Immune landscape of the affected brain in Rasmussen encephalitis

Immune landscape of the affected brain in Rasmussen encephalitis | 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 Article Immune landscape of the affected brain in Rasmussen encephalitis Giovanni Quinones-Valdez, Julia W. Chang, Shino D. Magaki, Harry V. Vinters, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8584794/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Background Rasmussen encephalitis (RE) is a rare neuroinflammatory disease characterized by intractable seizures and progressive brain atrophy that is usually confined to one cerebral hemisphere. Disease management involves anti-inflammatory, immune modulatory and anti-epileptic drugs, although surgical resection remains the only effective treatment option to achieve seizure freedom. The presence of clonally expanded resident memory T cells in brain tissue removed to control seizures suggests the involvement of an autoimmune response in the etiology of the disease. Methods Blocks of fresh brain tissue were obtained from three RE surgery cases (ages 5, 8, and 26 years at the time of surgery) and immune cells were isolated. Single cell RNA sequencing was used to define the types of immune cells present in the affected brain tissue and potential crosstalk between them, along with multiplex immunofluorescence immunostaining of sections from the same specimens. We matched T cell receptor sequences to T cell phenotypes and used ViralTrack software to search for evidence of activation of latent viruses in the immune cells. Results The immune cells isolated from the three RE cases comprised primarily activated microglia and resident memory CD8 T cells with fewer CD4 T cells, NK cells and monocyte-derived macrophages and dendritic cells. The majority of CD8 T cells expressed killer cell lectin-like receptors, and a virus responsive gene signature that included XCL1, TNFRSF9 and CRTAM, but also the exhaustion markers LAG3 and TIM3. Microglia expressed transcripts found in disease-associated microglia and transcripts associated with NLRP3 inflammasomes. We found no evidence for active latent viruses; however, we found endogenous HERV-K retrovirus sequences that were transcribed from multiple provirus insertion sites. Conclusions Our analysis highlights the complexity of the immune landscape in brain areas affected by RE and supports a central role for clonally expanded antigen experienced resident memory CD8 T cells. From the RNA sequencing data, we conclude that there is extensive cross talk between T cells and activated microglia, and monocyte-derived macrophages and dendritic cells that may regulate T cell activity. Biological sciences/Immunology Health sciences/Neurology Biological sciences/Neuroscience Rasmussen encephalitis T cells microglia single cell RNA sequencing endogenous retrovirus Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Background Rasmussen encephalitis (RE) is a rare neuroinflammatory disease characterized by seizures and progressive brain atrophy that is usually confined to one cerebral hemisphere ( 1 – 3 ), although bilateral cases have been described ( 4 ). In a recent retrospective study of 160 RE cases, the median age of clinical onset was seven years (range 1 to 53 years) with the left cerebral hemisphere more often affected ( 5 ). Disease management involves anti-inflammatory, immune modulatory, and anti-epileptic drugs (AEDs), although surgical resection remains the only effective treatment option to achieve seizure freedom ( 6 , 7 ). The presence of clonally focused T cells in resected involved brain tissue suggests that RE may have an autoimmune component ( 8 – 11 ), however, to date no T cell specific self-antigens have been identified. Comorbidity with several autoimmune diseases has indicated a potential heritable predisposition ( 12 ), and, in support of this idea, we identified HLA class I and II alleles in a cohort of 24 RE surgical cases that have been linked to susceptibility to autoimmune diseases ( 10 ). HLA class II nonsynonymous single nucleotide polymorphisms (SNP) were also found to be enriched in RE cases as well as several SNPs in other genes including those involved in an immune response ( 13 , 14 ). Takahashi et al. ( 15 ) identified SNPs in CTLA4 and PDCD1 that were overrepresented in a cohort of Japanese RE patients, which may negatively affect the function of these immune regulatory genes. Wang et al. identified a SNP in IFITM3 in Chinese RE patients that may promote human cytomegalovirus (HHV5) persistence in the brain ( 16 ). Epstein Barr virus (HHV4) has been proposed as a viral trigger for Multiple sclerosis (MS) ( 17 ), and HHV4, HHV5, and HHV6 sequences have been reported in some RE brain specimens but not in others ( 18 – 23 ). In support of a possible viral etiology, early treatment of two presumptive RE cases with ganciclovir was reported to stop seizures and resolve neurological deficits ( 23 , 24 ). We found that many of the expanded T cell clonotypes in RE brain specimens were public suggesting that they may have been directed against a common infectious agent ( 10 ). On the other hand, autoreactive T cell clonotypes may be present in healthy individuals ( 25 ). Co-existence of RE and focal cortical dysplasia (FCD) pathologies has also been documented in RE surgical cases suggesting that FCD could be involved in triggering the disease ( 26 – 28 ). FCD is caused by somatic mutations in neural progenitors that occur during early brain development resulting in focal seizures ( 29 ). It has been established that seizures can promote brain inflammation ( 30 , 31 ), suggesting that seizure activity per se could trigger an inflammatory cascade leading to an autoimmune response. From the recent large cohort retrospective study of RE cases ( 5 ), it was found that the number of RE patients with twin siblings was higher compared with the general population, leading the authors to suggest that the risks of complications associated with preterm births, which are more prevalent in twin pregnancies ( 32 ), may also be causative. For example, preterm birth is a risk factor for perinatal arterial ischemic stroke, which may in turn be linked to neuroinflammation and seizures ( 33 , 34 ). RNA sequencing of nuclei isolated from resected brain tissue confirmed earlier findings ( 35 , 36 ) and revealed the heterogeneity of activated microglia in affected brain areas ( 37 ). Based on the pathological stages of RE described by Pardo et al. ( 38 ), activation of microglia was shown to occur prior to T cell infiltration ( 39 ). In the present study we have sequenced RNA from individual immune cells isolated from three RE surgery cases and characterized the transcriptomes of both myeloid and lymphocyte populations including T cell receptors. We also present evidence for the reactivation of the endogenous retrovirus HERV-K in these cells. Methods and Materials This study was approved by the UCLA Institutional Review Board (IRB no. 18-001048). The patients or their parents or legal guardians provided informed consent for the use of the surgical remnant and blood for research purposes according to the Declaration of Helsinki. There were no exclusion criteria, and no reported comorbidities. All specimens were collected using the same standard operating procedures. De-identified patient information including age at seizure onset, age at surgery, and gender was collected with informed consent. Single cell RNA sequencing Immune cells were isolated from blocks of fresh resected brain tissue as previously described ( 40 ). In brief, brain tissue collected from the operating room in ice cold Hibernate® containing penicillin/streptomycin (120U/ml and 100ug/ml respectively, ThermoFisher, Carlsbad, CA) was transferred to magnesium calcium free Hanks Balanced Salt Solution containing HEPES (20mM), glucose (5mM) and penicillin/streptomycin (ThermoFisher), and finely minced on ice with spring scissors. Tissue fragments were transferred to RPM1 (ThermoFisher) containing 10% human serum (Phoenix Scientific, San Marcos, CA) and HEPES (20mM) and digested overnight at room temperature with Type IV collagenase (~ 800U/ml) (Worthington Biochemical Corp. Lakewood, NJ) followed by fractionation on a 70%:30% Percoll® (Millipore Sigma, St. Louis MO) step gradient. The immune cells were collected from the interface between the two Percoll® steps, and two Chromium single cell gene expression libraries (5’ gene expression and TCR V(D)J) were prepared (10X Genomics, Pleasanton, CA) and sequenced on a NovaSeq 6000 instrument in a SP flow cell (2x50 bp) (Illumina Inc., San Diego, CA). Reads were demultiplexed and aligned to the Genome Reference Consortium Human Reference 38 (GRCh Build 38) and VDJ reference data (based on Ensembl 94 release) using Cell Ranger count and Cell Ranger VDJ pipelines (10X Genomics). The single cell RNA sequencing (scRNA-seq) data from the 5’ expression library were analyzed using the R package SingCellaR with the default settings ( 41 ). The scRNA-seq data from each RE sample were combined using the integration function from Seurat ( 42 ). Clusters of putative T cells and NK cells (n = 10,971) were then extracted, reintegrated using Harmony ( 43 ), and re-clustered. Clusters of putative myeloid cells (n = 12,877) were also extracted, reintegrated and re-clustered. The R package Monocle 3.0 ( 44 ) was used to perform a trajectory analysis of the myeloid cells. The all_contig_annotations file from the Cell Ranger VDJ pipeline was used to assess the clonotype diversity in each RE sample, and to assign T cell phenotypes to the clonotypes based on the SingCellaR workflow. Potential cell: cell interactions based on the relative expression of known receptor ligand pairs were analyzed using the NATMI python-based toolkit ( 45 ). For each ligand: receptor pair a specificity estimate is calculated as the average expression of the ligand in the sending cell type divided by the average expression of the ligand in all cells multiplied by the average expression of the corresponding receptor by the receiving cell type divided by the average expression of the receptor in all cells. Spectratyping was performed with the immunarch R package ( 46 ), and diversity estimates were calculated using the iNEXT R package ( 47 ). Chord diagrams were made with the R package chorddiag ( 48 ). Barcodes of the most abundant clonotype(s) in each sample were integrated into the SingCellaR object for visualization in UMAP projections. Identification of Viral RNA using ViralTrack For each sample, processing of the raw RNA sequencing reads was carried out using Umitools ( 49 ). Initially, a whitelist of acceptable barcodes was generated, allowing for a Hamming distance of 1. Subsequently, the barcode sequences were extracted from the reads and incorporated into the read names. The next step involved mapping the reads to both the human genome (GRCh Build 38) and to a comprehensive viral genome database, encompassing over 14,000 distinct viral genomes. This mapping was performed utilizing the ViralTrack software package ( 50 ). Only viral genomes meeting specific criteria were retained for further analysis. These criteria included a minimum coverage of at least 50 reads covering at least 10% of the entire viral genome sequence, and a sequence complexity of 1.3, as indicated by base composition entropy. Viral hits meeting these criteria were then assembled into contigs, which were considered to represent active regions of transcription within the viral genome. Subsequently, only those reads that mapped to contigs with a length of 200 bases or longer were retained. A custom Python script was employed to parse the alignments to both the human and viral genomes. Reads that mapped to both the human genome and a viral genome were assigned to the organism showing the highest alignment score. Finally, the read count of viral hits was quantified for each barcode, and barcodes were categorized as expressing the virus if they contained three or more reads mapped to a viral genome. The resulting viral expression data were integrated into the SingCellaR object for visualization in UMAP projections. Identification of HERV-K provirus insertion sites We obtained the genomic coordinates of HERV-K provirus insertions from two previously published sources ( 51 , 52 ). Coordinates from the Xue et al. study, originally reported in the hg19 (GRCh37) genome build, were converted to the hg38 (GRCh38) assembly using the liftover Python package (v1 .3.2 ). Coordinates from both studies were then merged and manually curated to remove redundancies. To annotate the genomic context of each insertion, we used GENCODE v44 gene annotations. For each insertion, we recorded overlapping features including gene name, and region type (e.g., exon, intron, intergenic). Using a custom Python script, we quantified read coverage across the merged HERV-K insertion regions using the scRNA-seq data from the three RE patients, mapped to the reference hg38 (GRCh38) genome build. Reads were considered overlapping if they intersected a given region by at least 10 base pairs. We further filtered reads to retain only those also aligning to a reference HERV-K genome (NC_0022518) as determined from the prior ViralTrack analysis. Additional filtering criteria were applied to reduce mapping artifacts: retained reads were required to have an entropy score ≥ 1.0, an average Phred-scaled mapping quality ≥ 25, and a minimum aligned segment length of 20 bases. Chromosomal maps showing the sites of provirus insertion were made using the R package RIdeogram ( 53 ). B cell receptor sequencing Bulk genomic DNA was isolated from a frozen block of fresh brain tissue from RE patient 738, and from whole blood (Monarch® genomic DNA purification kit, New England Biolabs, Ipswich, MA). BCR sequences were obtained using the ImmunoSEQ® assay (Adaptive Biotechnologies, Seattle, WA), and clonal analysis was performed using the interactive web tool, ViCloD ( 54 ). Immunocytochemistry Paraffin embedded sections (5 um) were stained using Opal Fluorophore reagents (Akoya Biosciences, Marlborough, MA) and a Leica Bond RX auto-stainer (Leica Biosystems, Vista, CA) following antigen retrieval. The following antibodies were used CD20 (Opal 520), HLA-DR (Opal 570), CD4 (Opal 620) CD3 (Opal 690), and CD8 (Opal 780). Images were collected using a Leica Aperio Versa 200 Slide Scanning Microscope equipped with a 16-bit 5.5-megapixel fluorescence camera, and images were captured using Phenochart 2.0 (Akoya Biosciences), and colors were re-assigned. Images, as Tiff files, were imported into PHOTO-PAINT (Corel Corporation, Ottawa, Canada), to upscale the resolution to 600 dpi, and adjust tone curves. All figures were prepared in CorelDraw (Corel Corporation). Results Clinical and Pathological Description Three RE patients underwent surgery to control their drug-resistant seizures (Table 1 and Fig. S1 ). Post-operatively all three patients were seizure free although still taking AEDs. We obtained blood and fresh brain tissue from the surgical resections. Immune cells were isolated from each brain specimen and single cell cDNA libraries were constructed targeting 10,000 cells. Both whole transcriptome and T cell receptor sequences were obtained, and the transcriptomic data from each patient were integrated using canonical correlation analysis (Fig. S2 ). The identity of the cells in each cluster was assigned based on known marker genes (Fig. S2 ), and from these assignments, the proportions of the different cell types from the three brain specimens were calculated (Fig. 1 ). Myeloid cells mainly comprised microglia and were clearly distinguishable from T cells and NK cells, except for a mixed cluster derived primarily from patient 769, which we attribute to cell doublets (Cluster 9, Fig. S2 ). A high proportion of plasma cells was only evident in the brain specimen from patient 738, which we confirmed by multiplex immunostaining (Fig. S3 ). Table 1 Patient data. Case ID Gender Age at seizure onset (yr) Age at surgery (yr) Seizure frequency Affected hemisphere AEDS 738 male 21 26 ~ 15 per day right gabapentin lamotrigine levetiracetam 754 male 5 8 ~ 1 per day left brivaracetam clobazam clonazepam lacosamide 769 female 2 5 ~ 24 per day right lacosamide levetiracetam We used NATMI to estimate the potential cell: cell interactions between the different immune cell types. Interactions involving the small number of vascular cells were excluded. We filtered the results by limiting the number of interactions to cell types in which > 25% of the sending cells expressed the ligand and > 25% of the receiving cells expressed the corresponding receptor. A threshold of > 0.1 was set for the specificity estimate for both the ligand and the receptor (Fig. 2 ). This analysis predicted specific reciprocal interactions between myeloid cells and lymphocytes involving different ligand: receptor pairs. Based on the thresholds imposed on the data, TNFSF-13 signaling between macrophages and B cells, and IL1 signaling between microglia and macrophages were the most specific. T cell phenotypes The putative T cells and Natural Killer (NK) cells were extracted from the integrated dataset, and re-clustered to further resolve cell phenotypes (Fig. 3 A and B). The majority of cells in all 12 clusters expressed CD69 a marker of resident memory T cells (T RM ) ( 55 ); a variable number of cells in each cluster also expressed ITGAE (CD103), a second T RM marker ( 55 ) (Fig. 3 C). Based on differentially expressed genes, three of the 12 clusters contained CD4 T cells, of which cluster 5 was defined by the expression of IL7R and KLRB1, cluster 7 by FOXP3 expression and cluster 9 by KLRB1 and CXCL13 transcripts (Fig. 3 C). In agreement with this interpretation GSEA using gene signatures for different T cell phenotypes ( 56 ) indicated that cluster 5 contained CD4 memory T cells and cluster 7 comprised regulatory T cells (Tregs) (Table 2 ). Clusters 1–4, 6 and 10 comprised predominantly CD8 T cells, of which cluster 1 was further defined by high levels of XCL1 transcripts, cluster 3 by GMZK expression, and cluster 6 by high expression of heat shock proteins including HSP1A1 (Fig. 3 C). GSEA with gene signatures for T cell phenotypes ( 56 ), exhausted T cells ( 57 ), and for virus responsive T cells ( 58 ) indicated that cluster 1 CD8 T cells had a cytokine producing phenotype and were responsive to a virus whereas cluster 2 and 4 CD8 T cells were likely exhausted. Gamma delta T cells were found within the CD8 T cells clusters; presumably, their overall gene expression profiles were too similar to the alpha beta T cells (Fig. S4 ). Cluster 8 was defined by a higher level of SKI transcripts, a suppressor of CD103 expression in mice ( 59 ), cluster 11 by FCGR3A (CD16) expression, and cluster 12 by XCL1 expression. GSEA with a published Natural killer (NK) cell gene signature ( 60 ) indicated that clusters 11 and 12 contained NK cells (Table 2 ). In addition to the putative NK cells, the majority of CD8 T cells expressed KLRD1 (CD94) mRNA, which encodes an NK cell C-type lectin that forms heterodimers with NKG2 molecules to either inhibit (NKG2A/B) or activate (NKG2C and NKG2E) NK and CD8 T cells by binding to HLA-E molecules ( 61 ). Higher levels of transcripts encoding the activating receptors were found in the majority of NK cells and CD8 T cells compared with transcripts encoding the inhibitory receptor (Fig. S5 ). Similarly, KLRK1 encoding NKG2D, another activating NK cell C-type lectin-like receptor, was also highly expressed whereas KLRG1 encoding another inhibitory receptor was not (Fig. S5 ). Table 2 Gene set enrichment analysis. Cluster Gene signature Padj NES 1 CD8 cytokine 0.0261 1.8833 virus responsive 0.0041 1.8937 2 T exhausted cells 0.0021 1.8203 4 CD8 cytokine 0.0355 1.6735 T exhausted cells 0.0340 1.5864 5 CD4 naive/central memory 0.0015 1.8320 CD4/CD8 0.0034 1.7994 6 virus responsive 0.0490 1.8565 7 T reg cells 0.0017 1.6103 11 NK cells 0.0002 1.6981 CD8 cytotoxic 0.0005 1.6877 12 NK cells 0.0148 1.8039 Interrogating the dataset for expression of transcripts encoding co-stimulatory and co-inhibitory genes revealed that in contrast to CD4 T cells, LAG3 was the dominant co-inhibitory gene expressed by CD8 T cells (Fig. S6 ); far fewer cells expressed the co-inhibitory gene PDCD1, which were largely confined to cluster 9 CD4 T cells. CD4 Tregs (cluster 7) expressed CTLA4 and TIGIT, as well as the costimulatory genes ICOS , TNFRSF9 (4-1BB), CD27 and CD28 (Fig. S6 ). By contrast CD8 T cells expressed the co-stimulatory genes ICOS , CRTAM and TNFRSF9 , but essentially no CD27 or CD28 (Fig. S6 ). We also examined the expression of selected transcription factors. More cells expressed RUNX3, ZEB2, PRDM1 (Blimp), and ZNF683 (Hobit) mRNAs than Tox, TBX21 (Tbet) or TCF7 mRNAs (Fig. S7 ). However, fewer cells assigned to the CD4 T cell clusters ( 5 , 7 and 9 ) expressed ZEB2 and ZNF683 mRNA compared with cells assigned to the CD8 T cell clusters. Very few cells in the NK cell clusters (11 and 12) expressed BATF (Fig. S7 ). Clonally focused CD8 T cells We determined the number of different T cell