Molecular Tools For The Characterization of Seizure Susceptibility in Genetic Rodent Models of Epilepsy

Epilepsy & Behavior 121 (2021) 106594
Contents lists available at ScienceDirect
Epilepsy & Behavior
journal homepage: www.elsevier.com/locate/yebeh
Molecular tools for the characterization of seizure susceptibility in

genetic rodent models of epilepsy
José Ramón Bosque a,b, Ricardo Gómez-Nieto a,b,c, Sebastián Hormigo a,d, M.Javier Herrero-Turrión a,e,
Elena Díaz-Casado a,b, Consuelo Sancho a,b, Dolores E. López a,b,c,⁎
a
Institute for Neuroscience of Castilla y León (INCyL), University of Salamanca, Salamanca, Spain
b
Salamanca Institute for Biomedical Research (IBSAL), Spain
c
Department of Cell Biology and Pathology, School of Medicine, University of Salamanca, Salamanca, Spain
d
Department of Neurobiology and Anatomy, Drexel University College of Medicine, United States of America
e
INCYL Neurological Tissue Bank (BTN-INCYL), Spain
a r t i c l e i n f o a b s t r a c t
Article history: Epilepsy is a chronic neurological disorder characterized by abnormal neuronal activity that arises from imbal-
Received 26 August 2019 ances between excitatory and inhibitory synapses, which are highly correlated to functional and structural
Revised 20 September 2019 changes in specific brain regions. The difference between the normal and the epileptic brain may harbor genetic
Accepted 25 September 2019
alterations, gene expression changes, and/or protein alterations in the epileptogenic nucleus. It is becoming in-
Available online 1 November 2019
creasingly clear that such differences contribute to the development of distinct epilepsy phenotypes. The current
Keywords:
major challenges in epilepsy research include understanding the disease progression and clarifying epilepsy clas-
DNA analysis sifications by searching for novel molecular biomarkers. Thus, the application of molecular techniques to carry
Gene expression studies out comprehensive studies at deoxyribonucleic acid, messenger ribonucleic acid, and protein levels is of utmost
Immunodetection importance to elucidate molecular dysregulations in the epileptic brain. The present review focused on the great
Post-translational protein modifications diversity of technical approaches available and new research methodology, which are already being used to study
Proteomic molecular alterations underlying epilepsy. We have grouped the different techniques according to each step in
the flow of information from DNA to RNA to proteins, and illustrated with specific examples in animal models
of epilepsy, some of which are our own. Separately and collectively, the genomic and proteomic techniques,
each with its own strengths and limitations, provide valuable information on molecular mechanisms underlying
seizure susceptibility and regulation of neuronal excitability. Determining the molecular differences between ge-
netic rodent models of epilepsy and their wild-type counterparts might be a key in determining mechanisms of
seizure susceptibility and epileptogenesis as well as the discovery and development of novel antiepileptic agents.
This article is part of the Special Issue “NEWroscience 2018"

