Circulating cell-free DNA methylation mirrors alterations in cerebral patterns in epilepsy

Circulating cell-free DNA methylation mirrors alterations in cerebral patterns in epilepsy | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Circulating cell-free DNA methylation mirrors alterations in cerebral patterns in epilepsy Ricardo Martins-Ferreira, Bárbara Leal, João Chaves, Laura Ciudad, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1940501/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 28 Dec, 2022 Read the published version in Clinical Epigenetics → Version 1 posted 10 You are reading this latest preprint version Abstract Background: DNA methylation profiling of circulating cell-free DNA (cfDNA) has rapidly become a promising strategy for biomarker identification and development. The cell-type-specific nature of DNA methylation patterns and the direct relationship between cfDNA and apoptosis can potentially be used non-invasively to predict local alterations. In addition, direct detection of altered DNA methylation patterns performs well as a biomarker. In a previous study, we demonstrated marked DNA methylation alterations in brain tissue from patients with mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS) patients. Experimental Design: We performed DNA methylation profiling in cfDNA isolated from serum of MTLE patients and healthy controls using beadchip arrays followed by systematic bioinformatic analysis including deconvolution analysis and integration with DNase accessibility datasets. Results: Differential cfDNA methylation analysis showed overrepresentation of gene ontology terms and transcription factors related to central nervous system function and regulation. Deconvolution analysis of the DNA methylation datasets ruled out the possibility that the observed differences were due to changes in the proportional contribution of cortical neurons in cfDNA. Moreover, we found no overrepresentation of neuron- or glia-specific patterns in the described cfDNA methylation patterns. However, the MTLE-HS cfDNA methylation patterns featured significant overrepresentation of the epileptic DNA methylation alterations previously observed in hippocampus. Conclusions: Our results support the use of cfDNA methylation profiling as a rational approach to seeking non-invasive and reproducible epilepsy biomarkers. Cell-free DNA DNA methylation epilepsy biomarker Figures Figure 1 Figure 2 Figure 3 Background Cell-free DNA (cfDNA) consists of small DNA fragments released into the peripheral blood, predominantly as a result of apoptosis ( 1 ). This is substantiated by the consistent correspondence between the length of human circulating cfDNA (167 bp) and the length of DNA wrapped around a nucleosome (~ 147 bp) plus linker regions, which suggest the action of endonucleases. Apoptotic DNA degradation is mediated by caspase-activated DNase (CAD), which lacks exonuclease activity, and so can only fragment DNA in inter-nucleosomal regions ( 1 , 2 ). The evaluation of DNA methylation of cfDNA has been used to estimate tissue or cell of origin, and to non-invasively track ongoing cell death occurring anywhere in the body, based on the cell-specific nature of DNA methylation ( 3 ). The use of this approach is spreading in cancer studies ( 4 – 8 ). Direct detection of altered DNA methylation patterns under pathological conditions, regardless of cell or tissue contribution, has been thoroughly examined to determine its value as a strategy for searching for biomarkers ( 9 – 14 ). The identification and development of epilepsy and epileptogenesis biomarkers is of inherent interest ( 15 ), but progress has been slower than in other settings, including other neurodegenerative pathologies like Alzheimer’s disease (AD). This can be attributed to the low accessibility to pathological tissue and the complexity and variability within the spectrum of epilepsy syndromes. Nevertheless, nucleic acid-based biomarkers, predominantly microRNAs, are promising ( 16 ). Here, we obtained the DNA methylation profiles of serum cfDNA samples from mesial temporal lobe epilepsy (MTLE) patients and compared them with those obtained from healthy controls. MTLE is commonly associated with severe neuronal cell death, termed hippocampal sclerosis (HS) ( 17 ). We hypothesized that the analysis of the cfDNA methylome in MTLE-HS patients could serve as a predictive, diagnostic or prognostic tool of neuronal cell death estimation. Furthermore, direct comparison of the DNA methylation profile between patients and healthy controls could potentially fill the gap in peripheral biomarker development in epilepsy. Our results were not able to identify a significant increase in the proportion of cfDNA derived from brain. However, the cfDNA methylomes reflect a significant enrichment of epileptic patterns overlapping with those described in the hippocampus of MTLE patients. Methods Study population The MTLE patients included in this study were followed at the Reference Epilepsy Research Centre of Hospital de Santo António – Centro Hospitalar e Universitário do Porto (HSA-CHUP) (Table 1 and Supplementary Table 1). The diagnosis was based on clinical and electrophysiological data (electroencephalogram (EEG) and/or video-EEG monitoring) and brain MRI (minimum 1.5T), as defined by Wieser ( 18 ). A definition of HS by brain MRI required the detection of atrophy, T2 hyperintensity signal and altered internal structure on one or both hippocampi, associated or not with other imaging criteria such as ipsilateral fornix atrophy, ipsilateral mamillary body atrophy or ipsilateral entorhinal abnormalities. Visual and/or verbal memory impairment were not considered exclusion criteria. However, patients with other neurological abnormalities were not included. At the time of the study, all patients were receiving pharmacological treatment (monotherapy or polytherapy). The control population comprised healthy individuals who were ethnically matched and from the same geographical area, and who had been voluntarily recruited from blood donors. Individuals with any neurological condition or a positive family history were excluded. Table 1 Clinicodemographic characterization of the studied populations. Controls MTLE p n 11 12 - % Female (n) 81.8 (9) 75.0 (9) 1.000 Age, years (mean ± SD) 38.9 ± 8.4 44.8 ± 11.4 0.2063 Age of onset, years (mean ± SD) - 14.0 ± 13.6 - Epilepsy duration (mean ± SD) - 30.8 ± 14.5 - % of pharmacorresistant (n) - 83.3 (10) - % of FS history - 75.0 (9) - MTLE, Mesial Temporal Lobe Epilepsy FS, Febrile seizures SD, Standard deviation Ethics statement This work was approved by the ethical committee of HSA-CHUP (2018.051(047-DEFI/047-CES)). All individuals gave their written consent in accordance with principles of the Declaration of Helsinki. Serum collection and DNA extraction Peripheral blood was collected in Vacuette ® tubes without anticoagulant and centrifuged at 490 g for 20 minutes. Collected serum aliquots were stored at -20°C. Only samples processed within 4 h of collection were included. DNA was extracted from approximately 1 mL of serum using the QIAmp® MinElute® ccfDNA Mini Kit (Qiagen), following the manufacturer’s instructions. DNA methylation profiling Extracted genomic DNA was quantified using a Qubit DNA Assay Kit (Cat. No. 10146592) in a Qubit 2.0 Fluorometer (Life Technologies, CA, USA). CfDNA samples from twelve MTLE patients (8F, 4M; 44.8 ± 11.4 years old) and eleven controls (9F, 2M; 38.9 ± 8.4 years old) were profiled. All DNA extracted from each sample (44.72–238.95 ng) was bisulphite-converted using the EZ DNA Methylation-Gold™ Kit (Zymo Research, Irvine, CA, USA), following the manufacturer’s instructions. Converted DNA was hybridized in Infinium MethylationEPIC BeadChip arrays (Illumina), following the manufacturer’s instructions. The arrays encompass > 850,000 single-nucleotide methylation sites and cover 99% of the annotated reference sequence (RefSeq) genes. Fluorescence intensities were imaged using a BeadArray Reader (Illumina), and images were processed and intensities measured as previously described ( 19 ). A combination of the Cy3 and Cy5 fluorescence intensities of the methylated and unmethylated alleles was used to obtain each methylation data point. Background intensity was computed from a set of negative controls and subtracted from each data point. Beta values were used to illustrate methylation. Values can range between zero (0% methylation) and one (100% methylation) and represent the ratio of the methylated probe intensity to the overall intensity (sum of the methylated and unmethylated probe intensities). M values were calculated as the log 2 ratio of the intensities of the methylated and unmethylated probes. M values were used for statistical purposes, because beta values are heteroskedastic for highly methylated and unmethylated CpGs ( 20 ). Methylation data were analysed in the R statistical environment. The shinyÉpico web interface ( 21 ), based on minfi ( 22 ) and limma ( 23 ) pipelines, was used for array processing, normalization and differential methylation calculation. Deconvolution of cell/tissue of origin The cell or tissue of origin was estimated with the meth_atlas deconvolution algorithm ( https://github.com/nloyfer/meth_atlas ), which estimates the proportion of origin for a total of 25 tissue and cell types based on 450k and EPIC data ( 8 ). DNA methylation profiles from serum cfDNA of epileptic patients and controls were processed in accordance with the original study. Normalization was performed with the preprocessIllumina function. Probes with a detection significance of p > 0.01 were excluded, as were those mapping to sex chromosomes. Deconvolution was performed using the Python-based method in relation to the supplied reference atlas matrix composed of 7890 sites. Calculation of cfDNA differentially methylated regions (DMRs) Methylation data were normalized with the Noob + Quantile functions to assess the differentials. Probes with a detection p < 0.01 were filtered out, as were positions located in the X and Y chromosomes and/or overlapping with SNPs. CpHs were retained, based on evidence that neurons, unlike other CNS cells, present CpH as their dominant DNA methylation mark ( 24 ). A total of 783,351 individual positions were obtained after normalization and filtering. DMRs were calculated using the mCSEA (methylated CpGs Set Enrichment Analysis) package ( 25 ). The eBayes results of limma , sorted by t -statistics, were used as input. The limma eBayes-moderated t- test was carried out using M values and included no covariates. DMRs with at least five CpGs and an FDR < 0.05 were considered statistically significant. Gene ontology (GO) and transcription factor (TF) enrichment analysis Gene ontology (GO) evaluation was performed using the GREAT online tool ( http://great.stanford.edu/public/html ) ( 26 ), with the two nearest genes settings. Motif enrichment was analysed using the findMotifsGenome.pt tool of the HOMER motif discovery application ( 27 ), considering a window of ± 50 bp from each DMR. The total annotated DMRs from the mCSEA analysis were used as a background in both analyses. Human Protein Atlas (HPA) gene expression and expression cluster data We used public data from the HPA to evaluate potential cerebral regional activity of enriched TF associated with cfDNA DMRs. To evaluate the expression levels of the genes associated with those factors, we used the RNA consensus tissue gene data , which consists of transcript expression levels summarized per gene in 55 tissues obtained consensually from RNA-seq data from the HPA and Genotype-Tissue Expression (GTEx) projects. We also accessed data from the HPA consisting of the levels of confidence of protein-coding genes across tissue clusters and across single-cell clusters. For the tissue clusters, the HPA used RNA expression data from 53 tissues to classify genes into 87 expression clusters. A total of 144 cell types or cell lines were used to construct 68 single-cell expression clusters. For both approaches, Louvain clustering was performed based on gene-to-gene distances calculated from the Spearman correlation of gene expression across multiple samples. Clustering was performed 100 times to accommodate stochasticity. The confidence of the gene-to-cluster value, which varies between 0 and 1, corresponds to the proportion of times that a gene was assigned to a cluster. Clusters were identified manually through recourse to functional annotation tools. Overlap of cfDNA DMRs with epileptic, neuron-specific and glia-specific DMRs We extracted the lists of DMRs significantly altered in hippocampal and neocortical tissue in MTLE patients relative to autopsied non-epileptic controls, obtained in a previous study by our group ( 28 ). We extracted raw IDAT files from 33 NeuN + fractions of prefrontal cortex of healthy individuals (GSE112179), 23 NeuN- fractions (glia) (GSE166207) and 37 blood-cell-type samples, which included B cells, CD8 + T cells, CD4 + T cells, monocytes, neutrophils and NK cells (GSE110555). Data were processed and DMRs calculated as described above, all NeuN or glia samples being compared with all blood samples, without discrimination for blood-cell types, and without considering any covariates. Since the mCSEA package was used to calculate DMRs across all settings, we overlapped the lists of DMRs based on their annotated identification. The GeneOverlap function, which is based on Fisher’s exact test, was used to calculate the significance of the overlap. The total number of DMRs considered by mCSEA was used as background (26,208 for promoter DMRs; 23,772 for gene DMRs; 27,187 for CGI DMRs). DNase-seq data analysis DNase hypersensitivity bigwig files with the human GRCh37 assembly from brain tissue and blood cells were obtained from ENCODE (Supplementary Table 4). For each setting, the mean function of Wiggletools was used to aggregate the multiple files ( 29 ). The outputted wig file was reconverted to bigwig format using ucsc-wigtobigwig ( 30 ). Heatmaps, DMR visualization and plots All heatmaps were developed using the R gplots and ComplexHeatmap ( 31 ) packages. Row dendrogram clustering was carried out with complete-linkage hierarchical clustering. The overlaps between lists of DMRs was represented using the UpSet function of ComplexHeatmap . To visualize individual DMRs, along with genomic location and epigenetic and chromatin accessibility marks, we used the functions available in the gviz package ( 32 ). All additional plots were generated with the ggplot2 package ( 33 ). Statistical analysis All statistical analyses were done using R v4.0.2. or IBM SPSS Statistics version 27 (Armonk, NY, USA). All graphs were created in R. Group medians were compared using the Mann-Whitney test. Fisher’s exact test was used to calculate the significance of non-random association between two categorical variables (e.g., sex distribution in patients and controls). The levels of significance were: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Role of the funding source The funding bodies played no role in the