Abstract
Background and Objectives:
The adenosine pathway in drug-resistant epilepsy (DRE) due to hippocampal sclerosis (HS) is largely unexplored. The present study aims to provide new insights into the role of adenosine pathway in patients with DRE associated with HS (DRE-HS).
Methods:
Quantitative polymerase chain reaction and immunohistochemistry were used to analyze 10 genes involved in adenosine pathway (ADARB1, ADK, NT5E, ADORA1, C-FOS, C-MYC, CREB1, C-JUN, NF-kB1, and MAPK) in surgically resected sclerosed hippocampi (n = 37) and compare their expressions with control hippocampi (n = 38) obtained from the autopsy. Expression analyses were also carried out in peripheral blood of 20 matched patients and 30 healthy controls.
Results:
C-JUN, NT5E, C-FOS, ADARB1, and ADORA1 were significantly upregulated in the hippocampus, whereas in blood, C-JUN, NT5E, C-FOS, and ADORA1 were significantly downregulated. ADARB1, NF-kB1, MAPK1, C-FOS, and CREB1 showed the reverse direction of expressions in post-surgery blood samples. On clinico-genetic analysis, MAPK1 and ADARB1 correlated with neuronal dispersion in the dentate gyrus (P = 0.02 and 0.03, respectively) and C-JUN correlated with neuronal loss in CA1 (P = 0.01) and CA4 (P = 0.04) areas. In blood, MAPK1, NF-kB1, and C-FOS expressions correlated negatively with the age of onset of seizures (P = 0.02, 0.03, and 0.03, respectively). In addition, ADARB1 expression correlated with neuronal loss in CA3 and CA4 areas (P ≤ 0.001 and 0.04, respectively).
Conclusion:
Genes with similar expression patterns in the brain and peripheral blood are potential biomarkers in DRE-HS. Significant clinico-genetic correlations warrant further studies for developing novel therapeutic interventions.
Keywords: Adenosine, drug-resistant epilepsy, gene expression, hippocampus, surgery
Introduction
Epilepsy is a common neurologic disorder characterized by an enduring predisposition of the brain to generate epileptic seizures.[1] Despite the availability of effective antiepileptic drugs (AEDs), in 30% of people with epilepsy, the seizures are not controlled with conventional medications.[2] These epilepsies are called drug-resistant epilepsy (DRE). There are varied causes of DRE, but hippocampal sclerosis (HS) is the most common pathology seen in patients with DRE, characterized by neuronal loss and gliosis in the hippocampus and adjacent temporal cortex.[3] The molecular mechanisms underlying DRE associated with HS (DRE-HS) are poorly understood. There are various hypotheses underlying the pathogenesis of DRE-HS, which include dysregulation of multidrug transporter genes and AED metabolizing enzyme genes,[4] drug target genes,[5] neuronal excitability genes,[6] and genes involved in intrinsic neuronal properties,[7] but the exact mechanism involved in DRE-HS remains poorly explored.
Large-scale gene expression studies on epileptic brain tissues have proposed several gene pathways that could be involved in DRE.[8] Among these, the adenosine pathway seems promising as adenosine is a potent endogenous anticonvulsant.[9] Briefly, adenosine is synthesized from adenosine triphosphate by ectonucleotidases (encoded by NT5E),[9] and it is degraded into inosine and adenosine monophosphate (AMP) by adenosine deaminase (encoded by ADARB1) and adenosine kinase (encoded by ADK), respectively.[10] Adenosine activates several intracellular signaling molecules like NF-KB1 (nuclear factor-kappa B1), MAPK1 (mitogen-activated protein kinase), C-JUN, C-FOS, CREB1 (cyclic AMP response element binding-1), and C-MYC on binding to the A1 receptor in the hippocampus (encoded by ADORA1)[11,12,13] [Figure 1 and Table 1]. Several studies suggest the neuromodulatory role of the adenosine pathway in seizures and epilepsy.[9,10,14] Adenosine augmentation therapy has been shown to reduce seizures and prevent epileptogenesis in mice models of epilepsy.[12] In a double-blind, placebo-controlled trial in patients with DRE, an adenosine agonist allopurinol reduced seizure frequency by increasing adenosine concentration in the brain.[14] Likewise, genetic variations in the adenosine pathway predispose patients with traumatic brain injury (TBI) to develop posttraumatic epilepsy in a shorter time following TBI.[9] Similarly, the role of adenosine has been implicated in various other forms of epilepsies[15] and sudden unexpected death in epilepsy patients.[16] The neuroprotective role of the adenosine pathway is well studied in animal models of epilepsy and other epilepsies, but whether it has a role in human DRE-HS remains unexplored.
Figure 1.
Schematic representation of adenosine signaling pathway
Table 1.
Functions of genes involved in adenosine signaling pathway and their role in epilepsy
| Gene | Function | Role in epilepsy |
|---|---|---|
| ADARB1[10,11] | i) Reduces the concentration of adenosine ii) Increases the influx of calcium and death |
Unedited RNA caused increase Ca2+ influx leads to neuronal death |
| ADK[15,17] | Reduces the concentration of adenosine | Overexpression of ADK triggers seizures in transgenic animals and increases interictal activity |
| ADORA1[18] | Substrate for adenosine | Activation of ADORA1 reduces epileptiform activity |
| NT5E[9] | Synthesizes adenosine | NT5E activation leads to inflammation in endothelial cells |
| NF-kB1[12,13] | i) Transcription factor ii) Inflammation and apoptosis |
Activation of NF-kB1 has been observed in surgically resected hippocampus from DRE patients |
| MAPK1[12,13] | Cell growth, adhesion, survival, and differentiation | Activation of MAPK1 leads to neuronal excitation and seizure activity in in vivo and in vitro experiments |
| C-MYC, C-FOS, C-JUN[11,19] | Transcription factor and signal transduction | Upregulation of kainic acid/quinolinic acid |
| CREB1[19] | Transcription factor | CREB1 activation in the hippocampus in epilepsy patients |
DRE: drug-resistant epilepsy
Considering that most studies on DRE are based on mouse models, the present study aims to provide new insights into the role of the adenosine pathway in human DRE-HS by studying (a) the expression levels of genes involved in the adenosine signaling pathway in the hippocampal tissues and peripheral blood of patients with DRE-HS and (b) the correlation between gene expressions and clinical phenotypes of patients with DRE-HS.