clonotypes, defined by the Vbeta chain third complementarity region (CDR3) sequence, in the three RE brain specimens (Table S1 ). From Hill plots, TRBV and TRBJ gene usage, and CDR3 lengths it was clear that T cells in the brains of the three patients were clonally focused particularly in the brain specimens from patients 754 and 769 (Fig. S8 ). Matching the barcodes from the TCR libraries to those comprising the T cell clusters showed that the most frequent clonotypes (defined as > 1%) were found in the CD8 T cell clusters with one exception (Fig. 3 D). In patient 769 the most frequent clonotype could be ascribed to cluster 9 corresponding to a CD4 T cell subset defined by the expression of KLRB1 and CXCL13 (Fig. 3 E). In patients 738 and 754 the dominant clonotypes were CD8 T cells (Fig. 3 E). We confirmed the phenotypes of 49 of the 54 most frequent clonotypes by extracting the CD3D + / CD3E + CD4 + and CD8 + cells (≥ 1 normalized UMI) from the sparse matrix file of the reclustered T cell and NK cells and matching the barcodes to those of the top clonotypes (Table 1 S). To partition the top 1% into public and private clonotypes we compared them to Adaptive Biotechnologies’ immuneACCESS database (Table S1 ). As shown in Fig. 4 more than half of the abundant clonotypes were public. Vbeta chain CDR3 amino acid sequence and TRBV and TRBJ genes of a rare clonotype in the blood of a single donor from large scale covid study ( 62 ) was identical to the most abundant clonotype in patient 754 (Table S2 ). The CDR3 regions differed by three nucleotides indicating convergent selection ( 63 ). Likewise, the most abundant clonotype in patient 769 was found in five individuals from an unpublished study of celiac disease available in the immuneACCESS database, although TRBV gene usage differed (Table S2 ). We compared all the Vbeta CDR3 amino acid sequences to the VDJdb database ( 64 ) and found that there were exact matches including TRBV and TRBJ gene usage to TCRs that recognize common viral epitopes. All these matching clonotypes were rare with one exception (Table S3 ). We implemented ClusTCR ( 65 ) to identify clonotypes that may recognize the same epitopes. As shown in Table S4 , several clonotypes were found that likely recognize the same antigen, notably HHV4 (IVTDFSVIK), and HHV5 (KLGGALQK). Expansion of B cell clone s We found higher numbers of B cells and antibody-producing plasma cells in the brain specimen from patient 738 (Fig. 1 and Fig. S2 ). Bulk immunoglobulin heavy chain (IGH) sequences were obtained from a fresh frozen piece of tissue from the same surgical specimen, and from whole blood collected at the time of the surgery. Repertoire analysis of IGH sequences identified the six most frequent clones (> 1%) in the brain and showed the extent of intra-clonal diversity (Fig. 5 ). The most abundant clone comprised a single IGH sequence that was not detected in the sample of IGH sequences from the blood suggesting local clonal expansion (Fig. S9 ). Myeloid cell phenotypes Re-clustering the myeloid cells alone generated 14 clusters that were present in different proportions in each patient sample (Fig. 6 A and B). We interpreted 11 of the clusters as microglia based on the expression of the homeostatic glial marker genes TMEM119 , P2RY12 , OFML3 , SLC2A5 , and CXC3CR1 ( 66 – 68 ) (Fig. 6 C). From the analysis of differentially expressed genes, cluster 9 appeared to define a population of macrophages as evidenced by high expression of CD163 and FCGR2B (CD32) mRNAs by the majority of cells in the cluster ( 69 , 70 ) (Fig. 6 D). Cells in clusters 10 may comprise a dendritic cell population based on the expression of CD1c, CLEC10A and LGALS2 (Galectin-2) whereas cells in cluster 11 appeared to be a monocyte or monocyte-derived population based on VCAN (versican) expression ( 71 ). The majority of the cells comprising clusters 9–11 were derived from patient 738 (Fig. 6 B). The microglia populations shared genes expressed by disease-associated microglia, specifically, TREM2 , APOE , CD83 , and SPP1 (osteopontin) ( 68 , 72 , 73 ), as well as genes directly associated with the NPLR3 inflammasome pathway (Fig. 6 E) ( 35 , 74 ). All myeloid clusters expressed high levels of HLA II molecules (Fig. 6 E; see also Fig. S3 ). Cluster 3 microglia made up the largest fraction of myeloid cells in the brain specimen from patient 754, whereas cluster 2 microglia were the most abundant myeloid cells in patient 769 (Fig. 6 B). From the differential gene expression analysis, it appeared that cluster 3 microglia expressed higher levels DNAJB1 and CCL3 transcripts than cluster 2 microglia, whereas cluster 2 microglia expressed higher levels of AIF1 (IB1A), and FCGR3A (CD16) mRNA (Fig. 6 F). Clustering myeloid cells and trajectory analysis using Monocle 3.0 indicated that microglia cells may transition from CD83 − cells expressing more IBA1 and CD16 to CD83 + cells expressing cytokines and heat shock proteins (Figure S10 ). The NATMI analysis identified a number of ligand: receptor pairings between myeloid cells and T cells (Fig. 2 ). To determine whether any of these mapped to specific T cell subsets, we generated UMAPs comprising the normalized expression of the genes encoding these ligand: receptor pairs in the re-clustered populations of myeloid and T cells. As shown in Fig. 7 , Annexin A1 mRNA was expressed in all T cells except CD4 Tregs, whereas interferon gamma mRNA was found almost exclusively in CD8 T cells. ANXA1 transcripts were also expressed by cells in the macrophage/monocyte clusters 9–11 (Fig. S11 ). Formyl protein receptor 1 (FPR1), the cognate receptor was highly expressed in microglial cells except for cluster 12 microglia. FPR2 and FPR3 mRNAs were not expressed in the majority of T cells and myeloid cells (Fig. S11 ). The pairing of IL15 and its cognate receptor also reached the threshold of specificity that we imposed on the NATMI analysis. CD86 transcripts were found in almost all myeloid cells, whereas, the receptor, CTLA4, was only expressed by Tregs and some CD8 T cells (see also Fig. S6 ). A potential interaction between a second T cell co-inhibitory molecule LAG3, and LGALS3 (Galectin 3) on macrophages/monocytes also met the criteria we imposed on the NATMI analysis (Fig. 7 ). The transcript levels in myeloid cells of the ligands that bind the co-stimulatory and co-inhibitory receptors expressed by T cells are shown in Figure S12 . The interaction between TIM3 (HAVCR2) on T cells and Galectin 9 (LGALS9) on myeloid cells did not meet the criteria we imposed on the NATMI analysis. Although not identified in the NATMI analysis, it appeared that a subset of T cells, primarily assigned to cluster 5, expressed the G protein coupled receptor GPR183 (EB12), which binds an oxidized derivative of cholesterol, 7a, 25-dihydroxycholesterol. This oxysterol has been found to be a chemoattractant for Th17 CD4 T cells in MS ( 75 ). Formation of the oxysterol from cholesterol involves hydroxylation of cholesterol by cholesterol-25-hydroxylase (CH25H) ( 76 ). CH25H transcripts were present in microglia, primarily in clusters 3 and 8 (Fig. 7 ). Endogenous HERVK activation We used ViralTrack ( 50 ) to search for viral transcripts in the scRNA-seq data from each RE brain specimen. Based on the filtering steps applied in the analysis (see Methods and Materials) we did not find evidence for any active exogenous viruses in the dataset. However, we did find evidence for activity of the endogenous retrovirus, HERV-K. The filtered HERV-K reads from each RE case could be uniquely mapped to previously identified sites of provirus insertion ( 51 , 52 ) (Fig. 8 ; Table S5 ), indicating that multiple copies of HERV-K were transcriptionally active. Out of a total of 49 proviruses, 11 were in common between the three RE cases, and 13 were previously shown to be active in normal brain ( 52 ). Associating reads with individual barcodes and imposing a threshold of at least three reads per barcode showed that HERV-K was active in more immune cells from patient 754 compared with the other two patients (Fig. S13 ). As shown in Fig. 8 , the highest number of active proviruses was found in patient 754. Discussion We have used scRNA-seq to characterize the immune landscape in brain areas affected by RE from three surgery cases. Based on CD69 and CD103 antibody staining of T cells from other RE cases we previously reported that the majority of T cells in resected brain tissue were T RMs indicating that they had entered the brain at an earlier time point before the surgery in response to an immune challenge, and then remained there ( 77 ). In the present study we found that essentially all CD4 and CD8 T cells and NK cells expressed CD69 and variable levels of ITGAE (CD103) mRNA. We also documented the expression RUNX3, PRDM1 (Blimp-1), and ZNF683 (Hobit) transcripts, transcription factors that have been implicated in establishing T RMs ( 78 , 79 ). The bioinformatics pipeline employed identified distinct populations of CD4 and CD8 T cells. The major CD8 T cell populations, which accounted for most of the high frequency clonotypes, comprised cells that expressed genes associated with an effector phenotype, as evidenced by the expression of the granzyme gene GZMB , and the NK cell activating C-type lectin-like genes KRLD2 , KLRKC2 , KLRC3 and KLRK1 . Detection of IFNG mRNA in some CD8T cells was consonant with this interpretation. We also detected a population of CD8 T cells characterized by high levels of GZMK transcripts. Unlike Granzyme B, Granzyme K can induce non-caspase dependent cell death by cleaving both cytotoxic and non-cytotoxic intracellular proteins ( 80 ). GZMK + CD8 T cells have been found in tissues and lesions associated with several different autoimmune diseases and are assumed to be autoreactive ( 81 , 82 ). Exhausted T cells have been extensively described in chronic infections and cancers ( 83 ) and are characterized by high expression of co-inhibitory molecules including PD1, LAG3, TIM3, and TIGIT ( 84 ). The persistence of disease-triggering self-antigens in autoimmune diseases might also be expected to generate an exhausted or dysfunctional state. GSEA indicated that there were exhausted T cells among the CD8 T cells found in affected RE brain tissue; notably LAG3 transcripts were detected in about half the CD8 cells. LAG3 + T cells were recently documented in RE brain parenchyma by multiplex immunolabeling ( 85 ). Compared with LAG3 these investigators found that the number of T cells expressing PD1 was significantly higher in tissue sections from later clinical stages compared with early stages of the disease reflecting an increase in the proportion of exhausted T cells. This difference between clinical stages could explain why we found few T cells expressing PDCD1 transcripts in our dataset. In a mouse model of type 1 diabetes, it was shown that LAG3 + intra-islet CD8 T cells had an exhausted-like phenotype ( 86 ). Deleting Lag3 accelerated the development of diabetes, strongly implicating LAG3 + CD8 T cells in the disease ( 87 ). LAG3 binds MHC class II molecules, and also binds to other ligands including the product of the LGALS3 gene, Galectin 3 ( 88 ). We found MHC class II transcripts in essentially all myeloid cells, and NATMI predicted a specific interaction between LAG + CD8 T cells and LGALS3 + macrophage/monocyte cells, which suggests that the effector function of CD8 T cells might be held in check by these interactions. A LAG3 agonist antibody has been developed as a potential treatment for autoimmune diseases ( 89 ). NATMI predicted a potential ligand: receptor interaction between CD8 T cells and myeloid cells that involved VCAM-1 expression by CD8 T cells. VCAM1 can bind an integrin composed of integrin beta 2 (ITGB2) and integrin alpha X (ITGAX), both of which were expressed by myeloid cells suggesting direct binding to CD8 T cells. VCAM1 expression by CD8 T cells may reflect chronic exposure to antigen. It has recently been reported that VCAM1 is expressed by exhausted T cells and suppresses T cell activation in vitro by cis-binding TCR/CD3 complexes in the endomembrane system ( 90 ). Likewise, LAG3 has also been reported to bind TCR/CD3 complexes and inhibit T cell activation ( 91 ). We also found a potential interaction between Annexin A1 on T cells (except Tregs) and a N-formyl peptide receptor, FPR1, on myeloid cells, the only FRP that was highly expressed in myeloid cells (Fig. S11 ). Annexin A1 transcripts were also detected in macrophages (Fig. S11 ). Higher Annexin A1 expression has been documented in peripheral T cells from Rheumatoid arthritis and Systemic lupus erythematosus patients compared with healthy controls ( 92 , 93 ), and impaired activation of T cells and reduced severity of experimental autoimmune encephalitis was reported in Anxa1 −/− mice ( 94 ). It has been proposed that Annexin A1 may act in an autocrine manner in T cells ( 95 ), although likely not through FPRs in RE, since they were essentially absent from T cells (Fig. S11 ). Full length Annexin A1 and the N-terminal peptide of the protein, Ac2-26, are ligands for FPR1 although binding is associated with resolution of an inflammatory response ( 96 – 98 ). On the other hand, FPR1 binding to N-formyl peptides derived from bacterial and mitochondrial proteins is associated with an inflammatory response by innate immune cells including microglia ( 98 , 99 ). Microglia in affected areas of the brain may therefore be poised to be pro-resolving or pro-inflammatory depending on the ligands they are exposed to. There are a number of reports documenting the presence of active HHV4 and HHV5 in brain areas affected in RE ( 18 , 20 , 22 ). We found exact HLA-matched TRB sequences in the three RE brain specimens that have been shown to recognize immunodominant peptides from HHV4 and HHV5 indicating that the three patients may have been exposed to these viruses leading to a subclinical immune response and viral latency in circulating monocytes and B cells ( 100 – 102 ). Although the HHV4 and HHV5-specific T cell clonotypes did not correspond to the most abundant clonotypes in each surgery case, we considered the possibility that a reactivated herpes virus was present in peripheral immune cells present in the brain ( 103 ). We used ViralTrack to search for viral transcripts in our dataset but did not find any herpes virus transcripts. We cannot exclude the possibility of an active virus in other brain cells or that infected cells had already been eliminated leaving behind resident memory T cells ( 104 ). Over half of the frequent clonotypes (> 1%) were public suggesting that some of the expanded T cells in the brain might result from exposure to a common infectious agent, and raising the possibility of molecular mimicry if these T cells also recognize self-antigens ( 105 ). In patient 769 the Vbeta CDR3 amino acid sequence of the most abundant clonotype only matched a rare clonotype in the blood of seven individuals diagnosed with celiac disease, and in patient 754 the Vbeta CDR3 sequence and TRB and TRJ genes exactly matched a rare clonotype in the blood of a single uninfected individual from a recent SARS-CoV-2 study (Table S2 ) ( 62 ). Given the limited sharing of these two clonotypes with other individuals it is possible that they may have escaped peripheral tolerance mechanisms and be directly autoreactive. In our search for viral transcripts, we found HERV-K sequences in our scRNA-seq data indicating that this endogenous retrovirus was activated in immune cells from each RE patient. Mapping the virus contigs in individual cells showed that more microglia expressed HERV-K in Patient 754 than in the other two patients. It has been reported that the presence of HERV-K transcripts in glioblastomas, teratoid rhabdoid tumors and in spinal cord neurons of ALS patients has potential disease-modifying effects ( 106 – 108 ). Among the eleven proviruses active in all three patients, the provirus at Chr7p22.1a, ERK-6 (HERV-K108) contains full length open reading frames for the gagpol and envelope (env) genes; a mutation in the reverse transcriptase domain renders the virus defective ( 109 ). The envelope gene from this provirus has been shown to be translated and expressed on the cell surface of transfected cells ( 110 ). Whether the env protein is expressed in affected brain areas in RE remains to be determined. It has been reported that a GU-rich sequence in the env gene RNA can trigger TLR8-dependent death of neurons indicating that any pathological consequence of HERV-K activation might not depend on a translated open reading frame ( 111 ). We should note that HERV-K sequences, including 13 we identified in the RE specimens, have also been found in transcript data from normal brain samples in the GTEx database including ERK-6 ( 52 , 112 ). CD4 T cells could be assigned to three different phenotypes Tregs, IL7R + memory cells, and a population of T cells distinguished by the expression of CXCL13 and KLRB1 (CD161) transcripts. CXCL13 is a chemoattractant for B cells, and CD4 cells producing this cytokine have been associated with the formation of tertiary lymphoid structures (TLS) ( 113 ). In Figure S3 , a cluster of CD4 + T cells surrounded by HLA-DR + myeloid cells could correspond to this CD4 T cell subtype and define a developing TLS. Such ectopic structures have been found in affected tissues in autoimmune diseases ( 114 ), and CD4 T cells producing CXCL13 are present in synovial fluid in Rheumatoid arthritis patients ( 115 ). In patient 769 the most frequent clonotype was in fact this CD4 T cell subtype (Table S2 ). On the other hand, we found the highest number of B cells and plasma cells in patient 738. Sequencing immunoglobulin heavy chains revealed that these cells comprised expanded clones implying a specific humoral response to an antigen(s) in the brain of this patient. Patient 738 was the oldest surgery case at 26 years old, and seizure onset was documented five years earlier according to his clinical history (Table 1 ). However, the patient also experienced seizures when he was four years old and was treated with AEDs for two years. We speculate that the onset of seizures later in life may reflect a reactivation of resident T cells and a de novo humoral response. Autoantibodies directed against synaptic proteins have been found in the blood from some RE patients ( 116 – 118 ), which has led to the use of an anti-CD20 antibody to try to control the disease ( 119 – 122 ). TIGIT expression by Tregs suggested that they were likely to be highly suppressive effector Tregs ( 123 ). Therefore, their presence in the affected areas of the brain may play a pivotal role in suppressing the activity of the clonally focused CD8 T cells found in the same brain areas. It has been proposed that Treg dysfunction contributes to the etiology of autoimmune diseases ( 124 , 125 ), and ways to increase Treg numbers and efficacy in autoimmune diseases are being actively pursued ( 125 – 127 ). In addition to TIGIT, our analysis indicated that Tregs were regulated by several other co-stimulatory and co-inhibitory proteins including ICOS, CD27, CD28, and CTLA4 (Figure S6 ). CTLA4 is constitutively expressed by Tregs and competes for binding to CD86 and CD80 on antigen presenting cells to suppress conventional T cells by cell intrinsic and extrinsic mechanisms ( 128 , 129 ), that may also involve engagement of the TCR ( 130 ). Compared with ligands for other costimulatory and co-inhibitory genes including CD80, we detected CD86 transcripts in the majority of myeloid cells suggesting that the myeloid cells may regulate Treg cells present in the affected brain area via CTLA4 and CD28 binding to CD86 (Figure S12 ) ( 131 ). Most of myeloid cells that we isolated from affected brain tissue were microglia and were distinguished from smaller populations of macrophages and monocyte-derived immune cells. Based on the default resolution parameter used, ten clusters of microglia were generated by the Louvain algorithm, although all cells expressed known homeostatic microglia marker genes, and genes associated with activated microglia. Trajectory analysis indicated that microglia transitioned from cells expressing higher transcript levels of genes associated with phagocytosis, FCGR3A (CD16) and AIF1 (Ib1a), to cells expressing higher levels of proinflammatory cytokines, CCL3/4 and IL1B, and CD83 to cells expressing CD83 and heat shock protein transcripts, although the latter might be attributable to stress induced during ex vivo dissociation of the brain tissue ( 132 ). On the other hand, a subpopulation of microglia characterized by CD83 and heat shock protein gene expression was found to be selectively depleted in the substantia nigra of Parkinson disease patients implying a protective function that may also pertain to RE ( 133 ). Most microglia expressed HLA class II molecules, ITAGX (CD11c), and CD86 transcripts indicating competency as antigen presenting cells ( 134 , 135 ). As previously discussed, microglia could regulate CD8 T cell activity via LAG3 binding to HLA class II molecules, which is independent of the bound peptide ( 136 ), and via LGALS3 binding on