBS. All animal groups in experiment 2 replicated the seizure scores reported in in Experiment 1 (Fig. 2B) during the kindling protocol. After 7 days of washout, a statistically significant between group effect was observed (F [
] = 26.30, P < 0.001). During the PTZ challenge, the mean seizure score for the DBS-ON group (2.7 ± 0.2) was significantly lower than for both the DBS-OFF (4.3 ± 0.1) and PTZ (4.5 ± 0.1) groups, while these latter two groups did not differ from each other (P = 0.67). It is worth noting that of the 3 animals from the DBS-ON group that achieved seizure scores ≥3 during the three last consecutive kindling days, none exhibited further seizure severity increases during the PTZ challenge.
On KD 18, 94% of the PTZ and 74% of the DBS-OFF groups had mean seizure scores ≥3, whereas only 21% of the DBS-ON group animals achieved similar scores (P < 0.001; Fig. 2C). Our data show that ANT-DBS disrupts the kindling acquisition process, as demonstrated by reduced seizure scores which further persisted across the one-week period without stimulation, supporting the idea of a sustained anti-kindling effect of ANT-DBS.
Because the desynchronization of brain activity caused by DBS can be a powerful modulator of epileptic activity [
], we used the fast Fourier transform to analyze seizure stages ≥3 (Fig. 3A). This revealed that seizures under the ANT-DBS therapy have a lower frequency (6 ± 1.8 Hz; P = 0.0004) than in the PTZ group (18.4 ± 1.9 Hz), which was similar to the DBS-OFF group (13.5 ± 4.1 Hz; P = 0.35). When the mean power was compared among the three groups, at each specific frequency interval, an 8 Hz peak amplitude for PTZ group was revealed, reaching 67.3 ± 11.7 μV2 as compared to 29.4 ± 4.4 μV2 observed in the DBS-ON group (P = 0.02). There was no difference between the DBS-OFF group (41.8 ± 12.0) and both other groups. At a frequency of 20 Hz the mean peak power in DBS-treated animals was even more reduced (PTZ = 292.3 ± 9.4 μV2vs. DBS-ON = 42.2 ± 11.5 μV2; P = 0,003; Fig. 3B and C). Although the DBS-OFF group had, at 20 Hz, a mean power peak of 138 ± 74.6 μV2, as at 8 Hz it also did not differ from the other groups. Desynchronization of neural networks after ANT-DBS has a strong correlation with both alpha and theta band activity, and it was within these bandwidths that the specific frequency interval peaks of the DBS-ON group differed from both DBS-OFF and PTZ groups.
Fig. 3EEG traces, spectral and FFT power analysis of stage 3 seizures from DBS-ON and PTZ animals.
A. Representative electroencephalographic (EEG) traces of stage 3 seizures from the PTZ and DBS-ON animals with the respective heatmaps indicating the frequency peak (Hz) as well as the fast Fourier transform (FFT) analysis of peak power (μV). B. FFT traces representing power in each frequency band during both seizures presented in A, for PTZ and DBS-ON. C. Differences in mean power (μV2) were observed between PTZ and DBS-ON animals on KD 15 for the 8 Hz and 20 Hz frequencies. No differences were observed between PTZ and DBS-OFF groups.
We found prominent tissue atrophy in the hippocampus in the PTZ group (Fig. 4) that was prevented by ANT-DBS. Neuronal cell counting in the hippocampus of the DBS-ON group (n = 7) revealed a significant preservation of Nissl positive cells compared to the PTZ (n = 6) and naïve (N = 5) groups (F[9, 82, 898] = 13.81; P = 0.000), as observed in all studied hippocampal subfields: CA1 (DBS-ON: 48.1 ± 0.4 vs PTZ: 41.9 ± 1.8, P = 0.0001; vs naïve: 63.7 ± 1.4, P = 0.0001), CA3 (DBS-ON: 47 ± 5.8 vs PTZ: 28.2 ± 0.8; P = 0.001; vs naïve: 50.2 ± 2.4, P = 0.0003) and hilus (DBS-ON: 5.7 ± 0.1 vs PTZ: 4.4 ± 0.3; P = 0.01; vs naïve: 7.3 ± 0.5, P = 0.0001) (Fig. 4B). No difference was detected between the PTZ and DBS-OFF groups (analysis not shown).