Methods
Study design
This was a prospective exploratory study wherein biological samples were collected from a university teaching hospital and molecular studies were carried out in a university human molecular genetics laboratory in South India. The expression levels of 10 genes (ADK, ADARB1, NT5E, ADORA1, NF-KB1, MAPK1, C-JUN, C-FOS, CREB1, and C-MYC) of the adenosine pathway were examined in (a) the surgically resected hippocampi and peripheral blood obtained from patients with DRE-HS planned for epilepsy surgery and (b) hippocampal samples obtained from the autopsy of healthy road accident victims (control tissues). The expression levels of genes in the patients’ blood were compared to those obtained from blood samples of healthy subjects (control blood). In addition, the gene expression levels in pre-surgery blood samples were compared to those in post-surgery blood samples obtained from the same patients after 1 year of surgery. The 10 genes were selected based on their role in the adenosine pathway, that is, synthesis (NT5E) and degradation (ADK and ADARB1) of adenosine, receptor gene (ADORA1), and various intracellular transcription factor genes (NF-KB1, MAPK1, C-JUN, C-FOS, CREB1, and C-MYC).[9,10,11,12,13]
Patients and controls
The International League Against Epilepsy (ILAE) definitions for epilepsy and DRE were used to diagnose the patients in the study.[20] Keeping in view the site feasibility and limited availability of control brain tissues, a convenient sample of 30 was chosen. Using standard pre-surgical investigations, patients diagnosed with mesial temporal lobe epilepsy and planned for anterior temporal lobectomy with amygdalohippocampectomy were included in the study. The whole hippocampal tissue was processed for molecular studies after confirming sclerosis on histology using the international consensus classification of HS.[3] The clinical data regarding the age of onset of seizures, frequency of seizures, semiology, AEDs, and family history of seizures or any comorbid illness were recorded from the patients. Similarly, data on demographic profile, history of any systemic/neurologic disorders, family history of any disorder, and medications were obtained from the control group. The institutional ethics committee approved this study (approval number: Sl. no. 6/Clinical Neurosciences/NIMHANS/DO/SUB-COMMITTEE/2013), and written informed consent was obtained from the patients/relatives and healthy subjects.
The hippocampal tissues were immediately transferred post-surgery on saline gauze to the neuropathology lab, sliced fresh in the coronal plane. Adequate fixed tissue was reserved for histopathologic examination, and slices of fresh tissue were frozen at -80°C in RNAlater®, preserving neuroanatomical coordinates. The same procedure was followed for autopsy samples. Venous blood samples were collected from patients pre-surgery and from age- and gender-matched healthy volunteers. Both tissue and blood samples were transported on ice to the human molecular genetics lab for RNA isolation, complementary DNA (cDNA) preparation, and Real Time Polypmerase Chain Reaction (RT-PCR).
RNA extraction and isolation from tissues and blood was done using the RNeasy Lipid Tissue mini Kit and QIAamp RNA Blood Mini Kit, respectively. RNA concentration and quality were determined using a Nanodrop® ND1000 spectrophotometer. RNA was treated with DNase and converted to cDNA using the high-capacity cDNA reverse transcription kit. Gene expression was analyzed using qRT-PCR on a StepOnePlus™ analyzer with Taqman® Universal PCR Master Mix and gene-specific primers. Experiments included duplicates for hippocampal and blood samples, using Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a control. Fold change for each gene was calculated as Log2 ΔΔCT. Immunohistochemistry on hippocampal sections was carried out using the BenchmarkXT® IHC/ISH machine. Sections were stained with hematoxylin and eosin, Luxol fast blue, and Nissl stains. NeuN and Glial Fibrillary Acidic Protein (GFAP) antibodies assessed neuronal density and dysmorphic neurons, respectively. Neuronal loss in the hippocampus was evaluated and classified as per the ILAE classification[3] and graded semi-quantitatively in subfields CA1–CA4 and the dentate gyrus.
Statistical analysis and interpretation
The gene expression and clinical data were analyzed using IBM® SPSS® statistical software (version 25) and GraphPad Prism v8. Descriptive statistics were used to calculate the mean and median values for the gene expressions and clinical characteristics. Based on the number and the type of clinical variables, Spearman’s correlation, Mann–Whitney U, and Kruskal–Wallis H tests were used to analyze the clinico-genetic correlations. Wilcoxon rank-sum test was used to analyze the gene expressions between pre- and post-surgery blood samples. A mean fold change (MFC) of more than two in the gene expression levels and correlation coefficient (ρ) with a p value less than or equal to 0.05 in the clinico-genetic analysis were considered significant. Bonferroni’s correction was used to assess the effects of multiple comparisons. The adjusted significance level for multiple comparisons was calculated by dividing the original significance level by the number of tests performed.
Results
Clinical characteristics
The hippocampal tissues were collected from 37 patients [males: 20 (54.1%) and females: 17 (45.9%)] and 38 autopsy samples [males: 28 (73.7%) and females: 10 (26.3%)] from otherwise healthy road accident victims over a period of 2 years. The mean age of onset of seizures and the duration of epilepsy before surgery were 12.6 ± 8.21 years (median: 11.0 years, range: 0.7–29.0 years) and 15.09 ± 8.75 years (median: 13.0 years, range: 2.0–37.0 years), respectively. The mean frequency of seizures (per month) was 5.27 ± 5.41 (median: 4.0, range: 1.0–30.0). There was no history of seizures or febrile seizures reported in the autopsy records of the retrieved samples. The blood samples were collected from 20 patients [males: 10 (50.0%) and females: 10 (50.0%)] and 30 age- and sex-matched control subjects [males: 16 (53.3%) and females: 14 (46.7%)]. Post-surgery blood samples were collected from six patients [males: 3 (50.0%) and females: 3 (50.0%)] after a mean time gap of 1.1 ± 0.28 years of surgery.