the macrophages. A higher number of macrophages and monocyte-derived immune cells were present in the sample of brain tissue from patient 738, the adult surgery case in which we also found the highest number of plasma cells possibly indicting a greater de novo involvement of the peripheral immune system at the time of the surgery (Fig. 1 ). Conclusion Our analysis highlights the complexity of the immune landscape in brain areas affected by RE and supports the involvement of clonally expanded antigen experienced resident memory CD8 T cells in the etiology of RE. Crosstalk between T cells and activated microglia, and monocyte-derived macrophages and dendritic cells is predicted from the scRNA-seq data. The expression of heat shock proteins and CD83 transcripts in some microglia indicates a potential protective function. The presence of resident Tregs in the affected brain area suggests that they may be playing a role in restraining the activity of the conventional T cells. Likewise, microglia and monocyte-derived cells may regulate T cell activity via CD86, Galectin 3 and Galectin 9. Activation of HERV-K proviruses in the affected brain area may be unrelated to the etiology of RE although it is conceivable that translation of viral proteins from this endogenous virus could generate neoantigens to which the patient’s immune system reacts. Abbreviations RE, Rasmussen encephalitis FCD, Focal cortical dysplasia MS, Multiple sclerosis AED, Anti-epileptic drug scRNA-seq, single cell RNA sequencing UMI, Unique Molecular Identifiers UMAP, Uniform Manifold Approximation and Projection NATMI, Network Analysis Toolkit for Multicellular Interactions HLA, Human Leukocyte Antigen HHV, Human herpes virus TCR, T cell receptor CDR3, Complementarity determining region 3 BCR, B cell receptor IgH, Immunoglobulin heavy chain Declarations Ethics approval and consent to participate This study was approved by the UCLA Institutional Review Board (IRB no. 18-001048). The patients or their parents or legal guardians provided informed consent for the use of the surgical remnant and blood for research purposes according to the Declaration of Helsinki. De-identified patient information including age at seizure onset, age at surgery, and gender was collected with informed consent. Consent for publication The patients or their parents or legal guardians provided informed consent for future publication or presentation at conferences of de-identified research results. Availability of data and materials NCBI Gene Expression Omnibus accession number GSE312319 Competing interests The authors declare that they have no competing interests. Funding This work was funded by the RE Children’s Project to GCO and in part by a DOD/CDMRP Rare Cancers Research Program Idea Award (RA220179), The Lindonlight Collective grant (GR-240004), and an NIH/NCI award (P50 CA211015) to ACW. GQ-V received support as a member of The Collaboratory at the UCLA Institute for Quantitative and Computational Biosciences. Authors' contributions GCO designed the study, GCO and GQ-V analyzed the data and wrote the paper, JWC processed tissue and blood specimens, AF provided the surgical specimens, HVV and SDM provided pathologically evaluated tissue sections, NS provided MRI scans, and ACW provided additional funding for the study. All authors reviewed a draft of the manuscript. Acknowledgements Single cell RNA sequencing was performed by the Technology Center for Genomics and Bioinformatics, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, and multiplex immunofluorescence staining was carried out by the Translational Pathology Core Laboratory, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine. References Bien, C. G. et al. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis: a European consensus statement. Brain 128 (Pt 3), 454–471 (2005). Olson, H. E. et al. Clinical application and evaluation of the Bien diagnostic criteria for Rasmussen encephalitis. Epilepsia 54 (10), 1753–1760 (2013). Varadkar, S. et al. Rasmussen's encephalitis: clinical features, pathobiology, and treatment advances. Lancet Neurol. 13 (2), 195–205 (2014). Peariso, K. et al. Presentation, diagnosis and treatment of bilateral Rasmussen's encephalitis in a 12-year-old female. Epileptic Disord . 15 (3), 324–332 (2013). Fauser, S., Elger, C. E., Woermann, F. & Bien, C. G. Rasmussen encephalitis: Predisposing factors and their potential role in unilaterality. Epilepsia 63 (1), 108–119 (2022). Orsini, A. et al. Rasmussen's encephalitis: From immune pathogenesis towards targeted-therapy. Seizure 81 , 76–83 (2020). Harris, W. B. et al. Seizure outcomes in children with Rasmussen's encephalitis undergoing resective or hemispheric epilepsy surgery: an individual participant data meta-analysis. J. Neurosurg. Pediatr. 25 (3), 274–283 (2020). Schwab, N. et al. CD8 + T-cell clones dominate brain infiltrates in Rasmussen encephalitis and persist in the periphery. Brain 132 (Pt 5), 1236–1246 (2009). Schneider-Hohendorf, T. et al. CD8(+) T-cell pathogenicity in Rasmussen encephalitis elucidated by large-scale T-cell receptor sequencing. Nat. Commun. 7 , 11153 (2016). Dandekar, S. et al. Shared HLA Class I and II Alleles and Clonally Restricted Public and Private Brain-Infiltrating alphabeta T Cells in a Cohort of Rasmussen Encephalitis Surgery Patients. Front. Immunol. 7 , 608 (2016). Al Nimer, F. et al. Phenotypic and functional complexity of brain-infiltrating T cells in Rasmussen encephalitis. Neurol. Neuroimmunol. Neuroinflamm . 5 (1), e419 (2018). Amrom, D. et al. Rasmussen encephalitis and comorbid autoimmune diseases: A window into disease mechanism? Neurology 83 (12), 1049–1055 (2014). Leitner, D. F. et al. Brain molecular mechanisms in Rasmussen encephalitis. Epilepsia 64 (1), 218–230 (2023). Ai, J. et al. Genetic Factors in Rasmussen's Encephalitis Characterized by Whole-Exome Sequencing. Front. Neurosci. 15 , 744429 (2021). Takahashi, Y. et al. Genetic variations of immunoregulatory genes associated with Rasmussen syndrome. Epilepsy Res. 107 (3), 238–243 (2013). Wang, Y. S. et al. HCMV infection and IFITM3 rs12252 are associated with Rasmussen's encephalitis disease progression. Ann. Clin. Transl Neurol. 8 (3), 558–570 (2021). Bjornevik, K., Munz, C., Cohen, J. I. & Ascherio, A. Epstein-Barr virus as a leading cause of Multiple sclerosis: mechanisms and implications. Nat. Rev. Neurol. 19 (3), 160–171 (2023). Power, C., Poland, S. D., Blume, W. T., Girvin, J. P. & Rice, G. P. Cytomegalovirus and Rasmussen's encephalitis. Lancet 336 (8726), 1282–1284 (1990). Farrell, M. A., Cheng, L., Cornford, M. E., Grody, W. W. & Vinters, H. V. Cytomegalovirus and Rasmussen's encephalitis. Lancet 337 (8756), 1551–1552 (1991). Vinters, H. V., Wang, R. & Wiley, C. A. Herpesviruses in chronic encephalitis associated with intractable childhood epilepsy. Hum. Pathol. 24 (8), 871–879 (1993). Atkins, M. R., Terrell, W. & Hulette, C. M. Rasmussen's syndrome: a study of potential viral etiology. Clin. Neuropathol. 14 (1), 7–12 (1995). Liu, D. et al. Detection of EBV and HHV6 in the Brain Tissue of Patients with Rasmussen's Encephalitis. Virol. Sin . 33 (5), 402–409 (2018). McLachlan, R. S., Levin, S. & Blume, W. T. Treatment of Rasmussen's syndrome with ganciclovir. Neurology 47 (4), 925–928 (1996). McLachlan, R. S., Diosy, D. & Levin, S. Early treatment of a progressive Rasmussen's like syndrome with ganciclovir. Can. J. Neurol. Sci. 38 (2), 296–298 (2011). Eugster, A. et al. Physiological and pathogenic T cell autoreactivity converge in type 1 diabetes. Nat. Commun. 15 (1), 9204 (2024). Hart, Y. M. et al. Double pathology in Rasmussen's syndrome: a window on the etiology? Neurology 50 (3), 731–735 (1998). Takei, H. et al. Dual pathology in Rasmussen's encephalitis: a study of seven cases and review of the literature. Neuropathology 30 (4), 381–391 (2010). Galev, G. & Prayson, R. A. Focal cortical dysplasia is a frequent coexistent pathology in patients with Rasmussen's encephalitis. Ann. Diagn. Pathol. 68 , 152224 (2024). Bast, T., Ramantani, G., Seitz, A. & Rating, D. Focal cortical dysplasia: prevalence, clinical presentation and epilepsy in children and adults. Acta Neurol. Scand. 113 (2), 72–81 (2006). Vezzani, A., Aronica, E., Mazarati, A. & Pittman, Q. J. Epilepsy and brain inflammation. Exp. Neurol. 244 , 11–21 (2013). Kumar, P. et al. Single-cell transcriptomics and surface epitope detection in human brain epileptic lesions identifies pro-inflammatory signaling. Nat. Neurosci. 25 (7), 956–966 (2022). Tingleff, T. et al. Different pathways for preterm birth between singleton and twin pregnancies: a population-based registry study of 481 176 nulliparous women. BJOG 130 (4), 387–395 (2023). Hagberg, H. et al. The role of inflammation in perinatal brain injury. Nat. Rev. Neurol. 11 (4), 192–208 (2015). Rattani, A. et al. Incidence of Epilepsy and Associated Risk Factors in Perinatal Ischemic Stroke Survivors. Pediatr. Neurol. 90 , 44–55 (2019). Ramaswamy, V. et al. Inflammasome induction in Rasmussen's encephalitis: cortical and associated white matter pathogenesis. J. Neuroinflammation . 10 , 152 (2013). Wirenfeldt, M. et al. Increased activation of Iba1 + microglia in pediatric epilepsy patients with Rasmussen's encephalitis compared with cortical dysplasia and tuberous sclerosis complex. Neurobiol. Dis. 34 (3), 432–440 (2009). Westfall, J. J. et al. Molecular and spatial heterogeneity of microglia in Rasmussen encephalitis. Acta Neuropathol. Commun. 10 (1), 168 (2022). Pardo, C. A. et al. The pathology of Rasmussen syndrome: stages of cortical involvement and neuropathological studies in 45 hemispherectomies. Epilepsia 45 (5), 516–526 (2004). Troscher, A. R. et al. Microglial nodules provide the environment for pathogenic T cells in human encephalitis. Acta Neuropathol. 137 (4), 619–635 (2019). Owens, G. C. et al. Evidence for the involvement of gamma delta T cells in the immune response in Rasmussen encephalitis. J. Neuroinflammation . 12 (1), 134 (2015). Wang, G. et al. Processing single-cell RNA-seq datasets using SingCellaR. STAR. Protoc. 3 (2), 101266 (2022). Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184 (13), 3573–87e29 (2021). Korsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat. Methods . 16 (12), 1289–1296 (2019). Cao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. Nature 566 (7745), 496–502 (2019). Hou, R., Denisenko, E., Ong, H. T., Ramilowski, J. A. & Forrest, A. R. R. Predicting cell-to-cell communication networks using NATMI. Nat. Commun. 11 (1), 5011 (2020). Team, I. immunarch: An R Package for Painless Bioinformatics Analysis of T-Cell and B-Cell Immune Repertoires 2019 [Available from: http://doi.org/10.5281/zenodo.3367200 Hsieh, T. C., Ma, K. H. & Chao, A. iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods Ecol. Evol. 7 (12), 1451–1456 (2016). Flor, M. R interface to D3 chord diagrams 2020 [Available from: https://github.com/mattflor/chorddiag Smith, T., Heger, A. & Sudbery, I. UMI-tools: modeling sequencing errors in Unique Molecular Identifiers to improve quantification accuracy. Genome Res. 27 (3), 491–499 (2017). Bost, P. et al. Host-Viral Infection Maps Reveal Signatures of Severe COVID-19 Patients. Cell 181 (7), 1475–88e12 (2020). Xue, B. et al. Identification of the distribution of human endogenous retroviruses K (HML-2) by PCR-based target enrichment sequencing. Retrovirology 17 (1), 10 (2020). Burn, A., Roy, F., Freeman, M. & Coffin, J. M. Widespread expression of the ancient HERV-K (HML-2) provirus group in normal human tissues. PLoS Biol. 20 (10), e3001826 (2022). Hao, Z. et al. RIdeogram: drawing SVG graphics to visualize and map genome-wide data on the idiograms. PeerJ Comput. Sci. 6 , e251 (2020). Jeusset, L. et al. ViCloD, an interactive web tool for visualizing B cell repertoires and analyzing intraclonal diversities: application to human B-cell tumors. NAR Genom Bioinform . 5 (2), lqad064 (2023). Christo, S. N., Park, S. L., Mueller, S. N. & Mackay, L. K. The Multifaceted Role of Tissue-Resident Memory T Cells. Annu. Rev. Immunol. 42 (1), 317–345 (2024). Szabo, P. A. et al. Single-cell transcriptomics of human T cells reveals tissue and activation signatures in health and disease. Nat. Commun. 10 (1), 4706 (2019). Zhang, C. et al. Prioritizing exhausted T cell marker genes highlights immune subtypes in pan-cancer. iScience 26 (4), 106484 (2023). Fuchs, Y. F. et al. Gene Expression-Based Identification of Antigen-Responsive CD8(+) T Cells on a Single-Cell Level. Front. Immunol. 10 , 2568 (2019). Wu, B. et al. The SKI proto-oncogene restrains the resident CD103(+)CD8(+) T cell response in viral clearance. Cell. Mol. Immunol. 18 (10), 2410–2421 (2021). Shembrey, C., Foroutan, M. & Hollande, F. A new natural killer cell-specific gene signature predicting recurrence in colorectal cancer patients. Front. Immunol. 13 , 1011247 (2022). Middleton, D., Curran, M. & Maxwell, L. Natural killer cells and their receptors. Transpl. Immunol. 10 (2–3), 147–164 (2002). Gittelman, R. M. et al. Longitudinal analysis of T cell receptor repertoires reveals shared patterns of antigen-specific response to SARS-CoV-2 infection. JCI Insight ; 7 (10). (2022). Venturi, V., Price, D. A., Douek, D. C. & Davenport, M. P. The molecular basis for public T-cell responses? Nat. Rev. Immunol. 8 (3), 231–238 (2008). Bagaev, D. V. et al. VDJdb in 2019: database extension, new analysis infrastructure and a T-cell receptor motif compendium. Nucleic Acids Res. 48 (D1), D1057–D62 (2020). Valkiers, S., Van Houcke, M., Laukens, K. & Meysman, P. ClusTCR: a Python interface for rapid clustering of large sets of CDR3 sequences with unknown antigen specificity. Bioinformatics (2021). Patir, A., Shih, B., McColl, B. W. & Freeman, T. C. A core transcriptional signature of human microglia: Derivation and utility in describing region-dependent alterations associated with Alzheimer's disease. Glia 67 (7), 1240–1253 (2019). Sankowski, R. et al. Mapping microglia states in the human brain through the integration of high-dimensional techniques. Nat. Neurosci. 22 (12), 2098–2110 (2019). Martins-Ferreira, R. et al. The Human Microglia Atlas (HuMicA) unravels changes in disease-associated microglia subsets across neurodegenerative conditions. Nat. Commun. 16 (1), 739 (2025). Kim, W. K. et al. CD163 identifies perivascular macrophages in normal and viral encephalitic brains and potential precursors to perivascular macrophages in blood. Am. J. Pathol. 168 (3), 822–834 (2006). Wen, W., Cheng, J. & Tang, Y. Brain perivascular macrophages: current understanding and future prospects. Brain 147 (1), 39–55 (2024). Villar, J. & Segura, E. Decoding the Heterogeneity of Human Dendritic Cell Subsets. Trends Immunol. 41 (12), 1062–1071 (2020). Holtman, I. R. et al. Induction of a common microglia gene expression signature by aging and neurodegenerative conditions: a co-expression meta-analysis. Acta Neuropathol. Commun. 3 , 31 (2015). Masuda, T., Sankowski, R., Staszewski, O. & Prinz, M. Microglia Heterogeneity in the Single-Cell Era. Cell. Rep. 30 (5), 1271–1281 (2020). Paik, S. et al. Updated insights into the molecular networks for NLRP3 inflammasome activation. Cell. Mol. Immunol. 22 (6), 563–596 (2025). Wanke, F. et al. EBI2 Is Highly Expressed in Multiple Sclerosis Lesions and Promotes Early CNS Migration of Encephalitogenic CD4 T Cells. Cell. Rep. 18 (5), 1270–1284 (2017). Kurschus, F. C. & Wanke, F. EBI2 - Sensor for dihydroxycholesterol gradients in neuroinflammation. Biochimie 153 , 52–55 (2018). Owens, G. C. et al. Evidence for Resident Memory T Cells in Rasmussen Encephalitis. Front. Immunol. 7 , 64 (2016). Mackay, L. K. et al. Hobit and Blimp1 instruct a universal transcriptional program of tissue residency in lymphocytes. Science 352 (6284), 459–463 (2016). Fonseca, R. et al. Runx3 drives a CD8(+) T cell tissue residency program that is absent in CD4(+) T cells. Nat. Immunol. 23 (8), 1236–1245 (2022). Bouwman, A. C., van Daalen, K. R., Crnko, S., Ten Broeke, T. & Bovenschen, N. Intracellular and Extracellular Roles of Granzyme K. Front. Immunol. 12 , 677707 (2021). Arazi, A. et al. The immune cell landscape in kidneys of patients with lupus nephritis. Nat. Immunol. 20 (7), 902–914 (2019). Jonsson, A. H. et al. Granzyme K(+) CD8 T cells form a core population in inflamed human tissue. Sci. Transl Med. 14 (649), eabo0686 (2022). McLane, L. M., Abdel-Hakeem, M. S. & Wherry, E. J. CD8 T Cell Exhaustion During Chronic Viral Infection and Cancer. Annu. Rev. Immunol. 37 , 457–495 (2019). Baessler, A. & Vignali, D. A. A. T Cell Exhaustion. Annu. Rev. Immunol. 42 (1), 179–206 (2024). Mair, K. M. et al. The dynamics of brain T cell populations during the course of rasmussen encephalitis: from expansion to exhaustion. J. Neuroinflammation . 22 (1), 155 (2025). Grebinoski, S. et al. Autoreactive CD8(+) T cells are restrained by an exhaustion-like program that is maintained by LAG3. Nat. Immunol. 23 (6), 868–877 (2022). Bettini, M. et al. Cutting edge: accelerated autoimmune diabetes in the absence of LAG-3. J. Immunol. 187 (7), 3493–3498 (2011). Burnell, S. E. A. et al. Seven mysteries of LAG-3: a multi-faceted immune receptor of increasing complexity. Immunother Adv. 2 (1), ltab025 (2022). Angin, M., Brignone, C. & Triebel, F. A LAG-3-Specific Agonist Antibody for the Treatment of T Cell-Induced Autoimmune Diseases. J. Immunol. 204 (4), 810–818 (2020). Naoi, Y. et al. CD106 in Tumor-Specific Exhausted CD8 + T Cells Mediates Immunosuppression by Inhibiting TCR Signaling. Cancer Res. 84 (13), 2109–2122 (2024). Guy, C. et al. LAG3 associates with TCR-CD3 complexes and suppresses signaling by driving co-receptor-Lck dissociation. Nat. Immunol. 23 (5), 757–767 (2022). Mihaylova, N. et al. Suppression of autoreactive T and B lymphocytes by anti-annexin A1 antibody in a humanized NSG murine model of systemic lupus erythematosus. Clin. Exp. Immunol. 199 (3), 278–293 (2020). D'Acquisto, F. et al. Annexin-1 modulates T-cell activation and differentiation. Blood 109 (3), 1095–1102 (2007). Paschalidis, N. et al. Modulation of experimental autoimmune encephalomyelitis by endogenous annexin A1. J. Neuroinflammation . 6 , 33 (2009). D'Acquisto, F. On the adaptive nature of annexin-A1. Curr. Opin. Pharmacol. 9 (4), 521–528 (2009). Ernst, S. et al. An annexin 1 N-terminal peptide activates leukocytes by triggering different members of the formyl peptide receptor family. J. Immunol. 172 (12), 7669–7676 (2004). Gavins, F. N. & Hickey, M. J. Annexin A1 and the regulation of innate and adaptive immunity. Front. Immunol. 3 , 354 (2012). Napolitano, F. & Montuori, N. The N-formyl peptide receptors: much more than chemoattractant receptors. Relevance in health and disease. Front. Immunol. 16 , 1568629 (2025). Li, Z. et al. Formyl peptide receptor 1 signaling potentiates inflammatory brain injury. Sci. Transl Med. 13 , 605 (2021). Sinclair, J. Human cytomegalovirus: Latency and reactivation in the myeloid lineage. J. Clin. Virol. 41 (3), 180–185 (2008). Savard, M. et al. Infection of primary human monocytes by Epstein-Barr virus. J. Virol. 74 (6), 2612–2619 (2000). Babcock, G. J., Decker, L. L., Volk, M. & Thorley-Lawson, D. A. EBV persistence in memory B cells in vivo. Immunity 9 (3), 395–404 (1998). Rocamonde, B., Hasan, U., Mathieu, C. & Dutartre, H. Viral-induced neuroinflammation: Different mechanisms converging to similar exacerbated glial responses. Front. Neurosci. 17 , 1108212 (2023). Wakim, L. M., Woodward-Davis, A. & Bevan, M. J. Memory T cells persisting within the brain after local infection show functional adaptations to their tissue of residence. Proc. Natl. Acad. Sci. U S A . 107 (42), 17872–17879 (2010). Takahashi, Y. et al. Vaccination and infection as causative factors in Japanese patients with Rasmussen syndrome: molecular mimicry and HLA class I. Clin. Dev. Immunol. 13 (2–4), 381–387 (2006). Pasternack, N. et al. Endogenous retroviruses are dysregulated in ALS. iScience 27 (7), 110147 (2024). Shah, A. H. et al. Human endogenous retrovirus K contributes to a stem cell niche in glioblastoma. J. Clin. Invest. ; 133 (13). (2023). Doucet-O'Hare, T. T. et al. SMARCB1 deletion in atypical teratoid rhabdoid tumors results in human endogenous retrovirus K (HML-2) expression. Sci. Rep. 11 (1), 12893 (2021). Mayer, J. et al. An almost-intact human endogenous retrovirus K on human chromosome 7. Nat. Genet. 21 (3), 257–258 (1999). Dewannieux, M., Blaise, S. & Heidmann, T. Identification of a functional envelope protein from the HERV-K family of human endogenous retroviruses. J. Virol. 79 (24), 15573–15577 (2005). Dembny, P. et al. Human endogenous retrovirus HERV-K(HML-2) RNA causes neurodegeneration through Toll-like receptors. JCI Insight ; 5 (7). (2020). Nevalainen, T., Autio, A. & Hurme, M. Human endogenous retroviruses of the HERV-K (HML-2) family are expressed in the brain of healthy individuals and modify the composition of the brain-infiltrating immune cells. Heliyon 9 (11), e21283 (2023). Ukita, M. et al. CXCL13-producing CD4 + T cells accumulate in the early phase of tertiary lymphoid structures in ovarian cancer. JCI Insight ; 7 (12). (2022). Pipi, E. et al. Tertiary Lymphoid Structures: Autoimmunity Goes Local. Front. Immunol. 