Fig. 4ANT-DBS is neuroprotective during PTZ-kindling in rats.
A. Bilateral electrode placement in the anterior thalamic nucleus (arrows). Calibration bar: 100 μm. B. Hippocampal histological analysis of Nissl-stained sections (number of cells/5000 μm2) from both groups reveal extensive neuroprotection in the ANT-DBS group. C. Nissl-stained sections depicting hippocampal fields in representative animals from the PTZ, DBS-OFF, DBS-ON, and naïve groups. Results are expressed as mean ± SE, P ≤ 0.005. ∗ indicates differences between DBS-ON and PTZ groups; # indicates differences DBS-ON and DBS-OFF; & indicates differences between DBS-ON and naïve groups.
Fig. 5Hippocampal ADK levels. A. Western blot for ADK and α-tubulin expression in the hippocampus of three representative animals (PTZ, DBS-ON and naïve groups). Decreased ADK expression observed in DBS-ON animals when compared to PTZ animals and to treatment and surgery naïve controls. ADK values were normalized to α-tubulin. B. Immunofluorescence analysis depicting CA1, CA3, and hilar HUR-positive neurons (green) of two representative animals from the PTZ and DBS-ON groups confirm the reduction of ADK-stained cells (red) in the hippocampus of the DBS-ON treated group. The number of ADK-labeled cells is reduced in all hippocampal fields of the DBS-ON group when compared with the PTZ group. Calibration bar: 100 μm ∗ indicates differences between DBS-ON and PTZ groups. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
], and because DBS is known to affect adenosine levels in the brain, we quantified the expression of ADK in our experimental animals through quantitative immunoblotting (Fig. 5A). Multiple comparison tests indicated a significant group effect (χ2 = 9.935, P < 0.007). Importantly, a significant decrease in the hippocampal expression of ADK was observed in rats from the DBS-ON group (0.4 ± 0.1) when compared to those from the naïve (0.7 ± 0.06, P = 0.05) and PTZ (0.7 ± 0.05, P = 0.03) groups, indicating that DBS might attenuate the epileptogenic process by downregulation of pro-epileptogenic ADK. In contrast, pairwise comparisons revealed no significant difference in hippocampal ADK expression between PTZ and naïve (P = 0.9) groups.
Double labeling immunofluorescence analysis indicated that DBS-ON group had less ADK-stained cells in in CA1 (DBS-ON: 9.8 ± 0.9 vs PTZ: 18.6 ± 0.9; P < 0.0001), CA3 (DBS-ON: 6.0 ± 0.7 vs PTZ: 14.2 ± 0.6; P < 0.0001) and hilus (DBS-ON: 15.2 ± 0.6 vs PTZ: 30 ± 1.5; P < 0.0001). Fig. 5B show that only few neurons were positively stained for ADK, which was predominant in astrocytes in all studied groups.