© 2019 Elsevier Inc. All rights reserved.
Abbreviations: 2D-DIGE, 2D Fluorescence Difference Gel Electrophoresis; aCGH, array comparative genomic hybridization; Acsm3, coenzyme acyl-CoA synthetase; ATP5H, ATP
synthase subunit d, mitochondrial protein; C1q, complement C1q chain; Camkk2, calcium/calmodulin-dependent protein kinase 2; CDKL5, cyclin-dependent kinase like 5; cDNA,
complementary DNA; CGH, comparative genomic hybridization; ChIP-Seq, chromatin immunoprecipitation sequencing; CNVs, copy number variations; CSF, cerebrospinal fluid;
Egr1,2,3, early growth response protein 1, 2 or 3; EWCE, expression-weighted cell-type enrichment; FISSEQ, fluorescent in situ RNA sequencing; FN1, encodes fibronectin; GABA(A)R,
GABAA receptor; GAERS, Genetic Absence Epilepsy Rat from Strasbourg; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GASH/Sal, genetic audiogenic seizure hamster from
Salamanca; GFAP, glial fibrillary acidic protein; Gpr126, G-protein coupled receptor 126; Gria2, glutamate ionotropic receptor AMPA type subunit 2; Grin2a, glutamate ionotropic
receptor NMDA type subunit 2A; H3/H4, histone H3/H4; HDAC, histone deacetylase; HIF-1, hypoxia inducible factor 1; HPLC, high-performance liquid chromatography; Il10rb,
interleukin 10 receptor subunit beta; IC, inferior colliculus; iTRAQ, isobaric tags for a relative and absolute quantification; JAK/STAT signaling pathway, Janus kinases (JAK)/signal
transducer and activator of transcription proteins (STATs); KIF5A, kinesin heavy chain isoform 5A; LCM, laser capture microdissection; MALDI, matrix-assisted laser desorption/
ionization; MAPK, mitogen-activated protein kinase; MDH1, malate dehydrogenase; Methyl-Seq, massive methylation sequencing; miRNA, micro RNA; MS, mass spectrometry; MTLE,
mesial temporal lobe epilepsy; MTLE-HS, mesial temporal lobe epilepsy with hippocampal sclerosis; mTOR, mammalian target of rapamycin; NEFL, neurofilament light polypeptide;
NGS, next generation sequencing; PCDH19, protocadherin 19; PK, piruvate kinase; PNPO, pyridoxamine 5′-phosphate oxidase; PRDX2, peroxyredoxin-2; Qdpr, enzyme dihidopterina
reductase quinoide; RNA-Seq, RNA-sequencing; RT-qPCR, quantitative real time PCR; Rtel, regulator of telomere elongation; SBL, in situ sequencing by ligation; SC, superior colliculus;
scRNA, seq: single-cell RNA sequencing; SE, status epilepticus; seqFISH, sequential fluorescence in situ hybridization; SINEs, short interspersed nuclear elements; SLC25A22, solute
carrier family 25, member 22; Slc6a1, solute carrier family 6, member 1; smFISH, single-molecule RNA fluorescence in situ hybridization; Stat3, signal transducer and activator of
transcription 3; STXBP1, syntaxin-binding protein 1; Timp-1, tissue inhibitor of metalloproteinases 1; TIVA, the transcriptome in vivo analysis; TLE, temporal lobe epilepsy; TOF, time-
of-flight; WAG/Rij, Wistar Albino Glaxo from Rijswijk; WAR, Wistar Audiogenic Rat; WES, whole exome sequencing; WGS, whole genome sequencing.
⁎ Corresponding author at: Institute for Neuroscience of Castilla y León (INCyL), Laboratory 12, C/ Pintor Fernando Gallego 1., 37007, Salamanca, Spain.
E-mail address: [email protected] (D.E. López).
https://doi.org/10.1016/j.yebeh.2019.106594
1525-5050/© 2019 Elsevier Inc. All rights reserved.
2 J.R. Bosque et al. / Epilepsy & Behavior 121 (2021) 106594
1. Introduction the Wistar Albino Glaxo from Rijswijk (WAG/Rij; [26]), the Wistar Au-
diogenic Rat (WAR; [27]), and the genetic audiogenic seizure hamster
Currently, epilepsy affects around 50 million people in the world [1], from Salamanca (GASH/Sal; [7]). Activation of auditory pathways are
and its heterogeneous etiology includes structural alterations due to required for the onset and progression of seizures in all audiogenic sei-
traumatic brain injury [2], genetic disorders [3], or idiopathic epilepsy zure models, and many studies pointed out the inferior colliculus (IC), a
with no certain cause [4]. Epileptogenesis is a dynamic and gradual pro- critical integration center in the auditory midbrain pathway, as the ep-
cess that involves changes in the brain after a precipitating injury or in- ileptogenic nucleus [6]. However, the genetic alterations underlying
sult that results in the development of spontaneous recurrent seizure epileptogenesis in the IC are not fully known. With the increase in avail-
activity or epilepsy [5]. The complexity of this neurological disorder re- ability of genome-editing techniques that assess the relationship be-
quires models for exploring different aspects of epilepsy [6]. Among the tween mutation genotype and phenotype, nonrodent species like the
most used and well-characterized in vivo genetic models of epilepsy are zebrafish are also used to optimize and expedite genetic testing as
those genetically predisposed animal species, in which seizures occur in well as development of antiepileptic drugs [28].
response to high intensity acoustic stimulation, the so-called genetic au- The multiple molecular approaches and the many options for link-
diogenic seizure models [7,8]. Genetic and protein alterations have been age of different analysis may detect a number of mutations and DNA
described as possible contributors to the development of epilepsy phe- modifications involved in epilepsy. Fig. 1 depicts the analytic tools
notypes in the audiogenic seizure models [9–11]. Thus, studies focused used in genome-wide studies of epilepsy as well as a brief description
on molecular disruptions of the epileptogenic nucleus are of utmost in- of their corresponding outcomes, showing possible genetic alterations.
terest, as they provide insights into the mechanisms involved in seizure It should be noted that no single genetic testing has all the requisite ca-
susceptibility and regulation of neuronal excitability. Challenging these pacity to contend with the complexities of all genetic mechanisms, and
key assumptions may lead to discovery of molecular biomarkers, which in some cases, the genetic variants identified through the variety of ap-
can be used for the epilepsy diagnosis and prognosis as well as for proaches are not fully understood [29].
assessing the efficacy of new antiepileptic treatments. In the present
paper, we reviewed the molecular techniques used for the understand- 2.1. Comparative genomic hybridization (CGH)
ing of seizure susceptibility in genetic models of epilepsy, particularly in
genetically epilepsy-prone rodents. We also compared the results ob- The comparative genomic hybridization (CGH) and, lately, arrays of
tained from the main models of epilepsy and showed some our own CGH (aCGH) are the simplest approaches to study DNA variations, par-
findings to highlight the strengths and weaknesses of different molecu- ticularly if used species with known genome such as rat and mouse. The
lar approaches. aCGH method allows reliable detection of DNA sites in multiple genome
loci by comparing the relative amounts of DNA from two genomes, the
2. DNA studies control and the sample to be tested. Both samples are labeled with dif-
ferent fluorochromes that bind with DNA fragments of known se-
The need to identify a large set of genes involved in complex human quences or “probes”, fixed to a slide or glass holder (Fig. 1A). The color
diseases has led to an increased use of molecular techniques in biomed- of the fluorescence at each point of the aCGH informs about the relative
ical research and clinical practice. The discovery of genes associated amount of each DNA and allows inferring the presence of gains or losses
with epilepsy, excluding those involved in metabolic disorders and in- in specific regions of the genome [30]. The aCGH method is used to eval-
tellectual disability that may present high prevalence of epilepsy, has uate targeted regions throughout the chromosomes for copy number
exponentially increased from 30 to 90 genes in the last decade [12]. variations (CNVs). There are commercial rodent microarrays that con-
However, most pathologic genetic variants were identified in mono- tain 720,000 probes/oligonucleotides of 50–75 mer (average size) and
genic epilepsy and represent a small subset of all epilepsy types [13]. an average distance of 3537 pb. Those with less than 10 consecutive ol-
All of these achievements have been possible because of the develop- igonucleotides are considered as an aberration. In the GASH/Sal model,
ment and cost reduction of next generation sequencing (NGS)-based we used commercial arrays of mice and found genomic alterations at
panel tests that allowed the advent of multigene panels, exome, and ge- the level of 17qE5. The gain and loss detected in this chromosome re-
nome-scale sequencing. Additionally, neurophysiological and molecular gion could be indicative of CVNs, involving the neurexin I gene; how-
studies of mutations underlying seizure susceptibility might also con- ever, no association with epilepsy was found (unpublished data).
tribute to the successful development of new therapeutic targets and Mullen et al. [31] demonstrated that CNVs are overrepresented in pa-
antiepileptic drugs. A large percentage of the mutations identified in ep- tients with genetic generalized epilepsy, pointing out CNVs as an impor-
ilepsy involves altered genes that encode calcium, potassium, sodium, tant risk factor. Taking into account the large pattern of genetic and
and chloride ion channels, making up a family of channelopathies phenotypic heterogeneity in epileptic syndromes, the applicability of
[14], neurotransmitter receptors (e.g., γ-aminobutyric acid (GABA) re- CGH microarrays to evaluate the CVNs becomes controversial [32,33].
ceptors) [15], several structural proteins [16,17], mammalian target of Olson et al. [34] suggested assessing CVNs through conventional chro-
rapamycin (mTOR) pathway genes, modulation of synaptic vesicle mosomal microarray to shed light upon the idiopathic epilepsy, but
docking and release (e.g., syntaxin-binding protein 1 [STXBP1]) [18], the greatest impediment for CNVs detection might be the limit size of
cell signaling (e.g., cyclin-dependent kinase like 5 [CDKL5]) [19], cell– 10 kb. In this regard, more recent studies argue in favor of using other
cell adhesion (e.g., protocadherin 19 [PCDH19]) [20], transcription techniques such as the whole exome sequencing (WES, see below)
(e.g., the aristaless-related homeobox gene (ARX)) [21], DNA repair that is able to identify small CNVs less than 10 kB in size [35].
the bifunctional polynucleotide phosphatase/kinase gene (PNKP) [22],
mitochondrial glutamate symporter (e.g., solute carrier family 25, mem- 2.2. Whole genome sequencing (WGS)
ber 22 [SLC25A22]) [23], and enzymes involved in metabolic pathways
(e.g., pyridoxamine 5′-phosphate oxidase [PNPO]) [24]. The feasibility of implementing gene sequencing methods has made
Despite these significant advances, much less is known about the al- tremendous contributions to science advancement in general, and espe-
tered genes in animal models of epilepsy, with the exception of those cially in epilepsy research. Although sequencing technologies, some
related to monogenic epilepsies. Animal models of epilepsy arose from years ago, were relatively expensive and time consuming, nowadays, a
artificial selection of seizure-susceptible strains over many generations rapid development is taking place. Thus, novel sequencing platforms
that resulted in high predisposition to epilepsy, as is the case of audio- and supporting technologies involved in processes such as targeting
genic epilepsy models. Examples of rodent audiogenic seizure models and data analysis have emerged and are used routinely to reduce costs
are the Genetic Absence Epilepsy Rat from Strasbourg (GAERS; [25]), and determine the complete genome sequences of different species.
J.R. Bosque et al. / Epilepsy & Behavior 121 (2021) 106594 3
Fig. 1. Overview of techniques used for determining DNA alterations. A. Diagram of the microarray-based comparative genomic hybridization processes. Samples are labeled with
fluorescent dyes and applied to the microarray. Complementary nucleotide sequences are bound to the microarray wells and the microarray scanner measures the fluorescent signals
to generate plots. B. Workflow of genome, exome, or panel gene sequencing. The genomic DNA is cut into short fragments to generate of DNA library by PCR. Using NGS, the library is
sequenced, and the corresponding data are analyzed through bioinformatics analysis. Finally, SANGER sequencing is necessary to confirm the mutations. C. Workflow of histone
modifications analysis by Methyl-Seq. Genomic DNA is bisulfite-treated and nonmethylated cytosine residues are converted to uracils, then library is performed and sequenced in NGS
platform, and are further analyzed using bioinformatics tools.
When whole genome sequencing (WGS) is applied to animal models of is necessary to create a genomic library. To do this, the oligonucleotides
epilepsy, the identification of gene variants associated with epilepsy is are amplified via polymerase chain reaction (PCR), then transcribed
possible by comparing differences between the WGS on the wild type in vitro in the presence of biotinylated uridine triphosphate (UTP) to
and the epileptic strain (Fig. 1B). generate single-stranded RNA “bait.” Genomic DNA is sheared, ligated
As an example, the complete genome sequence of the GAERS was to Illumina sequencing adapters, and selected for lengths between 200
performed bringing promising results [36]. The GAERS model shows a and 350 bp. A quality control of the sequences is carried out with the
genetic generalized epilepsy that is analogue to the human absence ep- FASTQC program [40], assembling those that pass the checking with
ilepsy as verified by electroencephalic, pharmacological, and behavioral the SPAdes program [41]. As a result, the genome can be compared
results, and like the human condition, this model appears to have a with the sequences of the reference genome, identifying point muta-
complex genetic architecture. The GAERS genome study identified tions, structural variations, and genomic reorganizations by using the
1.12 million single nucleotide variants, 296.5 K short insertion–dele- GATK, the Lumpy, and the MAUVE software, respectively [42–44].
tions, and 354 putative copy number variants that result in complete After a bioinformatics analysis, the variants with higher or lower impact
or partial loss/duplication of 41 genes. Among the variants with high can be obtained [45]. Although the WES data analysis is a difficult task,
quality criteria, 25 variants gain a stop codon, 56 have putative essential variant catalogs such as ClinVar [132] and Human Gene Mutation Data-
splice sites, and 56 were predicted to result in a frameshift mutation. base (HGMD) [133] are suitable for guiding variant classifications
The information obtained from the GAERS genome study was too exten- among known disease genes, which facilitate the interpretation of a
sive and varied that the analysis is still in progress, and it is difficult to novel missense variant. Despite this, the final variant interpretation
make a correlation with genetic modifications responsible for seizure can be a highly subjective process, contributing to the variability re-
susceptibility [36]. The WGS approach can also be applied to identify ported in variant classification. For this reason, the American College
susceptible genes in human, which may allow high rates of molecular of Medical Genetics and Genomics (ACMG) has drastically reduced var-
diagnosis in patients with epilepsy and increase the identification of ep- iability in the classifications by systematizing the variant interpretation
ilepsy genes [37,38]. process.
The WES has also some limitations, for example, it is not able to de-
2.3. Whole exome sequencing (WES) tect alterations in noncoding regions or methylation abnormalities, and
hence, the use of other techniques is required [46,47].
The WES, a genomic technique for sequencing the entire protein- Several studies performed different techniques in a complementary
coding region of genes in a genome, may be an excellent option when way to identify deleterious genetic variants in epilepsy and to detect the
other methods fail (Fig. 1B). Allen et al. [39] used WES to investigate responsible genes for a given type of epilepsy. For example, a combina-