9 , 1952 (2018). Kobayashi, S. et al. A distinct human CD4 + T cell subset that secretes CXCL13 in Rheumatoid synovium. Arthritis Rheum. 65 (12), 3063–3072 (2013). Yang, R. et al. Autoimmunity to munc-18 in Rasmussen's encephalitis. Neuron 28 (2), 375–383 (2000). Alvarez-Baron, E. et al. Autoantibodies to Munc18, cerebral plasma cells and B-lymphocytes in Rasmussen encephalitis. Epilepsy Res. 80 (1), 93–97 (2008). Nibber, A. et al. Antibodies to AMPA receptors in Rasmussen's encephalitis. Eur. J. Paediatr. Neurol. 20 (2), 222–227 (2016). Thilo, B. et al. A case of Rasmussen encephalitis treated with rituximab. Nat. Rev. Neurol. 5 (8), 458–462 (2009). Timarova, G., Lisa, I. & Kukumberg, P. Long-term effect of rituximab in a case with late-onset Rasmussen s encephalitis with anti-ganglioside IgGQ1b and anti-GAD antibodies positivity. Case Report. Neuro Endocrinol. Lett. 37 (3), 179–183 (2016). Jagtap, S. A. et al. Rituximab in Rasmussen's encephalitis: A single center experience and review of the literature. Epilepsy Behav. Rep. 19 , 100540 (2022). Erdogan, S., Aliyeva, G., Kizil, T. E. O. & Sener, O. Rituximab-responsive late-onset Rasmussen's encephalitis. Seizure 123 , 57–59 (2024). Joller, N. et al. Treg cells expressing the coinhibitory molecule TIGIT selectively inhibit proinflammatory Th1 and Th17 cell responses. Immunity 40 (4), 569–581 (2014). Sumida, T. S., Cheru, N. T. & Hafler, D. A. The regulation and differentiation of regulatory T cells and their dysfunction in autoimmune diseases. Nat. Rev. Immunol. 24 (7), 503–517 (2024). Mohr, A., Atif, M., Balderas, R., Gorochov, G. & Miyara, M. The role of FOXP3(+) regulatory T cells in human autoimmune and inflammatory diseases. Clin. Exp. Immunol. 197 (1), 24–35 (2019). Selck, C. & Dominguez-Villar, M. Antigen-Specific Regulatory T Cell Therapy in Autoimmune Diseases and Transplantation. Front. Immunol. 12 , 661875 (2021). Schloder, J., Shahneh, F., Schneider, F. J. & Wieschendorf, B. Boosting regulatory T cell function for the treatment of autoimmune diseases - That's only half the battle! Front. Immunol. 13 , 973813 (2022). Wing, K., Yamaguchi, T. & Sakaguchi, S. Cell-autonomous and -non-autonomous roles of CTLA-4 in immune regulation. Trends Immunol. 32 (9), 428–433 (2011). Vignali, D. A., Collison, L. W. & Workman, C. J. How regulatory T cells work. Nat. Rev. Immunol. 8 (7), 523–532 (2008). Dikiy, S. & Rudensky, A. Y. Principles of regulatory T cell function. Immunity 56 (2), 240–255 (2023). Halliday, N. et al. CD86 Is a Selective CD28 Ligand Supporting FoxP3 + Regulatory T Cell Homeostasis in the Presence of High Levels of CTLA-4. Front. Immunol. 11 , 600000 (2020). Marsh, S. E. et al. Dissection of artifactual and confounding glial signatures by single-cell sequencing of mouse and human brain. Nat. Neurosci. 25 (3), 306–316 (2022). Chatila, Z. K. et al. RNA- and ATAC-sequencing Reveals a Unique CD83 + Microglial Population Focally Depleted in Parkinson's Disease. bioRxiv. (2023). Schetters, S. T. T., Gomez-Nicola, D., Garcia-Vallejo, J. J. & Van Kooyk, Y. Neuroinflammation: Microglia and T Cells Get Ready to Tango. Front. Immunol. 8 , 1905 (2017). Chhatbar, C. & Prinz, M. The roles of microglia in viral encephalitis: from sensome to therapeutic targeting. Cell. Mol. Immunol. 18 (2), 250–258 (2021). MacLachlan, B. J. et al. Molecular characterization of HLA class II binding to the LAG-3 T cell co-inhibitory receptor. Eur. J. Immunol. 51 (2), 331–341 (2021). Additional Declarations No competing interests reported. Supplementary Files Table1S.xlsx Table2S.pdf Tables3Sand4S.pdf Table5S.xlsx Fig.1S.pdf Fig.2S.pdf Fig.3S.pdf Fig.4S.pdf Fig.5S.pdf Fig.6S.pdf Fig.7S.pdf Fig.8S.pdf Fig.9S.pdf Fig.10S.pdf Fig.11S.pdf Fig.12S.pdf Fig.13S.pdf Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 16 Feb, 2026 Reviews received at journal 11 Feb, 2026 Reviewers agreed at journal 26 Jan, 2026 Reviews received at journal 24 Jan, 2026 Reviewers agreed at journal 23 Jan, 2026 Reviewers agreed at journal 23 Jan, 2026 Reviewers invited by journal 23 Jan, 2026 Editor invited by journal 16 Jan, 2026 Editor assigned by journal 13 Jan, 2026 Submission checks completed at journal 13 Jan, 2026 First submitted to journal 12 Jan, 2026 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-8584794","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":580339048,"identity":"63b897da-002e-4ec3-8814-51b354c8efc1","order_by":0,"name":"Giovanni Quinones-Valdez","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Giovanni","middleName":"","lastName":"Quinones-Valdez","suffix":""},{"id":580339049,"identity":"f2944a14-0979-4039-93e9-b82a752c390e","order_by":1,"name":"Julia W. Chang","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Julia","middleName":"W.","lastName":"Chang","suffix":""},{"id":580339050,"identity":"0a6b7e5f-3c55-4c2d-be4b-5f6fad248566","order_by":2,"name":"Shino D. Magaki","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Shino","middleName":"D.","lastName":"Magaki","suffix":""},{"id":580339051,"identity":"4d56a7fc-64cf-400e-8d5a-250539bf8ad8","order_by":3,"name":"Harry V. Vinters","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Harry","middleName":"V.","lastName":"Vinters","suffix":""},{"id":580339052,"identity":"8d9df85a-6183-49c8-a62a-3a72ac929038","order_by":4,"name":"Noriko Salamon","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Noriko","middleName":"","lastName":"Salamon","suffix":""},{"id":580339053,"identity":"b9760825-4c2f-46dd-a7ae-aad52ea122ca","order_by":5,"name":"Anthony C. Wang","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Anthony","middleName":"C.","lastName":"Wang","suffix":""},{"id":580339054,"identity":"6b278458-43d6-499c-834b-e38def9d441e","order_by":6,"name":"Aria Fallah","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Aria","middleName":"","lastName":"Fallah","suffix":""},{"id":580339055,"identity":"ebc09521-0f5d-4326-b0e2-d88d843a6d32","order_by":7,"name":"Geoffrey C. Owens","email":"data:image/png;base64,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","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":true,"prefix":"","firstName":"Geoffrey","middleName":"C.","lastName":"Owens","suffix":""}],"badges":[],"createdAt":"2026-01-12 18:53:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8584794/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8584794/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101364259,"identity":"8fe77c50-aef7-4c3b-806f-c4f1492eb631","added_by":"auto","created_at":"2026-01-29 00:47:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":137781,"visible":true,"origin":"","legend":"\u003cp\u003ePie charts depict the proportion of different cell types in the three RE surgery specimens after integration of the scRNA-seq data from each library.\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/40c5ab8b58817bf7fbd37679.png"},{"id":101364234,"identity":"66aa9066-286f-4477-b27f-0d2669c3dea4","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":334284,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted ligand receptor interactions between immune cells based on the NAMTI algorithm.\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/110128c01123968ef5e24ab7.png"},{"id":101364236,"identity":"3df4f6b5-3706-4511-8e8f-3144ea44fa42","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1692952,"visible":true,"origin":"","legend":"\u003cp\u003ePhenotypes of the T cells and NK cells in the immune cell fractions isolated from the three RE surgery specimens. A) UMAP of cell clusters based on the normalized counts of the most variable genes. B) UMAPs showing the proportion of T cells and NK cells in the three surgical specimens and identification of the clusters comprising CD8 and CD4 T cells. C) Bubble plot of selected genes that distinguish between the different T cell and NK cell clusters. D) Frequency histograms showing the distribution of T cells clonotypes from the three surgical specimens among the T cell clusters. In the upper histogram the frequent clonotypes (\u0026gt;1%) are plotted as a percentage of the total number of frequent clonotypes from each RE case. In the lower histogram the infrequent clonotypes (\u0026lt;1%) are plotted in the same way. D) UMAPS showing that the most frequent clonotypes in RE cases 738 and 754 are found in several CD8 clusters whereas the most frequent clonotype in RE case 769 is almost exclusively found in a single CD4 T cell cluster (see also Table S1).\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/119c6f07b5ac2e752778ea86.png"},{"id":101364251,"identity":"1093d3de-3b59-4d1c-a1cd-9eb315dbd3be","added_by":"auto","created_at":"2026-01-29 00:47:13","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":115066,"visible":true,"origin":"","legend":"\u003cp\u003ePie charts depict the proportion of the frequent CD4 and CD8 T clonotypes (\u0026gt;1%) that are either public or private (patient-specific).\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/f30a8239ee0e71a3e9159a0d.png"},{"id":101364258,"identity":"c4826f17-e30f-44f4-8e35-060feea2f470","added_by":"auto","created_at":"2026-01-29 00:47:14","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":441089,"visible":true,"origin":"","legend":"\u003cp\u003eB cell clonotypes in RE case 738. A) Frequency distribution of the CDR3 lengths in the sample of IgH sequences from sequencing genomic DNA isolated from part of the surgical specimen. B) Chord plot showing IgH V and J gene usage. C) Proportion of clonotypes that comprise the most frequent B cell clones (\u0026gt;1%) resolved using ViCloD.\u003c/p\u003e","description":"","filename":"Fig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/0390ca2616616950316f402f.png"},{"id":101364238,"identity":"7c9c2830-b58c-49f8-afd0-79c267d221c8","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1257495,"visible":true,"origin":"","legend":"\u003cp\u003ePhenotypes of the myeloid cells in the three RE surgical specimens. A) UMAP showing the clusters of myeloid cells based on the expression of the most variable genes after integration of the scRNA-seq data. B) Histogram showing the distribution of myeloid cell types among the three RE cases. C) Bubble plot of selected genes that define microglia. D) Bubble plot of genes that define monocyte-derived macrophages and dendritic cells. E) Bubble plot of genes associated with activated microglia. F) Violin plots of selected genes that distinguish between clusters of microglia.\u003c/p\u003e","description":"","filename":"Fig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/39f72e27502854e50afde229.png"},{"id":101398078,"identity":"6aa526c6-5e10-4428-8915-360af37aa16c","added_by":"auto","created_at":"2026-01-29 09:39:30","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":2555629,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted crosstalk between T cells and microglia. UMAPs showing normalized counts of transcripts encoding ligand and receptors predicted to interact according to the NATMI toolkit. The potential signaling between microglia and CD4 T cells by oxysterols produced by microglia was not predicted using the NATMI toolkit.\u003c/p\u003e","description":"","filename":"Fig.7.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/98bf57e24644e8a0911489cd.png"},{"id":101364240,"identity":"6087b753-3609-4a7c-9692-e513dd0a3d35","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":384057,"visible":true,"origin":"","legend":"\u003cp\u003eLocations of active HERV-K proviruses. A) HERV-K transcripts from each brain specimen were mapped to known sites of HERV-K integration and displayed in an ideogram of the human karyotype showing gene density on each chromosome. Purple pins mark the patient-specific sites, and red pins mark the sites common to the three patients. B) Venn diagrams showing the overlap between the three patients, the overlap with mapped sites of HERV-K integration from two published studies, and the overlap with active proviruses previously identified in normal brain cortex.\u003c/p\u003e","description":"","filename":"Fig.8.png","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/c360840fcaef1ec3c25ffd7f.png"},{"id":101755975,"identity":"49935854-158d-48ee-b523-d7ce6486084e","added_by":"auto","created_at":"2026-02-03 10:55:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6566926,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/a7cb6dfc-fce8-4f27-8913-47d23126407d.pdf"},{"id":101398334,"identity":"bc639cbe-a9d9-4452-b7d4-d3f3e449263e","added_by":"auto","created_at":"2026-01-29 09:41:00","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":107434,"visible":true,"origin":"","legend":"","description":"","filename":"Table1S.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/0a0c30ed7613145a4fd7e1e2.xlsx"},{"id":101364235,"identity":"5e459fdd-5ed3-4886-94ce-bcb847db6b1c","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":124786,"visible":true,"origin":"","legend":"","description":"","filename":"Table2S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/c24c1fa690b254430384dc6a.pdf"},{"id":101364252,"identity":"aa26b68b-7226-4191-8d38-9d3ac080e418","added_by":"auto","created_at":"2026-01-29 00:47:13","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":69812,"visible":true,"origin":"","legend":"","description":"","filename":"Tables3Sand4S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/91f622000d95e4e931d1a609.pdf"},{"id":101398573,"identity":"8d66b615-0561-4952-b03f-38cde7f94ea0","added_by":"auto","created_at":"2026-01-29 09:42:30","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":23517,"visible":true,"origin":"","legend":"","description":"","filename":"Table5S.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/53ba0856efefe03615765269.xlsx"},{"id":101364241,"identity":"bb026863-0f45-415b-99d4-703b99cd7f3d","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":2175512,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.1S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/c98aa831656bcace6f169f40.pdf"},{"id":101364255,"identity":"ea23aea7-ccc4-4df0-91b8-dd21798e84f8","added_by":"auto","created_at":"2026-01-29 00:47:14","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":10689814,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.2S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/82b477bb34a6889681693264.pdf"},{"id":101398035,"identity":"a5398b4f-c577-4214-b142-46844d999b86","added_by":"auto","created_at":"2026-01-29 09:39:12","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":5928438,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.3S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/77a6140258501c6b935741bc.pdf"},{"id":101364242,"identity":"d88b308e-8ba4-47e4-9693-5ec5b2aef423","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":1751307,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.4S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/c9fcb95bcd4a18d5653ef644.pdf"},{"id":101364243,"identity":"be0e94ad-de9c-44d3-abd7-cbdc73107a6b","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":4563605,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.5S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/c988b76914f34219ffb82b70.pdf"},{"id":101364249,"identity":"09fd0396-c3ca-4277-a8d3-513f5b13fd49","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":6419667,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.6S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/b1062fce146cde8fde11a9f7.pdf"},{"id":101398637,"identity":"ccfd0676-9d27-42c3-8a5b-aa6b69900411","added_by":"auto","created_at":"2026-01-29 09:43:21","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":5541028,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.7S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/aaffae9bc6c72ed3bc39c771.pdf"},{"id":101364245,"identity":"582b1ccb-16a8-4d18-984d-341dbc50febe","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":188442,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.8S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/c2c510ec6aa6896473261192.pdf"},{"id":101364246,"identity":"691a040f-8a78-4b77-a3cf-f3a507808acc","added_by":"auto","created_at":"2026-01-29 00:47:12","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":46070,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.9S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/867741aeb35fc915c2a31c3d.pdf"},{"id":101364256,"identity":"841eefa8-98ad-440a-85c8-af5dd8897be5","added_by":"auto","created_at":"2026-01-29 00:47:14","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":3015603,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.10S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/fe0682c4bd3ad06e258b44c5.pdf"},{"id":101364254,"identity":"ccfb03fe-94c9-44c3-abfa-b7b79e952e08","added_by":"auto","created_at":"2026-01-29 00:47:13","extension":"pdf","order_by":16,"title":"","display":"","copyAsset":false,"role":"supplement","size":5244552,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.11S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/fb2a537dc231860fe3441e70.pdf"},{"id":101751192,"identity":"3d09dea9-18c6-4cf0-bfc7-2d35e99d7096","added_by":"auto","created_at":"2026-02-03 10:18:01","extension":"pdf","order_by":17,"title":"","display":"","copyAsset":false,"role":"supplement","size":7178458,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.12S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/38aaad46af76582a9228c971.pdf"},{"id":101364250,"identity":"43fd64f8-d509-4eb5-8efc-3278e40b07e2","added_by":"auto","created_at":"2026-01-29 00:47:13","extension":"pdf","order_by":18,"title":"","display":"","copyAsset":false,"role":"supplement","size":8234064,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.13S.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8584794/v1/13c64ea481ccca40caba8ed2.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Immune landscape of the affected brain in Rasmussen encephalitis","fulltext":[{"header":"Background","content":"\u003cp\u003eRasmussen encephalitis (RE) is a rare neuroinflammatory disease characterized by seizures and progressive brain atrophy that is usually confined to one cerebral hemisphere (\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e), although bilateral cases have been described (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). In a recent retrospective study of 160 RE cases, the median age of clinical onset was seven years (range 1 to 53 years) with the left cerebral hemisphere more often affected (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Disease management involves anti-inflammatory, immune modulatory, and anti-epileptic drugs (AEDs), although surgical resection remains the only effective treatment option to achieve seizure freedom (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). The presence of clonally focused T cells in resected involved brain tissue suggests that RE may have an autoimmune component (\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), however, to date no T cell specific self-antigens have been identified. Comorbidity with several autoimmune diseases has indicated a potential heritable predisposition (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e), and, in support of this idea, we identified HLA class I and II alleles in a cohort of 24 RE surgical cases that have been linked to susceptibility to autoimmune diseases (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). HLA class II nonsynonymous single nucleotide polymorphisms (SNP) were also found to be enriched in RE cases as well as several SNPs in other genes including those involved in an immune response (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Takahashi et al. (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) identified SNPs in \u003cem\u003eCTLA4\u003c/em\u003e and \u003cem\u003ePDCD1\u003c/em\u003e that were overrepresented in a cohort of Japanese RE patients, which may negatively affect the function of these immune regulatory genes. Wang et al. identified a SNP in IFITM3 in Chinese RE patients that may promote human cytomegalovirus (HHV5) persistence in the brain (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Epstein Barr virus (HHV4) has been proposed as a viral trigger for Multiple sclerosis (MS) (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), and HHV4, HHV5, and HHV6 sequences have been reported in some RE brain specimens but not in others (\u003cspan additionalcitationids=\"CR19 CR20 CR21 CR22\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). In support of a possible viral etiology, early treatment of two presumptive RE cases with ganciclovir was reported to stop seizures and resolve neurological deficits (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). We found that many of the expanded T cell clonotypes in RE brain specimens were public suggesting that they may have been directed against a common infectious agent (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). On the other hand, autoreactive T cell clonotypes may be present in healthy individuals (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCo-existence of RE and focal cortical dysplasia (FCD) pathologies has also been documented in RE surgical cases suggesting that FCD could be involved in triggering the disease (\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). FCD is caused by somatic mutations in neural progenitors that occur during early brain development resulting in focal seizures (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). It has been established that seizures can promote brain inflammation (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e), suggesting that seizure activity per se could trigger an inflammatory cascade leading to an autoimmune response.