Focal onset seizure was the most common type of seizure in 29 patients (78.4%), followed by focal seizures with bilateral convulsive or tonic–clonic seizures in eight (21.6%) patients. Clinical history of febrile seizures and family history of epilepsy were present in 14 (37.8%) and three (8.1%) patients, respectively. Associated psychogenic non-epileptic seizures were present in two (5.4%) patients. The majority of patients showed neuropsychological deficits in multiple lobes (n = 26, 70%), followed by nine patients (24%) showing deficits in a single lobe. One patient each (2.7%) showed mental retardation and no neuropsychological deficit. Neuropsychological assessment showed concordance with the side of lesion in 31 (83.8%) patients and discordance with the side of lesion in six (16.2%) patients. Temporal and frontal regions showed concordance with the side of the lesion (n = 18, 48.6%). The most common interictal and ictal electroencephalogram patterns observed were sharp waves (n = 19, 51.4%) and theta activity (n = 29, 78.4%), respectively. The most common finding on 3 T magnetic resonance imaging was reduced hippocampus size with increased signal intensities seen in 34 (91.9%) patients. The majority of patients (n = 22, 59.5%) were taking at least two AEDs at the time of their recruitment in the study. ILAE HS type 1 was the most common pathology observed (n = 33, 89.2%). Severe neuronal loss (≥71%) was apparent mostly in CA1 (n = 27, 73.0%) followed by CA4 (n = 14, 37.8%) subfields of the hippocampus. ILAE HS type 2 (isolated CA1 loss) was seen in three patients (8.1%), while HS type 3 was seen in one patient (2.7%).
After surgery, the mean follow-up duration was 3.15 ± 2.4 years (median: 2.0 years, range: 0.6–8.9 years). All patients except two had Engel outcomes Ia (ILAE class 1) (94.6%) at the last follow-up. These two patients had Engel outcomes 4b and 3a at the end of 6 months and 2 years of follow-up, respectively.
Altered gene expressions in the hippocampal tissues and peripheral blood from patients with DRE-HS
On analyzing the differential gene expressions in the hippocampal and control tissues, it was found that the expression levels of five genes (C-JUN, NT5E, C-FOS, ADARB1, and ADORA1) were significantly higher in the sclerotic hippocampi relative to control tissues. MFC for the genes was as follows: 2.43 for C-JUN, 2.48 for NT5E, 2.9 for C-FOS, 2.23 for ADARB1, and 2.17 for ADORA1. On performing the test for multiple comparisons (Bonferroni’s corrections), C-JUN (P < 0.001), NT5E (P < 0.02), and C-FOS (P < 0.001)) showed upregulation. None of the genes showed significant downregulation in the gene expression levels. Maximum interindividual variation in the gene expression level was observed in C-FOS [Figure 2a, b and Table 2]. The gene expression data was validated by using immunohistochemistry (IHC) on the hippocampal tissues [Figure 3]. All genes except ADORA1 and NT5E showed a similar expression pattern on IHC (data not shown).
Figure 2.
Relative gene expression levels in the hippocampus from MTS patients. Represented here is a heat map with the expression levels of 10 genes (column) in the hippocampus (n = 37) (a) and peripheral blood (n = 20) (c). All positive values are presented as red and negative values as blue in the heat map. Graphs with violin plot depicting individual variation in the expression levels across genes in the hippocampus (b) and peripheral blood (d). The median levels are in dotted lines in the violins. All the values are relative to controls. The data are plotted in GraphPad Prism v 8.1. n: number of patients with significant fold change. MFC: mean fold change, MTS: mesial temporal sclerosis
Table 2.
Gene expression in the hippocampus (H) (n=37) and peripheral blood (B) (n=20) of DRE-HS patients
| Gene | Tissue | Fold change (Between ≤+2 and ≥-2) |
Up-regulation (fold change ≥2) |
||||
|---|---|---|---|---|---|---|---|
| n | Mean±SD | Median [IQR] | n | Mean±SD | Median [IQR] | ||
| ADK | H | 37 | 0.37±0.53 | 0.12 [0.74] | 0 | - | - |
| B | 20 | -0.56±0.27 | -0.57 [0.16] | 0 | - | - | |
| ADORA1 | H | 34 | 0.73±0.49 | 0.66 [0.90] | 3 | 2.17±0.10 | 2.22 [0.18] |
| B | 19 | 0.39±0.58 | 0.38 [0.69] | 0 | - | - | |
| MAPK1 | H | 37 | 0.53±0.66 | 0.42 [0.51] | 0 | - | - |
| B | 20 | 0.34±0.45 | 0.52 [0.42] | 0 | - | - | |
| C-JUN | H | 9 | 1.71±0.22 | 1.74 [0.23] | 28 | 2.43±0.34 | 2.34 [0.61] |
| B | 19 | -0.50±0.33 | -0.38 [0.30] | 0 | - | - | |
| NT5E | H | 9 | 1.85±0.10 | 1.81 [0.14] | 28 | 2.48±0.29 | 2.58 [0.47] |
| B | 12 | 1.70±0.18 | 1.76 [0.28] | 7 | 2.23±0.16 | 2.24 [0.27] | |
| ADARB1 | H | 32 | 0.97±0.32 | 0.96 [0.39] | 5 | 2.23±0.17 | 2.24 [0.17] |
| B | 20 | -0.43±0.31 | -0.40 [0.25] | 0 | - | - | |
| CREB1 | H | 37 | 1.17±0.18 | 1.19 [0.30] | 0 | - | - |
| B | 20 | 0.59±0.49 | 0.81 [0.45] | 0 | - | - | |
| NFkB1 | H | 37 | -0.41±0.59 | -0.51 [0.85] | 0 | - | - |
| B | 20 | 0.26±0.20 | 0.34 [0.15] | 0 | - | - | |
| C-FOS | H | 14 | 1.07±0.46 | 1.09 [0.66] | 23 | 2.9±0.58 | 2.81 [0.77] |
| B | 19 | -0.78±0.65 | -0.76 [1.22] | 0 | - | - | |
| C-MYC | H | 37 | 1.00±0.26 | 1.00 [0.33] | 0 | - | - |
| B | 20 | 2.00±0.25 | 2.00 [0.21] | 0 | - | - | |
Downregulation (fold change ≥2) of ADORA1 (-3.48), C-JUN (-2.18), NT5E (-2.07), and C-FOS (-2.58) in blood was seen in only one case. n: number of patients, DRE-HS: drug-resistant epilepsy associated with hippocampal sclerosis, IQR: interquartile range, SD: standard deviation
Figure 3.