mutations in epileptic encephalopathies and found 329 de novo mutation of aCGH, targeted sequencing, and WES successfully allowed the
tions; some of them have a clear statistical evidence of association identification of de novo mutations that affected the maturation and de-
with epileptic encephalopathy, such as GABRB3 (that encodes the sub- velopment of neural networks [48].
units beta 3 of the GABA type A receptors) and ALG13 (that encodes The WES results must be validated, ensuring that the mutations are
the ALG13 UDP-N-Acetylglucosaminyltransferase Subunit). found in all tested animals. To do so, primers near the mutated regions
The WES, despite its high potential, is not commonly used in exper- are designed to subsequently amplify those regions and perform the
imental models of epilepsy. However, it was seemingly a good choice in SANGER sequencing (Fig. 1C).
the first genetic studies of the GASH/Sal model, as the genome used as a The large number of gene variants detected in the GASH/Sal using
reference was incomplete. Once the DNA is isolated and fragmented, it WES analysis allowed us to identify high impact mutations in genes
that might be related to epilepsy (unpublished data). A first step to re- 3. Gene expression studies
duce such a big number of gene variants is to filter out based on the mu-
tation characteristics. Once the variants of interest have been identified, The fact that RNA-based approaches provide answers that cannot be
the next step will be to assess the implication of each gene variant in the revealed by conventional DNA-sequencing approaches have contrib-
seizure susceptibility. uted to resolve great conundrums in epilepsy research.
2.4. DNA modifications 3.1. Microarrays
The study of DNA modifications includes DNA methylation, specific The development of global gene expression profiling platforms has
chromatin changes such as histone modifications, and noncoding revolutionized research in molecular biology, allowing identification,
RNAs among other components that are tightly interconnected. All cataloging, and measuring of vast amounts of information at a single
these DNA modifications form integral part of the epigenetic machinery time. The corresponding genomic changes are consistent, and hence,
that regulates the gene expression [49,50]. Alterations in DNA methyla- these techniques give us a broader perspective for understanding the
tion of specific genomic regions may contribute to the development of disease process. The late 1990s heralded the development of several im-
complex disorders (e.g., Prader–Willi and Angelman syndromes), as portant technological advances in molecular biology, specifically the
they could prompt cells to gain or loss biological functions of genes typ- DNA microarrays that were used for comparing global gene expression
ically expressed only from the maternal or the paternal copy [51,52]. in experiments at different conditions. This has led to flourish a rela-
Recent studies show that epilepsy and epileptogenesis are tively new field of comparative genomics [61]. DNA microarrays gener-
associated with changes in each of the factors that affect the epigenome ate large libraries of mRNA sequences, enabling researchers to compile
[134]. Aberrant DNA methylation signatures link general differential gene expression lists in disease states by statistical compar-
pathomechanisms of epileptogenesis and epilepsy to epileptic brain tis- ison of transcript frequencies between two or more conditions (Fig. 2A).
sue in experimental animal models and humans. The repressive capac- It allows simultaneous monitoring of thousands of genes, thus providing
ity of cytosine DNA methylation is mediated by recruitment of silencing a functional aspect to sequence information in a given sample. There are
complexes, in which methyl-CpG binding domain proteins aid in the several types of microarray technologies currently in use, and among
mediation of gene silencing. There are standard procedures used to them, the oligonucleotide microarray technology is the most widely
study those modifications and patterns of methylations based on the bi- used for high-throughput quantitative studies of RNA expression [62].
sulfite conversion of genomic DNA, methylation-sensitive restriction As shown in Fig. 2A, RNA extracted from the material of interest is re-
enzyme digestions, and the immunoprecipitation of methylated DNA verse transcribed to complementary DNA (cDNA) and then incubated
using methyl CpG-specific antibodies [53]. The procedure consists of with a mixture of fluorescent markers on a microarray chip that con-
fragmenting the genomic DNA, enrichment of methylated DNA (using tains a predetermined set of genes. Quantification is then performed
commercial methylated DNA-binding enrichment kit), preparation of by computerized measurements of fluorescent intensities. Known limi-
the MethylC-sequencing library in NGS platforms (e.g., Illumina, tations of microarray technology are the large amounts of material at
SOLiD, or Ion Torrent), and finally, assessment using bioinformatics the initial stage and the unfeasibility to identify novel genes, as it relies
data analysis (Fig. 1C). These methods have been carried out in rodent on prior knowledge of the targeted genes. These have largely been ad-
models of epilepsy, suggesting that altered DNA methylation patterns dressed with a two-stage hybridization process or global RNA amplifica-
contribute to pathological mechanisms that induce genetic deregulation tion prior to running the array experiment [61,65]. The molecular
in epilepsy syndromes. In chronic epileptic rats, Kobow et al. [50] re- mechanisms underlying epileptogenesis are thought to be associated
ported an increase in DNA methylation, mainly confined to gene bodies. with altered expression of gene groups [66], and the microarray has
These altered methylation patterns were inversely correlated with the been the preferred platform for their gene expression analysis [61,67].
gene expression, so that an increase in gene body methylation was asso- In the last 20 years, there have been published more than 40 large-
ciated with gene silencing and, conversely, a decrease in methylation scale gene expression studies on epileptic tissue obtained from resec-
was associated with increased gene expression. The calcium/calmodu- tion of the epileptogenic zone [11,61,66–79]. Furthermore, mainly in
lin-dependent protein kinase 2 (Camkk2) gene that encodes a cal- the last decade, some works aimed to study the potential genes or path-
cium-dependent protein kinase was found hypermethylated and ways associated with epilepsy based on micro RNA (miRNA) expression
downregulated, which may be linked to aberrant neuronal activity profiles [80–84]. Numerous attempts have been made to employ tran-
and seizure. Moreover, interleukin 10 receptor subunit beta (Il10rb) scriptional expression profiles in animal models, as well as in patients
gene that encodes an interleukin receptor protein was found with temporal lobe epilepsy (TLE). Despite this, there is no consensus
hypermethylated and upregulated, which could be a compensatory regarding common transcriptional drivers of epileptogenesis [66]. Vari-
mechanism to limit brain damage following recurrent seizures. Admin- ability among animal species, epilepsy models, different sample size, tis-
istration of ketogenic diets to epileptic rats was found to have anticon- sue sampling time-points, array platforms, and normalization
vulsant activity, ameliorating DNA methylation changes [50]. algorithms, have resulted in only a few genes demonstrating a consis-
Consistently, similar studies also revealed increased methylation in tent expression change [66,68,70]. Thus, many studies reported the
gene bodies and hypomethylation at nongenic regions, supporting the same differentially expressed genes related to epilepsy, such as tissue
idea that altered DNA methylation is a general pathomechanism associ- inhibitor of metalloproteinases 1 (Timp-1), signal transducer and activa-
ated with epileptogenesis and epilepsy in these models [49,54]. Further, tor of transcription 3 (Stat3), complement C1q chain (C1q), solute car-
in vitro experiments correlated posttranscriptional modifications of his- rier family 6, member 1 (Slc6a1), and Pcdh19 [70]. All these genes
tones (H3/H4 acetylation and H3 phosphorylation) with downregula- share the common characteristic of belonging to different pathways,
tion in the expression of two potential epilepsy-associated genes, the some of which are related to epileptogenesis, such as the mitogen-acti-
excitatory glutamate receptor genes glutamate ionotropic receptor vated protein kinase (MAPK) signaling pathway, the Janus kinases/sig-
AMPA type subunit 2 (Gria2) and glutamate ionotropic receptor nal transducer and activator of transcription proteins (JAK/STAT)09
NMDA type subunit 2A (Grin2a) [55]. Also, increases in histone H4 acet- pathway, p53 signaling pathway, and the extracellular matrix (ECM)/
ylation and in phosphorylation of histone H3 have been reported in an- integrin signaling [70]. Interestingly, genes like MAPK, HIF, and JAK/
imal models of epilepsy [56–58]. Based on these results, several studies STAT are potentially associated and upregulated under a common hyp-
have been developed histone deacetylase (HDAC) inhibitors as a possi- oxic condition in a HIF-dependent manner [85].
ble therapy in epilepsy, since HDAC inhibition has a neuroprotective ef- By using microarrays, our research group analyzed changes in gene
fect [57], but without a principal anticonvulsant action [59,60]. expression of the IC in two audiogenic seizure strains, the WAR and
Fig. 2. General workflows for differential gene expression analysis using different techniques. Schemes show the standard workflow of microarray (A) (reproduced with permission from
[62]), RNA sequencing (B), RT-qPCR (C), and spatial transcriptome studies (D) [smFISH (reproduced with permission from [63]) scRNA-Seq (reproduced with permission from [64])].
the GASH/Sal models, and compared with age-matched controls under not depend on the limited probes present on tiled microarrays and
the same stimulation conditions. The microarray results showed that can, therefore, be used to challenge any location in the genome, includ-
the genomic profile of the WAR animals exhibited significant upregula- ing unannotated genes. Finally, sequencing is not limited by array hy-
tion of 38 genes and downregulation of 47 genes when compared with bridization chemistry, such as melting temperature, cross-
Wistar rats. Comparison of gene expression profiles between stimulated hybridization, and secondary structure concerns. In conclusion, RNA-
control and GASH/Sal hamsters revealed the upregulation of 10 genes Seq directly reveals sequence identity that is crucial for analysis of un-
and the downregulation of 5 genes. Finally, the comparison of the known genes, novel transcript isoforms, and genetic variations [87,88].
gene expression profiles between the two audiogenic seizure models Most RNA-Seq experiments are carried out on instruments that se-
showed only one common gene, the zinc finger immediate-early quence DNA molecules due to the technical maturity of commercial in-
growth response gene early growth response protein 3 (Egr3), which struments designed for DNA-based sequencing. Therefore, as shown in
was upregulated in both cases. We hypothesized that their overexpres- Fig. 2B, cDNA library preparation from RNA is a required step for RNA-
sion in both models might contribute to neuronal viability and develop- Seq. Each cDNA in an RNA library is composed of a cDNA insert of certain
ment of lymphoma in response to stress associated with audiogenic size (previously RNA or cDNA fragmentation), and flanked by adapter
seizures [11]. sequences, as required for amplification and sequencing on a specific
Microarray studies have also been shown to demonstrate significant platform. The cDNA library preparation method varies depending on
variations even under the same experimental conditions. As mentioned the RNA species under investigation, which can differ in size, sequence,
below, the validation of global expression profiling results by indepen- structural features, and abundance. Moreover, because of the detection
dent methods such as quantitative real-time polymerase chain reaction limit of most sequencers, several cDNA libraries need to be amplified by
(RT-qPCR), Northern/Western blotting, and immunodetection is essen- PCR before sequencing, and to correct for PCR amplification bias, using
tial [61]. On the other hand, microarray technology has certain limita- methods that eliminate PCR duplicates from sequencing results [88].
tions such as the low sensitivity and high false-positive rates [86]. One After, directly cDNA libraries, or subsequent PCR products, are se-
way of overcoming these drawbacks is to use meta-analysis methods quenced using NGS, producing millions of short sequence reads, typi-
that integrate the results of separate microarray studies and increase cally 30 to 500 bp in length, depending on the DNA-sequencing
both the sensitivity and the reliability of measurements of gene expres- technology used. Finally, the reads obtained after sequencing are either
sion changes [75,86]. aligned to a reference genome or transcriptome, or assembled de novo
without genomic sequence guidance, to create a genome-scale tran-
3.2. RNA-Seq scription map that provides both transcriptional structure and expres-
sion level for each gene. Thus, the number of reads mapped within a
The first decade of this millennium witnessed the advent of massive gene or an exon can then be used as a measure of its abundance in the
parallel sequencing, also known as deep sequencing or NGS. Thus, RNA- analyzed sample [87].
sequencing (RNA-Seq) is now the method of choice to study gene ex- Increasingly available data in RNA-Seq databases will help the neuro-
pression and identify novel RNA species. RNA-sequencing has several science community to make new discoveries at unprecedented speed and
advantages over microarrays. First, sequencing technology is much depth. However, many challenges still remain. For example, most of the
more sensitive and quantitative than arrays. Second, RNA-Seq provides read counts are produced by very highly expressed genes. Many genes
a larger dynamic range for detection of transcripts (N 9000-folds) com- of interest with low expression levels may have gone undetected because
pared with standard arrays. Third, sequence data are more specific of very low read counts. These low read counts can also be quite variable,
and have less background. Furthermore, sequencing experiments do which can cause biases in differential expression analyses [87]. Up to now,
RNA-Seq had been employed at least in 10 reports to profile the entire Skene and Grant [98] applied the expression-weighted cell-type enrich-
transcriptome across multiple phases of epileptogenesis in different ani- ment (EWCE), a method that uses single cell transcriptomes to generate
mal models of epilepsy, as well as changes in transcription at the chronic the probability distribution associated with a gene list that has an aver-
phase in humans [10,11,82,89–95]. age level of expression within a cell type. Also, these authors applied
Thus, the transcriptional profile of the corpus quadrigeminum, com- EWCE to human genetic data from cases of epilepsy among other neuro-
prised the inferior and superior colliculi, in WAR animals was recently logic pathologies [98]. By last, as mentioned below, a study describes a
described [10]. Transcriptomic analysis of WAR (versus Wistar rats) method called fluorescent in situ RNA sequencing (FISSEQ), which en-
after acoustic stimulation allowed us to identify a set of 62 and 16 differ- ables not only the study of the transcriptomes of single cells but also
entially regulated genes using two types of statistical packages (EdgeR the determination of the precise location of each transcript within the
and DESeq), such as a gene that encodes the coenzyme acyl-CoA synthe- cell [99].
tase (Acsm3), G-protein coupled receptor 126 (Gpr126, encodes a recep-
tor that could be associated with alterations in the acquisition or the 3.3. RT-qPCR
processing of acoustic information), regulator of telomere elongation