\u003c/p\u003e \u003cp\u003eFrom the recent large cohort retrospective study of RE cases (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e), it was found that the number of RE patients with twin siblings was higher compared with the general population, leading the authors to suggest that the risks of complications associated with preterm births, which are more prevalent in twin pregnancies (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e), may also be causative. For example, preterm birth is a risk factor for perinatal arterial ischemic stroke, which may in turn be linked to neuroinflammation and seizures (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRNA sequencing of nuclei isolated from resected brain tissue confirmed earlier findings (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e) and revealed the heterogeneity of activated microglia in affected brain areas (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). Based on the pathological stages of RE described by Pardo et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e), activation of microglia was shown to occur prior to T cell infiltration (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). In the present study we have sequenced RNA from individual immune cells isolated from three RE surgery cases and characterized the transcriptomes of both myeloid and lymphocyte populations including T cell receptors. We also present evidence for the reactivation of the endogenous retrovirus HERV-K in these cells.\u003c/p\u003e"},{"header":"Methods and Materials","content":"\u003cp\u003eThis study was approved by the UCLA Institutional Review Board (IRB no. 18-001048). The patients or their parents or legal guardians provided informed consent for the use of the surgical remnant and blood for research purposes according to the Declaration of Helsinki. There were no exclusion criteria, and no reported comorbidities. All specimens were collected using the same standard operating procedures. De-identified patient information including age at seizure onset, age at surgery, and gender was collected with informed consent.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSingle cell RNA sequencing\u003c/h2\u003e \u003cp\u003eImmune cells were isolated from blocks of fresh resected brain tissue as previously described (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). In brief, brain tissue collected from the operating room in ice cold Hibernate\u0026reg; containing penicillin/streptomycin (120U/ml and 100ug/ml respectively, ThermoFisher, Carlsbad, CA) was transferred to magnesium calcium free Hanks Balanced Salt Solution containing HEPES (20mM), glucose (5mM) and penicillin/streptomycin (ThermoFisher), and finely minced on ice with spring scissors. Tissue fragments were transferred to RPM1 (ThermoFisher) containing 10% human serum (Phoenix Scientific, San Marcos, CA) and HEPES (20mM) and digested overnight at room temperature with Type IV collagenase (~\u0026thinsp;800U/ml) (Worthington Biochemical Corp. Lakewood, NJ) followed by fractionation on a 70%:30% Percoll\u0026reg; (Millipore Sigma, St. Louis MO) step gradient. The immune cells were collected from the interface between the two Percoll\u0026reg; steps, and two Chromium single cell gene expression libraries (5\u0026rsquo; gene expression and TCR V(D)J) were prepared (10X Genomics, Pleasanton, CA) and sequenced on a NovaSeq 6000 instrument in a SP flow cell (2x50 bp) (Illumina Inc., San Diego, CA). Reads were demultiplexed and aligned to the Genome Reference Consortium Human Reference 38 (GRCh Build 38) and VDJ reference data (based on Ensembl 94 release) using Cell Ranger count and Cell Ranger VDJ pipelines (10X Genomics). The single cell RNA sequencing (scRNA-seq) data from the 5\u0026rsquo; expression library were analyzed using the R package SingCellaR with the default settings (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). The scRNA-seq data from each RE sample were combined using the integration function from Seurat (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). Clusters of putative T cells and NK cells (n\u0026thinsp;=\u0026thinsp;10,971) were then extracted, reintegrated using Harmony (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e), and re-clustered. Clusters of putative myeloid cells (n\u0026thinsp;=\u0026thinsp;12,877) were also extracted, reintegrated and re-clustered. The R package Monocle 3.0 (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e) was used to perform a trajectory analysis of the myeloid cells. The all_contig_annotations file from the Cell Ranger VDJ pipeline was used to assess the clonotype diversity in each RE sample, and to assign T cell phenotypes to the clonotypes based on the SingCellaR workflow. Potential cell: cell interactions based on the relative expression of known receptor ligand pairs were analyzed using the NATMI python-based toolkit (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e). For each ligand: receptor pair a specificity estimate is calculated as the average expression of the ligand in the sending cell type divided by the average expression of the ligand in all cells multiplied by the average expression of the corresponding receptor by the receiving cell type divided by the average expression of the receptor in all cells. Spectratyping was performed with the immunarch R package (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e), and diversity estimates were calculated using the iNEXT R package (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e). Chord diagrams were made with the R package chorddiag (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). Barcodes of the most abundant clonotype(s) in each sample were integrated into the SingCellaR object for visualization in UMAP projections.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIdentification of Viral RNA using ViralTrack\u003c/h3\u003e\n\u003cp\u003eFor each sample, processing of the raw RNA sequencing reads was carried out using Umitools (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e). Initially, a whitelist of acceptable barcodes was generated, allowing for a Hamming distance of 1. Subsequently, the barcode sequences were extracted from the reads and incorporated into the read names. The next step involved mapping the reads to both the human genome (GRCh Build 38) and to a comprehensive viral genome database, encompassing over 14,000 distinct viral genomes. This mapping was performed utilizing the ViralTrack software package (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e). Only viral genomes meeting specific criteria were retained for further analysis. These criteria included a minimum coverage of at least 50 reads covering at least 10% of the entire viral genome sequence, and a sequence complexity of 1.3, as indicated by base composition entropy. Viral hits meeting these criteria were then assembled into contigs, which were considered to represent active regions of transcription within the viral genome. Subsequently, only those reads that mapped to contigs with a length of 200 bases or longer were retained. A custom Python script was employed to parse the alignments to both the human and viral genomes. Reads that mapped to both the human genome and a viral genome were assigned to the organism showing the highest alignment score. Finally, the read count of viral hits was quantified for each barcode, and barcodes were categorized as expressing the virus if they contained three or more reads mapped to a viral genome. The resulting viral expression data were integrated into the SingCellaR object for visualization in UMAP projections.\u003c/p\u003e\n\u003ch3\u003eIdentification of HERV-K provirus insertion sites\u003c/h3\u003e\n\u003cp\u003eWe obtained the genomic coordinates of HERV-K provirus insertions from two previously published sources (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). Coordinates from the Xue et al. study, originally reported in the hg19 (GRCh37) genome build, were converted to the hg38 (GRCh38) assembly using the liftover Python package (v1\u003cem\u003e.3.2\u003c/em\u003e). Coordinates from both studies were then merged and manually curated to remove redundancies. To annotate the genomic context of each insertion, we used GENCODE v44 gene annotations. For each insertion, we recorded overlapping features including gene name, and region type (e.g., exon, intron, intergenic). Using a custom Python script, we quantified read coverage across the merged HERV-K insertion regions using the scRNA-seq data from the three RE patients, mapped to the reference hg38 (GRCh38) genome build. Reads were considered overlapping if they intersected a given region by at least 10 base pairs. We further filtered reads to retain only those also aligning to a reference HERV-K genome (NC_0022518) as determined from the prior ViralTrack analysis. Additional filtering criteria were applied to reduce mapping artifacts: retained reads were required to have an entropy score\u0026thinsp;\u0026ge;\u0026thinsp;1.0, an average Phred-scaled mapping quality\u0026thinsp;\u0026ge;\u0026thinsp;25, and a minimum aligned segment length of 20 bases. Chromosomal maps showing the sites of provirus insertion were made using the R package RIdeogram (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eB cell receptor sequencing\u003c/h3\u003e\n\u003cp\u003eBulk genomic DNA was isolated from a frozen block of fresh brain tissue from RE patient 738, and from whole blood (Monarch\u0026reg; genomic DNA purification kit, New England Biolabs, Ipswich, MA). BCR sequences were obtained using the ImmunoSEQ\u0026reg; assay (Adaptive Biotechnologies, Seattle, WA), and clonal analysis was performed using the interactive web tool, ViCloD (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eImmunocytochemistry\u003c/h3\u003e\n\u003cp\u003eParaffin embedded sections (5 um) were stained using Opal Fluorophore reagents (Akoya Biosciences, Marlborough, MA) and a Leica Bond RX auto-stainer (Leica Biosystems, Vista, CA) following antigen retrieval. The following antibodies were used CD20 (Opal 520), HLA-DR (Opal 570), CD4 (Opal 620) CD3 (Opal 690), and CD8 (Opal 780). Images were collected using a Leica Aperio Versa 200 Slide Scanning Microscope equipped with a 16-bit 5.5-megapixel fluorescence camera, and images were captured using Phenochart 2.0 (Akoya Biosciences), and colors were re-assigned. Images, as Tiff files, were imported into PHOTO-PAINT (Corel Corporation, Ottawa, Canada), to upscale the resolution to 600 dpi, and adjust tone curves. All figures were prepared in CorelDraw (Corel Corporation).\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eClinical and Pathological Description\u003c/h2\u003e \u003cp\u003eThree RE patients underwent surgery to control their drug-resistant seizures (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Post-operatively all three patients were seizure free although still taking AEDs. We obtained blood and fresh brain tissue from the surgical resections. Immune cells were isolated from each brain specimen and single cell cDNA libraries were constructed targeting 10,000 cells. Both whole transcriptome and T cell receptor sequences were obtained, and the transcriptomic data from each patient were integrated using canonical correlation analysis (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). The identity of the cells in each cluster was assigned based on known marker genes (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e), and from these assignments, the proportions of the different cell types from the three brain specimens were calculated (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Myeloid cells mainly comprised microglia and were clearly distinguishable from T cells and NK cells, except for a mixed cluster derived primarily from patient 769, which we attribute to cell doublets (Cluster 9, Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). A high proportion of plasma cells was only evident in the brain specimen from patient 738, which we confirmed by multiplex immunostaining (Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e).\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\u003ePatient data.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\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=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCase ID\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAge at seizure onset (yr)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAge at surgery (yr)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSeizure frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAffected hemisphere\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAEDS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003emale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e~\u0026thinsp;15 per day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eright\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003egabapentin lamotrigine\u003c/p\u003e \u003cp\u003elevetiracetam\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e754\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003emale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e~\u0026thinsp;1 per day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eleft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ebrivaracetam clobazam clonazepam lacosamide\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e769\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003efemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e~\u0026thinsp;24 per day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eright\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003elacosamide levetiracetam\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe used NATMI to estimate the potential cell: cell interactions between the different immune cell types. Interactions involving the small number of vascular cells were excluded. We filtered the results by limiting the number of interactions to cell types in which\u0026thinsp;\u0026gt;\u0026thinsp;25% of the sending cells expressed the ligand and \u0026gt;\u0026thinsp;25% of the receiving cells expressed the corresponding receptor. A threshold of \u0026gt;\u0026thinsp;0.1 was set for the specificity estimate for both the ligand and the receptor (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This analysis predicted specific reciprocal interactions between myeloid cells and lymphocytes involving different ligand: receptor pairs. Based on the thresholds imposed on the data, TNFSF-13 signaling between macrophages and B cells, and IL1 signaling between microglia and macrophages were the most specific.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eT cell phenotypes\u003c/h3\u003e\n\u003cp\u003eThe putative T cells and Natural Killer (NK) cells were extracted from the integrated dataset, and re-clustered to further resolve cell phenotypes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and B). The majority of cells in all 12 clusters expressed CD69 a marker of resident memory T cells (T\u003csub\u003eRM\u003c/sub\u003e) (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e); a variable number of cells in each cluster also expressed ITGAE (CD103), a second T\u003csub\u003eRM\u003c/sub\u003e marker (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Based on differentially expressed genes, three of the 12 clusters contained CD4 T cells, of which cluster 5 was defined by the expression of IL7R and KLRB1, cluster 7 by FOXP3 expression and cluster 9 by KLRB1 and CXCL13 transcripts (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). In agreement with this interpretation GSEA using gene signatures for different T cell phenotypes (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e) indicated that cluster 5 contained CD4 memory T cells and cluster 7 comprised regulatory T cells (Tregs) (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Clusters 1\u0026ndash;4, 6 and 10 comprised predominantly CD8 T cells, of which cluster 1 was further defined by high levels of XCL1 transcripts, cluster 3 by GMZK expression, and cluster 6 by high expression of heat shock proteins including HSP1A1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). GSEA with gene signatures for T cell phenotypes (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e), exhausted T cells (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e), and for virus responsive T cells (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e) indicated that cluster 1 CD8 T cells had a cytokine producing phenotype and were responsive to a virus whereas cluster 2 and 4 CD8 T cells were likely exhausted. Gamma delta T cells were found within the CD8 T cells clusters; presumably, their overall gene expression profiles were too similar to the alpha beta T cells (Fig. \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). Cluster 8 was defined by a higher level of SKI transcripts, a suppressor of CD103 expression in mice (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e), cluster 11 by FCGR3A (CD16) expression, and cluster 12 by XCL1 expression. GSEA with a published Natural killer (NK) cell gene signature (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e) indicated that clusters 11 and 12 contained NK cells (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In addition to the putative NK cells, the majority of CD8 T cells expressed KLRD1 (CD94) mRNA, which encodes an NK cell C-type lectin that forms heterodimers with NKG2 molecules to either inhibit (NKG2A/B) or activate (NKG2C and NKG2E) NK and CD8 T cells by binding to HLA-E molecules (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e). Higher levels of transcripts encoding the activating receptors were found in the majority of NK cells and CD8 T cells compared with transcripts encoding the inhibitory receptor (Fig. \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e). Similarly, \u003cem\u003eKLRK1\u003c/em\u003e encoding NKG2D, another activating NK cell C-type lectin-like receptor, was also highly expressed whereas \u003cem\u003eKLRG1\u003c/em\u003e encoding another inhibitory receptor was not (Fig. \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eGene set enrichment analysis.