Gene expression levels in the hippocampus and blood. (a) Comparative analysis of gene expression in both hippocampus and blood obtained from the same patients having DRE-HS. Represented here is a histogram showing the expression levels of 10 genes (column) in the hippocampus and blood obtained from the same patients. MFC is given in brackets. n = 17. (b) Effect of surgery on the gene expression levels in peripheral blood from patients with DRE-HS. Represented here is a graph with gene expression levels measured in blood cells pre- and post-surgery in patients with DRE-HS. All the values are relative to controls. The data are plotted in GraphPad Prism v 8.1 and presented as mean ± SEM of n = 6. B: blood, DRE-HS: drug-resistant epilepsy associated with hippocampal sclerosis, MFC: mean fold change, SEM: standard error of the mean, T: tissues (hippocampus)
In patients where surgery is not feasible, peripheral blood is a preferred medium to study the gene expressions. Hence, it was hypothesized that there would be significant differences in the gene expression levels in the patients’ blood with DRE-HS. Among 20 patients from whom blood samples were collected before surgery, it was found that NT5E showed significant upregulation in seven patients (MFC = 2.23). Four genes showed significantly lower expression in blood relative to control [(ADORA1 (MFC = -3.48), C-JUN (MFC = -2.18), NT5E (MFC = -2.07), and C-FOS (MFC = -2.58)] [Figure 2c, d and Table 2]. On using Bonferroni’s corrections, C-JUN and C-FOS showed significant downregulation (P < 0.001) and NT5E showed significant upregulation (P = 0.02).
Comparative analysis of gene expression in both hippocampus and blood obtained from the same DRE-HS patients
On comparative analysis of gene expression levels in the hippocampus and pre-surgery blood samples obtained from the same patients, it was found that ADORA1, MAPK1, NT5E, CREB1, and C-MYC showed the same pattern of expression (upregulation) in both the hippocampus and blood [Figure 4a], where the MFC values were as follows: ADORA1 (hippocampus: 0.59, blood: 0.08), MAPK1 (hippocampus: 0.63, blood: 0.29), NT5E (hippocampus: 1.99, blood: 1.62), CREB1 (hippocampus: 1.16, blood: 0.58), and C-MYC (hippocampus: 1.00, blood: 1.60). It was found that the expression levels were upregulated in the hippocampus for ADK (0.58), C-JUN (2.34), ADARB1 (1.47), and C-FOS (3.08), but downregulated in blood for the genes ADK (-0.63), C-JUN (-0.63), ADARB1 (-0.48), and C-FOS (-0.92)]. The expression of NF-kB1 showed a reverse pattern, that is, downregulation in the hippocampus (MFC = -0.19) and an upregulation in blood (MFC = 0.24). On Bonferroni’s corrections for multiple comparisons, C-JUN and C-FOS showed upregulation in the hippocampus (P < 0.001) and downregulation in blood (P = 0.03) [Figure 4c-h].
Figure 4.
Expression spectrum of genes on IHC. (a) ADK, PT3, dentate gyrus (glial cells positive); (b) ADARB1, PT31, dentate gyrus (glial cells nucleus and cytoplasm positive); (c) NFkB1, PT25, temporal cortex (neurons positive); (d) MAPK1, PT23, subiculum (glial and endothelial cells positive); (e) CJUN, PT5, hippocampus intense expression; (f) CFOS, PT5, hippocampus, intense synaptic expression; (g) CMYC, PT5, dentate gyrus (neurons lightly positive); (h) CREB1, PT31, temporal cortex (glial and endothelial cells positive) (magnification = scale bar). PT: patient tissue, IHC: Immunohistochemistry
Comparison of gene expressions in patient’s pre-and post-surgery blood
It was hypothesized that the patterns of expression of genes change after epilepsy surgery in the patients’ blood. To test this, the blood samples were collected post-surgery from six patients to test the effect of surgery on gene expression. It was found that the expression of three genes, that is, MAPK1 (P = 0.03), CREB1 (P = 0.05), and NF-kB1 (P = 0.03), was significantly downregulated compared to pre-surgery expression levels [Figure 4b]. The expression levels of ADARB1 (P = 0.03) and C-FOS (P = 0.03) were upregulated after the surgery. No significant changes were observed in C-JUN, NT5E, and C-MYC. On Bonferroni’s corrections for multiple comparisons, the changes in expressions of MAPK1 (P = 0.004), ADARB1 (P = 0.03), CREB1 (P = 0.02), and C-FOS (P = 0.007) were significant.
Clinico-genetic analysis
Clinico-genetic analysis was performed to find any correlation between clinical characteristics and gene expression levels in the hippocampus and peripheral blood of the patients (n = 17) [Table 3]. The expression of CREB1 in the hippocampus correlated with the age of onset of seizures in patients (ρ = 0.49, P = 0.05). The expression of MAPK1 correlated with dispersion in the dentate gyrus (ρ = 0.55, P = 0.02) and with neuronal loss in CA3 (ρ = 0.49, P = 0.05) and CA4 (ρ = 0.47, P = 0.05) subfields in the hippocampus. Similar correlations were found for the expression of ADARB1 and dispersion in the dentate gyrus (ρ = 0.51, P = 0.03) and neuronal loss in CA2 (ρ = 0.50, P = 0.04) and CA4 (ρ = 0.52, P = 0.03) subfields in the hippocampus. In addition, C-JUN expression correlated with neuronal loss in CA1 (ρ = 0.59, P = 0.01) and CA4 (ρ = 0.49, P = 0.04) hippocampus subfields. The expression of C-MYC was negatively correlated with pathological findings in the amygdala (ρ = -0.52, P = 0.03).
Table 3.