study design, data collection, analysis and interpretation, or writing of the manuscript. Results Estimation of cell of origin proportions in serum of MTLE and controls based on cfDNA methylation First, we generated the DNA methylation profiles of serum cfDNA samples of MTLE and healthy controls using beadchip arrays. To estimate the cell of origin, we used the deconvolution algorithm designed for this purpose by Moss et al. ( 8 ), in relation to a 7890-CpG reference matrix, accounting for 25 tissue and cell types (Fig. 1 A, Supplementary Fig. 1 and Supplementary Table 2). In agreement with other studies, including the original source of the algorithm, hematopoietic cells were the main contributors. We also noted contributions from non-blood cell or tissue types, including bladder, breast, vascular endothelial cells and cortical neurons. Significant differences in the cell-of-origin proportion between controls and MTLE patients were only observed for vascular endothelial cells ( p = 0.02, Wilcoxon test) (Supplementary Fig. 1B). No significant differences in the cortical neuronal origin were observed between patients and controls ( p = 0.54, Wilcoxon) (Fig. 1 B). We noted a striking predominance of neutrophil contribution in both groups, with a mean percentage contribution of approximately 92% across all samples (Fig. 1 A and Supplementary Fig. 1); this is a higher value than that reported by Moss and colleagues. This might be due to the use of serum samples, instead of plasma samples, as they used. We can speculate that the predominance of neutrophil contribution is associated with increased coagulation-related NETosis during sample collection. Differentially methylated regions (DMRs) in cfDNA We then determined DMRs, using the mCSEA algorithm, between MTLE and healthy control cfDNA. Three sets of DMRs, located in promoters, gene bodies and CpG islands (CGIs), were calculated. We identified 873 significant promoter-DMRs (744 hypomethylated and 129 hypermethylated in MTLE relative to controls) (Fig. 2 A and Supplementary Table 3). Gene ontology (GO) analysis of the hypermethylated and hypomethylated DMR clusters demonstrated enrichment of CNS-related terms, including some associated with GABAergic pathways, synaptic transmission, microglia activation and neurotrophin receptor binding. A thorough inspection of the DMRs associated with these GO categories included biologically relevant promoters such as those associated with the GABRG3 and CDH9 genes (hypermethylated), and those at the GABRA1, GABRA2, GABRG2 and BDNF genes (hypomethylated) (Fig. 2 B). GABAergic receptors represent key constituents of the CNS. γ-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the cerebral cortex and disruption in the excitatory/inhibitory balance has long been associated with seizure development ( 34 ). On this basis, genetic variability related to GABAergic subunits has shown potentially causal epileptogenic effects ( 35 ). BDNF encodes the brain-derived neurotrophic factor, one of the most prominent members of the neurotrophin family, which has a wide range of functions, encompassing regulation of neuronal development and synaptic plasticity ( 36 ). The regulation of the BDNF gene in neurons has long been associated with DNA methylation-related mechanisms ( 37 ). It has been described as being overexpressed in epilepsy ( 38 – 42 ). Moreover, the DNA methylation status of its promoter regions has been explored ( 43 – 45 ), revealing a marked tendency towards demethylation. We also described 327 gene-body DMRs (65 hypomethylated and 262 hypermethylated in MTLE compared with healthy controls) (Fig. 2 C and Supplementary Table 3). Multiple CNS-related GO categories were also found to be associated with obtained DMRs. These included terms associated with synaptic assembly and organization (hypermethylated) and febrile seizures (hypomethylated) (Fig. 2 D). We found 550 significant CGI-DMRs (334 hypomethylated and 216 hypermethylated) (Fig. 2 E and Supplementary Table 3). GO analysis was consistent with that described above, with enrichment of multiple mechanisms of potential relevance in neuropathology (Fig. 2 F). We also examined the TF binding motif enrichment of the generated DMRs (Fig. 2 G). To analyse potential CNS-specific activity of the enriched TFs, we used data from the HPA repository. Many of the enriched TFs were consistently expressed across different brain regions (Supplementary Fig. 2A). Moreover, we searched for associations of the genes coding the enriched TFs with HPA tissue clusters and HPA single-cell clusters. The most prominent results were the high confidence levels of ZNF528 and RARB in Cluster 73 – Brain: Transcription regulation (RNA HPA tissue expression cluster) (Supplementary Fig. 2B) and Cluster 8 – Neurons & Oligodendrocytes: Synaptic function (RNA HPA single-cell expression cluster) (Supplementary Fig. 2C), respectively. This suggests possible brain specificity in the establishment of these DNA methylation patterns in cfDNA of MTLE patients. cfDNA DMRs in MTLE show enrichment of brain epileptic DMRs but not of neuron- or glia-specific methylation patterns Previously, our group described significantly altered DNA methylation patterns in the hippocampus and neocortex of MTLE-HS patients compared with autopsied controls without neuropathology ( 28 ). These include 2650 DMRs in the hippocampus (736 promoter, 1111 gene and 803 CGI) (Supplementary Fig. 3A) and 2950 DMRs in the adjacent neocortex (785 promoter, 1264 gene and 901 CGI) (Supplementary Fig. 3B). The overlap of cfDNA DMRs with hippocampal DMRs in MTLE patients was significant (Fig. 3 A and 3 B). The coincidence of the direction of methylation change (hypermethylated or hypomethylated in both cfDNA and hippocampus) reinforced the relevance of the overlap of DMRs. There was a marked enrichment between hypermethylated DMRs, but we found no significant overlap between hypomethylated DMRs (Fig. 3 C and 3 D). We considered whether the dimensionality reduction achieved by genome-wide deconvolution algorithms was too broad and overlooked specific DNA methylation changes that could indicate differences in the cell-specific contribution. Cell of origin may be represented by a small subset of target regions. Therefore, we also attempted to determine whether the DMRs obtained in cfDNA of MTLE patients are enriched in neuron- or glia-specific DNA methylation patterns. We used public EPIC data consisting of neuron and glia samples and blood cells (monocytes, neutrophils, B cells, CD4 + T cells, CD8 + T cells, NK cells). We identified 5943 neuron-specific DMRs (1504 promoter, 2174 gene, 2265 CGI) (Supplementary Fig. 3C) and 3523 glia-specific DMRs (1038 promoter, 1444 gene, 1041 CGI) (Supplementary Fig. 3D). We did not find any significant overlap between cfDNA DMRs from the MTLE/CTR comparison or neuron-specific or glia-specific DMRs with concordant behaviour. In fact, greater enrichment was observed for DMRs with opposite behaviour (e.g., hypermethylated promoter cfDNA in MTLE cfDNA and hypomethylated promoter neuron-specific DMRs) (Supplementary Fig. 3E-H). These results therefore imply that the differences in DNA methylation observed in cfDNA of patients compared with controls do not appear to be a consequence of increased circulating neuronal or glial DNA. DNA accessibility may provide clues about the origin of cfDNA DMRs At this point, we have demonstrated that DNA methylation alterations in the MTLE brain are, to some extent, replicable in circulating cfDNA. However, we are yet to determine which cells are the source of the fragments carrying such differentially methylated patterns. We next investigated the potential origin of the fragments carrying the identified cfDNA DMRs by using public chromatin accessibility data, namely DNase-seq. It can be assumed that euchromatic regions (open chromatin) would be highly degraded and the resulting fragments would be too small to be detected in circulating cfDNA ( 46 ). We were able to individualize DMRs with higher chromatin accessibility in blood cells than in brain tissue (Fig. 3 E-F). We found lower DNA methylation levels in the promoter of the PM20D1 gene in both cfDNA and hippocampus of MTLE patients. This hypomethylation pattern is neither neuron- nor glia-specific. DNA accessibility, however, is high in blood cells but low in brain tissue (Fig. 3 E). We identified two more DMRs, localized in CpG islands located at the PLEC and AUTS2 genes, that show hypermethylation in cfDNA and hippocampus of patients compared with controls. Although the DMR does not show overall significant hypermethylation in neurons or glia, individual CpG probes that overlap with the blood euchromatin region have higher DNA methylation levels in neurons and glia than in blood cells. Both these regions also present higher accessibility in blood cells than in brain tissue (Fig. 3 F-G). Discussion Only two previous studies have analysed cfDNA in the context of epilepsy, and they only measured total cfDNA concentration levels ( 47 , 48 ). MTLE is the most incident-focal epilepsy in adults, which, together with its high pharmacoresistance rates, makes it one of the most widely studied epilepsy syndromes. DNA methylation has been thoroughly explored within epileptic brain tissue. In a recent study by our group ( 28 ), we described major DNA methylation alterations in the hippocampus, which is the focus of the epilepsy and the region of the lesion ( 17 ), and in the adjacent neocortical areas of MTLE patients, comparing these with the state in non-epileptic controls. Our present study demonstrates the presence of DNA methylation alterations in cfDNA of MTLE patients. There was a level of representation of CNS-related genes within those alterations and, in fact, we observed a significant overlap of epileptic patterns observed in the pathological tissue, mainly hippocampus biopsies. The functional enrichment analysis of the distinct cfDNA methylation patterns observed in MTLE suggests its possible origin by highlighting multiple genes, pathways and regulatory transcription factors associated with the CNS paradigm. We found that the epileptic patterns described in hippocampus and neocortex tissue were also significantly enriched in circulating cfDNA. However, a clearer overlap was demonstrated in hippocampus, which is further evidence of the primary pathological nature of this region. This concurs with the findings of previous studies that cfDNA from serum and plasma can contain tissue-pathological DNA methylation alterations ( 49 – 51 ). One aspect that remains unresolved is the source of the cfDNA fragments bearing those epileptic patterns. Studies have shown that the methylation status of DNA from blood cells also differs between controls and epilepsy patients ( 52 , 53 ). Furthermore, the DNA methylation profiles of peripheral tissues, including blood, positively correlate with brain tissue in epilepsy patients ( 54 ). It has been proposed that chromatin accessibility states may help to infer the origin of cfDNA ( 46 ). Highly accessible euchromatic regions are likely to be highly degraded, and the resulting fragment would be too small to be detected in circulation. In the three examples of DMR-containing sequences that we have presented, chromatin accessibility is high in blood cells whereas it shows a more compact state in brain tissue. We may speculate that the cfDNA fragments carrying those epileptic patterns, coincident in cfDNA and hippocampus, are more likely to originate in brain tissue rather than in blood cells. PM20D1 codes an N-fatty acyl amino acid (NAAs) synthase/hydrolase, and is located within a Parkinson’s susceptibility locus ( 55 ). It has been demonstrated that PM20D1 is both an expression and methylation quantitative trait locus in AD, with direct influence on molecular and behaviour pathological features ( 56 ). In AD blood samples, a U-shaped model has been proposed in which DNA methylation of this region is decreased in early stages and reverses with progression towards late AD ( 57 ). ( 57 ). Plectin ( PLEC ) is a plakin responsible for linking elements of the cytoskeleton. In the CNS, PLEC has been shown to be predominantly expressed in pia/glia and endothelia/glia junctions ( 58 ), where it is paramount for the structural and functional integrity of the BBB and the pial surface ( 59 ). In TLE, plectin is upregulated in astrocytes located at the sclerotic hippocampus ( 60 ). AUTS2 is a well-known risk gene for autism spectrum (ASD) but also for other neurodevelopmental disorders, including epilepsy ( 61 ). The autism susceptibility candidate 2 (AUTS2) gene links with PRC1 (Polycomb Repressive Complex 1), a known epigenetic regulator, and together are responsible for transcription activation of genes associated with neurodevelopment ( 62 ). Moreover, cytoplasmatic AUTS2 contributes to the reorganization of the cytoskeleton, since it is related to neuronal motility and morphogenesis ( 61 ). The potential use of brain-cell-derived cfDNA in neurodegenerative diseases has only recently begun to be explored. The most prominent results have originated from targeted studies of a small group of specific DNA methylation positions that show an exclusive pattern in brain cells. Using such an approach, increased brain cell-derived cfDNA was demonstrated in multiple sclerosis (MS) patients, cardiac arrest patients with brain damage and traumatic brain injury (TBI) patients ( 63 ), cancer patients with brain metastasis ( 64 ) and schizophrenic patients with a first psychotic episode ( 65 ). More complex mathematical deconvolution algorithms have been developed to estimate the simultaneous percentages of contributions of multiple cells and/or tissue types. Moss et al. described a DNA methylation-based deconvolution algorithm, based on the non-negative least-squares linear regression of a reference matrix encompassing 25 tissue and cell types, including cortical neurons ( 8 ). This tool was subsequently used to predict the proportion of neuronal contribution to cfDNA in pituitary neuroendocrine tumours ( 50 ) and glioma ( 51 ). The direct relation between neuronal death (HS) and cfDNA corroborates the suitability of cfDNA methylation-based deconvolution of brain origin as a logical stream towards biomarker identification in MTLE. Monitoring brain-derived cfDNA could contribute to the early detection of HS and help control the progression of neuroanatomical damage over the course of the disease ( 66 ). Nevertheless, we did not observe any enhancement of cortical neuron-derived cfDNA in our patients. One must consider the potential lack of precision of the current deconvolution tools in estimating the contribution of brain cells. Moss et al., for instance, used three cortical neuron samples to develop their reference matrix. It is of inherent interest to develop more precise algorithms which would account for, as far as possible, the whole complexity of the CNS spectrum by including multiple cell types (e.g., excitatory and inhibitory neurons, oligodendrocytes, OPCs, astrocytes, microglia) and also take regional variability