\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=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCluster\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGene signature\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePadj\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNES\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCD8 cytokine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0261\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.8833\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003evirus responsive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0041\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.8937\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT\u003csub\u003eexhausted\u003c/sub\u003e cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.8203\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCD8 cytokine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0355\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.6735\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT\u003csub\u003eexhausted\u003c/sub\u003e cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0340\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.5864\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCD4 naive/central memory\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.8320\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCD4/CD8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.7994\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003evirus responsive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0490\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.8565\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT\u003csub\u003ereg\u003c/sub\u003e cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0017\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.6103\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNK cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.6981\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCD8 cytotoxic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.6877\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNK cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.8039\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eInterrogating the dataset for expression of transcripts encoding co-stimulatory and co-inhibitory genes revealed that in contrast to CD4 T cells, LAG3 was the dominant co-inhibitory gene expressed by CD8 T cells (Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e); far fewer cells expressed the co-inhibitory gene PDCD1, which were largely confined to cluster 9 CD4 T cells. CD4 Tregs (cluster 7) expressed CTLA4 and TIGIT, as well as the costimulatory genes \u003cem\u003eICOS\u003c/em\u003e, \u003cem\u003eTNFRSF9\u003c/em\u003e (4-1BB), \u003cem\u003eCD27\u003c/em\u003e and \u003cem\u003eCD28\u003c/em\u003e (Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e). By contrast CD8 T cells expressed the co-stimulatory genes \u003cem\u003eICOS\u003c/em\u003e, \u003cem\u003eCRTAM\u003c/em\u003e and \u003cem\u003eTNFRSF9\u003c/em\u003e, but essentially no \u003cem\u003eCD27\u003c/em\u003e or \u003cem\u003eCD28\u003c/em\u003e (Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e). We also examined the expression of selected transcription factors. More cells expressed RUNX3, ZEB2, PRDM1 (Blimp), and ZNF683 (Hobit) mRNAs than Tox, TBX21 (Tbet) or TCF7 mRNAs (Fig. \u003cspan refid=\"MOESM7\" class=\"InternalRef\"\u003eS7\u003c/span\u003e). However, fewer cells assigned to the CD4 T cell clusters (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e and \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e) expressed ZEB2 and ZNF683 mRNA compared with cells assigned to the CD8 T cell clusters. Very few cells in the NK cell clusters (11 and 12) expressed BATF (Fig. \u003cspan refid=\"MOESM7\" class=\"InternalRef\"\u003eS7\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eClonally focused CD8 T cells\u003c/h2\u003e \u003cp\u003eWe determined the number of different T cell clonotypes, defined by the Vbeta chain third complementarity region (CDR3) sequence, in the three RE brain specimens (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). From Hill plots, TRBV and TRBJ gene usage, and CDR3 lengths it was clear that T cells in the brains of the three patients were clonally focused particularly in the brain specimens from patients 754 and 769 (Fig. \u003cspan refid=\"MOESM8\" class=\"InternalRef\"\u003eS8\u003c/span\u003e). Matching the barcodes from the TCR libraries to those comprising the T cell clusters showed that the most frequent clonotypes (defined as \u0026gt;\u0026thinsp;1%) were found in the CD8 T cell clusters with one exception (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). In patient 769 the most frequent clonotype could be ascribed to cluster 9 corresponding to a CD4 T cell subset defined by the expression of KLRB1 and CXCL13 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). In patients 738 and 754 the dominant clonotypes were CD8 T cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). We confirmed the phenotypes of 49 of the 54 most frequent clonotypes by extracting the CD3D\u003csup\u003e+\u003c/sup\u003e/ CD3E\u003csup\u003e+\u003c/sup\u003e CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e cells (\u0026ge;\u0026thinsp;1 normalized UMI) from the sparse matrix file of the reclustered T cell and NK cells and matching the barcodes to those of the top clonotypes (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003eS). To partition the top 1% into public and private clonotypes we compared them to Adaptive Biotechnologies\u0026rsquo; immuneACCESS database (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e more than half of the abundant clonotypes were public. Vbeta chain CDR3 amino acid sequence and TRBV and TRBJ genes of a rare clonotype in the blood of a single donor from large scale covid study (\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e) was identical to the most abundant clonotype in patient 754 (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). The CDR3 regions differed by three nucleotides indicating convergent selection (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e). Likewise, the most abundant clonotype in patient 769 was found in five individuals from an unpublished study of celiac disease available in the immuneACCESS database, although TRBV gene usage differed (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe compared all the Vbeta CDR3 amino acid sequences to the VDJdb database (\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e) and found that there were exact matches including TRBV and TRBJ gene usage to TCRs that recognize common viral epitopes. All these matching clonotypes were rare with one exception (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). We implemented ClusTCR (\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e) to identify clonotypes that may recognize the same epitopes. As shown in Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e, several clonotypes were found that likely recognize the same antigen, notably HHV4 (IVTDFSVIK), and HHV5 (KLGGALQK).\u003c/p\u003e \u003cp\u003e \u003cem\u003eExpansion of B cell clone\u003c/em\u003es\u003c/p\u003e \u003cp\u003eWe found higher numbers of B cells and antibody-producing plasma cells in the brain specimen from patient 738 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig.\u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). Bulk immunoglobulin heavy chain (IGH) sequences were obtained from a fresh frozen piece of tissue from the same surgical specimen, and from whole blood collected at the time of the surgery. Repertoire analysis of IGH sequences identified the six most frequent clones (\u0026gt;\u0026thinsp;1%) in the brain and showed the extent of intra-clonal diversity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The most abundant clone comprised a single IGH sequence that was not detected in the sample of IGH sequences from the blood suggesting local clonal expansion (Fig. \u003cspan refid=\"MOESM9\" class=\"InternalRef\"\u003eS9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eMyeloid cell phenotypes\u003c/h2\u003e \u003cp\u003eRe-clustering the myeloid cells alone generated 14 clusters that were present in different proportions in each patient sample (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and B). We interpreted 11 of the clusters as microglia based on the expression of the homeostatic glial marker genes \u003cem\u003eTMEM119\u003c/em\u003e, \u003cem\u003eP2RY12\u003c/em\u003e, \u003cem\u003eOFML3\u003c/em\u003e, \u003cem\u003eSLC2A5\u003c/em\u003e, and \u003cem\u003eCXC3CR1\u003c/em\u003e (\u003cspan additionalcitationids=\"CR67\" citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). From the analysis of differentially expressed genes, cluster 9 appeared to define a population of macrophages as evidenced by high expression of CD163 and FCGR2B (CD32) mRNAs by the majority of cells in the cluster (\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e, \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). Cells in clusters 10 may comprise a dendritic cell population based on the expression of CD1c, CLEC10A and LGALS2 (Galectin-2) whereas cells in cluster 11 appeared to be a monocyte or monocyte-derived population based on VCAN (versican) expression (\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e). The majority of the cells comprising clusters 9\u0026ndash;11 were derived from patient 738 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe microglia populations shared genes expressed by disease-associated microglia, specifically, \u003cem\u003eTREM2\u003c/em\u003e, \u003cem\u003eAPOE\u003c/em\u003e, \u003cem\u003eCD83\u003c/em\u003e, and \u003cem\u003eSPP1\u003c/em\u003e (osteopontin) (\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e, \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e), as well as genes directly associated with the NPLR3 inflammasome pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE) (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e). All myeloid clusters expressed high levels of HLA II molecules (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE; see also Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). Cluster 3 microglia made up the largest fraction of myeloid cells in the brain specimen from patient 754, whereas cluster 2 microglia were the most abundant myeloid cells in patient 769 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). From the differential gene expression analysis, it appeared that cluster 3 microglia expressed higher levels \u003cem\u003eDNAJB1\u003c/em\u003e and \u003cem\u003eCCL3\u003c/em\u003e transcripts than cluster 2 microglia, whereas cluster 2 microglia expressed higher levels of \u003cem\u003eAIF1\u003c/em\u003e (IB1A), and \u003cem\u003eFCGR3A\u003c/em\u003e (CD16) mRNA (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF). Clustering myeloid cells and trajectory analysis using Monocle 3.0 indicated that microglia cells may transition from CD83\u003csup\u003e\u0026minus;\u003c/sup\u003e cells expressing more IBA1 and CD16 to CD83\u003csup\u003e+\u003c/sup\u003e cells expressing cytokines and heat shock proteins (Figure \u003cspan refid=\"MOESM10\" class=\"InternalRef\"\u003eS10\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe NATMI analysis identified a number of ligand: receptor pairings between myeloid cells and T cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). To determine whether any of these mapped to specific T cell subsets, we generated UMAPs comprising the normalized expression of the genes encoding these ligand: receptor pairs in the re-clustered populations of myeloid and T cells. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e, Annexin A1 mRNA was expressed in all T cells except CD4 Tregs, whereas interferon gamma mRNA was found almost exclusively in CD8 T cells. ANXA1 transcripts were also expressed by cells in the macrophage/monocyte clusters 9\u0026ndash;11 (Fig. \u003cspan refid=\"MOESM11\" class=\"InternalRef\"\u003eS11\u003c/span\u003e). Formyl protein receptor 1 (FPR1), the cognate receptor was highly expressed in microglial cells except for cluster 12 microglia. FPR2 and FPR3 mRNAs were not expressed in the majority of T cells and myeloid cells (Fig. \u003cspan refid=\"MOESM11\" class=\"InternalRef\"\u003eS11\u003c/span\u003e). The pairing of IL15 and its cognate receptor also reached the threshold of specificity that we imposed on the NATMI analysis. CD86 transcripts were found in almost all myeloid cells, whereas, the receptor, CTLA4, was only expressed by Tregs and some CD8 T cells (see also Fig. \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e). A potential interaction between a second T cell co-inhibitory molecule LAG3, and LGALS3 (Galectin 3) on macrophages/monocytes also met the criteria we imposed on the NATMI analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The transcript levels in myeloid cells of the ligands that bind the co-stimulatory and co-inhibitory receptors expressed by T cells are shown in Figure \u003cspan refid=\"MOESM12\" class=\"InternalRef\"\u003eS12\u003c/span\u003e. The interaction between TIM3 (HAVCR2) on T cells and Galectin 9 (LGALS9) on myeloid cells did not meet the criteria we imposed on the NATMI analysis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAlthough not identified in the NATMI analysis, it appeared that a subset of T cells, primarily assigned to cluster 5, expressed the G protein coupled receptor GPR183 (EB12), which binds an oxidized derivative of cholesterol, 7a, 25-dihydroxycholesterol. This oxysterol has been found to be a chemoattractant for Th17 CD4 T cells in MS (\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e). Formation of the oxysterol from cholesterol involves hydroxylation of cholesterol by cholesterol-25-hydroxylase (CH25H) (\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e). CH25H transcripts were present in microglia, primarily in clusters 3 and 8 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eEndogenous HERVK activation\u003c/h2\u003e \u003cp\u003eWe used ViralTrack (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e) to search for viral transcripts in the scRNA-seq data from each RE brain specimen. Based on the filtering steps applied in the analysis (see Methods and Materials) we did not find evidence for any active exogenous viruses in the dataset. However, we did find evidence for activity of the endogenous retrovirus, HERV-K. The filtered HERV-K reads from each RE case could be uniquely mapped to previously identified sites of provirus insertion (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e; Table \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e), indicating that multiple copies of HERV-K were transcriptionally active. Out of a total of 49 proviruses, 11 were in common between the three RE cases, and 13 were previously shown to be active in normal brain (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). Associating reads with individual barcodes and imposing a threshold of at least three reads per barcode showed that HERV-K was active in more immune cells from patient 754 compared with the other two patients (Fig. \u003cspan refid=\"MOESM13\" class=\"InternalRef\"\u003eS13\u003c/span\u003e). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e, the highest number of active proviruses was found in patient 754.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe have used scRNA-seq to characterize the immune landscape in brain areas affected by RE from three surgery cases. Based on CD69 and CD103 antibody staining of T cells from other RE cases we previously reported that the majority of T cells in resected brain tissue were T\u003csub\u003eRMs\u003c/sub\u003e indicating that they had entered the brain at an earlier time point before the surgery in response to an immune challenge, and then remained there (\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e). In the present study we found that essentially all CD4 and CD8 T cells and NK cells expressed CD69 and variable levels of ITGAE (CD103) mRNA. We also documented the expression RUNX3, PRDM1 (Blimp-1), and ZNF683 (Hobit) transcripts, transcription factors that have been implicated in establishing T\u003csub\u003eRMs\u003c/sub\u003e (\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e, \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe bioinformatics pipeline employed identified distinct populations of CD4 and CD8 T cells. The major CD8 T cell populations, which accounted for most of the high frequency clonotypes, comprised cells that expressed genes associated with an effector phenotype, as evidenced by the expression of the granzyme gene \u003cem\u003eGZMB\u003c/em\u003e, and the NK cell activating C-type lectin-like genes \u003cem\u003eKRLD2\u003c/em\u003e, \u003cem\u003eKLRKC2\u003c/em\u003e, \u003cem\u003eKLRC3\u003c/em\u003e and \u003cem\u003eKLRK1\u003c/em\u003e. Detection of IFNG mRNA in some CD8T cells was consonant with this interpretation. We also detected a population of CD8 T cells characterized by high levels of GZMK transcripts. Unlike Granzyme B, Granzyme K can induce non-caspase dependent cell death by cleaving both cytotoxic and non-cytotoxic intracellular proteins (\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e). GZMK\u003csup\u003e+\u003c/sup\u003e CD8 T cells have been found in tissues and lesions associated with several different autoimmune diseases and are assumed to be autoreactive (\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e, \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eExhausted T cells have been extensively described in chronic infections and cancers (\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e) and are characterized by high expression of co-inhibitory molecules including PD1, LAG3, TIM3, and TIGIT (\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e). The persistence of disease-triggering self-antigens in autoimmune diseases might also be expected to generate an exhausted or dysfunctional state. GSEA indicated that there were exhausted T cells among the CD8 T cells found in affected RE brain tissue; notably LAG3 transcripts were detected in about half the CD8 cells. LAG3\u003csup\u003e+\u003c/sup\u003e T cells were recently documented in RE brain parenchyma by multiplex immunolabeling (\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e). Compared with LAG3 these investigators found that the number of T cells expressing PD1 was significantly higher in tissue sections from later clinical stages compared with early stages of the disease reflecting an increase in the proportion of exhausted T cells. This difference between clinical stages could explain why we found few T cells expressing PDCD1 transcripts in our dataset. In a mouse model of type 1 diabetes, it was shown that LAG3\u003csup\u003e+\u003c/sup\u003e intra-islet CD8 T cells had an exhausted-like phenotype (\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e). Deleting \u003cem\u003eLag3\u003c/em\u003e accelerated the development of diabetes, strongly implicating LAG3\u003csup\u003e+\u003c/sup\u003e CD8 T cells in the disease (\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e). LAG3 binds MHC class II molecules, and also binds to other ligands including the product of the \u003cem\u003eLGALS3\u003c/em\u003e gene, Galectin 3 (\u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e88\u003c/span\u003e). We found MHC class II transcripts in essentially all myeloid cells, and NATMI predicted a specific interaction between LAG\u003csup\u003e+\u003c/sup\u003e CD8 T cells and LGALS3\u003csup\u003e+\u003c/sup\u003e macrophage/monocyte cells, which suggests that the effector function of CD8 T cells might be held in check by these interactions. A LAG3 agonist antibody has been developed as a potential treatment for autoimmune diseases (\u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e89\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNATMI predicted a potential ligand: receptor interaction between CD8 T cells and myeloid cells that involved VCAM-1 expression by CD8 T cells. VCAM1 can bind an integrin composed of integrin beta 2 (ITGB2) and integrin alpha X (ITGAX), both of which were expressed by myeloid cells suggesting direct binding to CD8 T cells. VCAM1 expression by CD8 T cells may reflect chronic exposure to antigen. It has recently been reported that VCAM1 is expressed by exhausted T cells and suppresses T cell activation in vitro by cis-binding TCR/CD3 complexes in the endomembrane system (\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e). Likewise, LAG3 has also been reported to bind TCR/CD3 complexes and inhibit T cell activation (\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e91\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe also found a potential interaction between Annexin A1 on T cells (except Tregs) and a N-formyl peptide receptor, FPR1, on myeloid cells, the only FRP that was highly expressed in myeloid cells (Fig. \u003cspan refid=\"MOESM11\" class=\"InternalRef\"\u003eS11\u003c/span\u003e). Annexin A1 transcripts were also detected in macrophages (Fig. \u003cspan refid=\"MOESM11\" class=\"InternalRef\"\u003eS11\u003c/span\u003e). Higher Annexin A1 expression has been documented in peripheral T cells from Rheumatoid arthritis and Systemic lupus erythematosus patients compared with healthy controls (\u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e, \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e93\u003c/span\u003e), and impaired activation of T cells and reduced severity of experimental autoimmune encephalitis was reported in \u003cem\u003eAnxa1\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;/\u0026minus;\u003c/em\u003e\u003c/sup\u003e mice (\u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e94\u003c/span\u003e). It has been proposed that Annexin A1 may act in an autocrine manner in T cells (\u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e95\u003c/span\u003e), although likely not through FPRs in RE, since they were essentially absent from T cells (Fig. \u003cspan refid=\"MOESM11\" class=\"InternalRef\"\u003eS11\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFull length Annexin A1 and the N-terminal peptide of the protein, Ac2-26, are ligands for FPR1 although binding is associated with resolution of an inflammatory response (\u003cspan additionalcitationids=\"CR97\" citationid=\"CR96\" class=\"CitationRef\"\u003e96\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e98\u003c/span\u003e). On the other hand, FPR1 binding to N-formyl peptides derived from bacterial and mitochondrial proteins is associated with an inflammatory response by innate immune cells including microglia (\u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e98\u003c/span\u003e, \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e99\u003c/span\u003e). Microglia in affected areas of the brain may therefore be poised to be pro-resolving or pro-inflammatory depending on the ligands they are exposed to.