Clinicopathological correlation coefficients (P-values) with gene expression levels in the hippocampi (H) and peripheral blood (B) of DRE-HS patients (n=17), with P values in brackets
| Clinical feature | Tissue | ADK | ADORA1 | MAPK1 | C-JUN | NT5E | ADARB1 | CREB1 | NF-kB1 | C-FOS | C-MYC |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Age of onset of seizures | H | 0.09 (0.73) | 0.14 (0.59) | -0.17 (0.52) | 0.06 (0.81) | -0.04 (0.87) | -0.24 (0.36) | 0.49 (0.05) | 0.25 (0.33) | -0.01 (0.97) | 0.26 (0.32) |
| B | 0.05 (0.85) | -0.44 (0.08) | -0.56 (0.02) | -0.22 (0.39) | -0.29 (0.25) | -0.28 (0.27) | -0.33 (0.19) | -0.52 (0.03) | -0.51 (0.03) | -0.12 (0.64) | |
| Total duration of seizures | H | 0.34 (0.18) | 0.3 (0.24) | 0.26 (0.32) | 0.06 (0.83) | 0.07 (0.8) | 0.3 (0.24) | -0.09 (0.73) | 0.03 (0.9) | -0.32 (0.21) | 0 (0.99) |
| B | 0.55 (0.02) | 0.01 (0.96) | 0.18 (0.49) | -0.07 (0.79) | 0.43 (0.09) | 0.03 (0.9) | 0.39 (0.12) | 0.28 (0.28) | 0.24 (0.36) | 0.34 (0.18) | |
| Frequency of seizures (per month) | H | -0.3 (0.24) | -0.05 (0.86) | -0.14 (0.6) | -0.12 (0.66) | -0.15 (0.57) | -0.08 (0.76) | -0.39 (0.12) | -0.22 (0.4) | 0.22 (0.4) | -0.2 (0.44) |
| B | -0.36 (0.15) | 0.31 (0.22) | -0.06 (0.82) | -0.02 (0.94) | 0.12 (0.64) | 0.38 (0.13) | -0.11 (0.68) | -0.04 (0.88) | 0.02 (0.94) | 0 (1) | |
| Seizure classification | H | -0.36 (0.15) | -0.32 (0.22) | -0.19 (0.47) | -0.25 (0.33) | 0.09 (0.72) | -0.13 (0.63) | -0.16 (0.55) | -0.33 (0.19) | 0.22 (0.4) | 0.27 (0.3) |
| B | 0.16 (0.55) | 0.22 (0.4) | 0.13 (0.63) | 0.19 (0.47) | -0.28 (0.27) | 0.09 (0.72) | 0.16 (0.55) | 0.19 (0.47) | -0.19 (0.47) | -0.22 (0.4) | |
| History of febrile seizures | H | 0.16 (0.53) | -0.09 (0.74) | -0.26 (0.31) | -0.28 (0.28) | 20.03 (0.92) | -0.28 (0.28) | 0.29 (0.26) | -0.01 (0.96) | 0.05 (0.85) | 0.13 (0.63) |
| B | -0.2 (0.44) | -0.05 (0.85) | -0.18 (0.5) | 0.05 (0.85) | 0.1 (0.7) | 0.2 (0.44) | -0.3 (0.24) | -0.2 (0.44) | -0.23 (0.38) | 0 (1) | |
| Interictal epileptiform discharges | H | -0.18 (0.48) | -0.14 (0.6) | -0.14 (0.59) | -0.36 (0.16) | 0.31 (0.23) | -0.15 (0.58) | 0.14 (0.59) | -0.23 (0.38) | -0.13 (0.61) | -0.05 (0.86) |
| B | -0.12 (0.66) | -0.31 (0.22) | -0.22 (0.4) | -0.21 (0.43) | 0.04 (0.86) | 0.08 (0.75) | -0.19 (0.48) | -0.05 (0.84) | -0.28 (0.28) | -0.31 (0.23) | |
| Number of AEDs | H | 0.02 (0.93) | -0.05 (0.85) | 0.1 (0.71) | -0.24 (0.35) | 0.29 (0.26) | -0.12 (0.65) | 0.04 (0.89) | -0.12 (0.65) | -0.12 (0.65) | -0.36 (0.15) |
| B | -0.05 (0.85) | -0.26 (0.3) | -0.17 (0.52) | -0.19 (0.46) | 0.02 (0.93) | 0.26 (0.3) | -0.26 (0.3) | -0.02 (0.93) | -0.41 (0.1) | -0.31 (0.22) | |
| Dispersion in the dentate gyrus | H | 0.22 (0.39) | 0.07 (0.79) | 0.55 (0.02) | 0.32 (0.21) | -0.25 (0.33) | 0.51 (0.03) | -0.21 (0.41) | 0.23 (0.38) | -0.3 (0.25) | -0.14 (0.59) |
| B | 0.18 (0.49) | 0.15 (0.57) | 0.15 (0.56) | 0.47 (0.06) | 0.49 (0.05) | 0.17 (0.52) | 0.22 (0.39) | 0.35 (0.16) | 0.3 (0.25) | 0.58 (0.01) | |
| CA1 neuronal loss | H | 0.09 (0.74) | 0.16 (0.54) | 0.39 (0.12) | 0.59 (0.01) | -0.2 (0.44) | 0.46 (0.06) | -0.05 (0.85) | 0.32 (0.2) | 0.18 (0.48) | -0.17 (0.52) |
| B | -0.06 (0.82) | 0.22 (0.39) | 0.11 (0.67) | 0.21 (0.42) | 0.08 (0.77) | -0.24 (0.36) | -0.11 (0.69) | -0.06 (0.82) | 0.19 (0.47) | 0.27 (0.3) | |
| CA2 neuronal loss | H | -0.06 (0.83) | 0.27 (0.29) | 0.33 (0.19) | 0.42 (0.09) | 0.02 (0.93) | 0.50 (0.04) | -0.27 (0.3) | 0.11 (0.68) | 0.01 (0.97) | -0.4 (0.11) |
| B | -0.08 (0.76) | -0.03 (0.92) | 0.06 (0.82) | -0.09 (0.73) | -0.13 (0.62) | -0.25 (0.32) | 0.04 (0.87) | 0.09 (0.74) | 0.18 (0.49) | -0.02 (0.94) | |
| CA3 neuronal loss | H | 0.16 (0.54) | 0.31 (0.23) | 0.49 (0.05) | 0.44 (0.08) | -0.03 (0.92) | 0.43 (0.09) | -0.1 (0.71) | 0.16 (0.54) | -0.43 (0.08) | -0.45 (0.07) |
| B | -0.23 (0.38) | -0.24 (0.35) | -0.17 (0.51) | -0.22 (0.39) | -0.28 (0.28) | -0.66 (≤0.001) | -0.14 (0.59) | -0.23 (0.38) | 0.13 (0.63) | -0.22 (0.4) | |
| CA4 neuronal loss | H | 0.12 (0.63) | 0.38 (0.13) | 0.47 (0.05) | 0.49 (0.04) | -0.07 (0.78) | 0.52 (0.03) | -0.17 (0.51) | 0.24 (0.35) | -0.27 (0.29) | -0.38 (0.13) |
| B | -0.15 (0.57) | -0.12 (0.65) | -0.1 (0.69) | -0.26 (0.31) | -0.17 (0.51) | -0.52 (0.04) | -0.04 (0.88) | -0.14 (0.59) | 0.17 (0.51) | -0.07 (0.8) | |
| Amygdala pathology | H | 0.15 (0.55) | 0.14 (0.58) | 0.28 (0.28) | -0.1 (0.7) | 0.37 (0.15) | 0.16 (0.55) | -0.16 (0.55) | 0 (1) | -0.32 (0.22) | -0.52 (0.03) |