(Rtel, encodes a helicase that acts in the protection, stability, and elonga- After data mining obtained by microarrays and RNA-Seq, the results
tion of telomeres), and enzyme dihidopterina reductase quinoide (Qdpr, need to be validated with highly reliable biotechniques that allow pre-
encodes the enzyme dihidopterin reductase quinoid that is responsible cise quantitation of transcriptional abundance of the identified genes.
for recycling of BH4, a molecule that acts as enzyme cofactor and is in- The RT-qPCR technology is central to biomarker validation where
volved in the synthesis of serotonin, dopamine, and nitric oxide). potential markers need to be measured with greater accuracy and pre-
Gene Ontology (GO) enrichment analysis showed that “catalytic activ- cision in larger sample sets. RT-qPCR represents the method of choice
ity” and “metabolic processes” categories were among the most repre- for analyzing gene expression of a moderate number of genes in any-
sented functional categories of the genes differentially regulated in where from a small number to thousands of samples [100–102]. Quan-
WAR, suggesting that the model presents metabolic alterations [10]. titative real time PCR allows quantification of PCR products in “real
On the other hand, our research group analyzed also the transcriptome time” during each PCR cycle, yielding a quantitative measurement of
of IC of the GASH/Sal model in comparison with their control under the PCR products accumulated during the course of the reaction. Real-
same stimulation conditions. We previously used the mouse probes for time reactions are carried out in a thermocycler that permits measure-
the microarray analyses of gene expression (GeneChip® Mouse Gene ST ment of a fluorescent detector molecule, which decreases
Array) in hamsters as the Syrian hamster (Mesocricetus auratus) probes postprocessing steps and minimizes experimental error (Fig. 2C). Quan-
were not available at that time. To confirm these results, we employed Chi- titative real time PCR is most commonly achieved through the use of
nese hamster probes (Cricetulus griseus) via transcriptomic analysis, com- fluorescence-based technologies such as probe sequences that fluoresce
paring the stimulated controls with the stimulated GASH/Sal. Upon using upon hydrolysis (TaqMan) or hybridization (LightCycler), fluorescent
these probes, the number of differentially expressed genes between hairpins, or intercalating dyes (SYBR Green) [103]. The majority of anal-
GASH/Sal and control animals was increased [11]. yses of RT-qPCR data use relative quantitation that is easier to measure
In other study, Dixit et al. [91] performed transcriptome analysis of and are of primary interest to researchers examining disease states as
hippocampal tissues resected from patients with mesial TLE with hippo- epilepsy. In epilepsy research, the most common method is the 2−ΔCT
campal sclerosis (MTLE-HS), using the RNA-seq approach. Differential method that relies on two assumptions [104]. The first is that the reac-
gene expression analysis revealed 56 significantly regulated genes in tion is occurring with 100% efficiency; in other words, with each cycle of
MTLE (vs. controls without epilepsy) and their possible association PCR, the amount of product doubles. The second is that there is a gene
with epileptogenesis and/or pharmacoresistance in MTLE-HS. Gene (or genes) that is expressed at a constant level between the samples.
cluster analysis identified three important hubs of genes mostly linked This endogenous, housekeeping, or reference gene will be used to cor-
to the following: 1) neuroinflammation and innate immunity; 2) synap- rect for any difference in sample loading [103]. Housekeeping genes
tic transmission (as a gene encodes kinesin heavy chain isoform 5A such as β-actin, glyceraldehyde 3-phosphate dehydrogenase,
[KIF5A], involved in GABA(A)R trafficking, and whose deletion causes cyclophilin, or tubulin are commonly used since they are ubiquitously
epilepsy); and 3) neuronal network modulation (as a gene encodes fi- expressed in cells and tissues. In a model of TLE, Crans et al. [105]
bronectin [FN1] and proposed cerebrospinal fluid [CSF]-serum bio- have recently validated two rodent-specific short interspersed nuclear
marker for epilepsy), which are supportive of intrinsic severity elements (SINEs) as reference genes, comparing these with other nine
hypothesis of pharmacoresistance [91]. Furthermore, by using this genes and using three types of algorithms: geNorm, NormFinder, and
same technique, a previous study reported that the status epilepticus rank aggregation. Our research group, as mentioned above, investigated
(SE) induced changes in the hippocampal RNA expression in an animal the comparison of the gene expression profiles between two audiogenic
model of pilocarpine-evoked SE [93]. These authors performed whole seizure models using the microarray data, showing upregulation of Egr3
transcriptome profiling to identify differentially expressed mRNAs at in WAR and GASH/Sal animals [11]. By using RT-qPCR studies, we con-
12 h, 10 days, and 6 weeks after evoking experimental SE, which corre- firmed the differential expression of the Egr3 as well as in the two
spond with the distinct phases of the epileptogenic process (acute, sei- other early growth response proteins 1 and 2 (Egr1 and Egr2), which
zure silent, and spontaneous-seizure phases). were also upregulated in both models. Differences between microarray
Recently, Jehi et al. [96] also proposed that the maturation of a new and RT-qPCR data might occur for the following reasons: 1) the use of
epileptic focus may explain late seizure recurrences. Using RNA-seq, different probes in the microarray and RT-qPCR experiments, which
they identify 29 differentially expressed genes between late-recurrence may lead to capture differential expression in splice variants; 2) differ-
and seizure-free patient samples, concluding that late recurrences after ences in the methods used for the normalization of expression data;
epilepsy surgery may be influenced partly by differences in gene expres- and 3) possible false positive outcomes of expression changes. In addi-
sion in neuroinflammatory and brain healing/remodeling pathways [96]. tion, lower correlations between RT-qPCR and microarray results were
Finally, it is noteworthy to highlight the fact that transcriptome anal- consistently reported for genes exhibiting small degrees of changes
ysis can also be performed at the level of single cells. Unlike classical [106].
methods, which consider mixtures of heterogeneous cell populations,
single-cell RNA sequencing (scRNA-Seq) provides a much more detailed 3.4. Spatial transcriptome
view of transcription dynamics [135]. For example, the analysis of
transcriptomes from single cells has revealed the substantial transcrip- The advance in sequencing technologies makes possible the study of
tional heterogeneity among seemingly identical cells [97]. Recently, the genomic and transcriptomic of single cells and tissues. The
complexity of multicellular organisms requires the design of high- 4.1. Two-dimensional electrophoresis (2DE)
throughput measurements that preserve spatial information about the
tissue context or subcellular localization of the analyzed nucleic acids. The two-dimensional electrophoresis (2DE) is the typical method to
Some applications of this approach are intended to measure gene ex- separate proteins in two dimensions according to their isoelectric point,
pression and activity for thousands of genes in multiple regions of the molecular mass, and solubility, using a polyacrylamide gel. Then, the gel
brain [107–110]. is stained using different dyes — such as Coomassie blue or silver nitrate.
The single-molecule RNA fluorescence in situ hybridization Another way is to use fluorescent reporters to identify the target sam-
(smFISH) technique represents a powerful means of gauging individual ples as in the 2D Fluorescence Difference Gel Electrophoresis (2D-
RNA expression values from single-cell transcriptome-wide measure- DIGE) (Fig. 4A). This technique has the advantage to directly display
ments. However, the number of transcripts that can be visualized simul- the relative abundances and patterns of protein isoforms in the samples.
taneously in the same sample is small (usually 1–3) because of the In this case, protein spots are considered as differentially expressed with
limited availability of fluorophores with nonoverlapping spectra, statistical significance when 1) they are present at least in 70% of the gel
which prevents highly multiplex measurements. An improvement of images, 2) a 1.5-fold change as a threshold average ratio, and 3) p-
this method is the so-called sequential fluorescence in situ hybridiza- values lower than 0.05. For this analysis, different software programs
tion (seqFISH), which implies increasing multiplexing by the use of spa- are employed for image capturing (i.e., EttanTM DIGE Imager, GE
tial or sequential barcoding that is combined with super-resolution Healthcare), image analysis (i.e., DeCyder TM Differential Analysis Soft-
microscopy. Thus, each transcript is identified through multiple cycles ware, GE Healthcare), and image quantification and selection of quanti-
of hybridization, imaging, and probe removal (Fig. 2D). tative changes [113].
Other alternative method to visualize RNA molecules in situ is based In the samples separated by simple 2DE and DIGE, the protein spots
on smFISH that uses padlock probes and rolling circle amplification as of interest are manually excised from the stained gels, de-stained, and
well as branched DNA probes [107,109,110]. This strategy combines subsequently subjected to in-gel digestion with trypsin. After protein
smFISH and scRNA-Seq, allowing visualization and quantification of extraction, they are further analyzed with a mass spectrometer [high-
the transcription with spatial resolution in tissue sections [64]. performance liquid chromatography (HPLC)/mass spectrometry
Also, laser capture microdissection (LCM) is a powerful technology (MS)–MS or by matrix-assisted laser desorption/ionization-time-of
that enables the isolation of cells or small tissue regions from defined flight (MALDI-TOF) mass spectrometer for fluorescent samples] (see
anatomical locations. Data sets derived from these samples correlate di- next section for more information). Different spectra are analyzed
rectly with the known original location, thus preserving spatial informa- with specific software (i.e., FlexAnalysis 3.3.65, Bruker Daltonics),
tion. Nucleic acids can be extracted from LCM-captured cells and used in using different algorithms (i.e., MASCOT - Matrix Science, London, UK
a variety of downstream applications, including gene expression micro- or X! Tandem http://www.thegpm.org/tandem/).
arrays and RNA-seq. There are many databases that can be used to identify the protein se-
Other approach for spatially resolved transcriptomics based on RNA quences in molecular studies of rodent models of epilepsy. Among
extraction from discrete tissue regions is the serial microtomy sequenc- them, http://kr.expasy.org or database of SwissProt 2015_06, in which
ing. In this method, RNA is extracted from single thin tissue cryosections 1612 Rodentia sequences are available.
and subjected to sequencing. To improve spatial resolution and quanti- This technique has been used to analyze changes in brain tissue of
fication of this technique, a modified approach named RNA tomography the WAG/Rij model after induced seizure enhancement, showing 16 dif-
sequencing method (Tomo-seq) has been developed [107,110]. ferentially expressed proteins in the frontoparietal cortex and 35 pro-
The transcriptome in vivo analysis (TIVA) based on RNA extraction teins in the thalamus [113]. Four of them were found in both brain
and sequencing of selected neurons within tissue slices allows charac- areas and were related to metabolism [ATP synthase subunit d, mito-
terization of gene expression profiles in accurate brain locations chondrial protein (ATP5H) and glyceraldehyde-3-phosphate dehydro-
[107,110,111]. genase (GAPDH)] and glial activation [glial fibrillary acidic protein
(GFAP) and neurofilament light polypeptide (NEFL)], which are able
4. Proteomic to modulate voltage-gated calcium channels [114]. Consistently, 2DE
performed in the epileptogenic nucleus of naïve GASH/Sal animals
Proteomics refers to the large-scale analysis of proteins and deals also showed alterations in proteins related to metabolism, with a signif-
with the study of protein compositions, including those derived from icant decrease of ATP5H as well as increases in GAPDH and malate dehy-
the posttranslational modifications, as well as protein interactions. The drogenase (MDH1). In addition, 2D-DIGE analyses were carried out to
proteomic technologies also involve the identification and quantifica- compare the proteomic profile of baseline and postseizure states, show-
tion of overall proteins, and hence, are well-suited for understanding ing nonsignificant differences in the epileptogenic nucleus of the GASH/
the changes that occur in the epileptogenic tissue after seizures. One Sal (unpublished data). The most striking differences were found be-
of the major applications of proteomics is to find useful markers for tween control and GASH/Sal at baseline conditions, showing an incre-
the diagnosis, treatment, and monitoring of various clinical entities. In ment in the metabolism protein pyruvate kinase (PK) and ATP
this regard, a principal objective of the epilepsy biomarker discovery is synthase beta subunit (ATP5F1B), along with an elevated concentration
to identify “hot lesions” in epileptogenic brain tissues that express a of peroxyredoxin-2 (PRDX2) in the IC. Additionally, the induction of ep-
highly specific and sensitive biomarker signature [5]. For this reason, de- ileptic seizures made the PRDX2 to switch from a 9.62-fold increase to a
termining the proteomic profile differences in the epileptogenic nuclei 4.05-fold decrease, which suggests an impairment of the antioxidant
of the audiogenic seizure models provides crucial information about systems. These data are in agreement with other results in Lafora Dis-
the mechanisms involved in seizure susceptibility and regulation of ease mice models [115].
neuronal excitability.
Unfortunately, the results from proteomics studies are very variable 4.2. Mass spectrometry and MALDI-TOF
as they depend on the sample preparation method, the separation be-
tween nuclear and cytoplasmic components, the type of protein extrac- Mass spectrometry is an analytical technique that measures the
tion, and the high biological variation between individuals. Finally, mass-to-charge (m/z) ratio of ions. In this procedure, the samples are
unlike the DNA stability, the proteome (and transcriptome) varies in ionized into charged molecules and then subjected to an electric field
space and time, thereby contributing to the great heterogeneity of re- in the mass analyzer, which determines the acceleration of the sample
sults that can be found in the literature. The existing process templates toward a detector, so that the ratio of their m/z can be measured. In
for large-scale study of proteins are shown in Fig. 4. the MALDI-TOF MS, the ion source is the matrix-assisted laser
desorption/ionization (MALDI), and the mass analyzer is time-of-flight peptide or protein mixtures prior to their detection and data analysis.
(TOF) analyzer (Fig. 4B). It is the technique that must be used when This step was done in the past using HPLC–MALDI-MS that performs
the samples are attached to a fluorescent (i.e., DIGE). In proteomics, offline peptide separation and spotting or fraction collection. The fact
one of the most important and difficult step is the separation of complex that this technique involves time-consuming manual inspection for
Table 1
Massive analysis in rodent genetic models of epilepsy shows dysregulated proteins and their corresponding functional pathway.
Reference Model organism Epilepsy Anatomical area Proteins upregulated (symbol) Proteins downregulated Classification/function
subtype of study (symbol)
Penot et Kainate injection in MTLE Hippocampus IL-1beta; IL-1Ra; Cytokines, inflammatory