\u003c/p\u003e \u003cp\u003eThere are a number of reports documenting the presence of active HHV4 and HHV5 in brain areas affected in RE (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). We found exact HLA-matched TRB sequences in the three RE brain specimens that have been shown to recognize immunodominant peptides from HHV4 and HHV5 indicating that the three patients may have been exposed to these viruses leading to a subclinical immune response and viral latency in circulating monocytes and B cells (\u003cspan additionalcitationids=\"CR101\" citationid=\"CR100\" class=\"CitationRef\"\u003e100\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e102\u003c/span\u003e). Although the HHV4 and HHV5-specific T cell clonotypes did not correspond to the most abundant clonotypes in each surgery case, we considered the possibility that a reactivated herpes virus was present in peripheral immune cells present in the brain (\u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e103\u003c/span\u003e). We used ViralTrack to search for viral transcripts in our dataset but did not find any herpes virus transcripts. We cannot exclude the possibility of an active virus in other brain cells or that infected cells had already been eliminated leaving behind resident memory T cells (\u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e104\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOver half of the frequent clonotypes (\u0026gt;\u0026thinsp;1%) were public suggesting that some of the expanded T cells in the brain might result from exposure to a common infectious agent, and raising the possibility of molecular mimicry if these T cells also recognize self-antigens (\u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e105\u003c/span\u003e). In patient 769 the Vbeta CDR3 amino acid sequence of the most abundant clonotype only matched a rare clonotype in the blood of seven individuals diagnosed with celiac disease, and in patient 754 the Vbeta CDR3 sequence and TRB and TRJ genes exactly matched a rare clonotype in the blood of a single uninfected individual from a recent SARS-CoV-2 study (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e) (\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e). Given the limited sharing of these two clonotypes with other individuals it is possible that they may have escaped peripheral tolerance mechanisms and be directly autoreactive.\u003c/p\u003e \u003cp\u003eIn our search for viral transcripts, we found HERV-K sequences in our scRNA-seq data indicating that this endogenous retrovirus was activated in immune cells from each RE patient. Mapping the virus contigs in individual cells showed that more microglia expressed HERV-K in Patient 754 than in the other two patients. It has been reported that the presence of HERV-K transcripts in glioblastomas, teratoid rhabdoid tumors and in spinal cord neurons of ALS patients has potential disease-modifying effects (\u003cspan additionalcitationids=\"CR107\" citationid=\"CR106\" class=\"CitationRef\"\u003e106\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e108\u003c/span\u003e). Among the eleven proviruses active in all three patients, the provirus at Chr7p22.1a, ERK-6 (HERV-K108) contains full length open reading frames for the gagpol and envelope (env) genes; a mutation in the reverse transcriptase domain renders the virus defective (\u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e109\u003c/span\u003e). The envelope gene from this provirus has been shown to be translated and expressed on the cell surface of transfected cells (\u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e110\u003c/span\u003e). Whether the env protein is expressed in affected brain areas in RE remains to be determined. It has been reported that a GU-rich sequence in the env gene RNA can trigger TLR8-dependent death of neurons indicating that any pathological consequence of HERV-K activation might not depend on a translated open reading frame (\u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e111\u003c/span\u003e). We should note that HERV-K sequences, including 13 we identified in the RE specimens, have also been found in transcript data from normal brain samples in the GTEx database including ERK-6 (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR112\" class=\"CitationRef\"\u003e112\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCD4 T cells could be assigned to three different phenotypes Tregs, IL7R\u0026thinsp;+\u0026thinsp;memory cells, and a population of T cells distinguished by the expression of CXCL13 and KLRB1 (CD161) transcripts. CXCL13 is a chemoattractant for B cells, and CD4 cells producing this cytokine have been associated with the formation of tertiary lymphoid structures (TLS) (\u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e113\u003c/span\u003e). In Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e, a cluster of CD4\u0026thinsp;+\u0026thinsp;T cells surrounded by HLA-DR\u0026thinsp;+\u0026thinsp;myeloid cells could correspond to this CD4 T cell subtype and define a developing TLS. Such ectopic structures have been found in affected tissues in autoimmune diseases (\u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e114\u003c/span\u003e), and CD4 T cells producing CXCL13 are present in synovial fluid in Rheumatoid arthritis patients (\u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e115\u003c/span\u003e). In patient 769 the most frequent clonotype was in fact this CD4 T cell subtype (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). On the other hand, we found the highest number of B cells and plasma cells in patient 738. Sequencing immunoglobulin heavy chains revealed that these cells comprised expanded clones implying a specific humoral response to an antigen(s) in the brain of this patient. Patient 738 was the oldest surgery case at 26 years old, and seizure onset was documented five years earlier according to his clinical history (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). However, the patient also experienced seizures when he was four years old and was treated with AEDs for two years. We speculate that the onset of seizures later in life may reflect a reactivation of resident T cells and a de novo humoral response. Autoantibodies directed against synaptic proteins have been found in the blood from some RE patients (\u003cspan additionalcitationids=\"CR117\" citationid=\"CR116\" class=\"CitationRef\"\u003e116\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR118\" class=\"CitationRef\"\u003e118\u003c/span\u003e), which has led to the use of an anti-CD20 antibody to try to control the disease (\u003cspan additionalcitationids=\"CR120 CR121\" citationid=\"CR119\" class=\"CitationRef\"\u003e119\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR122\" class=\"CitationRef\"\u003e122\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTIGIT expression by Tregs suggested that they were likely to be highly suppressive effector Tregs (\u003cspan citationid=\"CR123\" class=\"CitationRef\"\u003e123\u003c/span\u003e). Therefore, their presence in the affected areas of the brain may play a pivotal role in suppressing the activity of the clonally focused CD8 T cells found in the same brain areas. It has been proposed that Treg dysfunction contributes to the etiology of autoimmune diseases (\u003cspan citationid=\"CR124\" class=\"CitationRef\"\u003e124\u003c/span\u003e, \u003cspan citationid=\"CR125\" class=\"CitationRef\"\u003e125\u003c/span\u003e), and ways to increase Treg numbers and efficacy in autoimmune diseases are being actively pursued (\u003cspan additionalcitationids=\"CR126\" citationid=\"CR125\" class=\"CitationRef\"\u003e125\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR127\" class=\"CitationRef\"\u003e127\u003c/span\u003e). In addition to TIGIT, our analysis indicated that Tregs were regulated by several other co-stimulatory and co-inhibitory proteins including ICOS, CD27, CD28, and CTLA4 (Figure \u003cspan refid=\"MOESM6\" class=\"InternalRef\"\u003eS6\u003c/span\u003e). CTLA4 is constitutively expressed by Tregs and competes for binding to CD86 and CD80 on antigen presenting cells to suppress conventional T cells by cell intrinsic and extrinsic mechanisms (\u003cspan citationid=\"CR128\" class=\"CitationRef\"\u003e128\u003c/span\u003e, \u003cspan citationid=\"CR129\" class=\"CitationRef\"\u003e129\u003c/span\u003e), that may also involve engagement of the TCR (\u003cspan citationid=\"CR130\" class=\"CitationRef\"\u003e130\u003c/span\u003e). Compared with ligands for other costimulatory and co-inhibitory genes including CD80, we detected CD86 transcripts in the majority of myeloid cells suggesting that the myeloid cells may regulate Treg cells present in the affected brain area via CTLA4 and CD28 binding to CD86 (Figure \u003cspan refid=\"MOESM12\" class=\"InternalRef\"\u003eS12\u003c/span\u003e) (\u003cspan citationid=\"CR131\" class=\"CitationRef\"\u003e131\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMost of myeloid cells that we isolated from affected brain tissue were microglia and were distinguished from smaller populations of macrophages and monocyte-derived immune cells. Based on the default resolution parameter used, ten clusters of microglia were generated by the Louvain algorithm, although all cells expressed known homeostatic microglia marker genes, and genes associated with activated microglia. Trajectory analysis indicated that microglia transitioned from cells expressing higher transcript levels of genes associated with phagocytosis, FCGR3A (CD16) and AIF1 (Ib1a), to cells expressing higher levels of proinflammatory cytokines, CCL3/4 and IL1B, and CD83 to cells expressing CD83 and heat shock protein transcripts, although the latter might be attributable to stress induced during \u003cem\u003eex vivo\u003c/em\u003e dissociation of the brain tissue (\u003cspan citationid=\"CR132\" class=\"CitationRef\"\u003e132\u003c/span\u003e). On the other hand, a subpopulation of microglia characterized by CD83 and heat shock protein gene expression was found to be selectively depleted in the substantia nigra of Parkinson disease patients implying a protective function that may also pertain to RE (\u003cspan citationid=\"CR133\" class=\"CitationRef\"\u003e133\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMost microglia expressed HLA class II molecules, ITAGX (CD11c), and CD86 transcripts indicating competency as antigen presenting cells (\u003cspan citationid=\"CR134\" class=\"CitationRef\"\u003e134\u003c/span\u003e, \u003cspan citationid=\"CR135\" class=\"CitationRef\"\u003e135\u003c/span\u003e). As previously discussed, microglia could regulate CD8 T cell activity via LAG3 binding to HLA class II molecules, which is independent of the bound peptide (\u003cspan citationid=\"CR136\" class=\"CitationRef\"\u003e136\u003c/span\u003e), and via LGALS3 binding on the macrophages. A higher number of macrophages and monocyte-derived immune cells were present in the sample of brain tissue from patient 738, the adult surgery case in which we also found the highest number of plasma cells possibly indicting a greater de novo involvement of the peripheral immune system at the time of the surgery (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOur analysis highlights the complexity of the immune landscape in brain areas affected by RE and supports the involvement of clonally expanded antigen experienced resident memory CD8 T cells in the etiology of RE. Crosstalk between T cells and activated microglia, and monocyte-derived macrophages and dendritic cells is predicted from the scRNA-seq data. The expression of heat shock proteins and CD83 transcripts in some microglia indicates a potential protective function. The presence of resident Tregs in the affected brain area suggests that they may be playing a role in restraining the activity of the conventional T cells. Likewise, microglia and monocyte-derived cells may regulate T cell activity via CD86, Galectin 3 and Galectin 9. Activation of HERV-K proviruses in the affected brain area may be unrelated to the etiology of RE although it is conceivable that translation of viral proteins from this endogenous virus could generate neoantigens to which the patient\u0026rsquo;s immune system reacts.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eRE, Rasmussen encephalitis\u003c/p\u003e\n\u003cp\u003eFCD, Focal cortical dysplasia\u003c/p\u003e\n\u003cp\u003eMS, Multiple sclerosis\u003c/p\u003e\n\u003cp\u003eAED, Anti-epileptic drug\u003c/p\u003e\n\u003cp\u003escRNA-seq, single cell RNA sequencing\u003c/p\u003e\n\u003cp\u003eUMI, Unique Molecular Identifiers\u003c/p\u003e\n\u003cp\u003eUMAP, Uniform Manifold Approximation and Projection\u003c/p\u003e\n\u003cp\u003eNATMI, Network Analysis Toolkit for Multicellular Interactions\u003c/p\u003e\n\u003cp\u003eHLA, Human Leukocyte Antigen\u003c/p\u003e\n\u003cp\u003eHHV, Human herpes virus\u003c/p\u003e\n\u003cp\u003eTCR, T cell receptor\u003c/p\u003e\n\u003cp\u003eCDR3, Complementarity determining region 3\u003c/p\u003e\n\u003cp\u003eBCR, B cell receptor\u003c/p\u003e\n\u003cp\u003eIgH, Immunoglobulin heavy chain\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the UCLA Institutional Review Board (IRB no. 18-001048). The patients or their parents or legal guardians provided informed consent for the use of the surgical remnant and blood for research purposes according to the Declaration of Helsinki. De-identified patient information including age at seizure onset, age at surgery, and gender was collected with informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe patients or their parents or legal guardians provided informed consent for future publication or presentation at conferences of de-identified research results.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNCBI Gene Expression Omnibus accession number GSE312319\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCompeting interests\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis work was funded by the RE Children\u0026rsquo;s Project to GCO and in part by a DOD/CDMRP Rare Cancers Research Program Idea Award (RA220179), The Lindonlight Collective grant (GR-240004), and an NIH/NCI award (P50 CA211015)\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eto ACW. GQ-V received support as a member of The Collaboratory at the UCLA Institute for Quantitative and Computational Biosciences.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003cem\u003eAuthors\u0026apos; contributions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eGCO designed the study, GCO and GQ-V analyzed the data and wrote the paper, JWC processed tissue and blood specimens, AF provided the surgical specimens, HVV and SDM provided pathologically evaluated tissue sections, NS provided MRI scans, and ACW provided additional funding for the study. All authors reviewed a draft of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAcknowledgements\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSingle cell RNA sequencing was performed by the Technology Center for Genomics and Bioinformatics, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, and multiplex immunofluorescence staining was carried out by the Translational Pathology Core Laboratory, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBien, C. G. et al. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis: a European consensus statement. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e128\u003c/b\u003e (Pt 3), 454\u0026ndash;471 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlson, H. E. et al. Clinical application and evaluation of the Bien diagnostic criteria for Rasmussen encephalitis. \u003cem\u003eEpilepsia\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e (10), 1753\u0026ndash;1760 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVaradkar, S. et al. Rasmussen's encephalitis: clinical features, pathobiology, and treatment advances. \u003cem\u003eLancet Neurol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e (2), 195\u0026ndash;205 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeariso, K. et al. Presentation, diagnosis and treatment of bilateral Rasmussen's encephalitis in a 12-year-old female. \u003cem\u003eEpileptic Disord\u003c/em\u003e. \u003cb\u003e15\u003c/b\u003e (3), 324\u0026ndash;332 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFauser, S., Elger, C. E., Woermann, F. \u0026amp; Bien, C. G. Rasmussen encephalitis: Predisposing factors and their potential role in unilaterality. \u003cem\u003eEpilepsia\u003c/em\u003e \u003cb\u003e63\u003c/b\u003e (1), 108\u0026ndash;119 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOrsini, A. et al. Rasmussen's encephalitis: From immune pathogenesis towards targeted-therapy. \u003cem\u003eSeizure\u003c/em\u003e \u003cb\u003e81\u003c/b\u003e, 76\u0026ndash;83 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris, W. B. et al. Seizure outcomes in children with Rasmussen's encephalitis undergoing resective or hemispheric epilepsy surgery: an individual participant data meta-analysis. \u003cem\u003eJ. Neurosurg. Pediatr.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e (3), 274\u0026ndash;283 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwab, N. et al. CD8\u0026thinsp;+\u0026thinsp;T-cell clones dominate brain infiltrates in Rasmussen encephalitis and persist in the periphery. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e132\u003c/b\u003e (Pt 5), 1236\u0026ndash;1246 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider-Hohendorf, T. et al. CD8(+) T-cell pathogenicity in Rasmussen encephalitis elucidated by large-scale T-cell receptor sequencing. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 11153 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDandekar, S. et al. Shared HLA Class I and II Alleles and Clonally Restricted Public and Private Brain-Infiltrating alphabeta T Cells in a Cohort of Rasmussen Encephalitis Surgery Patients. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 608 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl Nimer, F. et al. Phenotypic and functional complexity of brain-infiltrating T cells in Rasmussen encephalitis. \u003cem\u003eNeurol. Neuroimmunol. Neuroinflamm\u003c/em\u003e. \u003cb\u003e5\u003c/b\u003e (1), e419 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmrom, D. et al. Rasmussen encephalitis and comorbid autoimmune diseases: A window into disease mechanism? \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e83\u003c/b\u003e (12), 1049\u0026ndash;1055 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeitner, D. F. et al. Brain molecular mechanisms in Rasmussen encephalitis. \u003cem\u003eEpilepsia\u003c/em\u003e \u003cb\u003e64\u003c/b\u003e (1), 218\u0026ndash;230 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAi, J. et al. Genetic Factors in Rasmussen's Encephalitis Characterized by Whole-Exome Sequencing. \u003cem\u003eFront. Neurosci.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 744429 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakahashi, Y. et al. Genetic variations of immunoregulatory genes associated with Rasmussen syndrome. \u003cem\u003eEpilepsy Res.\u003c/em\u003e \u003cb\u003e107\u003c/b\u003e (3), 238\u0026ndash;243 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Y. S. et al. HCMV infection and IFITM3 rs12252 are associated with Rasmussen's encephalitis disease progression. \u003cem\u003eAnn. Clin. Transl Neurol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (3), 558\u0026ndash;570 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBjornevik, K., Munz, C., Cohen, J. I. \u0026amp; Ascherio, A. Epstein-Barr virus as a leading cause of Multiple sclerosis: mechanisms and implications. \u003cem\u003eNat. Rev. Neurol.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e (3), 160\u0026ndash;171 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePower, C., Poland, S. D., Blume, W. T., Girvin, J. P. \u0026amp; Rice, G. P. Cytomegalovirus and Rasmussen's encephalitis. \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e336\u003c/b\u003e (8726), 1282\u0026ndash;1284 (1990).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarrell, M. A., Cheng, L., Cornford, M. E., Grody, W. W. \u0026amp; Vinters, H. V. Cytomegalovirus and Rasmussen's encephalitis. \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e337\u003c/b\u003e (8756), 1551\u0026ndash;1552 (1991).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVinters, H. V., Wang, R. \u0026amp; Wiley, C. A. Herpesviruses in chronic encephalitis associated with intractable childhood epilepsy. \u003cem\u003eHum. Pathol.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e (8), 871\u0026ndash;879 (1993).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAtkins, M. R., Terrell, W. \u0026amp; Hulette, C. M. Rasmussen's syndrome: a study of potential viral etiology. \u003cem\u003eClin. Neuropathol.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (1), 7\u0026ndash;12 (1995).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, D. et al. Detection of EBV and HHV6 in the Brain Tissue of Patients with Rasmussen's Encephalitis. \u003cem\u003eVirol. Sin\u003c/em\u003e. \u003cb\u003e33\u003c/b\u003e (5), 402\u0026ndash;409 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcLachlan, R. S., Levin, S. \u0026amp; Blume, W. T. Treatment of Rasmussen's syndrome with ganciclovir. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e47\u003c/b\u003e (4), 925\u0026ndash;928 (1996).