| B | 0.11 (0.68) | -0.07 (0.79) | -0.47 (0.06) | 0.1 (0.71) | 0.11 (0.68) | -0.01 (0.96) | -0.50 (0.04) | -0.50 (0.04) | -0.33 (0.2) | 0.09 (0.73) | |
| ILAE HS classification | H | -0.1 (0.71) | -0.37 (0.15) | -0.31 (0.23) | -0.22 (0.4) | -0.02 (0.93) | -0.32 (0.21) | 0.2 (0.45) | -0.06 (0.82) | 0.44 (0.08) | 0.34 (0.18) |
| B | 0.12 (0.64) | 0.27 (0.3) | 0.17 (0.51) | 0.44 (0.08) | 0.27 (0.3) | 0.49 (0.05) | -0.05 (0.85) | 0.12 (0.64) | -0.12 (0.64) | 0.24 (0.35) |
AEDs: antiepileptic drugs, HS: hippocampal sclerosis, ILAE: International League Against Epilepsy. Numbers in bold: p ≤ 0.05
In blood, the expression of MAPK1, NF-kB1, and C-FOS correlated negatively with the age of onset of seizures in patients (ρ = -0.56, P = 0.02; ρ = -0.52, P = 0.03; and ρ = -0.51, P = 0.03, respectively). The total duration of seizures correlated with ADK expression (ρ =0.55, P = 0.02). ADARB1 expression correlated with neuronal loss in CA3 (ρ = -0.66, P ≤ 0.001) and CA4 (ρ = -0.52, P = 0.04) subfields and with the types of HS (ρ =0.49, P = 0.05). The degree of neuronal dispersion in the dentate gyrus correlated with the expressions of NT5E (ρ = 0.49, P = 0.05) and C-MYC (ρ = 0.58, P = 0.01). In addition, the pathological findings in the amygdala correlated negatively with the expressions of CREB1 (ρ = -0.50, P = 0.04) and NF-kB1 (ρ = -0.50, P = 0.04).
Discussion
Adenosine is a potent anticonvulsant and modulates neuronal excitability and seizures; hence, various genes involved in this pathway may have a role in determining the clinical characteristics of DRE-HS patients and can be potential therapeutic targets.[18] In the present study, the gene expression analysis of 10 genes of the adenosine pathway in the hippocampal tissues and peripheral blood of patients operated for DRE-HS was performed and correlated with their phenotype.
In the hippocampal tissues, the expression of five genes (ADARB1, ADORA1, NT5E, C-JUN, and C-FOS) out of 10 genes involved in adenosine pathway showed upregulation, while in blood, three genes (ADORA1, C-JUN, and C-FOS) showed downregulation and one gene (NT5E) showed upregulation. The underlying biological mechanisms for the differences in the pattern of gene regulation are difficult to understand as both brain tissues and peripheral blood are different tissues. While the blood–brain barrier may play a role in causing this differential gene expression, further studies are required to understand the underlying mechanisms. The expression of C-JUN, NT5E, C-FOS, ADARB1, and ADORA1 was significantly upregulated in the hippocampal tissues of patients with DRE-HS. The elevated expression of both NT5E and ADARB1, which increase and decrease the synthesis of adenosine, respectively, in the hippocampus, may suggest that both proepileptic and antiepileptic mechanisms operate simultaneously in DRE-HS patients to regulate seizure activity. This excitatory–inhibitory mechanism of seizures is well known in acute epilepsies,[21] but this mechanism, mediated by adenosine, was observed in DRE-HS in this study. Previous studies have shown elevated adenosine levels and increased activation of ADORA1 in patients having TBI with astrogliosis.[17,22] The upregulation of ADORA1 in the present study suggests elevated adenosine levels in sclerotic hippocampal tissues as A1 receptors are most abundantly found in the hippocampus.[12]
The expression pattern of the same genes in the brain and peripheral blood has been previously demonstrated in studies on rats[23] and schizophrenia patients.[24] This is probably the first report to show how the same genes express in the hippocampus and blood obtained from the same DRE-HS patients. As shown in the present study, ADORA1, MAPK1, NT5E, CREB1, and C-MYC have the potential to become blood markers for DRE-HS. Further studies are required on a larger sample size to study ADORA1, MAPK1, NT5E, CREB1, and C-MYC as blood biomarkers for DRE-HS. The expressions of MAPK1, CREB1, NF-kB1, and ADARB1 was downregulated and that of C-FOS was upregulated in post-surgery blood compared to pre-surgery blood. Previous studies have shown changes in the differential gene expressions of estrogen receptor and HER2 expression after resection surgery in patients with cancers, but this pre-and post-gene expression analysis has not been adequately studied in human epilepsies.[25] The reversal of expression of genes in the present study suggests that these genes might be sensitive to surgical resection of the hippocampus.