al. hippocampus of mouse SOCS3; COX-2 processes
[136] cPLA(2)α apoptosis
Danis et Genetically modified rat AbE Cortex ATP5F1D; TMEM70 Energy generation
al. GAERS Thalamus MBP; MIF membrane conductance
[137] Hippocampus MIF; HBB inflammatory processes
Oxygen transport
Huang et SD rats injected TLE Hippocampus HSP27 PPIase Stress response
al. intraperitoneally with lithium TRIP6; LIMK1; TUBB; ACTB; INA Cytoskeleton
[138] chloride and then pilocarpine FBA, CK GADPH; SCOAL; ACO Metabolism
twice MBP Membrane conductance
SNX3 Intracellular trafficking
Jiang et Phenytoin in electrical RefE Full brain VDAC1 VDAC2 Membrane conductance
al. amygdala-kindled rats MT-ATP; GLUD1; MDH Energy generation
[139] ABCB10; NDUFA10; Mitochondrial transport
HSP-70 Stress response
GADPH; TPI1; ALDOA ACO; ALDH; metabolism
TUBB2 CFL1 cytoskeleton
Greene et Lithium pilocarpine-induced CStE Hippocampus HSP-27 Stress response
al. rat TUBA; EZR cytoskeleton
[140] Endocytosis production
DPYSL2 of neurotransmitters
DHPR
Junker et Kindled rats RefE Hippocampus UQCRFS1 Mitochondrial
al. respiratory chain
[141]
Walker et Rats, electrically-induced StE Parahippocampal HSP90B1 DNM3;IL1RAP Immune and
al. status epilepticus cortex HSPA1A TOLLIP; USP7 inflammatory responses
[142] IL1rapl1; ITGB2;RPS27A
Hippocampus HMGB1; HSP90B1; HSPA1A; TMED7;
TOLLIP
Bitsika et KA-MTLE mouse model MTLE Hippocampus TRF; GFAP; VIM; CLU; DHRS1; FLNA; MAP2; CTTN; PPP1R9B; Cytoskeletal proteins
al. HEXB; FLNA; APOD; TGM1; CAPG; CTSZ MAP1A; AAK1; BAIAP2; neuronal responses
[143] RPH3A; SLC6A1 Microglia/Astrocyte
Activation
Inflammatory
Response
Györffy WAG/Rij rats AbE Frontoparietal GFAP;GAPDH ATP5H; NEFL Cytoskeletal proteins
et al., cortex PSME1; UQCRC1 LMN; SUCLA2; GUK1 Motor proteins
[113] HNRNPA2B1 Chaperone/oxidative
Thalamus GFAP; CAPG GFAP; ACTA2 stress metabolism
INA; DPYSL2 MYH1; NEFL ATP
MYL1; TUBB ATP5H; GAPDH synthesis/respiratory
PRDX6;ENO1 LMN; TIP1 chain
MDH1; SEPT3 ATP6V1E1 MDH1;HNRNPH2 signal transduction
PARK7; APOE ATP6V1E1 transcription/translation
synaptic transmission
Liu et al. Pilocarpine rat model TLE Hippocampus NEFM; NRP2: ACTA1; TUBB2B; ACTB; Cytoskeletal proteins
[144] HSP90AA1; DDT HSPA1A; ANXA6; ENOSF1; PDHA1; ENO1;
VDAC2 SEMA6B; PKM TUBA1B; SYN2 ALDOA; PRPS1; Metabolism
HOMER2; INA ATP5F1B; SPTA1; NRGN HSPD1;
MSI1; COX6B1 SNAP25; STMN1; ARHGDIA;
UBE2E2; HBA1/2 AK1; Stress response
TAGLN3; NADPH; GSTM1; energy metabolism
HBB
Li et al. Pilocarpine mouse model TLE Dentate gyrus PFN1; UCHL1 ACTB; LDHB Cytoskeletal proteins
[145] ARHGDIA; CA2 PDXP; CRYM Metabolism
PPID; CRYAB SNCB; PRDX6 HNRNPDL; APOA1 Stress response
ENO2; LAMP2 GRHPR Apoptosis
VIM; CTSD synaptic transmission
HSPB1
Wu et al. Lithium pilocarpine-induced MTLE Hippocampus ACBP; CAH2 Cytoskeletal proteins
[146] rat treated CAH5A; CH10 synaptic function
COF1; CX6B1 FABP7; GADPH energy metabolism
GFAP; GLOD4 mitochondrial function
HBA; KAD1; KCRB molecular chaperones
NRGN; PDIA3; PPIA; PRDX1; TTR; UB2V2 signal regulation
subsequent MS analysis resulted in very low separation efficiencies glycerol uptake/transporter 1 (Hhatl), were identified as playing crucial
[116]. Currently, there are multiple automated liquid chromatography roles in cardiac injury during epilepsy [120].
(LC)/MS platforms that allow the separation of proteins and MS analysis
automatically [117]. The automation of procedure has yielded the sepa- 4.5. Protein modifications
ration of thousands of proteins in each sample and has become the reg-
ular method of choice when it is desired to produce a high-level Proteomics is capable of profiling posttranslational protein modifica-
proteomic profile in the sample. The main limitation of the LC/MS is tions such as phosphorylation, acetylation, hydroxylation, methylation,
the fractionation of the sample prior to the MS analysis, which makes glycosylation, AMPylation, lipidation, ubiquitination, and deamidation,
that the large amount of proteins produced by the LC separation can as well as splicing-dependent alterations in protein expression patterns.
quickly saturate the capacity of the spectrometer detector. Since these posttranslational mRNA and protein modifications define
unique functional features in a cell phenotype, they are highly relevant
and have to be considered in epilepsy research. Spectrophotometry
4.3. Massive analysis techniques have become very sophisticated in recent times and allow
visualizing the posttranslational changes. Mass spectrometry tech-
Another approach, especially useful for the analysis of complex pro- niques can be used to profile posttranslational modified (PTM) peptides
tein mixtures, is to carry out a global analysis, the so-called Shotgun pro- in a digested sample. The PTMScan® technology combines antibody en-
teomics. This technique refers to the use of bottom-up proteomics richment of PTM-containing peptides with HPLC–MS/MS and allows
techniques in identifying proteins in complex mixtures by using a com- identification and quantification of hundreds to thousands of even the
bination of LC/MS. The most common method of shotgun proteomics lowest abundance peptides (Fig. 4D). All in all, this technique provides
starts with the digestion of proteins in the mixture and subsequently a more focused approach to peptide enrichment than other strategies
the separation of the resulting peptides by LC. Tandem MS is then and makes possible to determine novel protein sites that are phosphor-
used to identify the peptides (tandem MS or MS–MS) (Fig. 4C). This se- ylated, ubiquitinated, acetylated, methylated, or protease cleaved (cell
quential MS provides detailed information on the sequence of the pep- signaling technology- https://www.cellsignal.com/contents//
tide, which is compiled, constituting the primary technique for the simplifying-proteomics/proteomics). Independently of the proteomics
genesis of the “peptide trace”. Using this technique, several authors de- tools used, all results need to be validated before any potential bio-
scribe a list of proteins related to changes in the brain of different epi- marker could be subjected to further analysis. The most common
lepsy models (Table 1). method for biomarker validation is the immunodetection.
In the GASH/Sal model, some dysregulated proteins are still un-
known, 5.63% are upregulated and 22.22% are downregulated. In gen-
eral, there is a great variability of proteins that could be classified in 5. Immnunodetection
different functional categories (Fig. 3) according to the PHANTHER
method [118]. Epilepsy research using rodents to model epilepsy has made signifi-
cant progress thanks to a variety of techniques that specifically detect
and identify antigens (or proteins) in epileptic samples. Common
4.4. Isobaric tags for relative and absolute quantitation (iTRAQ) methods of immunological detection in epilepsy research are western
blot analysis, immunostaining, and protein microarrays. All of them
Mass spectrometry analysis can also be used directly to quantita- are based on the existence of specific antibodies (Fig. 4E).
tively explore the proteome, applying isobaric tags for a relative and ab-
solute quantification (iTRAQ)-based proteomics approach to identify 5.1. Western blot analysis
differentially expressed proteins. This technique requires the “tagging”
in N-terminal position of previously digested protein samples. Each ex- This technique – also known as electroblotting [121] or immuno-
perimental group is tagged with a different reporter, then samples are blotting [122] – is a rapid and sensitive assay that combines the resolu-
intermixed, separated through LC, and analyzed with MS–MS. The N- tion of gel electrophoresis and the principles of immunological
terminal reporters are then ionized and relative-quantified. A funda- recognition of an antigen [123]. By using a western blot analysis, re-
mental disadvantage of this tool is the high cost due to the tagging pro- searchers are able to separate and identify specific proteins from a mix-
cess and the powerful spectrometers that are necessary [119]. Using this ture of proteins extracted from complex biological samples such as
technique, a series of cardiac tissue proteins from Wistar in acute and serum, cells, or tissue. In this technique, a mixture of proteins is sepa-
chronic epilepsy rat models have been described. Three of these pro- rated based on molecular weight, and by protein type, through gel elec-
teins, the receptor for activated protein kinase C1 (RACK1), the alde- trophoresis. The results are then transferred to a membrane producing a
hyde dehydrogenase 6 family member A1 (ALDH6A1), and the band for each protein. The membrane is then incubated with labeled
Notes to table 1
Epilepsy subtypes: AbE—absence epilepsy; CStE — convulsive status epilepticus; MTLE — mesiotemporal lobe epilepsy; RefE — refractory epilepsy; StE — status epilepticus; TLE — temporal
lobe epilepsy.
Abbreviations: AAK1 — AP2-associated protein kinase 1; ACBP — Acyl-CoA-binding protein; ACO — aconitate hydratase; ACTA1 — alpha actin-1; ACTA2 — alpha actin-2; ACTB — beta actin;
ALDH — Aldehyde dehydrogenase; AK1 — Adenylate kinase isoenzyme 1; ALDOA — Aldose A; ALK — Anaplastic lymphoma kinase; ANXA6 — Protein disulfideisomerase A6 precursor;
APOA1 — apolipoprotein A1; APOE — Apolipoprotein E; ARHGDIA — Rho GDP — Dissociation Inhibitor Alpha; ATP5F1B — ATP synthase subunit beta; ATP5F1D-
ATP Synthase F1 Subunit Delta; ATP5H — ATP synthase subunit D; ATP6V1E1 — ATPase H+ Transporting V1 Subunit E1; BAIAP2 — Brain-specific angiogenesis inhibitor 1-associated pro-
tein 2; CA2 — carbonic anhydrase 2; CAH5A — Carbonic anhydrase 5A; CAPG — Macrophage-capping protein; CD28 — CD28 Antigen; CFL1 — Cofilin 1; CH10 — 10 kDa heat shock protein;
CK — creatine kinase; CLU — Clusterin; COF1 — Cofilin-1; COX2 — cycloxygenase-2; COX6B1 — Cytochrome c oxidase subunit VIb isoform 1; cPLA(2)α — group IVA cytosolic phospholipase
α2; CRYAB — crystallin, alpha B; CRYM — crystallin, mu; CTSD — cathepsin D; CTSZ — Cathepsin Z; CTTN — Src substrate cortactin; CX6B1 — Cytochrome c oxidase subunit VIb isoform 1;
DDT — D-Dopachrome Tautomerase; DHPR — dihydropteridine reductase; DHRS1 — Dehydrogenase/reductase SDR family member 1; DNM3 — Dynamin 3; DPYSL2 —
dihydropyrimidinase-related protein-2; ENO1 — enolase 1; ENO2 — enolase 2; ENOSF1 — Enolase Superfamily Member 1; EZR — ezrin; FABP7 — Fatty acid-binding protein; FAS — Fas
Cell Surface Death Receptor; FBA — fructose-bisphosphate aldolase A; FLNA — Filamin A; GADPH — glyceraldehyde-3-phosphate dehydrogenase; GAERS — Genetic Absence Epilepsy
Rats from Strasbourg; GDNF — Glial cell line derived neurotrophic factor; GFAP — Glial fibrillary acidic protein; GLOD4 — Glyoxalase domain-containing protein 4; GLUD1 — Glutamate
dehydrogenase 1; GRHPR — glyoxylate reductase/hydroxypyruvate reductase; GSTM1 — Glutathione S-Transferase Mu 1; GUK1 — Guanylate kinase 1; HBA1/2 — Hemoglobin subunit
alpha-1/2;HBB — Hemoglobin subunit beta 1; HEXB — Beta-hexosaminidase subunit beta; HMGB1 — High Mobility Group Box 1; HNRNPA2B1 — Heterogeneous nuclear ribonucleoprotein
A2/B1; HNRNPDL — Heterogeneous Nuclear Ribonucleoprotein D Like; HNRNPH2 — Heterogeneous nuclear ribonucleoprotein H2; HOMER2 — Homer Scaffold Protein 2; HSP27 — heat
shock protein-27; HSP90AA1 — Heat shock protein HSP 90-beta; HSP90B1 — HSPA1A- Heat shock 70 kDa.
Fig. 3. Variation of protein expression between GASH/Sal and control animals in the epileptogenic nucleus (inferior colliculus) and the corresponding categorical classification. The plot
displays proteins with ±2-fold change compared with controls. https://www.ebi.ac.uk/reference_proteomes
antibodies specific to the protein of interest. Since the thickness of the distribution of EGR3 protein in the brain tissue of control and seizure-
band corresponds to the amount of protein present, western blot anal- prone animals. Therefore, comparison between different samples ob-
ysis provides a relative comparison of protein levels, but not an absolute tained from control and epilepsy-prone rodents is of paramount impor-
measure of quantity [123]. Detection, specificity, and sensitivity are cru- tance and, hence, needs to be based on objective data in order to obtain
cial for accurate and quantitative analysis in western blotting that accurate and reproducible results. As in western blot analysis, the quan-
strongly depends on the quality of the available antibodies. titative immunostaining performed on tissue sections by means of dig-
ital photomicroscopy and image analysis can be used to estimate the
5.2. Immunostaining number or density of immunopositive structures within the tissue sam-
ple as well as the immunostaining intensity across the region of interest.
Together with the western blot analysis, immunostaining is one of To compare multiple specimens, immunostaining and image acquisition
the gold standard protein analytical techniques that selectively identify should be performed in parallel for the entire set. Identical reagents and
proteins in cells (so-called immunocytochemistry) or in tissue sections processing should be used, with identical image acquisition settings and
(so-called immunohistochemistry) by exploiting the principles of im- exposure times. The most commonly used image-processing analysis
munological recognition of an antigen. Both immunostaining tech- software for quantitative immunostaining is the Java-based NIH ImageJ
niques are powerful microscopy-based techniques for the detection of software. ImageJ is in the public domain (available at http://rsb.info.nih.
specific proteins within individual cells and tissue samples and, hence, gov/nih-image), supports a wide image formats, and can be expanded
add valuable information about the immunolocalization and distribu- with the installation of more than 150 plugins that are additional tools
tion of a target protein in cell compartments and tissues. Immunostain- for facilitating scientific image analysis including quantitative immuno-
ing is developed following a variety of protocols that involve exposure staining. Fig. 5 shows an example of semiquantitative immunostaining
of fixed cells or tissues to primary antibodies directed against one or analysis of transthyretin (TTR) protein using the ImageJ software.
more proteins of interest. Bounded primary antibodies are then de- Mutant forms and overexpression of Ttr gene in brain tissue have
tected directly if they are conjugated to a label (direct detection been associated with seizure as well as abnormalities in behavior
method) or using commercially available secondary antibodies directed and movement [125]. Also, overexpression of Ttr could be directly re-
against the invariant portion of the primary antibody (indirect detection lated with membrane depolarization [126] and loss of GABA receptor
method). Two principal methodologies exist to visualize antigen–anti- activity [127] and thereby increasing the risk of seizures. As is the
body complexes: immunofluorescence using fluorophore-conjugated case of immediate early genes, an overexpression of Ttr gene might
antibodies or chemiluminescence using antibodies coupled to horserad- contribute to the altered brain gene expression profiles associated
ish peroxidase [124]. A good example of immunohistochemistry applied with seizure susceptibility in genetic rodent models of epilepsy.
in rodent models of epilepsy is the study of the association between ac- Since a gap between the mRNA and protein levels exists because of
tivation immediate early genes and seizure activity. The WAR and various levels of regulation, the results obtained by gene expression
GASH/Sal are two genetically epileptic prone rodents that showed over- (RT-qPCR) need to be correlated with the protein levels. The western
expression of the immediate-early growth response genes (Egr1, Egr2, blot and immunostaining analysis in the IC of the GASH/Sal model
and Egr3) in the IC (the epileptogenic focus) after an ictal event [11]. showed that TTR protein levels are higher in the GASH/Sal model
López-López et al. [11] used immunohistochemistry to complement than in age-matched controls, confirming the alterations in the
the results of RT-qPCR analysis of the Egr3 gene, determining the gene expression profiles.
Fig. 4. Experimental workflow for proteomic studies. A. Types of bidimensional electrophoresis. B. MALDI-TOF mass spectrometry (image from https://www.creative-proteomics.com/
technology/maldi-tof-mass-spectrometry.htm [112]). C. Conventional massive protein analysis: digestion of protein with trypsin, separation of peptides by LC followed by tandem
mass spectrometry. D. PTM-Scan. Posttransductional modifications (image obtained from https://www.cellsignal.com/contents/_/simplifying-proteomics/proteomics). E.
Immunodetection: Steps for protein immunodetection (wide arrows) and different techniques based on antigen–antibody recognition: Protein arrays (reproduced with permission of
R&D Systems), Western blotting and immunohistochemistry. Abbreviation: AB — antibody; DIGE — Fluorescence Difference Gel Electrophoresis; ID — Identity; LC — liquid
chromatography; MS — Mass spectrometry; PTM — posttranslational modified.
5.3. Protein microarrays relative levels of selected cytokines and chemokines (Mouse Cytokine
Array# ARY006, R&D Systems, Inc.; mouse Th1/Th2/Th17 cytometric
Protein microarray is a relatively new technology used for the char- bead array kit, BD Biosciences, San Diego, CA, USA). This type of protein
acterization of large numbers of proteins in parallel [128]. Protein mi- microarrays is very useful to determine modifications of certain inter-
croarrays include three classes: analytical, functional, and reverse- leukins associated with seizures as well as the effects of different anti-
phase protein microarrays that provide a powerful tool in quantifying convulsant compounds [131].
and profiling proteins. It is also a high-throughput method for tracking
the interactions and activities of proteins as well as determining their 6. Crosslinking
function and protein posttranslational modifications [129,130]. There
are multiple protein microarrays designed to identify a large number It is noteworthy to review the functional classification of genes or
of proteins as well as specific protein microarrays that are used for diag- proteins involved in epilepsy. The gene ontology classification of cellular
nostic assays in humans. Despite the considerable investments made by components, biological processes, and molecular functions has been
several companies, there are few commercial protein microarrays for used for functional classification by the majority of authors. Four func-
rodents. The few protein microarrays available so far are aimed at the tional groups have been described so far as the most representative in
identification of signaling pathways (Mouse AKT Pathway Phosphoryla- epilepsy (neuronal signaling, immunity/inflammation, transcriptional
tion Array C1, # AAH-AKT-1-2, RayBiotech, Inc.) or specific proteins, regulation, and signal transduction), and two pathways showed enrich-
such as inflammatory proteins. Among the latter, it should be noted ment: the chemokine signaling pathway and Toll-like receptor signaling
that the cytokine array is used for the parallel determination of the pathway. Moreover, additional processes and pathways that are linked
Fig. 5. Semiquantitative immunostaining analysis of transthyretin (TTR) protein in the inferior colliculus of the GASH/Sal. A. Western blot analysis of TTR protein with antibodies to TTR and
β-actin. The western blot band intensities were measured with ImageJ software. The histogram shows a semiquantitative analysis of TTR protein levels, which were normalized to β-actin
levels. The intensity (mean ± standard error) was normalized to wild-type values (control), which were set equal to 100%. Note that protein levels of TTR differed between control and
GASH/Sal animals. B. High-magnification photomicrographs illustrating TTR immunoreactivity in the central nucleus of the IC (scale bar represents 20 μm and is the same for all panels).
The RGB-color images corresponding to control and GASH/Sal animals were converted to 8-bit images that contained a grayscale of pixel intensities ranging from 0 (black) to 255 (white).
The densitometric procedure for the evaluation of TTR immunostaining was performed by using ImageJ software. The upper histogram shows percentages of variation (mean ± standard
error) for mean gray value of the immunostaining that were used as an indicator of changes in protein levels. The lower histogram shows the number of TTR-immunolabeled cells per mm2.
Note the increased density of immunostained cells in the GASH/Sal compared with the control animal.
to these two pathways include ubiquitin-mediated proteolysis, comple- designs in rodent models of epilepsy should incorporate complemen-
ment cascade, JAK–STAT signaling pathway, cellular growth and dif- tary analysis for verifying results obtained at each stage of the flow in-
ferentiation, cell survival, migration, apoptosis, Reactive Oxygen formation from DNA to RNA to protein levels. To achieve this goal, a
Species (ROS) production, Nitric oxide (NO) induction, MAPK signal- good experimental design should be developed with careful consider-
ing pathway, leukocyte transendothelial migration, and regulation of ation for determining the most appropriate approach, taking into ac-
cytoskeleton system [11,66]. There are several computer-based tools count that no one technique is applicable to every set objective, and
to establish relationships between differentially expressed genes or each has its own strengths and weaknesses.
proteins with epilepsy. These web interfaces integrate database Development of these research methods and their application to an-
that is used to determine co-occurrence between different annota- imal model of epilepsy might led to a rapid increase in our understand-
tions, identifying those that are statistically significant. These inter- ing of the complex mechanisms underlying epileptogenesis and seizure
actions are of special interest for proteins, because most of them generation in the immediate future. In this regard, comparative molec-
work in conjunction with other proteins. This is especially useful to ular approaches to determine the differences between rodent seizure
determine potential associations in cellular signaling pathways. models and their wild-type counterparts provide valuable information
Among the multiple tools on the web, Table 2 shows the most used on molecular events that contribute to seizure susceptibility and regula-
to search functional or structural associations between genes, pro- tion of neuronal excitability.
teins, and metabolic pathways. The rapid identification of these molecular alterations, as facilitated
by genomic and proteomic technologies, will shed light not only on
7. Conclusions the molecular pathways mediating epileptogenesis, but also on the de-
velopment of novel applications for the diagnosis, prognosis, and treat-
Advances in molecular tools have improved substantially the geno- ment of seizures.
mic and proteomic studies on specific genes and proteins associated
with epilepsy, allowing comprehensive research of rodent seizure Declaration of competing interest
models.
Since genetic variations, gene expression changes, and protein alter- The authors declare no conflicts of financial and nonfinancial
ations are causes of specific epilepsy phenotypes, the experimental interests.
Table 2
Web-based interfaces and database collections used to query functional and structural associations between genes, proteins and metabolic pathways.
Gene/Protein interactions Functional analysis.
http://akt.ucsf.edu/ http://amp.pharm.mssm.edu/Enrichr/
http://biocyc.org/ http://bioinfo.vanderbilt.edu/webgestalt/
http://david.abcc.ncifcrf.gov/ http://genecodis.cnb.csic.es
http://pid.nci.nih.gov/ http://gtlinker.cnb.csic.es/-
http://string-db.org http://mimi.ncibi.org/MimiWeb/main-page.jsp
http://www.csuchico.edu/ge/ http://www.pantherdb.org/
http://www.cuny.edu/pathways https://pathcards.genecards.org/
http://www.elsevier.com/online-tools/pathway-studio/biological-database https://www.ensembl.org/Multi/Search/New?db=core
http://www.genecards.org
http://www.genome.jp/kegg/
http://mimi.ncibi.org/MimiWeb/main-page.jsp
https://www.ncbi.nlm.nih.gov/
https://www.wikigenes.org
Pathological phenotypes pathways