into account. Additionally, the eventual release of brain-derived cfDNA in MTLE may be an acute event. In fact, Chatterton et al. reported an increase in the release of neuronal and glial cfDNA in plasma of entry personnel (breachers) during explosive training, on the day that participants were exposed to higher pressures, after which it promptly decreased ( 46 ). In MTLE, such events could occur immediately following the seizure. However, the interval between the time of the last seizure and serum collection was taken into account in the present study. The introduction of methods based on artificial intelligence and machine learning in biological research has significantly increased the assessment of biomarker performance. The development of predictive models has made it possible to potentiate the use of large omics datasets for biomarker identification. Such tools have been used to access diagnostic performance of cfDNA methylation modifications in neurodegenerative settings ( 50 , 51 , 67 ). We consider the use of predictive models to be an important next step in cfDNA methylation analysis. However, larger populations are required to develop robust and replicable models. Our study shows that the analysis of cfDNA methylation in epilepsy has predictive potential. To follow this up, complementary studies are needed that exploit this emerging field to its full potential. We consider cfDNA methylation to be a promising tool with which to pursue the ultimate goal of reducing patients’ burden through early diagnosis, proper monitoring of the progressive nature of the disease, and a better understanding of the causes of epileptic refractoriness. Abbreviations AD, Alzheimer’s disease; BDNF, brain-derived neurotrophic factor; CAD, caspase-activated DNase; cfDNA, cell-free DNA; CGIs, CpG islands; CTR, controls; DMR, differentially methylated region; EEG, electroencephalogram; GABA, γ-aminobutyric acid; GO, gene ontology; GTEx, genotype-tissue expression; HPA, Human Protein Atlas; HS, hippocampal sclerosis; mCSEA, methylated CpG set enrichment analysis; MS, multiple sclerosis; MTLE, mesial temporal lobe epilepsy; NET, neutrophil extracellular traps; OPCs, oligodendrocyte precursor cells; TBI, traumatic brain injury; TF, transcription factor. Declarations Acknowledgements We thank the CERCA Programme/Generalitat de Catalunya, the Josep Carreras Foundation and ICBAS-UP for institutional support. We acknowledge Professor Berta Martins da Silva and the Immunogenetics Laboratory of the Molecular Pathology and Immunology of the ICBAS-UP. We thank the other members of the Epigenetics and Immune Diseases group of the IJC for support and advice throughout the development of study, and Dr. João Lopes and Dr. João Ramalheira of the Neurophysiology Service of HSA-CHUP for insightful clinical guidance and continuous collaboration. We thank the patients and their families for their essential contributions. Funding EB is funded by the Spanish Ministry of Science and Innovation (MICINN) [PID2020117212RB-I00; AEI/10.13039/501100011033]. This work was partially supported by a BICE Tecnifar Grant. RM-F is funded by an FCT ( Fundação para a Ciência e Tecnologia )fellowship (SFRH/BD/137900/2018). UMIB is funded by FCT Portugal (UIDB/00215/2020 and UIDP/00215/2020), and ITR (LA/P/006/2020). Data availability EPIC DNA methylation array data have been deposited in the NCBI’s Gene Expression Omnibus database under accession code GSE208758. All additional data used in this study is available in this article or in the supplementary information. Ethics approval and consent to participate This work was approved by the ethical committee of HSA-CHUP (2018.051(047-DEFI/047-CES)). All participants provided written informed consent. Consent for publication All participants provided written informed consent for publication. Declarations of interest None of the authors report any conflict of interest. Contributors Study conception and design: RM-F, EB, BL. Recruitment of patients and collection of serum samples: JC, RS, BL, RM-F. Laboratory work: RM-F and LC. Bioinformatic analysis: RM-F. Verification and interpretation of the analysis: EB, BL, PPC, RM-F. Original drafting of the article: RM-F, EB, BL, PPC. All the remaining authors reviewed, edited and approved the final version of the manuscript. References Heitzer E, Auinger L, Speicher MR. Cell-Free DNA and Apoptosis: How Dead Cells Inform About the Living. Trends Mol Med. 2020 May;26(5):519–28. Snyder MW, Kircher M, Hill AJ, Daza RM, Shendure J. Cell-free DNA Comprises an In Vivo Nucleosome Footprint that Informs Its Tissues-Of-Origin. Cell. 2016 Jan;164(1–2):57–68. Dor Y, Cedar H. Principles of DNA methylation and their implications for biology and medicine. Lancet (London, England). 2018 Sep;392(10149):777–86. Guo S, Diep D, Plongthongkum N, Fung H-L, Zhang K, Zhang K. Identification of methylation haplotype blocks aids in deconvolution of heterogeneous tissue samples and tumor tissue-of-origin mapping from plasma DNA. Nat Genet. 2017 Apr;49(4):635–42. Li W, Li Q, Kang S, Same M, Zhou Y, Sun C, et al. CancerDetector: ultrasensitive and non-invasive cancer detection at the resolution of individual reads using cell-free DNA methylation sequencing data. Nucleic Acids Res. 2018 Sep;46(15):e89. Sun K, Jiang P, Chan KCA, Wong J, Cheng YKY, Liang RHS, et al. Plasma DNA tissue mapping by genome-wide methylation sequencing for noninvasive prenatal, cancer, and transplantation assessments. Proc Natl Acad Sci U S A. 2015 Oct;112(40):E5503-12. Kang S, Li Q, Chen Q, Zhou Y, Park S, Lee G, et al. CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA. Genome Biol. 2017 Mar;18(1):53. Moss J, Magenheim J, Neiman D, Zemmour H, Loyfer N, Korach A, et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free DNA in health and disease. Nat Commun. 2018 Nov;9(1):5068. Xu R-H, Wei W, Krawczyk M, Wang W, Luo H, Flagg K, et al. Circulating tumour DNA methylation markers for diagnosis and prognosis of hepatocellular carcinoma. Nat Mater. 2017 Nov;16(11):1155–61. Legendre C, Gooden GC, Johnson K, Martinez RA, Liang WS, Salhia B. Whole-genome bisulfite sequencing of cell-free DNA identifies signature associated with metastatic breast cancer. Clin Epigenetics. 2015;7(1):100. Gallardo-Gómez M, Moran S, Páez de la Cadena M, Martínez-Zorzano VS, Rodríguez-Berrocal FJ, Rodríguez-Girondo M, et al. A new approach to epigenome-wide discovery of non-invasive methylation biomarkers for colorectal cancer screening in circulating cell-free DNA using pooled samples. Clin Epigenetics. 2018;10:53. Gordevičius J, Kriščiūnas A, Groot DE, Yip SM, Susic M, Kwan A, et al. Cell-Free DNA Modification Dynamics in Abiraterone Acetate-Treated Prostate Cancer Patients. Clin cancer Res an Off J Am Assoc Cancer Res. 2018 Jul;24(14):3317–24. Lu H, Liu Y, Wang J, Fu S, Wang L, Huang C, et al. Detection of ovarian cancer using plasma cell-free DNA methylomes. Clin Epigenetics. 2022 Jun;14(1):74. Shen SY, Burgener JM, Bratman S V, De Carvalho DD. Preparation of cfMeDIP-seq libraries for methylome profiling of plasma cell-free DNA. Nat Protoc. 2019 Oct;14(10):2749–80. Engel JJ, Pitkänen A. Biomarkers for epileptogenesis and its treatment. Neuropharmacology. 2020 May;167:107735. Henshall DC, Hamer HM, Pasterkamp RJ, Goldstein DB, Kjems J, Prehn JHM, et al. MicroRNAs in epilepsy: pathophysiology and clinical utility. Lancet Neurol. 2016 Dec;15(13):1368–76. Blümcke I, Thom M, Aronica E, Armstrong DD, Bartolomei F, Bernasconi A, et al. International consensus classification of hippocampal sclerosis in temporal lobe epilepsy: a Task Force report from the ILAE Commission on Diagnostic Methods. Epilepsia. 2013 Jul;54(7):1315–29. Wieser H-G. ILAE Commission Report. Mesial temporal lobe epilepsy with hippocampal sclerosis. Vol. 45, Epilepsia. United States; 2004. p. 695–714. Garcia-Gomez A, Li T, Kerick M, Català-Moll F, Comet NR, Rodríguez-Ubreva J, et al. TET2- and TDG-mediated changes are required for the acquisition of distinct histone modifications in divergent terminal differentiation of myeloid cells. Nucleic Acids Res. 2017 Sep;45(17):10002–17. Du P, Zhang X, Huang C-C, Jafari N, Kibbe WA, Hou L, et al. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics. 2010 Nov;11:587. Morante-Palacios O, Ballestar E. shinyÉPICo: A graphical pipeline to analyze Illumina DNA methylation arrays. Bioinformatics. 2021 Jan; Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics. 2014 May;30(10):1363–9. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015 Apr;43(7):e47. Lister R, Mukamel EA, Nery JR, Urich M, Puddifoot CA, Johnson ND, et al. Global epigenomic reconfiguration during mammalian brain development. Science. 2013 Aug;341(6146):1237905. Martorell-Marugán J, González-Rumayor V, Carmona-Sáez P. mCSEA: detecting subtle differentially methylated regions. Bioinformatics. 2019 Sep;35(18):3257–62. McLean CY, Bristor D, Hiller M, Clarke SL, Schaar BT, Lowe CB, et al. GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol. 2010 May;28(5):495–501. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010 May;38(4):576–89. Martins-Ferreira R, Leal B, Chaves J, Li T, Ciudad L, Rangel R, et al. Epilepsy progression is associated with cumulative DNA methylation changes in inflammatory genes. Prog Neurobiol. 2021 Dec;209:102207. Zerbino DR, Johnson N, Juettemann T, Wilder SP, Flicek P. WiggleTools: parallel processing of large collections of genome-wide datasets for visualization and statistical analysis. Bioinformatics. 2014 Apr;30(7):1008–9. Kent WJ, Zweig AS, Barber G, Hinrichs AS, Karolchik D. BigWig and BigBed: enabling browsing of large distributed datasets. Bioinformatics. 2010 Sep;26(17):2204–7. Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016 Sep;32(18):2847–9. Hahne F, Ivanek R. Visualizing Genomic Data Using Gviz and Bioconductor. Methods Mol Biol. 2016;1418:335–51. Wickham H. ggplot2: Elegant Graphics for Data Analysis [Internet]. Springer-Verlag New York; 2016. Available from: https://ggplot2.tidyverse.org Treiman DM. GABAergic mechanisms in epilepsy. Epilepsia. 2001;42 Suppl 3:8–12. Maljevic S, Møller RS, Reid CA, Pérez-Palma E, Lal D, May P, et al. Spectrum of GABAA receptor variants in epilepsy. Curr Opin Neurol. 2019 Apr;32(2):183–90. Wang CS, Kavalali ET, Monteggia LM. BDNF signaling in context: From synaptic regulation to psychiatric disorders. Cell. 2022 Jan;185(1):62–76. Chen WG, Chang Q, Lin Y, Meissner A, West AE, Griffith EC, et al. Derepression of BDNF transcription involves calcium-dependent phosphorylation of MeCP2. Science. 2003 Oct;302(5646):885–9. Dugich-Djordjevic MM, Tocco G, Willoughby DA, Najm I, Pasinetti G, Thompson RF, et al. BDNF mRNA expression in the developing rat brain following kainic acid-induced seizure activity. Neuron. 1992 Jun;8(6):1127–38. Poulsen FR, Lauterborn J, Zimmer J, Gall CM. Differential expression of brain-derived neurotrophic factor transcripts after pilocarpine-induced seizure-like activity is related to mode of Ca2 + entry. Neuroscience. 2004;126(3):665–76. Zhang HN, Ko MC. Seizure activity involved in the up-regulation of BDNF mRNA expression by activation of central mu opioid receptors. Neuroscience. 2009 Jun;161(1):301–10. Murray KD, Isackson PJ, Eskin TA, King MA, Montesinos SP, Abraham LA, et al. Altered mRNA expression for brain-derived neurotrophic factor and type II calcium/calmodulin-dependent protein kinase in the hippocampus of patients with intractable temporal lobe epilepsy. J Comp Neurol. 2000 Mar;418(4):411–22. Takahashi M, Hayashi S, Kakita A, Wakabayashi K, Fukuda M, Kameyama S, et al. Patients with temporal lobe epilepsy show an increase in brain-derived neurotrophic factor protein and its correlation with neuropeptide Y. Brain Res. 1999 Feb;818(2):579–82. Ryley Parrish R, Albertson AJ, Buckingham SC, Hablitz JJ, Mascia KL, Davis Haselden W, et al. Status epilepticus triggers early and late alterations in brain-derived neurotrophic factor and NMDA glutamate receptor Grin2b DNA methylation levels in the hippocampus. Neuroscience. 2013 Sep;248:602–19. Martínez-Levy GA, Rocha L, Lubin FD, Alonso-Vanegas MA, Nani A, Buentello-García RM, et al. Increased expression of BDNF transcript with exon VI in hippocampi of patients with pharmaco-resistant temporal lobe epilepsy. Neuroscience. 2016 Feb;314:12–21. Bandeira IC, Giombelli L, Werlang IC, Abujamra AL, Secchi TL, Brondani R, et al. Methylation of BDNF and SLC6A4 Gene Promoters in Brazilian Patients With Temporal Lobe Epilepsy Presenting or Not Psychiatric Comorbidities. Front Integr Neurosci. 2021;15:764742. Chatterton Z, Mendelev N, Chen S, Carr W, Kamimori GH, Ge Y, et al. Bisulfite Amplicon Sequencing Can Detect Glia and Neuron Cell-Free DNA in Blood Plasma. Front Mol Neurosci. 2021;14:672614. Alapirtti T, Jylhävä J, Raitanen J, Mäkinen R, Peltola J, Hurme MA, et al. The concentration of cell-free DNA in video-EEG patients is dependent on the epilepsy syndrome and duration of epilepsy. Neurol Res. 2016 Jan;38(1):45–50. Liimatainen SP, Jylhävä J, Raitanen J, Peltola JT, Hurme MA. The concentration of cell-free DNA in focal epilepsy. Epilepsy Res. 2013 Aug;105(3):292–8. Shen SY, Singhania R, Fehringer G, Chakravarthy A, Roehrl MHA, Chadwick D, et al. Sensitive tumour detection and classification using plasma cell-free DNA methylomes. Nature. 2018 Nov;563(7732):579–83. Herrgott GA, Asmaro KP, Wells M, Sabedot TS, Malta TM, Mosella MS, et al. Detection of Tumor-specific DNA Methylation Markers in the Blood of Patients with Pituitary Neuroendocrine Tumors. Neuro Oncol. 2022 Feb; Sabedot TS, Malta TM, Snyder J, Nelson K, Wells M, deCarvalho AC, et al. A serum-based DNA methylation assay provides accurate detection of glioma. Neuro Oncol. 2021 Sep;23(9):1494–508. Long H-Y, Feng L, Kang J, Luo Z-H, Xiao W-B, Long L-L, et al. Blood DNA methylation pattern is altered in mesial temporal lobe epilepsy. Sci Rep. 2017 Mar;7:43810. Tao H, Chen Z, Wu J, Chen J, Chen Y, Fu J, et al. DNA Methylation Signature of Epileptic Encephalopathy-Related Pathogenic Genes Encoding Ion Channels in Temporal Lobe Epilepsy. Front Neurol. 2021;12:692412. Braun PR, Han S, Hing B, Nagahama Y, Gaul LN, Heinzman JT, et al. Genome-wide DNA methylation comparison between live human brain and peripheral tissues within individuals. Transl Psychiatry. 2019 Jan;9(1):47. Simón-Sánchez J, Schulte C, Bras JM, Sharma M, Gibbs JR, Berg D, et al. Genome-wide association study reveals genetic risk underlying Parkinson’s disease. Nat Genet. 2009 Dec;41(12):1308–12. Sanchez-Mut J V, Heyn H, Silva BA, Dixsaut L, Garcia-Esparcia P, Vidal E, et al. PM20D1 is a quantitative trait locus associated with Alzheimer’s disease. Nat Med. 2018 May;24(5):598–603. Wang Q, Chen Y, Readhead B, Chen K, Su Y, Reiman EM, et al. Longitudinal data in peripheral blood confirm that PM20D1 is a quantitative trait locus (QTL) for Alzheimer’s disease and implicate its dynamic role in disease progression. Clin Epigenetics. 2020 Dec;12(1):189. Lie AA, Schröder R, Blümcke I, Magin TM, Wiestler OD, Elger CE. Plectin in the human central nervous system: predominant expression at pia/glia and endothelia/glia interfaces. Acta Neuropathol. 