Previous in vitro and animal studies have shown that C-FOS, C-JUN, and MAPK1 mediate neuronal excitability and apoptosis.[26] It has been shown that severe kainic acid (KA)–induced seizures increased neuronal excitability and cell death in C-FOS mutant mice.[27] It has been demonstrated that there is a marked increase in the C-FOS mRNA level in the hippocampus following ethanol withdrawal seizures in mice.[27,28,29] Similarly, it has been shown that MAPK acts on RNA-binding proteins to regulate mRNAs, indicating a misregulation of protein expression at synapses in epilepsy.[29] Furthermore, upregulation of MAPK1 has been observed in the brain of mice following KA-induced seizures.[30] Present study’s findings of upregulation of C-FOS, C-JUN, and MAPK1 in the hippocampus of DRE-HS patients strengthens the role of these intracellular signaling molecules in human DRE. The downregulation of C-MYC has been observed in the frontal cortex of mice[19]; in the present study, downregulation of C-MYC was observed in the amygdala of DRE-HS patients.
Peripheral blood is a relatively easy medium to study gene expressions, especially in DRE patients where surgical treatment is not recommended. The patterns of gene expressions in peripheral blood can be used as surrogate markers to monitor the severity and progression of seizures; for example, in the present study, the expression of ADK decreased with an increase in the duration of seizures. This finding might suggest that the expression of ADK reduces in blood with the chronicity of seizures, which thereby decreases the availability of adenosine in blood. This finding might not correlate directly with seizures in patients because the expressions of genes in blood may not be same as in the brain. Similarly, the blood expressions of ADARB1, CREB1, NT5E, C-MYC, and NF-KB1 can be employed as surrogate markers for the loss of neurons in the hippocampus in DRE-HS patients. Previous studies have shown that with overexpression of ADK, spontaneous recurrent seizures occur[10,11]; the present study corroborates these findings, albeit in peripheral blood of DRE-HS patients. It is important to note that how accurately the gene expressions in blood reflect the brain pathology needs to be investigated.
Several limitations need to be noted while interpreting the results of this study. The whole hippocampal tissue was used for mRNA isolation; hence, the gene expression results might be confounded by the composition effects due to extensive reorganization of hippocampus in the patients. Since whole blood was used for mRNA isolation, the confounding effects of potential inflammation on the expression of investigated genes cannot be ruled out. The majority of patients (59.5%) were on two AEDs at the time of the study and the blood levels of AEDs were not available in the patients, hence the potential influence of AEDs on gene expressions needs has to be considered while interpreting the results of this study. The differences in the micro-structure of the hippocampal samples, effects of multiple cellular and humoral factors in blood, and effects of AEDs may cause differences in the pattern of expressions of genes in the hippocampal tissues and blood. Pre- and post-surgery gene expressions could be done in six samples; hence, a bigger sample size is required to validate these results. The postmortem hippocampal tissues were the best available control tissues for this study. To minimize the heterogeneity in the autopsy hippocampal tissues due to cause of death, postmortem delay, handling and storage of tissues, we used multiple autopsy tissues for comparisons with patients’ hippocampal tissues. The cause of death, postmortem delay, and handling and storage of tissues can affect the expression of genes and can potentially confound any differences observed in the results. Another important limitation is the wide range of the age of onset and duration of seizures before surgery, which could influence the pattern of gene expression.
Conclusion
This study highlights that the adenosine pathway plays a vital role in DRE-HS, as suggested by the differential gene expression in the hippocampus and blood of the patients. The genes with similar expression patterns in the brain and peripheral blood can be used as blood biomarkers for DRE-HS. In addition, the reversal of expressions of MAPK1, CREB1, NF-kB1, ADARB1, and C-FOS in post-surgery blood in patients with DRE-HS may suggest the sensitivity of these genes to surgery. Identifying clinically relevant gene correlations may pave the way for discovering new genes and pathways that are highly predictive of DRE-HS.
Conflicts of interest
There are no conflicts of interest.
Funding Statement
This study was supported by the National Institute of Mental Health and Neurosciences Project Fund.
References
- 1.Moshé SL, Perucca E, Ryvlin P, Tomson, T. Epilepsy: New advances. Lancet. 2014;385:884–98. doi: 10.1016/S0140-6736(14)60456-6. [DOI] [PubMed] [Google Scholar]
- 2.Duncan JS, Sander JW, Sisodiya SM, Walker MC. Adult epilepsy. Lancet 2006. 2006;367:1087–1100. doi: 10.1016/S0140-6736(06)68477-8. [DOI] [PubMed] [Google Scholar]
- 3.Blümcke I, Thom M, Aronica E, Armstrong DD, Bartolomei F, Bernasconi A, et al. International consensus classification of hippocampal sclerosis in temporal lobe epilepsy: A Task Force report from the ILAE commission on diagnostic methods. Epilepsia. 2013;54:1315–29. doi: 10.1111/epi.12220. [DOI] [PubMed] [Google Scholar]
- 4.Rogawski MA, Johnson MR. Intrinsic severity as a determinant of antiepileptic drug refractoriness. Epilepsy Curr. 2008;8:127–130. doi: 10.1111/j.1535-7511.2008.00272.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Engel T, Murphy BM, Schindler CK, Henshall DC. Elevated p53 and lower MDM2 expression in hippocampus from patients with intractable temporal lobe epilepsy. Epilepsy Res. 2007;77:151–6. doi: 10.1016/j.eplepsyres.2007.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Siebzehnrubl FA, Blumcke I. Neurogenesis in the human hippocampus and its relevance to temporal lobe epilepsies. Epilepsia. 2008;49:55–65. doi: 10.1111/j.1528-1167.2008.01638.x. [DOI] [PubMed] [Google Scholar]