https://ve.genecards.org/#input http://highered.colorado.gov/Academics/Transfers/gtPathways/
https://discovery.lifemapsc.com/gene-expression-signals/high-throughput-disease http://www.cuny.edu/pathways
http://www.informatics.jax.org/humanDisease.shtml http://www.ebi.ac.uk/biomodels-main/BIOMD0000000286
https://www.malacards.org/ http://www.elsevier.com/online-tools/pathway-studio/biological-database
https://www.genome.jp/kegg/pathway.html#disease http://www.genemania.org
http://www.genmapp.org/go_elite/
https://pathcards.genecards.org/
https://www.genome.jp/kegg/tool/map_pathway1.html
https://www.uniprot.org/
Acknowledgments [16] Helbig I, Lowenstein DH. Genetics of the epilepsies. Curr Opin Neurol 2013;26:
179–85. https://doi.org/10.1097/WCO.0b013e32835ee6ff.
[17] Heron SE, Grinton BE, Kivity S, Afawi Z, Zuberi SM, Hughes JN, et al. PRRT2 muta-
This study was supported by the Regional Government of Castilla tions cause benign familial infantile epilepsy and infantile convulsions with
and León [#SA070P17 (IP-López DE), and #GRS 1838/A/18 (IP- choreoathetosis syndrome. Am J Hum Genet 2012;90:152–60. https://doi.org/10.
1016/j.ajhg.2011.12.003.
Gonçalves)] and by SAF2016-78898-C2-2R (IP- Merchán). [18] Saitsu H, Kato M, Mizuguchi T, Hamada K, Osaka H, Tohyama J, et al. De novo mu-
tations in the gene encoding STXBP1 (MUNC18-1) cause early infantile epileptic
encephalopathy. Nat Genet 2008;40:782–8. https://doi.org/10.1038/ng.150.
References [19] Olson H, Poduri A. CDKL5 mutations in early onset epilepsy: case report and review
of the literature. J Pediatr Epilepsy 2012;1:151–9.
[1] World Health Organization (WHO), Updated 2017. Available in: http://www.who. [20] Marini C, Mei D, Parmeggiani L, Norci V, Calado E, Ferrari A, et al. Protocadherin 19
int/mediacentre/factsheets/fs999/en/ mutations in girls with infantile-onset epilepsy. Neurology 2010;75:646–53.
[2] Scheffer IE, Berkovic S, Capovilla G, Connolly MB, French J, Guilhoto L, et al. ILAE clas- https://doi.org/10.1212/WNL.0b013e3181ed9e67.
sification of the epilepsies: position paper of the ILAE Commission for Classification [21] Olivetti PR, Noebels JL. Interneuron, interrupted: molecular pathogenesis of ARX
and Terminology. Epilepsia 2017;58:512–21. https://doi.org/10.1111/epi.13709. mutations and X-linked infantile spasms. Curr Opin Neurobiol 2012;22:859–65.
[3] Steinlein OK. Genetics and epilepsy. Dialogs Clin Neurosci 2008;10:29–38. https://doi.org/10.1016/j.conb.2012.04.006.
[4] Caraballo RH, Dalla Bernardina B. Handb Clin Neurol. 2013;111:579–89. doi: [22] Shen J, Gilmore EC, Marshall CA, Haddadin M, Reynolds JJ, Eyaid W, et al. Mutations
https://doi.org/10.1016/B978-0-444-52,891-9.00060-9. in PNKP cause microcephaly, seizures and defects in DNA repair. Nat Genet 2010;
[5] Pitkänen A, Ekolle Ndode-Ekane X, Lapinlampi N, Puhakka N. Epilepsy biomarkers 42:245–9. https://doi.org/10.1038/ng.526.
— toward etiology and pathology specificity. Neurobiol Dis 2019;123:42–58. [23] Molinari F, Kaminska A, Fiermonte G, Boddaert N, Raas-Rothschild A, Plouin P, et al.
https://doi.org/10.1016/j.nbd.2018.05.007. Mutations in the mitochondrial glutamate carrier SLC25A22 in neonatal epileptic
[6] Kandratavicius L, Balista P, Lopes-Aguiar C, Ruggiero R, Umeoka E, Garcia-Cairasco encephalopathy with suppression bursts. Clin Genet 2009;76:188–94. https://doi.
N, et al. Animal models of epilepsy: use and limitations. Neuropsychiatr Dis Treat org/10.1111/j.1399-0004.2009.01236.x.
2014;10:1693. https://doi.org/10.2147/NDT.S50371. [24] Mills PB, Camuzeaux SSM, Footitt EJ, Mills KA, Gissen P, Fisher L, et al. Epilepsy due
[7] Muñoz LJ, Carballosa-Gautam MM, Yanowsky K, García-Atarés N, López DE. The ge- to PNPO mutations: genotype, environment and treatment affect presentation and
netic audiogenic seizure hamster from Salamanca: the GASH:Sal. Epilepsy Behav outcome. Brain 2014;137:1350–60. https://doi.org/10.1093/brain/awu051.
2017;71:181–92. https://doi.org/10.1016/j.yebeh.2016.03.002. [25] Depaulis A, David O, Charpier S. The genetic absence epilepsy rat from Strasbourg
[8] Ross K, Coleman J. Developmental and genetic audiogenic seizure models: behavior as a model to decipher the neuronal and network mechanisms of generalized idi-
and biological substrates. Neurosci Biobehav Rev 2000;24:639–53. https://doi.org/ opathic epilepsies. J Neurosci Methods 2016;260:159–74. https://doi.org/10.1016/
10.1016/S0149-7634(00)00029-4. j.jneumeth.2015.05.022.
[9] Coenen AML, Van Luijtelaar ELJM. The WAG/Rij rat model for absence epilepsy: age [26] van Luijtelaar G, Zobeiri M. Progress and outlooks in a genetic absence epilepsy
and sex factors. Epilepsy Res 1987;1:297–301. https://doi.org/10.1016/0920-1211 model (WAG/Rij). Curr Med Chem 2014;21(6):704–21. https://doi.org/10.2174/
(87)90005-2. 0929867320666131119152913.
[10] Damasceno S, de Menezes NB, Rocha C de S, de Matos AHB, Vieira AS, Moraes MFD, [27] Garcia-Cairasco N, Umeoka EHL, Cortes de Oliveira JA. The Wistar Audiogenic Rat
et al. Transcriptome of the Wistar audiogenic rat (WAR) strain following audio- (WAR) strain and its contributions to epileptology and related comorbidities: his-
genic seizures. Epilepsy Res 2018;147:22–31. https://doi.org/10.1016/j. tory and perspectives. Epilepsy Behav. 2017;71(Pt B):250–273. doi: https://doi.
eplepsyres.2018.08.010. org/10.1016/j.yebeh.2017.04.001.
[11] López-López D, Gómez-Nieto R, Herrero-Turrión MJ, García-Cairasco N, Sánchez- [28] Cunliffe VT. Building a zebrafish toolkit for investigating the pathobiology of epi-
Benito D, Ludeña MD, et al. Overexpression of the immediate-early genes Egr1, lepsy and identifying new treatments for epileptic seizures. J Neurosci Methods
Egr2, and Egr3 in two strains of rodents susceptible to audiogenic seizures. Epi- 2016;260:91–5. https://doi.org/10.1016/j.jneumeth.2015.07.015.
lepsy Behav 2017;71:226–37. https://doi.org/10.1016/j.yebeh.2015.12.020. [29] Beaudet AL. Which way for genetic-test regulation? Leave test interpretation to
[12] Wang J, Lin Z, Liu L, Xu H, Shi Y, Yi Y, et al. Epilepsy-associated genes. Seizure 2017; specialists. Nature 2010;466:816–7. https://doi.org/10.1038/466816a.
44:11–20. https://doi.org/10.1016/j.seizure.2016.11.030. [30] Castells-Sarret N, Cueto-Gonzáleza MA, Borreganc M, López-Grondona F, Miró R,
[13] Kearney H, Byrne S, Cavalleri GL, Delanty N. Tackling epilepsy with high-definition Tizzano E, et al. Comparative genomic hybridisation as a first option in genetic di-
precision medicine: a review. JAMA Neurol 2019. https://doi.org/10.1001/ agnosis: 1,000 cases and a cost–benefit analysis. An Pediatr (Barc) 2018;89(1):
jamaneurol.2019.2384. 3–11. https://doi.org/10.1016/j.anpedi.2017.07.011.
[14] Weber YG, Lerche H. Genetic mechanisms in idiopathic epilepsies. Dev Med Child [31] Mullen SA, Carvill GL, Bellows S, Bayly MA, Berkovic SF, Dibbens LM, et al. Copy
Neurol 2008;50:648–54. https://doi.org/10.1111/j.1469-8749.2008.03058.x. number variants are frequent in genetic generalized epilepsy with intellectual dis-
[15] Butler KM, Moody OA, Schuler E, Coryell J, Alexander JJ, Jenkins A, et al. De novo ability. Neurology 2013;81:1507–14. https://doi.org/10.1212/WNL.
variants in GABRA2 and GABRA5 alter receptor function and contribute to early- 0b013e3182a95829.
onset epilepsy. Brain 2018;141(8):2392–405. https://doi.org/10.1093/brain/ [32] Hartmann C, von Spiczak S, Suls A, Weckhuysen S, Buyse G, Vilain C, et al. Investi-
awy171. gating the genetic basis of fever-associated syndromic epilepsies using copy
number variation analysis. Epilepsia 2015;56:e26–32. https://doi.org/10.1111/epi. Epilepsy Res 2008;81(2–3):107–13. https://doi.org/10.1016/j.eplepsyres.2008.04.
12920. 019.
[33] Mefford HC, Muhle H, Ostertag P, von Spiczak S, Buysse K, Baker C, et al. Genome- [60] Younus I, Reddy DS. Epigenetic interventions for epileptogenesis: a new frontier for
wide copy number variation in epilepsy: novel susceptibility loci in idiopathic gen- curin epilepsy. Pharmacol Ther 2017;177:108–22. https://doi.org/10.1016/j.
eralized and focal epilepsies. PLoS Genet 2010;6:e1000962. https://doi.org/10. pharmthera.2017.03.002.
1371/journal.pgen.1000962. [61] Wang YY, Smith P, Murphy M, Cook M. Global expression profiling in
[34] Olson H, Shen Y, Avallone J, Sheidley BR, Pinsky R, Bergin AM, et al. Copy number epileptogenesis: does it add to the confusion? Brain Pathol 2010;20:1–16.
variation plays an important role in clinical epilepsy. Ann Neurol 2014;75: https://doi.org/10.1111/j.1750-3639.2008.00254.x.
943–58. https://doi.org/10.1002/ana.24178. [62] Jayapal M, Melendez AJ. DNA microarray technology for target identification and
[35] Tsuchida N, Nakashima M, Kato M, Heyman E, Inui T, Haginoya K, et al. Detection of validation. Clin Exp Pharmacol Physiol 2006;33:496–503. https://doi.org/10.1111/
copy number variations in epilepsy using exome data. Clin Genet 2018;93:577–87. j.1440-1681.2006.04398.x.
https://doi.org/10.1111/cge.13144. [63] Lubeck E, Cai L. Single-cell systems biology by super-resolution imaging and combi-
[36] Casillas-Espinosa PM, Powell KL, Zhu M, Campbell CR, Maia JM, Ren Z, et al. Evalu- natorial labeling. Nat Methods 2012;9:743–8. https://doi.org/10.1038/nmeth.2069.
ating whole genome sequence data from the Genetic Absence Epilepsy Rat from [64] Ståhl PL, Salmén F, Vickovic S, Lundmark A, Navarro JF, Magnusson J, et al. Visual-
Strasbourg and its related non-epileptic strain. PLoS One 2017;12(7):e0179924. ization and analysis of gene expression in tissue sections by spatial transcriptomics.
https://doi.org/10.1371/journal.pone.0179924. Science 2016;80(353):78–82. https://doi.org/10.1126/science.aaf2403.
[37] Epi4K Consortium, Epilepsy phenome/genome project. De novo mutations in epi- [65] Nygaard V, Hovig E. Options available for profiling small samples: a review of sam-
leptic encephalopathies. Nature 2013; 12;501(7466):217–21 doi: https://doi.org/ ple amplification technology when combined with microarray profiling. Nucleic
10.1038/nature12439. Acids Res 2006;34(3):996–1014. https://doi.org/10.1093/nar/gkj499.
[38] Traynelis J, Silk M, Wang Q, Berkovic SF, Liu L, Ascher DB, et al. Optimizing genomic [66] Sharma A. Genome-wide expression analysis in epilepsy: a synthetic review. Curr
medicine in epilepsy through a gene-customized approach to missense variant in- Top Med Chem 2012;12:1008–32. https://doi.org/10.2174/156802612800229189.
terpretation. Genome Res 2017;27(10):1715–29. https://doi.org/10.1101/g. [67] Mirza N, Vasieva O, Marson AG, Pirmohamed M. Exploring the genomic basis of
226589.117. pharmacoresistance in epilepsy: an integrative analysis of large-scale gene expres-
[39] Allen AS, Berkovic SF, Cossette P, Delanty N, Dlugos D, Eichler EE, et al. De novo mu- sion profiling studies on brain tissue from epilepsy surgery. Hum Mol Genet 2011;
tations in epileptic encephalopathies. Nature 2013;501:217–21. https://doi.org/10. 20:4381–94. https://doi.org/10.1093/hmg/ddr365.
1038/nature12439. [68] Dingledine R, Coulter DA, Fritsch B, Gorter JA, Lelutiu N, McNamara J, et al. Tran-
[40] Wingett SW, Andrews S. FastQ Screen: a tool for multi-genome mapping and qual- scriptional profile of hippocampal dentate granule cells in four rat epilepsy models.
ity control. F1000Res. 2018;7:1338. doi: 10.12688/f1000research.15931.2. Sci Data 2017;4:170061. https://doi.org/10.1038/sdata.2017.61.