1998 Sep;96(3):215–21. Potokar M, Jorgačevski J. Plectin in the Central Nervous System and a Putative Role in Brain Astrocytes. Cells. 2021 Sep;10(9). Lee T-S, Mane S, Eid T, Zhao H, Lin A, Guan Z, et al. Gene expression in temporal lobe epilepsy is consistent with increased release of glutamate by astrocytes. Mol Med. 2007;13(1–2):1–13. Hori K, Shimaoka K, Hoshino M. AUTS2 Gene: Keys to Understanding the Pathogenesis of Neurodevelopmental Disorders. Cells. 2021 Dec;11(1). Gao Z, Lee P, Stafford JM, von Schimmelmann M, Schaefer A, Reinberg D. An AUTS2-Polycomb complex activates gene expression in the CNS. Nature. 2014 Dec;516(7531):349–54. Lehmann-Werman R, Neiman D, Zemmour H, Moss J, Magenheim J, Vaknin-Dembinsky A, et al. Identification of tissue-specific cell death using methylation patterns of circulating DNA. Proc Natl Acad Sci U S A. 2016 Mar;113(13):E1826-34. Lubotzky A, Zemmour H, Neiman D, Gotkine M, Loyfer N, Piyanzin S, et al. Liquid biopsy reveals collateral tissue damage in cancer. JCI insight. 2022 Jan;7(2). Lubotzky A, Pelov I, Teplitz R, Neiman D, Smadja A, Zemmour H, et al. Elevated brain-derived cell-free DNA among patients with first psychotic episode - a proof-of-concept study. Elife. 2022 Jun;11. Caciagli L, Bernasconi A, Wiebe S, Koepp MJ, Bernasconi N, Bernhardt BC. A meta-analysis on progressive atrophy in intractable temporal lobe epilepsy: Time is brain? Neurology. 2017 Aug;89(5):506–16. Bahado-Singh RO, Radhakrishna U, Gordevičius J, Aydas B, Yilmaz A, Jafar F, et al. Artificial Intelligence and Circulating Cell-Free DNA Methylation Profiling: Mechanism and Detection of Alzheimer’s Disease. Cells. 2022 May;11(11). Additional Declarations No competing interests reported. Supplementary Files SupplementaryFigure1.pdf SupplementaryFigure2.pdf SupplementaryFigure3.pdf SupplementaryFigureLegends.docx SupplementaryTable1.xlsx SupplementaryTable2.xlsx SupplementaryTable3.xlsx SupplementaryTable4.xlsx Cite Share Download PDF Status: Published Journal Publication published 28 Dec, 2022 Read the published version in Clinical Epigenetics → Version 1 posted Editorial decision: Major revision 17 Nov, 2022 Reviews received at journal 17 Nov, 2022 Reviewers agreed at journal 04 Nov, 2022 Reviews received at journal 25 Oct, 2022 Reviewers agreed at journal 03 Oct, 2022 Reviewers agreed at journal 16 Sep, 2022 Reviewers invited by journal 16 Sep, 2022 Editor assigned by journal 08 Aug, 2022 Submission checks completed at journal 08 Aug, 2022 First submitted to journal 08 Aug, 2022 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-1940501","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":127274125,"identity":"2db23401-f73d-4732-9ed6-04f5f89a41c8","order_by":0,"name":"Ricardo Martins-Ferreira","email":"","orcid":"","institution":"Josep Carreras Research Institute (IJC)","correspondingAuthor":false,"prefix":"","firstName":"Ricardo","middleName":"","lastName":"Martins-Ferreira","suffix":""},{"id":127274128,"identity":"a224e679-ca80-414c-8bbd-0f60067dea11","order_by":1,"name":"Bárbara Leal","email":"","orcid":"","institution":"Universidade do Porto (ICBAS-UPorto)","correspondingAuthor":false,"prefix":"","firstName":"Bárbara","middleName":"","lastName":"Leal","suffix":""},{"id":127274130,"identity":"9b26b0ef-bc86-4822-8664-58e64f6c39af","order_by":2,"name":"João Chaves","email":"","orcid":"","institution":"Universidade do Porto (ICBAS-UPorto)","correspondingAuthor":false,"prefix":"","firstName":"João","middleName":"","lastName":"Chaves","suffix":""},{"id":127274133,"identity":"8e334c4d-f928-4c60-a589-dab9c42adfc5","order_by":3,"name":"Laura Ciudad","email":"","orcid":"","institution":"Josep Carreras Research Institute (IJC)","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"","lastName":"Ciudad","suffix":""},{"id":127274138,"identity":"e8b5ddb3-aee5-4ad5-9cdc-b1adfaa3c32c","order_by":4,"name":"Raquel Samões","email":"","orcid":"","institution":"Hospital de Santo António - Centro Hospitalar Universitário do Porto (HSA-CHUP)","correspondingAuthor":false,"prefix":"","firstName":"Raquel","middleName":"","lastName":"Samões","suffix":""},{"id":127274139,"identity":"c4f2aae0-c6f4-4769-a1cd-0ef077be5cea","order_by":5,"name":"António Martins da Silva","email":"","orcid":"","institution":"Universidade do Porto (ICBAS-UPorto)","correspondingAuthor":false,"prefix":"","firstName":"António","middleName":"Martins da","lastName":"Silva","suffix":""},{"id":127274141,"identity":"6f782523-4393-487c-b534-0751c4d3ad7b","order_by":6,"name":"Paulo Pinho Costa","email":"","orcid":"","institution":"Universidade do Porto (ICBAS-UPorto)","correspondingAuthor":false,"prefix":"","firstName":"Paulo","middleName":"Pinho","lastName":"Costa","suffix":""},{"id":127274146,"identity":"8304fc2c-3157-4c5c-b356-002271c2659f","order_by":7,"name":"Esteban Ballestar","email":"data:image/png;base64,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","orcid":"","institution":"Josep Carreras Research Institute (IJC)","correspondingAuthor":true,"prefix":"","firstName":"Esteban","middleName":"","lastName":"Ballestar","suffix":""}],"badges":[],"createdAt":"2022-08-08 09:14:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1940501/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1940501/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13148-022-01416-2","type":"published","date":"2022-12-28T18:07:27+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":25047088,"identity":"a5aa0bc7-4351-4c59-95b4-4792eefb7d04","added_by":"auto","created_at":"2022-08-10 15:48:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":563207,"visible":true,"origin":"","legend":"\u003cp\u003eEstimated percentage contribution of each cell and tissue type from the meth_atlas deconvolution tool for the 11 control (CTR) and 12 MTLE serum samples analysed. \u003cstrong\u003eA.\u003c/strong\u003e Pie chart of the mean estimated percentage contribution for each group. Percentage of cfDNA originating in cortical neurons is highlighted in bold and with an asterisk. \u003cstrong\u003eB. \u003c/strong\u003eBoxplot representation of the individual estimated proportion of cfDNA originated by cortical neurons within each study group (\u003cem\u003ep\u003c/em\u003e = 0.54, Wilcoxon test).\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/8f6779469c31d1a22da530ce.png"},{"id":25047093,"identity":"bdcf7295-a14f-4041-9e9d-8aa4ef0263d6","added_by":"auto","created_at":"2022-08-10 15:48:05","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1256081,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmap representation of DNA methylation of promoter \u003cstrong\u003e(A)\u003c/strong\u003e, gene \u003cstrong\u003e(C)\u003c/strong\u003e, and CGI-DMRs \u003cstrong\u003e(E)\u003c/strong\u003e in 12 MTLE patients compared with 11 controls. The DNA methylation value of each DMR corresponds to the mean beta value across all single-nucleotide positions encompassed by the DMR. Each individual is annotated with respect to age and sex. GO enrichment analysis of hypermethylated and hypomethylated DMRs across the promoter \u003cstrong\u003e(B)\u003c/strong\u003e, gene \u003cstrong\u003e(D)\u003c/strong\u003e, and CGI-DMR \u003cstrong\u003e(F)\u003c/strong\u003e types, showing the most biologically relevant terms. GO categories include Biological Process (BP), Molecular Function (MF), Cellular Component (CC) and Human Phenotype (HP). Enrichment is represented by \u003cem\u003ep \u003c/em\u003evalue and fold enrichment. For each GO term, the corresponding DMR hits were identified in a heatmap. NES and beta difference values (MTLE-CTR) for each gene were included to represent the degree of differential methylation. \u003cstrong\u003eG.\u003c/strong\u003e HOMER binding-motif enrichment of hypomethylated and hypermethylated DMRs. Colour depicts the transcription factor family; bubble size indicates the level of significance (\u003cem\u003ep\u003c/em\u003e).\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/bf953444bbd99acc279bb0b3.png"},{"id":25048430,"identity":"4123a98e-3a07-4086-b33f-c4286df1252d","added_by":"auto","created_at":"2022-08-10 15:58:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1166228,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eRepresentation of the overlap between cfDNA DMRs in MTLE and hippocampal DMRs in MTLE. The sets of DMRs overlapped separately in relation to the type of DMR (promoter, gene, CGI) and the methylation behaviour (hypermethylated and hypomethylated). Red bars and lines indicate the overlap of DMRs with coincidence in the direction of change. \u003cstrong\u003eB. \u003c/strong\u003eHeatmap matrix representation of the \u003cem\u003ep\u003c/em\u003e values associated with the Fisher’s exact test of the overlaps between DMRs in cfDNA and in the hippocampus. \u003cstrong\u003eC. \u003c/strong\u003eRepresentation of the overlap between cfDNA DMRs in MTLE and neocortical DMRs in MTLE. The sets of DMRs overlapped separately in relation to the type of DMR (promoter, gene, CGI) and the methylation behaviour (hypermethylated and hypomethylated). Red bars and lines indicate the overlap of DMRs with coincidence in the direction of change.\u003cstrong\u003e D. \u003c/strong\u003eHeatmap matrix representation of the \u003cem\u003ep\u003c/em\u003e values associated with the Fisher’s exact test of the overlaps between DMRs in cfDNA and in the neocortex. \u003cstrong\u003eE-G. \u003c/strong\u003eGraphical representation of the DNA methylation of the individual probes in cfDNA, hippocampus, neurons \u003cem\u003evs\u003c/em\u003e. blood cells, and glia \u003cem\u003evs.\u003c/em\u003e blood cells, and DNase-seq hypersensitivity in brain tissue and blood cells, corresponding to the DMRs located at the promoter of the \u003cem\u003ePM20D1\u003c/em\u003e gene (\u003cstrong\u003eE\u003c/strong\u003e), at the chr8:145008908-145009407 CGI (\u003cstrong\u003eF\u003c/strong\u003e) and at the chr7: 70254894-70255986 CGI (\u003cstrong\u003eG\u003c/strong\u003e). DNA methylation is presented as beta values. Beta diff is the mean difference of the beta values of all individual probes in the DMR for each comparison. FDR corresponds to the Bonferroni-adjusted \u003cem\u003ep\u003c/em\u003e value emerging from the \u003cem\u003emCSEA\u003c/em\u003e DMR calculation. The genomic location of each DMR is highlighted by a red line in the respective chromosome. The DMRs (green) and the individual probes (orange) are presented in relation to the annotated genes in the UCSC Ref Seq.\u003c/p\u003e\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/d6897ea7438123723a55208e.png"},{"id":44715529,"identity":"52ec3f65-a52e-4f29-877e-7fe5a25f290a","added_by":"auto","created_at":"2023-10-16 18:14:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1403680,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/6c26452a-eb18-4d3e-a7ee-09cc347da6f3.pdf"},{"id":25047722,"identity":"16c1db4f-3377-4ca4-8f48-ebe0a1fa8cea","added_by":"auto","created_at":"2022-08-10 15:53:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":71156,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigure1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/95cdf78ad6e1afdd01582e4d.pdf"},{"id":25047097,"identity":"5ce12092-4f92-488e-9d09-37879cf849ca","added_by":"auto","created_at":"2022-08-10 15:48:05","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":67060,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigure2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/1cf4aa64f4cd1891d6c9725c.pdf"},{"id":25047727,"identity":"8b410514-57e2-4c2a-98a5-db801637b2e0","added_by":"auto","created_at":"2022-08-10 15:53:05","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":1181960,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigure3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/4189e19b278392f59b19de1b.pdf"},{"id":25047090,"identity":"68a76cd4-31ff-43fa-9d37-f74d298d6180","added_by":"auto","created_at":"2022-08-10 15:48:05","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":14932,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigureLegends.docx","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/31e1969da80bc0227ef21725.docx"},{"id":25047725,"identity":"583f210d-a8af-43ff-9a36-b589bbc1c1e1","added_by":"auto","created_at":"2022-08-10 15:53:05","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":10763,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/29611f15e45c1a286d49df43.xlsx"},{"id":25047094,"identity":"8ae73a3c-b7a7-41e2-bb3b-71e22d78e8a9","added_by":"auto","created_at":"2022-08-10 15:48:05","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":11377,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/3411d7e3ab6b08736c12d82b.xlsx"},{"id":25047723,"identity":"1e088af7-5afc-4514-a923-dde6c4c59b43","added_by":"auto","created_at":"2022-08-10 15:53:05","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":12866,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/4150e2bdc6761da057866658.xlsx"},{"id":25049023,"identity":"0c927881-68cb-467a-90bb-279491460edc","added_by":"auto","created_at":"2022-08-10 16:03:05","extension":"xlsx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":374901,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1940501/v1/6e95a9154cb479138c3352f3.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Circulating cell-free DNA methylation mirrors alterations in cerebral patterns in epilepsy","fulltext":[{"header":"Background","content":"\u003cp\u003eCell-free DNA (cfDNA) consists of small DNA fragments released into the peripheral blood, predominantly as a result of apoptosis (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). This is substantiated by the consistent correspondence between the length of human circulating cfDNA (167 bp) and the length of DNA wrapped around a nucleosome (~\u0026thinsp;147 bp) plus linker regions, which suggest the action of endonucleases. Apoptotic DNA degradation is mediated by caspase-activated DNase (CAD), which lacks exonuclease activity, and so can only fragment DNA in inter-nucleosomal regions (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). The evaluation of DNA methylation of cfDNA has been used to estimate tissue or cell of origin, and to non-invasively track ongoing cell death occurring anywhere in the body, based on the cell-specific nature of DNA methylation (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). The use of this approach is spreading in cancer studies (\u003cspan additionalcitationids=\"CR5 CR6 CR7\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Direct detection of altered DNA methylation patterns under pathological conditions, regardless of cell or tissue contribution, has been thoroughly examined to determine its value as a strategy for searching for biomarkers (\u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe identification and development of epilepsy and epileptogenesis biomarkers is of inherent interest (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e), but progress has been slower than in other settings, including other neurodegenerative pathologies like Alzheimer\u0026rsquo;s disease (AD). This can be attributed to the low accessibility to pathological tissue and the complexity and variability within the spectrum of epilepsy syndromes. Nevertheless, nucleic acid-based biomarkers, predominantly microRNAs, are promising (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHere, we obtained the DNA methylation profiles of serum cfDNA samples from mesial temporal lobe epilepsy (MTLE) patients and compared them with those obtained from healthy controls. MTLE is commonly associated with severe neuronal cell death, termed hippocampal sclerosis (HS) (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). We hypothesized that the analysis of the cfDNA methylome in MTLE-HS patients could serve as a predictive, diagnostic or prognostic tool of neuronal cell death estimation. Furthermore, direct comparison of the DNA methylation profile between patients and healthy controls could potentially fill the gap in peripheral biomarker development in epilepsy. Our results were not able to identify a significant increase in the proportion of cfDNA derived from brain. However, the cfDNA methylomes reflect a significant enrichment of epileptic patterns overlapping with those described in the hippocampus of MTLE patients.