4. Discussion
In this study, we have demonstrated that ANT-DBS attenuates kindling in the rat PTZ model of epileptogenesis. Importantly, we demonstrated that ANT-DBS has lasting effects that go beyond mere seizure suppression and suggest possible antiepileptogenic effect of ANT-DBS. We also found that ANT-DBS reduced ADK expression and neuronal cell loss in the hippocampus of rats treated with PTZ. Because ADK is a driver of the epileptogenic process [
], we propose that ANT-DBS driven downregulation of ADK can act as mediator of its kindling suppression effects. Reduced expression of ADK following ANT-DBS is also consistent with the demonstrated elevation of adenosine following DBS stimulation [
] demonstrated the efficacy and safety of anterior thalamus deep brain stimulation to treat epilepsy. Based on this trial, in 2018, the U.S. Food and Drug Administration approved ANT-DBS therapy for epilepsy, with an indication as follows: “Bilateral stimulation of the anterior nucleus of the thalamus (ANT) for epilepsy is indicated as an adjunctive therapy for reducing the frequency of seizures in individuals 18 years of age or older diagnosed with epilepsy characterized by partial onset seizures with or without secondary generalization that are refractory to three or more antiepileptic drugs”. Since then, 23 clinical studies have demonstrated that ANT-DBS has beneficial long-term effects reducing seizures and improving the quality of life of persons with epilepsy (for review, see Ref. [
]). Clinically, the major focus of ANT-DBS therapy has been the goal of improved seizure control. However, the SANTE trial also suggests that ANT-DBS might have additional disease-modifying properties that go beyond mere seizure control. The long-term follow-up of the SANTE study showed a reduction for the most severe seizure type with an overall median seizure reduction of 70%. When improvements in seizure reduction were observed, they were either temporally related to the stimulation change or occurred after a cumulative effect of the stimulation. In terms of quality of life and positive cognitive outcomes, the percentage of subjects reported to be satisfied or greatly satisfied with the results of their therapy was 84% (54/64) [
Kindling induced by PTZ injections is a widely used model to screen for antiepileptic drugs and to study epileptogenesis in several animal species including mice, rats, guinea pigs and non-human primates [
]. The PTZ kindling model enables the tight experimental control of gradual seizure severity progression. The underlying principle is the old adage that “seizures beget seizures”. This means that interventions that interfere with a kindling-induced seizure also attenuate the progression of epilepsy, which is of potential therapeutic value. Here, we observed a robust attenuation of the kindling process in ANT-DBS treated animals, as more PTZ injections were necessary to induce seizures of similar severity in stimulated animals compared to non-stimulated ones (right-shift of kindling curve). These findings indicate that ANT-DBS attenuated kindling epileptogenesis. Using a similar experimental design, a previous study reported an anti-epileptogenic effect by pairing ketogenic diet treatment with PTZ kindling stimulations [
]. However, as we used ANT-DBS paired with PTZ injections, it can also be interpreted that the ANT-DBS-induced seizure attenuation per se interfered with the kindling stimulus and thereby suppressed kindling epileptogenesis. To determine whether ANT-DBS has additional disease modifying effects independent from direct seizure reduction, future studies would need to apply ANT-DBS stimulations during kindling-free intervals (e.g. ANT-DBS delivered after the cessation of recorded seizures, or on the alternate days when PTZ injections were not given). The maintenance of a higher seizure threshold after a week of ‘treatment washout’, without any further ANT-DBS stimulations, suggests lasting therapeutic benefits of ANT-DBS, although ADK expression was not measured at the end of experiment 2. The current results reinforce the importance of mechanistic studies to understand potential disease modifying effects of DBS.
Although high frequency stimulation of the ANT has been widely used to treat partial seizures [
] there are few studies on ANT DBS-related mechanisms of action. The variety of possible mechanisms that culminate with seizure reduction include reduction of cell death [
Deep brain stimulation of the anterior nucleus of the thalamus reverses the gene expression of cytokines and their receptors as well as neuronal degeneration in epileptic rats.
Anterior thalamic nuclei deep brain stimulation reduces disruption of the blood-brain barrier, albumin extravasation, inflammation and apoptosis in kainic acid-induced epileptic rats.
]. In line with those mechanisms, our current results demonstrate that ANT-DBS resulted in hippocampal neuroprotection and reduced expression of ADK. As studied extensively in the past, in the adult brain ADK is predominantly expressed in astrocytes, which serve as metabolic sink for the clearance of adenosine; any treatments which reduce ADK expression or activity increase adenosine and have antiictogenic and antiepileptogenic properties [
]. Considering that ADK expression increases only in SRS models with an inflammatory component, which includes glial activation and hippocampal sclerosis [
in: Noebels J. Avoli M. Rogawski M. Olsen R. Delgado-Escueta A. Jasper's basic mechanisms of the epilepsies. Oxford University Press,
Oxford2012: 1150-1160
], and that chemical or electrical kindling models do not affect endogenous ADK expression, the reduction of ADK expression observed here is likely a direct effect of ANT-DBS. Accordingly, we have previously demonstrated that control, non-epileptic rats triplicate the baseline hippocampal adenosine release under ANT-DBS [
]. This is in agreement with our results, where ADK expression was only reduced in the ANT-DBS group, whereas the PTZ group had normal ADK expression levels comparable to naïve control animals.