eCollection. [69] Floriano-Sánchez E, Brindis F, Ortega-Cuellar D, Ignacio-Mejía I, Moreno-Arriola E,
[41] Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: Romero-Morelos P, et al. Differential gene expression profile induced by valproic
a new genome assembly algorithm and its applications to single-cell sequencing. J acid (VPA) in pediatric epileptic patients. Genes (Basel) 2018;9:328. https://doi.
Comput Biol 2012;19:455–77. https://doi.org/10.1089/cmb.2012.0021. org/10.3390/genes9070328.
[42] Alkan C, Coe BP, Eichler EE. Genome structural variation discovery and genotyping. [70] Han C, Zhao X, Liu Y, Wang K, Chen N, Hu W, et al. Gene expression profiling of two
Nat Rev Genet 2011;12:363–76. https://doi.org/10.1038/nrg2958. epilepsy models reveals the ECM/integrin signaling pathway is involved in
[43] Darling AE, Mau B, Blattner FR, Perna NT. Mauve: multiple alignment of conserved epiletogenesis. Neuroscience 2019;396:187–99. https://doi.org/10.1016/j.neuro-
genomic sequence with rearrangements. Genome Res 2004;14(7):1394–403. science.2018.10.021.
https://doi.org/10.1101/g.2289704. [71] He K, Xiao W, Lv W. Comprehensive identification of essential pathways and tran-
[44] Layer RM, Chiang C, Quinlan AR, Hall IM. LUMPY: a probabilistic framework for scription factors related to epilepsy by gene set enrichment analysis on microarray
structural variant discovery. Genome Biol 2014;15:R84. https://doi.org/10.1089/ datasets. Int J Mol Med 2014;34:715–24. https://doi.org/10.3892/ijmm.2014.1843.
cmb.2012.0021. [72] Lukasiuk K, Pitkänen A. Large-scale analysis of gene expression in epilepsy re-
[45] Musunuru K, Pirruccello JP, Do R, Peloso GM, Guiducci C, Sougnez C, et al. Exome search: is synthesis already possible? Neurochem Res 2004;29:1169–78. https://
sequencing, ANGPTL3 mutations, and familial combined hypolipidemia. N Engl J doi.org/10.1023/B:NERE.0000023604.91584.6c.
Med 2010;363:2220–7. https://doi.org/10.1056/NEJMoa1002926. [73] Lukasiuk K, Dabrowski M, Adach A, Pitkänen A. Epileptogenesis-related genes
[46] Foo J, Liu J, Tan E. Whole-genome and whole-exome sequencing in neurological revisited. Prog Brain Res 2006;158:223–41. https://doi.org/10.1016/S0079-6123
diseases. Nat Rev Neurol 2012;8:508–17. https://doi.org/10.1038/nrneurol.2012. (06)58011-2.
148. [74] Pitkänen A, Lukasiuk K. Mechanisms of epileptogenesis and potential treatment
[47] Handel AE, Disanto G, Ramagopalan SV. Next-generation sequencing in under- targets. Lancet Neurol 2011;10:173–86. https://doi.org/10.1016/S1474-4422(10)
standing complex neurological disease. Expert Rev Neurother 2013;13:215–27. 70310-0.
https://doi.org/10.1586/ern.12.165. [75] Rogic S, Pavlidis P. Meta-analysis of kindling-induced gene expression changes in
[48] Michaud JL, Lachance M, Hamdan FF, Carmant L, Lortie A, Diadori P, et al. The ge- the rat hippocampus. Front Neurosci 2009;3:1–10. https://doi.org/10.3389/neuro.
netic landscape of infantile spasms. Hum Mol Genet 2014;23:4846–58. https:// 15.001.2009.
doi.org/10.1093/hmg/ddu199. [76] Salman MM, Sheilabi MA, Bhattacharyya D, Kitchen P, Conner AC, Bill RM, et al.
[49] Kobow K, Blümcke I. The methylation hypothesis: do epigenetic chromatin modi- Transcriptome analysis suggests a role for the differential expression of cerebral
fications play a role in epileptogenesis? Epilepsia 2011;52(Suppl. 4):15–9. aquaporins and the MAPK signalling pathway in human temporal lobe epilepsy.
https://doi.org/10.1111/j.1528-1167.2011.03145.x. Eur J Neurosci 2017;46:2121–32. https://doi.org/10.1111/ejn.13652.
[50] Kobow K, Daspi A, Harikrishnan KN, Kiese K, Ziemann M, Khurana I, et al. Deep se- [77] Theilhaber J, Rakhade SN, Sudhalter J, Kothari N, Klein P, Pollard J, et al. Gene ex-
quencing reveals increased DNA methylation in chronic rat epilepsy. Acta pression profiling of a hypoxic seizure model of epilepsy suggests a role for
Neuropathol 2013;126(5):741–56. https://doi.org/10.1007/s00401-013-1168-8. mTOR and Wnt signaling in epileptogenesis. PLoS One 2013;8:e74428. https://
[51] Buiting K. Prader–Willi syndrome and Angelman syndrome. Am J Med Genet Part C doi.org/10.1371/journal.pone.0074428.
Semin Med Genet 2010;154C:365–76. https://doi.org/10.1002/ajmg.c.30273. [78] Venugopal AK, Kumar GSS, Mahadevan A, Selvan N. Transcriptomic profiling of me-
[52] Weksberg R. Imprinted genes and human disease. Am J Med Genet Part C Semin dial temporal lobe epilepsy. J Proteomics Bioinform 2012;05:1–18. https://doi.org/
Med Genet 2010;154C:317–20. https://doi.org/10.1002/ajmg.c.30268. 10.4172/jpb.1000210.
[53] Subhash S, Kanduri M. Comprehensive DNA methylation analysis using a methyl- [79] von Deimling M, Häsler R, Steinbach V, Holterhus P, von Spiczak S, Stephani U, et al.
CpG-binding domain capture-based method in chronic lymphocytic leukemia pa- Gene expression analysis in untreated absence epilepsy demonstrates an inconsis-
tients. J Vis Exp 2017;124:e55773. https://doi.org/10.3791/55773. tent pattern. Epilepsy Res 2017;132:84–90. https://doi.org/10.1016/j.eplepsyres.
[54] Debski DJ, Pitkanen A, Puhakka N, Bot AM, Khurana I, Harikrishnan KN, et al. Etilogy 2017.02.008.
matters—genomic DNA methylation patterns in three rat models of acquired epi- [80] Araújo MA, Marques TE, Octacílio-Silva S, Arroxelas-Silva CL, Pereira MG, Peixoto-
lepsy. Sci Rep 2016;9(6):25668. https://doi.org/10.1038/srep25668. Santos JE, et al. Identification of microRNAs with dysregulated expression in status
[55] Kiese K, Jablonski J, Hackenbracht J, Wrosch JK, Groemer TW, Kornhuber J, et al. epilepticus induced epileptogenesis. PLoS One 2016;11(10):e0163855. https://doi.
Epigenetic control of epilepsy target genes contributes to a celular memory of org/10.1371/journal.pone.0163855.
epileptogenesis in cultured rat hippocampal neurons. Acta Neuropathol Commun [81] Henshall DC, Hamer HM, Pasterkamp RJ, Goldstein DB, Kjems J, Prehn JHM, et al.
2017;31(5(1)):79. https://doi.org/10.1186/s40478-017-0485-x. MicroRNAs in epilepsy: pathophysiology and clinical utility. Lancet Neurol 2016;
[56] Crosio C, Heitz E, Allis CD, Borrelli E, Sassone-Corsi P. Chromatin remodeling and 15:1368–76. https://doi.org/10.1016/S1474-4422(16)30246-0.
neuronal response: multiple signaling pathways induce specific histone H3 modi- [82] Kaalund SS, Venø MT, Bak M, Møller RS, Laursen H, Madsen F, et al. Aberrant ex-
fications and early gene expression in hippocampal neurons. J Cell Sci 2003;15(116 pression of miR-218 and miR-204 in human mesial temporal lobe epilepsy and
(Pt 24)):4905–14. https://doi.org/10.1242/jcs.00804. hippocampal sclerosis—convergence on axonal guidance. Epilepsia 2014;55:
[57] Huang Y, Doherty JJ, Dingledine R. Alterd histone acetylation at glutamate receptor 2017–27. https://doi.org/10.1111/epi.12839.
2 and brain-derived neurotrophic factor genes is an early event triggered by status [83] Meng F, You Y, Liu Z, Liu J, Ding H, Xu R. Neuronal calcium signaling pathways are
epilepticus. J Neurosci 2002;22(19):8422–8. https://doi.org/10.1523/JNeurosci.22- associated with the development of epilepsy. Mol Med Rep 2015;11:196–202.
19-08422.2002. https://doi.org/10.3892/mmr.2014.2756.
[58] Tsankova NM, Kumar A, Nestler EJ. Histone modifications at gene promoeter re- [84] Srivastava PK, Roncon P, Lukasiuk K, Gorter JA, Aronica E, Pitkänen A, et al. Meta-
gions in rat hippocampus after acute and chronic electroconvulsive seizures. J analysis of microRNAs dysregulated in the hippocampal dentate gyrus of animal
Neurosci 2004;16(24(24)):5603–10. https://doi.org/10.1523/JNEUROSCI.0589-04. models of epilepsy. Eneuro 2017;4. https://doi.org/10.1523/ENEURO.0152-17.
2004. 2017 ENEURO.0152-17.2017.
[59] Hoffmann K, Czapp M, Löscher W. Increase in antiepileptic efficacy during [85] Merelli A, Rodríguez JCG, Folch J, Regueiro MR, Camins A, Lazarowski A. Under-
prolonged treatment with valproic acid: role of inhibition of histone deacetylases? standing the role of hypoxia inducible factor during neurodegeneration for new
therapeutics opportunities. Curr Neuropharmacol 2018;16(10):1484–98. https:// peripheral lipopolysaccharide treatment. Brain Behav Immun 2014;35:86–95.
doi.org/10.2174/1570159X16666180110130253. https://doi.org/10.1016/j.bbi.2013.09.001.
[86] Rhodes DR, Barrette TR, Rubin MA, Ghosh D, Chinnaiyan AM. Meta-analysis of mi- [114] Fiordelisio T, Jiménez N, Baba S, Shiba K, Hernández-Cruz A. Immunoreactivity to
croarrays: interstudy validation of gene expression profiles reveals pathway dys- neurofilaments in the rodent anterior pituitary is associated with the expression
regulation in prostate cancer 1. Cancer Res 2002;62:4427–33. of α1A protein subunits of voltage-gated Ca 2+ channels. J Neuroendocrinol
[87] Dong X, You Y, Wu JQ. Building an RNA sequencing transcriptome of the central 2007;19:870–81. https://doi.org/10.1111/j.1365-2826.2007.01596.x.
nervous system. Neurosci 2016;22:579–92. https://doi.org/10.1177/ [115] Romá-Mateo C, Aguado C, García-Giménez JL, Ibáñez-Cabellos JS, Seco-Cervera M,
1073858415610541. Pallardó FV, et al. Increased oxidative stress and impaired antioxidant response in
[88] Hrdlickova R, Toloue M, Tian B. RNA-Seq methods for transcriptome analysis. Wiley Lafora disease. Mol Neurobiol 2015;51:932–46. https://doi.org/10.1007/s12035-
Interdiscip Rev RNA 2017;8:e1364. https://doi.org/10.1002/wrna.1364.ç. 014-8747-0.
[89] Bosco DB, Zheng J, Xu Z, Peng J, Eyo UB, Tang K, et al. RNAseq analysis of hippocam- [116] Mitulović G. New HPLC techniques for proteomics analysis: a short overview of lat-
pal microglia after kainic acid-induced seizures. Mol Brain 2018;11:34. https://doi. est developments. J Liq Chromatography Related Technol 2015;38(3):390–403.
org/10.1186/s13041-018-0376-5. https://doi.org/10.1080/10826076.2014.941266.
[90] Brueggeman L, Sturgeon ML, Martin RM, Grossbach AJ, Nagahama Y, Zhang A, et al. [117] Magdeldin S, Moresco JJ, Yamamoto T, Yates JR. Off-line multidimensional liquid
Drug repositioning in epilepsy reveals novel antiseizure candidates. Ann Clin Transl chromatography and auto sampling result in sample loss in LC/LC–MS/MS. J Prote-
Neurol 2018. https://doi.org/10.1002/acn3.703 acn3.703. ome Res 2014;13(8):3826–36. https://doi.org/10.1021/pr500530e.
[91] Dixit AB, Banerjee J, Srivastava A, Tripathi M, Sarkar C, Kakkar A, et al. RNA-seq [118] Thomas PD. PANTHER: a library of protein families and subfamilies indexed by
analysis of hippocampal tissues reveals novel candidate genes for drug refractory function. Genome Res 2003;13:2129–41. https://doi.org/10.1101/g.772403.
epilepsy in patients with MTLE-HS. Genomics 2016;107(5):178–88. https://doi. [119] Jara A, Kopchick JJ. Proteómica: una aproximación integral. An Pediatría 2013;78:
org/10.1016/j.ygeno.2016.04.001. 137–9. https://doi.org/10.1016/j.anpedi.2012.10.007.
[92] Griffin NG, Wang Y, Hulette CM, Halvorsen M, Cronin KD, Walley NM, et al. Differ- [120] Zhang P, Zhang L, Li Y, Zhu S, Zhao M, Ding S, et al. Quantitative proteomic analysis
ential gene expression in dentate granule cells in mesial temporal lobe epilepsy to identify differentially expressed proteins in myocardium of epilepsy using iTRAQ
with and without hippocampal sclerosis. Epilepsia 2016;57:376–85. https://doi. coupled with nano-LC–MS/MS. J Proteome Res 2018;17(1):305–14. https://doi.