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy population\u003c/h2\u003e \u003cp\u003eThe MTLE patients included in this study were followed at the Reference Epilepsy Research Centre of \u003cem\u003eHospital de Santo Ant\u0026oacute;nio \u0026ndash; Centro Hospitalar e Universit\u0026aacute;rio do Porto\u003c/em\u003e (HSA-CHUP) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Supplementary Table\u0026nbsp;1). The diagnosis was based on clinical and electrophysiological data (electroencephalogram (EEG) and/or video-EEG monitoring) and brain MRI (minimum 1.5T), as defined by Wieser (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). A definition of HS by brain MRI required the detection of atrophy, T2 hyperintensity signal and altered internal structure on one or both hippocampi, associated or not with other imaging criteria such as ipsilateral fornix atrophy, ipsilateral mamillary body atrophy or ipsilateral entorhinal abnormalities. Visual and/or verbal memory impairment were not considered exclusion criteria. However, patients with other neurological abnormalities were not included. At the time of the study, all patients were receiving pharmacological treatment (monotherapy or polytherapy). The control population comprised healthy individuals who were ethnically matched and from the same geographical area, and who had been voluntarily recruited from blood donors. Individuals with any neurological condition or a positive family history were excluded.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eClinicodemographic characterization of the studied populations.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eControls\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMTLE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003en\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003e% Female (n)\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e81.8 (9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.0 (9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAge, years (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e38.9\u0026thinsp;\u0026plusmn;\u0026thinsp;8.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44.8\u0026thinsp;\u0026plusmn;\u0026thinsp;11.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.2063\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAge of onset, years (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.0\u0026thinsp;\u0026plusmn;\u0026thinsp;13.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eEpilepsy duration (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.8\u0026thinsp;\u0026plusmn;\u0026thinsp;14.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003e% of pharmacorresistant (n)\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.3 (10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003e% of FS history\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.0 (9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eMTLE, Mesial Temporal Lobe Epilepsy\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eFS, Febrile seizures\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eSD, Standard deviation\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eEthics statement\u003c/h2\u003e \u003cp\u003e This work was approved by the ethical committee of HSA-CHUP (2018.051(047-DEFI/047-CES)). All individuals gave their written consent in accordance with principles of the Declaration of Helsinki.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eSerum collection and DNA extraction\u003c/h2\u003e \u003cp\u003ePeripheral blood was collected in \u003cem\u003eVacuette\u003c/em\u003e\u0026reg; tubes without anticoagulant and centrifuged at 490 g for 20 minutes. Collected serum aliquots were stored at -20\u0026deg;C. Only samples processed within 4 h of collection were included. DNA was extracted from approximately 1 mL of serum using the QIAmp\u0026reg; MinElute\u0026reg; ccfDNA Mini Kit (Qiagen), following the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eDNA methylation profiling\u003c/h2\u003e \u003cp\u003eExtracted genomic DNA was quantified using a Qubit DNA Assay Kit (Cat. No. 10146592) in a Qubit 2.0 Fluorometer (Life Technologies, CA, USA). CfDNA samples from twelve MTLE patients (8F, 4M; 44.8\u0026thinsp;\u0026plusmn;\u0026thinsp;11.4 years old) and eleven controls (9F, 2M; 38.9\u0026thinsp;\u0026plusmn;\u0026thinsp;8.4 years old) were profiled. All DNA extracted from each sample (44.72\u0026ndash;238.95 ng) was bisulphite-converted using the EZ DNA Methylation-Gold\u0026trade; Kit (Zymo Research, Irvine, CA, USA), following the manufacturer\u0026rsquo;s instructions. Converted DNA was hybridized in Infinium MethylationEPIC BeadChip arrays (Illumina), following the manufacturer\u0026rsquo;s instructions. The arrays encompass\u0026thinsp;\u0026gt;\u0026thinsp;850,000 single-nucleotide methylation sites and cover 99% of the annotated reference sequence (RefSeq) genes. Fluorescence intensities were imaged using a BeadArray Reader (Illumina), and images were processed and intensities measured as previously described (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). A combination of the Cy3 and Cy5 fluorescence intensities of the methylated and unmethylated alleles was used to obtain each methylation data point. Background intensity was computed from a set of negative controls and subtracted from each data point. Beta values were used to illustrate methylation. Values can range between zero (0% methylation) and one (100% methylation) and represent the ratio of the methylated probe intensity to the overall intensity (sum of the methylated and unmethylated probe intensities). M values were calculated as the log\u003csub\u003e2\u003c/sub\u003e ratio of the intensities of the methylated and unmethylated probes. M values were used for statistical purposes, because beta values are heteroskedastic for highly methylated and unmethylated CpGs (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Methylation data were analysed in the R statistical environment. The \u003cem\u003eshiny\u0026Eacute;pico\u003c/em\u003e web interface (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e), based on \u003cem\u003eminfi\u003c/em\u003e (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e) and \u003cem\u003elimma\u003c/em\u003e (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) pipelines, was used for array processing, normalization and differential methylation calculation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eDeconvolution of cell/tissue of origin\u003c/h2\u003e \u003cp\u003eThe cell or tissue of origin was estimated with the meth_atlas deconvolution algorithm (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/nloyfer/meth_atlas\u003c/span\u003e\u003cspan address=\"https://github.com/nloyfer/meth_atlas\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), which estimates the proportion of origin for a total of 25 tissue and cell types based on 450k and EPIC data (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). DNA methylation profiles from serum cfDNA of epileptic patients and controls were processed in accordance with the original study. Normalization was performed with the \u003cem\u003epreprocessIllumina\u003c/em\u003e function. Probes with a detection significance of p\u0026thinsp;\u0026gt;\u0026thinsp;0.01 were excluded, as were those mapping to sex chromosomes. Deconvolution was performed using the Python-based method in relation to the supplied reference atlas matrix composed of 7890 sites.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCalculation of cfDNA differentially methylated regions (DMRs)\u003c/h2\u003e \u003cp\u003eMethylation data were normalized with the \u003cem\u003eNoob\u0026thinsp;+\u0026thinsp;Quantile\u003c/em\u003e functions to assess the differentials. Probes with a detection \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01 were filtered out, as were positions located in the X and Y chromosomes and/or overlapping with SNPs. CpHs were retained, based on evidence that neurons, unlike other CNS cells, present CpH as their dominant DNA methylation mark (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). A total of 783,351 individual positions were obtained after normalization and filtering. DMRs were calculated using the \u003cem\u003emCSEA\u003c/em\u003e (methylated CpGs Set Enrichment Analysis) package (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). The eBayes results of \u003cem\u003elimma\u003c/em\u003e, sorted by \u003cem\u003et\u003c/em\u003e-statistics, were used as input. The \u003cem\u003elimma\u003c/em\u003e eBayes-moderated \u003cem\u003et-\u003c/em\u003etest was carried out using M values and included no covariates. DMRs with at least five CpGs and an FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered statistically significant.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eGene ontology (GO) and transcription factor (TF) enrichment analysis\u003c/h2\u003e \u003cp\u003eGene ontology (GO) evaluation was performed using the GREAT online tool (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://great.stanford.edu/public/html\u003c/span\u003e\u003cspan address=\"http://great.stanford.edu/public/html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e), with the \u003cem\u003etwo nearest genes\u003c/em\u003e settings. Motif enrichment was analysed using the findMotifsGenome.pt tool of the HOMER motif discovery application (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e), considering a window of \u0026plusmn;\u0026thinsp;50 bp from each DMR. The total annotated DMRs from the mCSEA analysis were used as a background in both analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eHuman Protein Atlas (HPA) gene expression and expression cluster data\u003c/h2\u003e \u003cp\u003e We used public data from the HPA to evaluate potential cerebral regional activity of enriched TF associated with cfDNA DMRs. To evaluate the expression levels of the genes associated with those factors, we used the \u003cem\u003eRNA consensus tissue gene data\u003c/em\u003e, which consists of transcript expression levels summarized per gene in 55 tissues obtained consensually from RNA-seq data from the HPA and Genotype-Tissue Expression (GTEx) projects.\u003c/p\u003e \u003cp\u003eWe also accessed data from the HPA consisting of the levels of confidence of protein-coding genes across tissue clusters and across single-cell clusters. For the tissue clusters, the HPA used RNA expression data from 53 tissues to classify genes into 87 expression clusters. A total of 144 cell types or cell lines were used to construct 68 single-cell expression clusters. For both approaches, Louvain clustering was performed based on gene-to-gene distances calculated from the Spearman correlation of gene expression across multiple samples. Clustering was performed 100 times to accommodate stochasticity. The confidence of the gene-to-cluster value, which varies between 0 and 1, corresponds to the proportion of times that a gene was assigned to a cluster. Clusters were identified manually through recourse to functional annotation tools.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eOverlap of cfDNA DMRs with epileptic, neuron-specific and glia-specific DMRs\u003c/h2\u003e \u003cp\u003eWe extracted the lists of DMRs significantly altered in hippocampal and neocortical tissue in MTLE patients relative to autopsied non-epileptic controls, obtained in a previous study by our group (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). We extracted raw IDAT files from 33 NeuN\u0026thinsp;+\u0026thinsp;fractions of prefrontal cortex of healthy individuals (GSE112179), 23 NeuN- fractions (glia) (GSE166207) and 37 blood-cell-type samples, which included B cells, CD8\u0026thinsp;+\u0026thinsp;T cells, CD4\u0026thinsp;+\u0026thinsp;T cells, monocytes, neutrophils and NK cells (GSE110555). Data were processed and DMRs calculated as described above, all NeuN or glia samples being compared with all blood samples, without discrimination for blood-cell types, and without considering any covariates. Since the \u003cem\u003emCSEA\u003c/em\u003e package was used to calculate DMRs across all settings, we overlapped the lists of DMRs based on their annotated identification. The \u003cem\u003eGeneOverlap\u003c/em\u003e function, which is based on Fisher\u0026rsquo;s exact test, was used to calculate the significance of the overlap. The total number of DMRs considered by mCSEA was used as background (26,208 for promoter DMRs; 23,772 for gene DMRs; 27,187 for CGI DMRs).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eDNase-seq data analysis\u003c/h2\u003e \u003cp\u003eDNase hypersensitivity bigwig files with the human GRCh37 assembly from brain tissue and blood cells were obtained from ENCODE (Supplementary Table\u0026nbsp;4). For each setting, the mean function of Wiggletools was used to aggregate the multiple files (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). The outputted wig file was reconverted to bigwig format using ucsc-wigtobigwig (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eHeatmaps, DMR visualization and plots\u003c/h2\u003e \u003cp\u003eAll heatmaps were developed using the R \u003cem\u003egplots\u003c/em\u003e and \u003cem\u003eComplexHeatmap\u003c/em\u003e (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e) packages. Row dendrogram clustering was carried out with complete-linkage hierarchical clustering. The overlaps between lists of DMRs was represented using the \u003cem\u003eUpSet\u003c/em\u003e function of \u003cem\u003eComplexHeatmap\u003c/em\u003e. To visualize individual DMRs, along with genomic location and epigenetic and chromatin accessibility marks, we used the functions available in the \u003cem\u003egviz\u003c/em\u003e package (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). All additional plots were generated with the \u003cem\u003eggplot2\u003c/em\u003e package (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses were done using R v4.0.2. or IBM SPSS Statistics version 27 (Armonk, NY, USA). All graphs were created in R. Group medians were compared using the Mann-Whitney test. Fisher\u0026rsquo;s exact test was used to calculate the significance of non-random association between two categorical variables (e.g., sex distribution in patients and controls). The levels of significance were: *, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; **, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01; ***, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eRole of the funding source\u003c/h2\u003e \u003cp\u003eThe funding bodies played no role in the study design, data collection, analysis and interpretation, or writing of the manuscript.