The reduction of hippocampal ADK expression reported here is in line with a previously proposed role of adenosine as a potential candidate to underlie the neuroprotective, anti-inflammatory [
]. The well-characterized anticonvulsant effects of ANT-DBS are likely at least in part mediated by activation of G-protein coupled adenosine A1 receptors (A1R), which inhibit the pre-synaptic release of glutamate and act by stabilization of the post-synaptic membrane potential [
As pointed out earlier, over 30% of all epilepsies are refractory for current treatments and there is a major unmet clinical need for the development of therapeutic strategies capable of preventing or modifying the development of epilepsy or its progression. The epileptogenic process involves multiple mechanisms including brain inflammation, microglial and astroglial activation, epigenetic reprogramming and dysregulation in adenosine metabolism [
]. The epileptogenic process is driven by acquired pathological overexpression of ADK in conjunction with astrogliosis, whereby overexpression of ADK deprives the brain of its endogenous anticonvulsant and seizure terminator adenosine [
]. Increased ADK expression during epileptogenesis drives increased DNA methylation, which independently has been shown to be implicated in the epileptogenic process [
]. Because pathological increases in ADK drive the epileptogenic process, therapeutic adenosine augmentation is a promising approach for epilepsy prevention. Adenosine's therapeutic benefits have been previously demonstrated in kindling and pilocarpine post status epilepticus (SE) rodent models [
] and adenosine released via silk-based brain implants prevented epilepsy progression in a systemic kainic acid model of temporal lobe epilepsy (TLE) in rats, based on an epigenetic mechanism, whereby adenosine reversed maladaptive DNA hypermethylation [
] and as suggested by our current data, ANT-DBS induced adenosine augmentation might have additional disease modifying benefits, which are not yet fully exploited in the clinic.
In conclusion, our data suggest the clinical utility of ANT-DBS beyond mere seizure control. The beneficial and potential antiepileptogenic effect described here might broaden the spectrum of candidate ANT-DBS device recipients to those at risk of epilepsy progression.
Author contributions
LC and DB conceived the study, designed the experiments, and drafted the manuscript. CG, MLMP and ED, performed the animal experiments, tissue collection and helped with the interpretation of the results. LC performed statistical analysis. ELH, LC and DB ofered advice on data analysis and manuscript preparation. LC supervised the project. All authors discussed and commented on the manuscript.
Financial Support
FAPESP ( 2019/25974-4 ), CAPES -PrInt ( 88881.310490/2018–01 ) and NIH grants R01 NS103740 , R01 NS065957 , and CURE Epilepsy Catalyst award to DB.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
The authors are grateful to the research support provided by FAPESP and CAPES -PrInt program and NIH grants R01 NS103740 , R01 NS065957 , and CURE Epilepsy Catalyst awards to DB.
References
Walsh S.
Donnan J.
Fortin Y.
Sikora L.
Morrissey A.
Collins K.
et al.
A systematic review of the risks factors associated with the onset and natural progression of epilepsy.
Progression of temporal lobe epilepsy in the rat is associated with immunocytochemical changes in inhibitory interneurons in specific regions of the hippocampal formation.
Deep brain stimulation of the anterior nucleus of the thalamus reverses the gene expression of cytokines and their receptors as well as neuronal degeneration in epileptic rats.
Anterior thalamic nuclei deep brain stimulation reduces disruption of the blood-brain barrier, albumin extravasation, inflammation and apoptosis in kainic acid-induced epileptic rats.
in: Noebels J. Avoli M. Rogawski M. Olsen R. Delgado-Escueta A. Jasper's basic mechanisms of the epilepsies. Oxford University Press,
Oxford2012: 1150-1160
00099-7&id=gr1.jpg){kind=link}
00099-7&id=gr2.jpg){kind=link}
00099-7&id=gr3.jpg){kind=link}
00099-7&id=gr4.jpg){kind=link}
00099-7&id=gr5.jpg){kind=link}