org/10.1111/epi.13305. org/10.1021/acs.jproteome.7b00579.
[93] Hansen KF, Sakamoto K, Pelz C, Impey S, Obrietan K. Profiling status epilepticus-in- [121] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from poly-
duced changes in hippocampal RNA expression using high-throughput RNA se- acrylamide gels to nitrocellulose sheets: procedure and some applications. Proc
quencing. Sci Rep 2014;4:6930. https://doi.org/10.1038/srep06930. Natl Acad Sci 1979;76:4350–4. https://doi.org/10.1073/pnas.76.9.4350.
[94] Hawkins NA, Kearney JA. Confirmation of an epilepsy modifier locus on mouse [122] Burnette WN. “Western Blotting”: electrophoretic transfer of proteins from sodium
chromosome 11 and candidate gene analysis by RNA-Seq. Genes, Brain Behav dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic
2012;11:452–60. https://doi.org/10.1111/j.1601-183X.2012.00790.x. detection with antibody and radioiodinated protein A. Anal Biochem 1981;112:
[95] Miller AR, Hawkins NA, Mccollom CE, Kearney JA. Mapping genetic modifiers of 195–203. https://doi.org/10.1016/0003-2697(81)90281-5.
survival in a mouse model of Dravet syndrome. Genes, Brain Behav 2014;13(2): [123] Mousseau DD, Raghavendra Rao VL. Western blot analysis. In: Boulton AA, Baker
163–72. https://doi.org/10.1111/gbb.12099.Mapping. GB, Bateson AN, editors. In vitro neurochemical techniques. Neuromethods, vol
[96] Jehi L, Yehia L, Peterson C, Niazi F, Busch R, Prayson R, et al. Preliminary report: late 34. Humana Press; 1999.
seizure recurrence years after epilepsy surgery may be associated with alterations [124] Maity B, Sheff D, Fisher RA. Immunostaining. Methods Cell Biol., vol. 113, Elsevier;
in brain tissue transcriptome. Epilepsia Open 2018;3:299–304. https://doi.org/10. 2013, p. 81–105. doi:https://doi.org/10.1016/B978-0-12-407,239-8.00005-7.
1002/epi4.12119. [125] Park J-W, Park E-S, Choi EN, Park H, Jung S. Altered brain gene expression profiles
[97] Shalek AK, Satija R, Shuga J, Trombetta JJ, Gennert D, Lu D, et al. Single-cell RNA-seq associated with the pathogenesis of phenylketonuria in a mouse model. Clin Chim
reveals dynamic paracrine control of cellular variation. Nature 2014;510:363–9. Acta 2009;401:90–9. https://doi.org/10.1016/j.cca.2008.11.019.
https://doi.org/10.1038/nature13437. [126] Gasperini RJ, Hou X, Parkington H, Coleman H, Klaver DW, Vincent AJ, et al. TRPM8
[98] Skene NG, Grant SGN. Identification of vulnerable cell types in major brain dis- and Nav1.8 sodium channels are required for transthyretin-induced calcium influx
orders using single cell transcriptomes and expression weighted cell type en- in growth cones of small-diameter TrkA-positive sensory neurons. Mol
richment. Front Neurosci 2016;10:1–11. https://doi.org/10.3389/fnins.2016. Neurodegener 2011;6:19–30. https://doi.org/10.1186/1750-1326-6-19.
00016. [127] Zhou L, Tang X, Li X, Bai Y, Buxbaum JN, Chen G. Identification of transthyretin as a
[99] Lee JH, Daugharthy ER, Scheiman J, Kalhor R, Yang JL, Ferrante TC, et al. Highly novel interacting partner for the δ subunit of GABAA receptors. PLoS ONE 2019;14
multiplexed subcellular RNA sequencing in situ. Science 2014;80(343):1360–3. (1):e0210094. https://doi.org/10.1371/journal.pone.0210094.
https://doi.org/10.1126/science.1250212. [128] Melton L. Proteomics in multiplex. Nature 2004;429:105–7. https://doi.org/10.
[100] Canales RD, Luo Y, Willey JC, Austermiller B, Barbacioru CC, Boysen C, et al. Evalu- 1038/429101a.
ation of DNA microarray results with quantitative gene expression platforms. Nat [129] Sutandy FR, Qian J, Chen C-S, Zhu H. Overview of protein microarrays. Curr Protoc
Biotechnol 2006;24:1115–22. https://doi.org/10.1038/nbt1236. Protein Sci 2013;72:27.1.1–27.1.16. https://doi.org/10.1002/0471140864.ps270 1
[101] Devonshire AS, Sanders R, Wilkes TM, Taylor MS, Foy CA, Huggett JF. Application of s72.
next generation qPCR and sequencing platforms to mRNA biomarker analysis. [130] Talapatra A, Rouse R, Hardiman G. Protein microarrays: challenges and promises.
Methods 2013;59:89–100. https://doi.org/10.1016/j.ymeth.2012.07.021. Pharmacogenomics 2002;3:527–36. https://doi.org/10.1517/14622416.3.4.527.
[102] Provenzano M, Mocellin S. Complementary techniques. Microarray Technol. Cancer [131] Vilela LR, Lima IV, Kunsch ÉB, Pinto HPP, de Miranda AS, Vieira ÉLM, et al. Anticon-
Gene Profiling. New York, NY: Springer New York; 2007. p. 66–73. https://doi.org/ vulsant effect of cannabidiol in the pentylenetetrazole model: pharmacological
10.1007/978-0-387-39978-2_7. mechanisms, electroencephalographic profile, and brain cytokine levels. Epilepsy
[103] VanGuilder HD, Vrana KE, Freeman WM. Twenty-five years of quantitative PCR for Behav 2017;75:29–35. https://doi.org/10.1016/j.yebeh.2017.07.014.
gene expression analysis. Biotechniques 2008;44:619–26. https://doi.org/10.2144/ [132] Landrum MJ, Lee JM, Benson M, Brown G, Chao C, Chitipiralla S, et al. ClinVar: pub-
000112776. lic archive of interpretations of clinically relevant variants. Nucleic Acids Res 2016;
[104] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time 44(D1):D862–8. https://doi.org/10.1093/nar/gkv1222.
quantitative PCR and the 2(−Delta Delta C(T)) Method. Methods 2001;25(4):402–8. [133] Stenson PD, Ball EV, Mort M, Phillips AD, Shiel JA, Thomas NS, et al. Human Gene
[105] Crans RAJ, Janssens J, Daelemans S, Wouters E, Raedt R, Van Dam D, et al. The val- Mutation Database (HGMD): 2003 update. Hum Mutat 2003 Jun;21(6):577–81.
idation of short interspersed nuclear elements (SINEs) as a RT-qPCR normalization https://doi.org/10.1002/humu.10212.
strategy in a rodent model for temporal lobe epilepsy. PLoS One 2019;14: [134] Hauser RM, Henshall DC, Lubin FD. The Epigenetics of Epilepsy and Its Progression.
e0210567. https://doi.org/10.1371/journal.pone.0210567. Neuroscientist 2018;24(2):186–200. https://doi.org/10.1177/1073858417705840.
[106] Morey JS, Ryan JC, Dolah FM. Microarray validation: factors influencing correlation [135] van Dijk EL, Auger H, Jaszczyszyn Y, Thermes C. Ten years of next-generation se-
between oligonucleotide microarrays and real-time PCR. Biol Proced Online 2006; quencing technology. Trends Genet 2014;30:418–26. https://doi.org/10.1016/j.tig.
8:175–93. https://doi.org/10.1251/bpo126. 2014.07.001.
[107] Crosetto N, Bienko M, van Oudenaarden A. Spatially resolved transcriptomics and [136] Pernot F, Heinrich C, Barbier L, Peinnequin A, Carpentier P, Dhote F, et al. Inflamma-
beyond. Nat Rev Genet 2015;16:57–66. https://doi.org/10.1038/nrg3832. tory changes during epileptogenesis and spontaneous seizures in a mouse model of
[108] Lein E, Borm LE, Linnarsson S. The promise of spatial transcriptomics for neurosci- mesiotemporal lobe epilepsy. Epilepsia 2011;52(12):2315–25. https://doi.org/10.
ence in the era of molecular cell typing. Science 2017;80(358):64–9. https://doi. 1111/j.1528-1167.2011.03273.x.
org/10.1126/science.aan6827. [137] Danis O, Demir S, Gunel A, Aker RG, Gulcebi M, Onat F, et al. Changes in intracellular
[109] Moor AE, Itzkovitz S. Spatial transcriptomics: paving the way for tissue-level sys- protein expression in cortex, thalamus and hippocampus in a genetic rat model of
tems biology. Curr Opin Biotechnol 2017;46:126–33. https://doi.org/10.1016/j. absence epilepsy. Brain Res Bull 2011;84(6):381–8. https://doi.org/10.1016/j.
copbio.2017.02.004. brainresbull.2011.02.002.
[110] Strell C, Hilscher MM, Laxman N, Svedlund J, Wu C, Yokota C, et al. Placing RNA in [138] Huang ZL, Zhou Y, Xiao B, Wu J, Wu XM, Yang P. Wu LY. Proteomic screening of
context and space — methods for spatially resolved transcriptomics. FEBS J 2019; postsynaptic density proteins related with temporal lobe epilepsy. Zhonghua Yi
286:1468–81. https://doi.org/10.1111/febs.14435. Xue Za Zhi 2008;88(45):3205–9. https://doi.org/10.3321/j.issn:0376-2491.2008.
[111] Lovatt D, Ruble BK, Lee J, Dueck H, Kim TK, Fisher S, et al. Transcriptome in vivo 45.010.
analysis (TIVA) of spatially defined single cells in live tissue. Nat Methods 2014; [139] Jiang W, Du B, Chi Z, Ma L, Wang S, Zhang X, et al. Preliminary explorations of the
11:190–6. https://doi.org/10.1038/nmeth.2804. role of mitochondrial proteins in refractory epilepsy: some findings from compar-
[112] Creative Proteomics, Recovered May 15, 2019, from https://www.creative- ative proteomics. J Neurosci Res 2007;85(14):3160–70. https://doi.org/10.1002/
proteomics.com/technology/maldi-tof-mass-spectrometry.htm jnr.21384.
[113] Györffy B, Kovács Z, Gulyássy P, Simor A, Völgyi K, Orbán G, et al. Brain protein ex- [140] Greene ND, Bamidele A, Choy M, de Castro SC, Wait R, Leung KY, et al. Proteome
pression changes in WAG/Rij rats, a genetic rat model of absence epilepsy after changes associated with hippocampal MRI abnormalities in the lithium
pilocarpine-induced model of convulsive status epilepticus. Proteomics 2007;7(8): Development. J Proteome Res 2016;15(5):1546–62. https://doi.org/10.1021/acs.
1336–44. https://doi.org/10.1002/pmic.200601027. jproteome.6b00003.
[141] Junker H, Spate K, Suofu Y, Walther R, Schwarz G, Kammer W, et al. Proteomic [144] Liu X, Yang J, Chen L, Zhang Y, Yang M. Comparative proteomics and correlated sig-
identification of the involvement of the mitochondrial rieske protein in epi- naling network of rat hippocampus in the pilocarpine model of temporal lobe ep-
lepsy. Epilepsia 2005;46(3):339–43. https://doi.org/10.1111/j.0013-9580. ilepsy 2008:582–603. doi:https://doi.org/10.1002/pmic.200700514
2005.46904.x. [145] Li A, Choi YS, Dziema H, Cao R, Cho HY, Jung YJ, et al. Own-regulated of the epileptic
[142] Walker A, Russmann V, Deeg CA, von Toerne C, Kleinwort KJH, Szober C, et al. Pro- dentate gyrus. Brain Pathol 2010;20(6):1077–89. https://doi.org/10.1111/j.1750-
teomic profiling of epileptogenesis in a rat model: Focus on inflammation. Brain 3639.2010.00414.x.
Behav Immun 2016;53:138–58. https://doi.org/10.1016/j.bbi.2015.12.007. [146] Wu L, Peng J, Wei C, Liu G. Characterization , using comparative proteomics , of dif-
[143] Bitsika V, Duveau V, Simon-Areces J, Mullen W, Roucard C, Makridaki M, et al. ferentially expressed proteins in the hippocampus of the mesial temporal lobe of
High-Throughput LC-MS/MS Proteomic Analysis of a Mouse Model of epileptic rats following treatment with valproate 2011:221–38. doi:https://doi.
Mesiotemporal Lobe Epilepsy Predicts Microglial Activation Underlying Disease org/10.1007/s00726-010-0638-8.

Molecular Tools For The Characterization of Seizure Susceptibility in Genetic Rodent Models of Epilepsy

Uploaded by

Copyright:

Available Formats

Molecular Tools For The Characterization of Seizure Susceptibility in Genetic Rodent Models of Epilepsy

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Molecular Tools For The Characterization of Seizure Susceptibility in Genetic Rodent Models of Epilepsy

Uploaded by

Copyright:

Available Formats

Epilepsy & Behavior 121 (2021) 106594

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

Molecular tools for the characterization of seizure susceptibility in

This article is part of the Special Issue “NEWroscience 2018"

2.4. DNA modiﬁcations 3.1. Microarrays

Penot et Kainate injection in MTLE Hippocampus IL-1beta; IL-1Ra; Cytokines, inﬂammatory

Gene/Protein interactions Functional analysis.

Pathological phenotypes pathways

You might also like