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cem\u003eEstimation of cell of origin proportions in serum of MTLE and controls based on cfDNA methylation\u003c/em\u003e \u003c/p\u003e \u003cp\u003eFirst, we generated the DNA methylation profiles of serum cfDNA samples of MTLE and healthy controls using beadchip arrays. To estimate the cell of origin, we used the deconvolution algorithm designed for this purpose by Moss et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e), in relation to a 7890-CpG reference matrix, accounting for 25 tissue and cell types (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA, Supplementary Fig.\u0026nbsp;1 and Supplementary Table\u0026nbsp;2). In agreement with other studies, including the original source of the algorithm, hematopoietic cells were the main contributors. We also noted contributions from non-blood cell or tissue types, including bladder, breast, vascular endothelial cells and cortical neurons. Significant differences in the cell-of-origin proportion between controls and MTLE patients were only observed for vascular endothelial cells (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02, Wilcoxon test) (Supplementary Fig.\u0026nbsp;1B). No significant differences in the cortical neuronal origin were observed between patients and controls (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.54, Wilcoxon) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). We noted a striking predominance of neutrophil contribution in both groups, with a mean percentage contribution of approximately 92% across all samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and Supplementary Fig.\u0026nbsp;1); this is a higher value than that reported by Moss and colleagues. This might be due to the use of serum samples, instead of plasma samples, as they used. We can speculate that the predominance of neutrophil contribution is associated with increased coagulation-related NETosis during sample collection.\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eDifferentially methylated regions (DMRs) in cfDNA\u003c/h2\u003e \u003cp\u003eWe then determined DMRs, using the mCSEA algorithm, between MTLE and healthy control cfDNA. Three sets of DMRs, located in promoters, gene bodies and CpG islands (CGIs), were calculated. We identified 873 significant promoter-DMRs (744 hypomethylated and 129 hypermethylated in MTLE relative to controls) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and Supplementary Table\u0026nbsp;3). Gene ontology (GO) analysis of the hypermethylated and hypomethylated DMR clusters demonstrated enrichment of CNS-related terms, including some associated with GABAergic pathways, synaptic transmission, microglia activation and neurotrophin receptor binding. A thorough inspection of the DMRs associated with these GO categories included biologically relevant promoters such as those associated with the \u003cem\u003eGABRG3\u003c/em\u003e and \u003cem\u003eCDH9\u003c/em\u003e genes (hypermethylated), and those at the \u003cem\u003eGABRA1, GABRA2, GABRG2\u003c/em\u003e and \u003cem\u003eBDNF\u003c/em\u003e genes (hypomethylated) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). GABAergic receptors represent key constituents of the CNS. γ-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the cerebral cortex and disruption in the excitatory/inhibitory balance has long been associated with seizure development (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). On this basis, genetic variability related to GABAergic subunits has shown potentially causal epileptogenic effects (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). \u003cem\u003eBDNF\u003c/em\u003e encodes the brain-derived neurotrophic factor, one of the most prominent members of the neurotrophin family, which has a wide range of functions, encompassing regulation of neuronal development and synaptic plasticity (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). The regulation of the \u003cem\u003eBDNF\u003c/em\u003e gene in neurons has long been associated with DNA methylation-related mechanisms (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). It has been described as being overexpressed in epilepsy (\u003cspan additionalcitationids=\"CR39 CR40 CR41\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). Moreover, the DNA methylation status of its promoter regions has been explored (\u003cspan additionalcitationids=\"CR44\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e), revealing a marked tendency towards demethylation.\u003c/p\u003e \u003cp\u003eWe also described 327 gene-body DMRs (65 hypomethylated and 262 hypermethylated in MTLE compared with healthy controls) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC and Supplementary Table\u0026nbsp;3). Multiple CNS-related GO categories were also found to be associated with obtained DMRs. These included terms associated with synaptic assembly and organization (hypermethylated) and febrile seizures (hypomethylated) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003eWe found 550 significant CGI-DMRs (334 hypomethylated and 216 hypermethylated) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE and Supplementary Table\u0026nbsp;3). GO analysis was consistent with that described above, with enrichment of multiple mechanisms of potential relevance in neuropathology (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003eWe also examined the TF binding motif enrichment of the generated DMRs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG). To analyse potential CNS-specific activity of the enriched TFs, we used data from the HPA repository. Many of the enriched TFs were consistently expressed across different brain regions (Supplementary Fig.\u0026nbsp;2A). Moreover, we searched for associations of the genes coding the enriched TFs with HPA tissue clusters and HPA single-cell clusters. The most prominent results were the high confidence levels of \u003cem\u003eZNF528\u003c/em\u003e and \u003cem\u003eRARB\u003c/em\u003e in \u003cem\u003eCluster 73 \u0026ndash; Brain: Transcription regulation\u003c/em\u003e (RNA HPA tissue expression cluster) (Supplementary Fig.\u0026nbsp;2B) and \u003cem\u003eCluster 8 \u0026ndash; Neurons \u0026amp; Oligodendrocytes: Synaptic function\u003c/em\u003e (RNA HPA single-cell expression cluster) (Supplementary Fig.\u0026nbsp;2C), respectively. This suggests possible brain specificity in the establishment of these DNA methylation patterns in cfDNA of MTLE patients.\u003c/p\u003e \u003cp\u003e \u003cem\u003ecfDNA DMRs in MTLE show enrichment of brain epileptic DMRs but not of neuron- or glia-specific methylation patterns\u003c/em\u003e \u003c/p\u003e \u003cp\u003ePreviously, our group described significantly altered DNA methylation patterns in the hippocampus and neocortex of MTLE-HS patients compared with autopsied controls without neuropathology (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). These include 2650 DMRs in the hippocampus (736 promoter, 1111 gene and 803 CGI) (Supplementary Fig.\u0026nbsp;3A) and 2950 DMRs in the adjacent neocortex (785 promoter, 1264 gene and 901 CGI) (Supplementary Fig.\u0026nbsp;3B). The overlap of cfDNA DMRs with hippocampal DMRs in MTLE patients was significant (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). The coincidence of the direction of methylation change (hypermethylated or hypomethylated in both cfDNA and hippocampus) reinforced the relevance of the overlap of DMRs. There was a marked enrichment between hypermethylated DMRs, but we found no significant overlap between hypomethylated DMRs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003eWe considered whether the dimensionality reduction achieved by genome-wide deconvolution algorithms was too broad and overlooked specific DNA methylation changes that could indicate differences in the cell-specific contribution. Cell of origin may be represented by a small subset of target regions. Therefore, we also attempted to determine whether the DMRs obtained in cfDNA of MTLE patients are enriched in neuron- or glia-specific DNA methylation patterns. We used public EPIC data consisting of neuron and glia samples and blood cells (monocytes, neutrophils, B cells, CD4\u0026thinsp;+\u0026thinsp;T cells, CD8\u0026thinsp;+\u0026thinsp;T cells, NK cells). We identified 5943 neuron-specific DMRs (1504 promoter, 2174 gene, 2265 CGI) (Supplementary Fig.\u0026nbsp;3C) and 3523 glia-specific DMRs (1038 promoter, 1444 gene, 1041 CGI) (Supplementary Fig.\u0026nbsp;3D). We did not find any significant overlap between cfDNA DMRs from the MTLE/CTR comparison or neuron-specific or glia-specific DMRs with concordant behaviour. In fact, greater enrichment was observed for DMRs with opposite behaviour (e.g., hypermethylated promoter cfDNA in MTLE cfDNA and hypomethylated promoter neuron-specific DMRs) (Supplementary Fig.\u0026nbsp;3E-H). These results therefore imply that the differences in DNA methylation observed in cfDNA of patients compared with controls do not appear to be a consequence of increased circulating neuronal or glial DNA.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eDNA accessibility may provide clues about the origin of cfDNA DMRs\u003c/h2\u003e \u003cp\u003eAt this point, we have demonstrated that DNA methylation alterations in the MTLE brain are, to some extent, replicable in circulating cfDNA. However, we are yet to determine which cells are the source of the fragments carrying such differentially methylated patterns. We next investigated the potential origin of the fragments carrying the identified cfDNA DMRs by using public chromatin accessibility data, namely DNase-seq.\u0026nbsp;It can be assumed that euchromatic regions (open chromatin) would be highly degraded and the resulting fragments would be too small to be detected in circulating cfDNA (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). We were able to individualize DMRs with higher chromatin accessibility in blood cells than in brain tissue (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE-F). We found lower DNA methylation levels in the promoter of the \u003cem\u003ePM20D1\u003c/em\u003e gene in both cfDNA and hippocampus of MTLE patients. This hypomethylation pattern is neither neuron- nor glia-specific. DNA accessibility, however, is high in blood cells but low in brain tissue (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). We identified two more DMRs, localized in CpG islands located at the \u003cem\u003ePLEC\u003c/em\u003e and \u003cem\u003eAUTS2\u003c/em\u003e genes, that show hypermethylation in cfDNA and hippocampus of patients compared with controls. Although the DMR does not show overall significant hypermethylation in neurons or glia, individual CpG probes that overlap with the blood euchromatin region have higher DNA methylation levels in neurons and glia than in blood cells. Both these regions also present higher accessibility in blood cells than in brain tissue (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF-G).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eOnly two previous studies have analysed cfDNA in the context of epilepsy, and they only measured total cfDNA concentration levels (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). MTLE is the most incident-focal epilepsy in adults, which, together with its high pharmacoresistance rates, makes it one of the most widely studied epilepsy syndromes. DNA methylation has been thoroughly explored within epileptic brain tissue. In a recent study by our group (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e), we described major DNA methylation alterations in the hippocampus, which is the focus of the epilepsy and the region of the lesion (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), and in the adjacent neocortical areas of MTLE patients, comparing these with the state in non-epileptic controls. Our present study demonstrates the presence of DNA methylation alterations in cfDNA of MTLE patients. There was a level of representation of CNS-related genes within those alterations and, in fact, we observed a significant overlap of epileptic patterns observed in the pathological tissue, mainly hippocampus biopsies. The functional enrichment analysis of the distinct cfDNA methylation patterns observed in MTLE suggests its possible origin by highlighting multiple genes, pathways and regulatory transcription factors associated with the CNS paradigm. We found that the epileptic patterns described in hippocampus and neocortex tissue were also significantly enriched in circulating cfDNA. However, a clearer overlap was demonstrated in hippocampus, which is further evidence of the primary pathological nature of this region. This concurs with the findings of previous studies that cfDNA from serum and plasma can contain tissue-pathological DNA methylation alterations (\u003cspan additionalcitationids=\"CR50\" citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOne aspect that remains unresolved is the source of the cfDNA fragments bearing those epileptic patterns. Studies have shown that the methylation status of DNA from blood cells also differs between controls and epilepsy patients (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). Furthermore, the DNA methylation profiles of peripheral tissues, including blood, positively correlate with brain tissue in epilepsy patients (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e). It has been proposed that chromatin accessibility states may help to infer the origin of cfDNA (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). Highly accessible euchromatic regions are likely to be highly degraded, and the resulting fragment would be too small to be detected in circulation. In the three examples of DMR-containing sequences that we have presented, chromatin accessibility is high in blood cells whereas it shows a more compact state in brain tissue. We may speculate that the cfDNA fragments carrying those epileptic patterns, coincident in cfDNA and hippocampus, are more likely to originate in brain tissue rather than in blood cells. \u003cem\u003ePM20D1\u003c/em\u003e codes an N-fatty acyl amino acid (NAAs) synthase/hydrolase, and is located within a Parkinson\u0026rsquo;s