- 7.Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med. 2000;342:314–9. doi: 10.1056/NEJM200002033420503. [DOI] [PubMed] [Google Scholar]
- 8.Mirza N, Vasieva O, Marson AG, Pirmohamed M. Exploring the genomic basis of pharmacoresistance in epilepsy: An integrative analysis of large-scale gene expression profiling studies on brain tissue from epilepsy surgery. Hum Mol Genet. 2011;20:4381–94. doi: 10.1093/hmg/ddr365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Diamond ML, Ritter AC, Jackson EK, Conley YP, Kochanek PM, Boison D, et al. Genetic variation in the adenosine regulatory cycle is associated with posttraumatic epilepsy development. Epilepsia. 2015;56:1198–206. doi: 10.1111/epi.13044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Boison D. Adenosine-based cell therapy approaches for pharmacoresistant epilepsies. Neurodegener Dis. 2007;4:28–33. doi: 10.1159/000100356. [DOI] [PubMed] [Google Scholar]
- 11.Boison D. Adenosine dysfunction and adenosine kinase in epileptogenesis. Open Neurosci J. 2010;4:93–101. doi: 10.2174/1874082001004020093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hamil NE, Cock HR, Walker MC. Acute down-regulation of adenosine A(1) receptor activity in status epilepticus. Epilepsia. 2011;53:177–88. doi: 10.1111/j.1528-1167.2011.03340.x. [DOI] [PubMed] [Google Scholar]
- 13.Siebel AM, Menezes FP, Capiotti KM, Kist LW, da Costa Schaefer I, Frantz JZ, et al. Role of adenosine signaling on pentylenetetrazole-induced seizures in zebrafish. Zebrafish. 2015;12:127–36. doi: 10.1089/zeb.2014.1004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Togha M, Akhondzadeh S, Motamedi M, Ahmadi B, Razeghi S. Allopurinol as adjunctive therapy in intractable epilepsy: A double-blind and placebo-controlled trial. Arch Med Res. 2007;38:313–16. doi: 10.1016/j.arcmed.2006.10.010. [DOI] [PubMed] [Google Scholar]
- 15.de Groot M, Iyer A, Zurolo E, Anink J, Heimans JJ, Boison D, et al. Overexpression of ADK in human astrocytic tumors and peritumoral tissue is related to tumor-associated epilepsy. Epilepsia. 2011;53:58–66. doi: 10.1111/j.1528-1167.2011.03306.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Purnell B, Murugan M, Jani R, Boison D. The good, the bad, and the deadly: Adenosinergic mechanisms underlying sudden unexpected death in epilepsy. Front Neurosci. 2021;15:708304. doi: 10.3389/fnins.2021.708304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fedele DE, Gouder N, Güttinger M, Gabernet L, Scheurer L, Rülicke T, et al. Astrogliosis in epilepsy leads to overexpression of adenosine kinase, resulting in seizure aggravation. Brain. 2005;128:2383–95. doi: 10.1093/brain/awh555. [DOI] [PubMed] [Google Scholar]
- 18.Muzzi M, Coppi E, Pugliese AM, Chiarugi, A. Anticonvulsant effect of AMP by direct activation of adenosine A1 receptor. Exp Neurol. 2013;250:189–93. doi: 10.1016/j.expneurol.2013.09.010. [DOI] [PubMed] [Google Scholar]
- 19.Jeon WJ, Kim SH, Seo MS, Kim Y, Kang UG, Juhnn YS, et al. Repeated electroconvulsive seizure induces c-Myc down-regulation and Bad inactivation in the rat frontal cortex. Exp Mol Med. 2008;40:435–44. doi: 10.3858/emm.2008.40.4.435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. ILAE official report: A practical clinical definition of epilepsy. Epilepsia. 2014;55:475–82. doi: 10.1111/epi.12550. [DOI] [PubMed] [Google Scholar]
- 21.Pitkänen A, Lukasiuk K. Mechanisms of epileptogenesis and potential treatment targets. Lancet Neurol. 2011;10:173–86. doi: 10.1016/S1474-4422(10)70310-0. [DOI] [PubMed] [Google Scholar]
- 22.Ke RH, Xiong J, Liu Y, Ye ZR. Adenosine A2a receptor induced gliosis via Akt/NF-kappaB pathway in vitro. Neurosci Res. 2009;65:280–285. doi: 10.1016/j.neures.2009.08.002. [DOI] [PubMed] [Google Scholar]
- 23.Witt SH, Sommer WH, Hansson AC, Sticht C, Rietschel M, Witt CC. Comparison of gene expression profiles in the blood, hippocampus and prefrontal cortex of rats. In Silico Pharmacol. 2013;1:15. doi: 10.1186/2193-9616-1-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Glatt SJ, Everall IP, Kremen WS, Corbeil J, Sásik R, Khanlou N, et al. Comparative gene expression analysis of blood and brain provides concurrent validation of SELENBP1 up-regulation in schizophrenia. Proc Natl Acad Sci U S A. 2005;102:15533–8. doi: 10.1073/pnas.0507666102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gonzalez-Angulo AM, Iwamoto T, Liu S, Chen H, Do KA, Hortobagyi GN, et al. Gene expression, molecular class changes, and pathway analysis after neoadjuvant systemic therapy for breast cancer. Clin Cancer Res. 2012;18:1109–9. doi: 10.1158/1078-0432.CCR-11-2762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Chawla MK, Penner MR, Olson KM, Sutherland VL, Mittelman-Smith MA, et al. Spatial behavior and seizure-induced changes in c-fos mRNA expression in young and old rats. Neurobiol Aging. 2012;34:1184–98. doi: 10.1016/j.neurobiolaging.2012.10.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Zhang J, Zhang D, McQuade JS, Behbehani M, Tsien JZ, Xu M. c-fos regulates neuronal excitability and survival. Nat Genet. 2002;30:416–20. doi: 10.1038/ng859. [DOI] [PubMed] [Google Scholar]
- 28.Dave JR, Tabakoff B, Hoffman PL. Ethanol withdrawal seizures produce increased c-fos mRNA in mouse brain. Mol Pharmacol. 1990;37:367–71. [PubMed] [Google Scholar]
- 29.Pernice HF, Schieweck R, Kiebler MA, Popper B. mTOR and MAPK: From localized translation control to epilepsy. BMC Neurosci. 2016;17:73. doi: 10.1186/s12868-016-0308-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Che Y, Yu YM, Han PL, Lee JK. Delayed induction of p38 MAPKs in reactive astrocytes in the brain of mice after KA-induced seizure. Brain Res Mol Brain Res. 2001;94:157–65. doi: 10.1016/s0169-328x(01)00233-9. [DOI] [PubMed] [Google Scholar]