susceptibility locus (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). It has been demonstrated that \u003cem\u003ePM20D1\u003c/em\u003e is both an expression and methylation quantitative trait locus in AD, with direct influence on molecular and behaviour pathological features (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). In AD blood samples, a U-shaped model has been proposed in which DNA methylation of this region is decreased in early stages and reverses with progression towards late AD (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e). (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e). Plectin (\u003cem\u003ePLEC\u003c/em\u003e) is a plakin responsible for linking elements of the cytoskeleton. In the CNS, \u003cem\u003ePLEC\u003c/em\u003e has been shown to be predominantly expressed in pia/glia and endothelia/glia junctions (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e), where it is paramount for the structural and functional integrity of the BBB and the pial surface (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e). In TLE, plectin is upregulated in astrocytes located at the sclerotic hippocampus (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e). \u003cem\u003eAUTS2\u003c/em\u003e is a well-known risk gene for autism spectrum (ASD) but also for other neurodevelopmental disorders, including epilepsy (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e). The autism susceptibility candidate 2 (AUTS2) gene links with PRC1 (Polycomb Repressive Complex 1), a known epigenetic regulator, and together are responsible for transcription activation of genes associated with neurodevelopment (\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e). Moreover, cytoplasmatic AUTS2 contributes to the reorganization of the cytoskeleton, since it is related to neuronal motility and morphogenesis (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe potential use of brain-cell-derived cfDNA in neurodegenerative diseases has only recently begun to be explored. The most prominent results have originated from targeted studies of a small group of specific DNA methylation positions that show an exclusive pattern in brain cells. Using such an approach, increased brain cell-derived cfDNA was demonstrated in multiple sclerosis (MS) patients, cardiac arrest patients with brain damage and traumatic brain injury (TBI) patients (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e), cancer patients with brain metastasis (\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e) and schizophrenic patients with a first psychotic episode (\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e). More complex mathematical deconvolution algorithms have been developed to estimate the simultaneous percentages of contributions of multiple cells and/or tissue types. Moss et al. described a DNA methylation-based deconvolution algorithm, based on the non-negative least-squares linear regression of a reference matrix encompassing 25 tissue and cell types, including cortical neurons (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). This tool was subsequently used to predict the proportion of neuronal contribution to cfDNA in pituitary neuroendocrine tumours (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e) and glioma (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e). The direct relation between neuronal death (HS) and cfDNA corroborates the suitability of cfDNA methylation-based deconvolution of brain origin as a logical stream towards biomarker identification in MTLE. Monitoring brain-derived cfDNA could contribute to the early detection of HS and help control the progression of neuroanatomical damage over the course of the disease (\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e). Nevertheless, we did not observe any enhancement of cortical neuron-derived cfDNA in our patients. One must consider the potential lack of precision of the current deconvolution tools in estimating the contribution of brain cells. Moss et al., for instance, used three cortical neuron samples to develop their reference matrix. It is of inherent interest to develop more precise algorithms which would account for, as far as possible, the whole complexity of the CNS spectrum by including multiple cell types (e.g., excitatory and inhibitory neurons, oligodendrocytes, OPCs, astrocytes, microglia) and also take regional variability into account. Additionally, the eventual release of brain-derived cfDNA in MTLE may be an acute event. In fact, Chatterton et al. reported an increase in the release of neuronal and glial cfDNA in plasma of entry personnel (breachers) during explosive training, on the day that participants were exposed to higher pressures, after which it promptly decreased (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). In MTLE, such events could occur immediately following the seizure. However, the interval between the time of the last seizure and serum collection was taken into account in the present study.\u003c/p\u003e \u003cp\u003eThe introduction of methods based on artificial intelligence and machine learning in biological research has significantly increased the assessment of biomarker performance. The development of predictive models has made it possible to potentiate the use of large omics datasets for biomarker identification. Such tools have been used to access diagnostic performance of cfDNA methylation modifications in neurodegenerative settings (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e). We consider the use of predictive models to be an important next step in cfDNA methylation analysis. However, larger populations are required to develop robust and replicable models.\u003c/p\u003e \u003cp\u003eOur study shows that the analysis of cfDNA methylation in epilepsy has predictive potential. To follow this up, complementary studies are needed that exploit this emerging field to its full potential. We consider cfDNA methylation to be a promising tool with which to pursue the ultimate goal of reducing patients\u0026rsquo; burden through early diagnosis, proper monitoring of the progressive nature of the disease, and a better understanding of the causes of epileptic refractoriness.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAD, Alzheimer\u0026rsquo;s disease; BDNF, brain-derived neurotrophic factor; CAD, caspase-activated DNase; cfDNA, cell-free DNA; CGIs, CpG islands; CTR, controls; DMR, differentially methylated region; EEG, electroencephalogram; GABA, \u0026gamma;-aminobutyric acid; GO, gene ontology; GTEx, genotype-tissue expression; HPA, Human Protein Atlas; HS, hippocampal sclerosis; mCSEA, methylated CpG set enrichment analysis; MS, multiple sclerosis; MTLE, mesial temporal lobe epilepsy; NET, neutrophil extracellular traps; OPCs, oligodendrocyte precursor cells; TBI, traumatic brain injury; TF, transcription factor.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the CERCA Programme/Generalitat de Catalunya, the Josep Carreras Foundation and ICBAS-UP for institutional support. We acknowledge Professor Berta Martins da Silva and the Immunogenetics Laboratory of the Molecular Pathology and Immunology of the ICBAS-UP. We thank the other members of the Epigenetics and Immune Diseases group of the IJC for support and advice throughout the development of study, and Dr. João Lopes and Dr. João Ramalheira of the Neurophysiology Service of HSA-CHUP for insightful clinical guidance and continuous collaboration. We thank the patients and their families for their essential contributions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEB is funded by the Spanish Ministry of Science and Innovation (MICINN) [PID2020117212RB-I00; AEI/10.13039/501100011033]. This work was partially supported by a BICE Tecnifar Grant. RM-F is funded by an FCT (\u003cem\u003eFundação para a Ciência e Tecnologia\u003c/em\u003e)fellowship (SFRH/BD/137900/2018). UMIB is funded by FCT Portugal (UIDB/00215/2020 and UIDP/00215/2020), and ITR (LA/P/006/2020).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEPIC DNA methylation array data have been deposited in the NCBI’s Gene Expression Omnibus database under accession code GSE208758. All additional data used in this study is available in this article or in the supplementary information.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was approved by the ethical committee of HSA-CHUP (2018.051(047-DEFI/047-CES)). All participants provided written informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll participants provided written informed consent for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclarations of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone of the authors report any conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy conception and design: RM-F, EB, BL. Recruitment of patients and collection of serum samples: JC, RS, BL, RM-F. Laboratory work: RM-F and LC. Bioinformatic analysis: RM-F. Verification and interpretation of the analysis: EB, BL, PPC, RM-F. Original drafting of the article: RM-F, EB, BL, PPC. All the remaining authors reviewed, edited and approved the final version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHeitzer E, Auinger L, Speicher MR. Cell-Free DNA and Apoptosis: How Dead Cells Inform About the Living. Trends Mol Med. 2020 May;26(5):519\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSnyder MW, Kircher M, Hill AJ, Daza RM, Shendure J. Cell-free DNA Comprises an In Vivo Nucleosome Footprint that Informs Its Tissues-Of-Origin. Cell. 2016 Jan;164(1\u0026ndash;2):57\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDor Y, Cedar H. Principles of DNA methylation and their implications for biology and medicine. Lancet (London, England). 2018 Sep;392(10149):777\u0026ndash;86.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo S, Diep D, Plongthongkum N, Fung H-L, Zhang K, Zhang K. Identification of methylation haplotype blocks aids in deconvolution of heterogeneous tissue samples and tumor tissue-of-origin mapping from plasma DNA. Nat Genet. 2017 Apr;49(4):635\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi W, Li Q, Kang S, Same M, Zhou Y, Sun C, et al. CancerDetector: ultrasensitive and non-invasive cancer detection at the resolution of individual reads using cell-free DNA methylation sequencing data. Nucleic Acids Res. 2018 Sep;46(15):e89.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun K, Jiang P, Chan KCA, Wong J, Cheng YKY, Liang RHS, et al. Plasma DNA tissue mapping by genome-wide methylation sequencing for noninvasive prenatal, cancer, and transplantation assessments. Proc Natl Acad Sci U S A. 2015 Oct;112(40):E5503-12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang S, Li Q, Chen Q, Zhou Y, Park S, Lee G, et al. CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA. Genome Biol. 2017 Mar;18(1):53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoss J, Magenheim J, Neiman D, Zemmour H, Loyfer N, Korach A, et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free DNA in health and disease. Nat Commun. 2018 Nov;9(1):5068.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu R-H, Wei W, Krawczyk M, Wang W, Luo H, Flagg K, et al. Circulating tumour DNA methylation markers for diagnosis and prognosis of hepatocellular carcinoma. Nat Mater. 2017 Nov;16(11):1155\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLegendre C, Gooden GC, Johnson K, Martinez RA, Liang WS, Salhia B. Whole-genome bisulfite sequencing of cell-free DNA identifies signature associated with metastatic breast cancer. Clin Epigenetics. 2015;7(1):100.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGallardo-G\u0026oacute;mez M, Moran S, P\u0026aacute;ez de la Cadena M, Mart\u0026iacute;nez-Zorzano VS, Rodr\u0026iacute;guez-Berrocal FJ, Rodr\u0026iacute;guez-Girondo M, et al. A new approach to epigenome-wide discovery of non-invasive methylation biomarkers for colorectal cancer screening in circulating cell-free DNA using pooled samples. Clin Epigenetics. 2018;10:53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGordevičius J, Kriščiūnas A, Groot DE, Yip SM, Susic M, Kwan A, et al. Cell-Free DNA Modification Dynamics in Abiraterone Acetate-Treated Prostate Cancer Patients. Clin cancer Res an Off J Am Assoc Cancer Res. 2018 Jul;24(14):3317\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu H, Liu Y, Wang J, Fu S, Wang L, Huang C, et al. Detection of ovarian cancer using plasma cell-free DNA methylomes. Clin Epigenetics. 2022 Jun;14(1):74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen SY, Burgener JM, Bratman S V, De Carvalho DD. Preparation of cfMeDIP-seq libraries for methylome profiling of plasma cell-free DNA. Nat Protoc. 2019 Oct;14(10):2749\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngel JJ, Pitk\u0026auml;nen A. Biomarkers for epileptogenesis and its treatment. Neuropharmacology. 2020 May;167:107735.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenshall DC, Hamer HM, Pasterkamp RJ, Goldstein DB, Kjems J, Prehn JHM, et al. MicroRNAs in epilepsy: pathophysiology and clinical utility. Lancet Neurol. 2016 Dec;15(13):1368\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBl\u0026uuml;mcke I, Thom M, Aronica E, Armstrong DD, Bartolomei F, Bernasconi A, et al. International consensus classification of hippocampal sclerosis in temporal lobe epilepsy: a Task Force report from the ILAE Commission on Diagnostic Methods. Epilepsia. 2013 Jul;54(7):1315\u0026ndash;29.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWieser H-G. ILAE Commission Report. Mesial temporal lobe epilepsy with hippocampal sclerosis. Vol.\u0026nbsp;45, Epilepsia. United States; 2004. p.\u0026nbsp;695\u0026ndash;714.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia-Gomez A, Li T, Kerick M, Catal\u0026agrave;-Moll F, Comet NR, Rodr\u0026iacute;guez-Ubreva J, et al. TET2- and TDG-mediated changes are required for the acquisition of distinct histone modifications in divergent terminal differentiation of myeloid cells. Nucleic Acids Res. 2017 Sep;45(17):10002\u0026ndash;17.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDu P, Zhang X, Huang C-C, Jafari N, Kibbe WA, Hou L, et al. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics. 2010 Nov;11:587.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorante-Palacios O, Ballestar E. shiny\u0026Eacute;PICo: A graphical pipeline to analyze Illumina DNA methylation arrays. Bioinformatics. 2021 Jan;\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics. 2014 May;30(10):1363\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015 Apr;43(7):e47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLister R, Mukamel EA, Nery JR, Urich M, Puddifoot CA, Johnson ND, et al. Global epigenomic reconfiguration during mammalian brain development. Science. 2013 Aug;341(6146):1237905.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartorell-Marug\u0026aacute;n J, Gonz\u0026aacute;lez-Rumayor V, Carmona-S\u0026aacute;ez P. mCSEA: detecting subtle differentially methylated regions. Bioinformatics. 2019 Sep;35(18):3257\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcLean CY, Bristor D, Hiller M, Clarke SL, Schaar BT, Lowe CB, et al. GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol. 2010 May;28(5):495\u0026ndash;501.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010 May;38(4):576\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartins-Ferreira R, Leal B, Chaves J, Li T, Ciudad L, Rangel R, et al. Epilepsy progression is associated with cumulative DNA methylation changes in inflammatory genes. Prog Neurobiol. 2021 Dec;209:102207.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZerbino DR, Johnson N, Juettemann T, Wilder SP, Flicek P. WiggleTools: parallel processing of large collections of genome-wide datasets for visualization and statistical analysis. Bioinformatics. 2014 Apr;30(7):1008\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKent WJ, Zweig AS, Barber G, Hinrichs AS, Karolchik D. BigWig and BigBed: enabling browsing of large distributed datasets. Bioinformatics. 2010 Sep;26(17):2204\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016 Sep;32(18):2847\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHahne F, Ivanek R. Visualizing Genomic Data Using Gviz and Bioconductor. Methods Mol Biol. 2016;1418:335\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H. ggplot2: Elegant Graphics for Data Analysis [Internet]. Springer-Verlag New York; 2016. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ggplot2.tidyverse.org\u003c/span\u003e\u003cspan address=\"https://ggplot2.tidyverse.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreiman DM. GABAergic mechanisms in epilepsy. Epilepsia. 2001;42 Suppl 3:8\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaljevic S, M\u0026oslash;ller RS, Reid CA, P\u0026eacute;rez-Palma E, Lal D, May P, et al. Spectrum of GABAA receptor variants in epilepsy. Curr Opin Neurol. 2019 Apr;32(2):183\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang CS, Kavalali ET, Monteggia LM. BDNF signaling in context: From synaptic regulation to psychiatric disorders. Cell. 2022 Jan;185(1):62\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen WG, Chang Q, Lin Y, Meissner A, West AE, Griffith EC, et al. Derepression of BDNF transcription involves calcium-dependent phosphorylation of MeCP2. Science. 2003 Oct;302(5646):885\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDugich-Djordjevic MM, Tocco G, Willoughby DA, Najm I, Pasinetti G, Thompson RF, et al. BDNF mRNA expression in the developing rat brain following kainic acid-induced seizure activity. Neuron. 1992 Jun;8(6):1127\u0026ndash;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoulsen FR, Lauterborn J, Zimmer J, Gall CM. Differential expression of brain-derived neurotrophic factor transcripts after pilocarpine-induced seizure-like activity is related to mode of Ca2 + entry. Neuroscience. 2004;126(3):665\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang HN, Ko MC. Seizure activity involved in the up-regulation of BDNF mRNA expression by activation of central mu opioid receptors. Neuroscience. 2009 Jun;161(1):301\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMurray KD, Isackson PJ, Eskin TA, King MA, Montesinos SP, Abraham LA, et al. Altered mRNA expression for brain-derived neurotrophic factor and type II calcium/calmodulin-dependent protein kinase in the hippocampus of patients with intractable temporal lobe epilepsy. J Comp Neurol. 2000 Mar;418(4):411\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakahashi M, Hayashi S, Kakita A, Wakabayashi K, Fukuda M, Kameyama S, et al. Patients with temporal lobe epilepsy show an increase in brain-derived neurotrophic factor protein and its correlation with neuropeptide Y. Brain Res. 1999 Feb;818(2):579\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRyley Parrish R, Albertson AJ, Buckingham SC, Hablitz JJ, Mascia KL, Davis Haselden W, et al. Status epilepticus triggers early and late alterations in brain-derived neurotrophic factor and NMDA glutamate receptor Grin2b DNA methylation levels in the hippocampus. Neuroscience. 2013 Sep;248:602\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMart\u0026iacute;nez-Levy GA, Rocha L, Lubin FD, Alonso-Vanegas MA, Nani A, Buentello-Garc\u0026iacute;a RM, et al. Increased expression of BDNF transcript with exon VI in hippocampi of patients with pharmaco-resistant temporal lobe epilepsy. Neuroscience. 2016 Feb;314:12\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBandeira IC, Giombelli L, Werlang IC, Abujamra AL, Secchi TL, Brondani R, et al. Methylation of BDNF and SLC6A4 Gene Promoters in Brazilian Patients With Temporal Lobe Epilepsy Presenting or Not Psychiatric Comorbidities. Front Integr Neurosci. 2021;15:764742.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChatterton Z, Mendelev N, Chen S, Carr W, Kamimori GH, Ge Y, et al. Bisulfite Amplicon Sequencing Can Detect Glia and Neuron Cell-Free DNA in Blood Plasma. Front Mol Neurosci. 2021;14:672614.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlapirtti T, Jylh\u0026auml;v\u0026auml; J, Raitanen J, M\u0026auml;kinen R, Peltola J, Hurme MA, et al. The concentration of cell-free DNA in video-EEG patients is dependent on the epilepsy syndrome and duration of epilepsy. Neurol Res. 2016 Jan;38(1):45\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiimatainen SP, Jylh\u0026auml;v\u0026auml; J, Raitanen J, Peltola JT, Hurme MA. The concentration of cell-free DNA in focal epilepsy. Epilepsy Res. 2013 Aug;105(3):292\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen SY, Singhania R, Fehringer G, Chakravarthy A, Roehrl MHA, Chadwick D, et al. Sensitive tumour detection and classification using plasma cell-free DNA methylomes. Nature. 2018 Nov;563(7732):579\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHerrgott GA, Asmaro KP, Wells M, Sabedot TS, Malta TM, Mosella MS, et al. Detection of Tumor-specific DNA Methylation Markers in the Blood of Patients with Pituitary Neuroendocrine Tumors. Neuro Oncol. 2022 Feb;\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSabedot TS, Malta TM, Snyder J, Nelson K, Wells M, deCarvalho AC, et al. A serum-based DNA methylation assay provides accurate detection of glioma. Neuro Oncol. 2021 Sep;23(9):1494\u0026ndash;508.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLong H-Y, Feng L, Kang J, Luo Z-H, Xiao W-B, Long L-L, et al. Blood DNA methylation pattern is altered in mesial temporal lobe epilepsy. Sci Rep. 2017 Mar;7:43810.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTao H, Chen Z, Wu J, Chen J, Chen Y, Fu J, et al. DNA Methylation Signature of Epileptic Encephalopathy-Related Pathogenic Genes Encoding Ion Channels in Temporal Lobe Epilepsy. Front Neurol. 2021;12:692412.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBraun PR, Han S, Hing B, Nagahama Y, Gaul LN, Heinzman JT, et al. Genome-wide DNA methylation comparison between live human brain and peripheral tissues within individuals. Transl Psychiatry. 2019 Jan;9(1):47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSim\u0026oacute;n-S\u0026aacute;nchez J, Schulte C, Bras JM, Sharma M, Gibbs JR, Berg D, et al. Genome-wide association study reveals genetic risk underlying Parkinson\u0026rsquo;s disease. Nat Genet. 2009 Dec;41(12):1308\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanchez-Mut J V, Heyn H, Silva BA, Dixsaut L, Garcia-Esparcia P, Vidal E, et al. PM20D1 is a quantitative trait locus associated with Alzheimer\u0026rsquo;s disease. Nat Med. 2018 May;24(5):598\u0026ndash;603.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Q, Chen Y, Readhead B, Chen K, Su Y, Reiman EM, et al. Longitudinal data in peripheral blood confirm that PM20D1 is a quantitative trait locus (QTL) for Alzheimer\u0026rsquo;s disease and implicate its dynamic role in disease progression. Clin Epigenetics. 2020 Dec;12(1):189.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLie AA, Schr\u0026ouml;der R, Bl\u0026uuml;mcke I, Magin TM, Wiestler OD, Elger CE. Plectin in the human central nervous system: predominant expression at pia/glia and endothelia/glia interfaces. Acta Neuropathol. 1998 Sep;96(3):215\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePotokar M, Jorgačevski J. Plectin in the Central Nervous System and a Putative Role in Brain Astrocytes. Cells. 2021 Sep;10(9).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee T-S, Mane S, Eid T, Zhao H, Lin A, Guan Z, et al. Gene expression in temporal lobe epilepsy is consistent with increased release of glutamate by astrocytes. Mol Med. 2007;13(1\u0026ndash;2):1\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHori K, Shimaoka K, Hoshino M. AUTS2 Gene: Keys to Understanding the Pathogenesis of Neurodevelopmental Disorders. Cells. 2021 Dec;11(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao Z, Lee P, Stafford JM, von Schimmelmann M, Schaefer A, Reinberg D. An AUTS2-Polycomb complex activates gene expression in the CNS. Nature. 2014 Dec;516(7531):349\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLehmann-Werman R, Neiman D, Zemmour H, Moss J, Magenheim J, Vaknin-Dembinsky A, et al. Identification of tissue-specific cell death using methylation patterns of circulating DNA. Proc Natl Acad Sci U S A. 2016 Mar;113(13):E1826-34.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLubotzky A, Zemmour H, Neiman D, Gotkine M, Loyfer N, Piyanzin S, et al. Liquid biopsy reveals collateral tissue damage in cancer. JCI insight. 2022 Jan;7(2).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLubotzky A, Pelov I, Teplitz R, Neiman D, Smadja A, Zemmour H, et al. Elevated brain-derived cell-free DNA among patients with first psychotic episode - a proof-of-concept study. Elife. 2022 Jun;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaciagli L, Bernasconi A, Wiebe S, Koepp MJ, Bernasconi N, Bernhardt BC. A meta-analysis on progressive atrophy in intractable temporal lobe epilepsy: Time is brain? Neurology. 2017 Aug;89(5):506\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBahado-Singh RO, Radhakrishna U, Gordevičius J, Aydas B, Yilmaz A, Jafar F, et al. Artificial Intelligence and Circulating Cell-Free DNA Methylation Profiling: Mechanism and Detection of Alzheimer\u0026rsquo;s Disease. Cells. 2022 May;11(11).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"clinical-epigenetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"clep","sideBox":"Learn more about [Clinical Epigenetics](http://clinicalepigeneticsjournal.biomedcentral.com/)","snPcode":"13148","submissionUrl":"https://submission.nature.com/new-submission/13148/3","title":"Clinical Epigenetics","twitterHandle":"@OAgenetics","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Cell-free DNA, DNA methylation, epilepsy, biomarker","lastPublishedDoi":"10.21203/rs.3.rs-1940501/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1940501/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003e\u003c/p\u003e\u003cp\u003eDNA methylation profiling of circulating cell-free DNA (cfDNA) has rapidly become a promising strategy for biomarker identification and development. The cell-type-specific nature of DNA methylation patterns and the direct relationship between cfDNA and apoptosis can potentially be used non-invasively to predict local alterations. In addition, direct detection of altered DNA methylation patterns performs well as a biomarker. In a previous study, we demonstrated marked DNA methylation alterations in brain tissue from patients with mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS) patients. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eExperimental Design: \u003c/strong\u003e\u003c/p\u003e\u003cp\u003eWe performed DNA methylation profiling in cfDNA isolated from serum of MTLE patients and healthy controls using beadchip arrays followed by systematic bioinformatic analysis including deconvolution analysis and integration with DNase accessibility datasets. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Differential cfDNA methylation analysis showed overrepresentation of gene ontology terms and transcription factors related to central nervous system function and regulation. Deconvolution analysis of the DNA methylation datasets ruled out the possibility that the observed differences were due to changes in the proportional contribution of cortical neurons in cfDNA. Moreover, we found no overrepresentation of neuron- or glia-specific patterns in the described cfDNA methylation patterns. However, the MTLE-HS cfDNA methylation patterns featured significant overrepresentation of the epileptic DNA methylation alterations previously observed in hippocampus. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eOur results support the use of cfDNA methylation profiling as a rational approach to seeking non-invasive and reproducible epilepsy biomarkers.\u003c/p\u003e","manuscriptTitle":"Circulating cell-free DNA methylation mirrors alterations in cerebral patterns in epilepsy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-08-10 15:48:03","doi":"10.21203/rs.3.rs-1940501/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2022-11-17T15:28:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2022-11-17T14:58:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"f39f52ee-7fb4-4d04-aed8-e4a07014b8a1","date":"2022-11-04T12:06:16+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2022-10-25T15:11:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"8a3f6de5-2105-47b5-b120-21ff72696f42","date":"2022-10-03T11:40:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"5091551b-5556-4864-bc58-8fb99853d25b","date":"2022-09-16T13:39:46+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2022-09-16T12:44:00+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2022-08-08T18:08:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2022-08-08T10:43:05+00:00","index":"","fulltext":""},{"type":"submitted","content":"Clinical Epigenetics","date":"2022-08-08T09:09:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"clinical-epigenetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"clep","sideBox":"Learn more about [Clinical Epigenetics](http://clinicalepigeneticsjournal.biomedcentral.com/)","snPcode":"13148","submissionUrl":"https://submission.nature.com/new-submission/13148/3","title":"Clinical Epigenetics","twitterHandle":"@OAgenetics","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0076c2b3-040b-4b78-86d3-602d0f3274e3","owner":[],"postedDate":"August 10th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2023-10-16T18:14:15+00:00","versionOfRecord":{"articleIdentity":"rs-1940501","link":"https://doi.org/10.1186/s13148-022-01416-2","journal":{"identity":"clinical-epigenetics","isVorOnly":false,"title":"Clinical Epigenetics"},"publishedOn":"2022-12-28 18:07:27","publishedOnDateReadable":"December 28th, 2022"},"versionCreatedAt":"2022-08-10 15:48:03","video":"","vorDoi":"10.1186/s13148-022-01416-2","vorDoiUrl":"https://doi.org/10.1186/s13148-022-01416-2","workflowStages":[]},"version":"v1","identity":"rs-1940501","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1940501","identity":"rs-1940501","version":["v1"]},"buildId":"J0_U0BvcaRcwD8yVFaRlm","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}