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcLachlan, R. S., Diosy, D. \u0026amp; Levin, S. Early treatment of a progressive Rasmussen's like syndrome with ganciclovir. \u003cem\u003eCan. J. Neurol. Sci.\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e (2), 296\u0026ndash;298 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEugster, A. et al. Physiological and pathogenic T cell autoreactivity converge in type 1 diabetes. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e (1), 9204 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHart, Y. M. et al. Double pathology in Rasmussen's syndrome: a window on the etiology? \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e50\u003c/b\u003e (3), 731\u0026ndash;735 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakei, H. et al. Dual pathology in Rasmussen's encephalitis: a study of seven cases and review of the literature. \u003cem\u003eNeuropathology\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e (4), 381\u0026ndash;391 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalev, G. \u0026amp; Prayson, R. A. Focal cortical dysplasia is a frequent coexistent pathology in patients with Rasmussen's encephalitis. \u003cem\u003eAnn. Diagn. Pathol.\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e, 152224 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBast, T., Ramantani, G., Seitz, A. \u0026amp; Rating, D. Focal cortical dysplasia: prevalence, clinical presentation and epilepsy in children and adults. \u003cem\u003eActa Neurol. Scand.\u003c/em\u003e \u003cb\u003e113\u003c/b\u003e (2), 72\u0026ndash;81 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVezzani, A., Aronica, E., Mazarati, A. \u0026amp; Pittman, Q. J. Epilepsy and brain inflammation. \u003cem\u003eExp. Neurol.\u003c/em\u003e \u003cb\u003e244\u003c/b\u003e, 11\u0026ndash;21 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar, P. et al. Single-cell transcriptomics and surface epitope detection in human brain epileptic lesions identifies pro-inflammatory signaling. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e (7), 956\u0026ndash;966 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTingleff, T. et al. Different pathways for preterm birth between singleton and twin pregnancies: a population-based registry study of 481 176 nulliparous women. \u003cem\u003eBJOG\u003c/em\u003e \u003cb\u003e130\u003c/b\u003e (4), 387\u0026ndash;395 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHagberg, H. et al. The role of inflammation in perinatal brain injury. \u003cem\u003eNat. Rev. Neurol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (4), 192\u0026ndash;208 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRattani, A. et al. Incidence of Epilepsy and Associated Risk Factors in Perinatal Ischemic Stroke Survivors. \u003cem\u003ePediatr. Neurol.\u003c/em\u003e \u003cb\u003e90\u003c/b\u003e, 44\u0026ndash;55 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamaswamy, V. et al. Inflammasome induction in Rasmussen's encephalitis: cortical and associated white matter pathogenesis. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e, 152 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWirenfeldt, M. et al. Increased activation of Iba1\u0026thinsp;+\u0026thinsp;microglia in pediatric epilepsy patients with Rasmussen's encephalitis compared with cortical dysplasia and tuberous sclerosis complex. \u003cem\u003eNeurobiol. Dis.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e (3), 432\u0026ndash;440 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestfall, J. J. et al. Molecular and spatial heterogeneity of microglia in Rasmussen encephalitis. \u003cem\u003eActa Neuropathol. Commun.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e (1), 168 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePardo, C. A. et al. The pathology of Rasmussen syndrome: stages of cortical involvement and neuropathological studies in 45 hemispherectomies. \u003cem\u003eEpilepsia\u003c/em\u003e \u003cb\u003e45\u003c/b\u003e (5), 516\u0026ndash;526 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTroscher, A. R. et al. Microglial nodules provide the environment for pathogenic T cells in human encephalitis. \u003cem\u003eActa Neuropathol.\u003c/em\u003e \u003cb\u003e137\u003c/b\u003e (4), 619\u0026ndash;635 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOwens, G. C. et al. Evidence for the involvement of gamma delta T cells in the immune response in Rasmussen encephalitis. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e (1), 134 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, G. et al. Processing single-cell RNA-seq datasets using SingCellaR. \u003cem\u003eSTAR. Protoc.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e (2), 101266 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHao, Y. et al. Integrated analysis of multimodal single-cell data. \u003cem\u003eCell\u003c/em\u003e \u003cb\u003e184\u003c/b\u003e (13), 3573\u0026ndash;87e29 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKorsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. \u003cem\u003eNat. Methods\u003c/em\u003e. \u003cb\u003e16\u003c/b\u003e (12), 1289\u0026ndash;1296 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao, J. et al. The single-cell transcriptional landscape of mammalian organogenesis. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e566\u003c/b\u003e (7745), 496\u0026ndash;502 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHou, R., Denisenko, E., Ong, H. T., Ramilowski, J. A. \u0026amp; Forrest, A. R. R. Predicting cell-to-cell communication networks using NATMI. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (1), 5011 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeam, I. immunarch: An R Package for Painless Bioinformatics Analysis of T-Cell and B-Cell Immune Repertoires 2019 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.5281/zenodo.3367200\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.3367200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsieh, T. C., Ma, K. H. \u0026amp; Chao, A. iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e (12), 1451\u0026ndash;1456 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlor, M. R interface to D3 chord diagrams 2020 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/mattflor/chorddiag\u003c/span\u003e\u003cspan address=\"https://github.com/mattflor/chorddiag\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith, T., Heger, A. \u0026amp; Sudbery, I. UMI-tools: modeling sequencing errors in Unique Molecular Identifiers to improve quantification accuracy. \u003cem\u003eGenome Res.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e (3), 491\u0026ndash;499 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBost, P. et al. Host-Viral Infection Maps Reveal Signatures of Severe COVID-19 Patients. \u003cem\u003eCell\u003c/em\u003e \u003cb\u003e181\u003c/b\u003e (7), 1475\u0026ndash;88e12 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXue, B. et al. Identification of the distribution of human endogenous retroviruses K (HML-2) by PCR-based target enrichment sequencing. \u003cem\u003eRetrovirology\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e (1), 10 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurn, A., Roy, F., Freeman, M. \u0026amp; Coffin, J. M. Widespread expression of the ancient HERV-K (HML-2) provirus group in normal human tissues. \u003cem\u003ePLoS Biol.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e (10), e3001826 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHao, Z. et al. RIdeogram: drawing SVG graphics to visualize and map genome-wide data on the idiograms. \u003cem\u003ePeerJ Comput. Sci.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, e251 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJeusset, L. et al. ViCloD, an interactive web tool for visualizing B cell repertoires and analyzing intraclonal diversities: application to human B-cell tumors. \u003cem\u003eNAR Genom Bioinform\u003c/em\u003e. \u003cb\u003e5\u003c/b\u003e (2), lqad064 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChristo, S. N., Park, S. L., Mueller, S. N. \u0026amp; Mackay, L. K. The Multifaceted Role of Tissue-Resident Memory T Cells. \u003cem\u003eAnnu. Rev. Immunol.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e (1), 317\u0026ndash;345 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSzabo, P. A. et al. Single-cell transcriptomics of human T cells reveals tissue and activation signatures in health and disease. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e (1), 4706 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, C. et al. Prioritizing exhausted T cell marker genes highlights immune subtypes in pan-cancer. \u003cem\u003eiScience\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e (4), 106484 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFuchs, Y. F. et al. Gene Expression-Based Identification of Antigen-Responsive CD8(+) T Cells on a Single-Cell Level. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 2568 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, B. et al. The SKI proto-oncogene restrains the resident CD103(+)CD8(+) T cell response in viral clearance. \u003cem\u003eCell. Mol. Immunol.\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e (10), 2410\u0026ndash;2421 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShembrey, C., Foroutan, M. \u0026amp; Hollande, F. A new natural killer cell-specific gene signature predicting recurrence in colorectal cancer patients. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 1011247 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiddleton, D., Curran, M. \u0026amp; Maxwell, L. Natural killer cells and their receptors. \u003cem\u003eTranspl. Immunol.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e (2\u0026ndash;3), 147\u0026ndash;164 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGittelman, R. M. et al. Longitudinal analysis of T cell receptor repertoires reveals shared patterns of antigen-specific response to SARS-CoV-2 infection. \u003cem\u003eJCI Insight\u003c/em\u003e ;\u003cb\u003e7\u003c/b\u003e(10). (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVenturi, V., Price, D. A., Douek, D. C. \u0026amp; Davenport, M. P. The molecular basis for public T-cell responses? \u003cem\u003eNat. Rev. Immunol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (3), 231\u0026ndash;238 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBagaev, D. V. et al. VDJdb in 2019: database extension, new analysis infrastructure and a T-cell receptor motif compendium. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e48\u003c/b\u003e (D1), D1057\u0026ndash;D62 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValkiers, S., Van Houcke, M., Laukens, K. \u0026amp; Meysman, P. ClusTCR: a Python interface for rapid clustering of large sets of CDR3 sequences with unknown antigen specificity. \u003cem\u003eBioinformatics\u003c/em\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatir, A., Shih, B., McColl, B. W. \u0026amp; Freeman, T. C. A core transcriptional signature of human microglia: Derivation and utility in describing region-dependent alterations associated with Alzheimer's disease. \u003cem\u003eGlia\u003c/em\u003e \u003cb\u003e67\u003c/b\u003e (7), 1240\u0026ndash;1253 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSankowski, R. et al. Mapping microglia states in the human brain through the integration of high-dimensional techniques. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e (12), 2098\u0026ndash;2110 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartins-Ferreira, R. et al. The Human Microglia Atlas (HuMicA) unravels changes in disease-associated microglia subsets across neurodegenerative conditions. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e (1), 739 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, W. K. et al. CD163 identifies perivascular macrophages in normal and viral encephalitic brains and potential precursors to perivascular macrophages in blood. \u003cem\u003eAm. J. Pathol.\u003c/em\u003e \u003cb\u003e168\u003c/b\u003e (3), 822\u0026ndash;834 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWen, W., Cheng, J. \u0026amp; Tang, Y. Brain perivascular macrophages: current understanding and future prospects. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e147\u003c/b\u003e (1), 39\u0026ndash;55 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillar, J. \u0026amp; Segura, E. Decoding the Heterogeneity of Human Dendritic Cell Subsets. \u003cem\u003eTrends Immunol.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e (12), 1062\u0026ndash;1071 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoltman, I. R. et al. Induction of a common microglia gene expression signature by aging and neurodegenerative conditions: a co-expression meta-analysis. \u003cem\u003eActa Neuropathol. Commun.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 31 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMasuda, T., Sankowski, R., Staszewski, O. \u0026amp; Prinz, M. Microglia Heterogeneity in the Single-Cell Era. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e (5), 1271\u0026ndash;1281 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaik, S. et al. Updated insights into the molecular networks for NLRP3 inflammasome activation. \u003cem\u003eCell. Mol. Immunol.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e (6), 563\u0026ndash;596 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWanke, F. et al. EBI2 Is Highly Expressed in Multiple Sclerosis Lesions and Promotes Early CNS Migration of Encephalitogenic CD4 T Cells. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e (5), 1270\u0026ndash;1284 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKurschus, F. C. \u0026amp; Wanke, F. EBI2 - Sensor for dihydroxycholesterol gradients in neuroinflammation. \u003cem\u003eBiochimie\u003c/em\u003e \u003cb\u003e153\u003c/b\u003e, 52\u0026ndash;55 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOwens, G. C. et al. Evidence for Resident Memory T Cells in Rasmussen Encephalitis. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 64 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMackay, L. K. et al. Hobit and Blimp1 instruct a universal transcriptional program of tissue residency in lymphocytes. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e352\u003c/b\u003e (6284), 459\u0026ndash;463 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFonseca, R. et al. Runx3 drives a CD8(+) T cell tissue residency program that is absent in CD4(+) T cells. \u003cem\u003eNat. Immunol.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (8), 1236\u0026ndash;1245 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBouwman, A. C., van Daalen, K. R., Crnko, S., Ten Broeke, T. \u0026amp; Bovenschen, N. Intracellular and Extracellular Roles of Granzyme K. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 677707 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArazi, A. et al. The immune cell landscape in kidneys of patients with lupus nephritis. \u003cem\u003eNat. Immunol.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e (7), 902\u0026ndash;914 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJonsson, A. H. et al. Granzyme K(+) CD8 T cells form a core population in inflamed human tissue. \u003cem\u003eSci. Transl Med.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (649), eabo0686 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcLane, L. M., Abdel-Hakeem, M. S. \u0026amp; Wherry, E. J. CD8 T Cell Exhaustion During Chronic Viral Infection and Cancer. \u003cem\u003eAnnu. Rev. Immunol.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, 457\u0026ndash;495 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaessler, A. \u0026amp; Vignali, D. A. A. T Cell Exhaustion. \u003cem\u003eAnnu. Rev. Immunol.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e (1), 179\u0026ndash;206 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMair, K. M. et al. The dynamics of brain T cell populations during the course of rasmussen encephalitis: from expansion to exhaustion. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e22\u003c/b\u003e (1), 155 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrebinoski, S. et al. Autoreactive CD8(+) T cells are restrained by an exhaustion-like program that is maintained by LAG3. \u003cem\u003eNat. Immunol.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (6), 868\u0026ndash;877 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBettini, M. et al. Cutting edge: accelerated autoimmune diabetes in the absence of LAG-3. \u003cem\u003eJ. Immunol.\u003c/em\u003e \u003cb\u003e187\u003c/b\u003e (7), 3493\u0026ndash;3498 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurnell, S. E. A. et al. Seven mysteries of LAG-3: a multi-faceted immune receptor of increasing complexity. \u003cem\u003eImmunother Adv.\u003c/em\u003e \u003cb\u003e2\u003c/b\u003e (1), ltab025 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAngin, M., Brignone, C. \u0026amp; Triebel, F. A LAG-3-Specific Agonist Antibody for the Treatment of T Cell-Induced Autoimmune Diseases. \u003cem\u003eJ. Immunol.\u003c/em\u003e \u003cb\u003e204\u003c/b\u003e (4), 810\u0026ndash;818 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaoi, Y. et al. CD106 in Tumor-Specific Exhausted CD8\u0026thinsp;+\u0026thinsp;T Cells Mediates Immunosuppression by Inhibiting TCR Signaling. \u003cem\u003eCancer Res.\u003c/em\u003e \u003cb\u003e84\u003c/b\u003e (13), 2109\u0026ndash;2122 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuy, C. et al. LAG3 associates with TCR-CD3 complexes and suppresses signaling by driving co-receptor-Lck dissociation. \u003cem\u003eNat. Immunol.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (5), 757\u0026ndash;767 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMihaylova, N. et al. Suppression of autoreactive T and B lymphocytes by anti-annexin A1 antibody in a humanized NSG murine model of systemic lupus erythematosus. \u003cem\u003eClin. Exp. Immunol.\u003c/em\u003e \u003cb\u003e199\u003c/b\u003e (3), 278\u0026ndash;293 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD'Acquisto, F. et al. Annexin-1 modulates T-cell activation and differentiation. \u003cem\u003eBlood\u003c/em\u003e \u003cb\u003e109\u003c/b\u003e (3), 1095\u0026ndash;1102 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaschalidis, N. et al. Modulation of experimental autoimmune encephalomyelitis by endogenous annexin A1. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, 33 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD'Acquisto, F. On the adaptive nature of annexin-A1. \u003cem\u003eCurr. Opin. Pharmacol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (4), 521\u0026ndash;528 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErnst, S. et al. An annexin 1 N-terminal peptide activates leukocytes by triggering different members of the formyl peptide receptor family. \u003cem\u003eJ. Immunol.\u003c/em\u003e \u003cb\u003e172\u003c/b\u003e (12), 7669\u0026ndash;7676 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGavins, F. N. \u0026amp; Hickey, M. J. Annexin A1 and the regulation of innate and adaptive immunity. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 354 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNapolitano, F. \u0026amp; Montuori, N. The N-formyl peptide receptors: much more than chemoattractant receptors. Relevance in health and disease. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, 1568629 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, Z. et al. Formyl peptide receptor 1 signaling potentiates inflammatory brain injury. \u003cem\u003eSci. Transl Med.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 605 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSinclair, J. Human cytomegalovirus: Latency and reactivation in the myeloid lineage. \u003cem\u003eJ. Clin. Virol.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e (3), 180\u0026ndash;185 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSavard, M. et al. Infection of primary human monocytes by Epstein-Barr virus. \u003cem\u003eJ. Virol.\u003c/em\u003e \u003cb\u003e74\u003c/b\u003e (6), 2612\u0026ndash;2619 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBabcock, G. J., Decker, L. L., Volk, M. \u0026amp; Thorley-Lawson, D. A. EBV persistence in memory B cells in vivo. \u003cem\u003eImmunity\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (3), 395\u0026ndash;404 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRocamonde, B., Hasan, U., Mathieu, C. \u0026amp; Dutartre, H. Viral-induced neuroinflammation: Different mechanisms converging to similar exacerbated glial responses. \u003cem\u003eFront. Neurosci.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 1108212 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWakim, L. M., Woodward-Davis, A. \u0026amp; Bevan, M. J. Memory T cells persisting within the brain after local infection show functional adaptations to their tissue of residence. \u003cem\u003eProc. Natl. Acad. Sci. U S A\u003c/em\u003e. \u003cb\u003e107\u003c/b\u003e (42), 17872\u0026ndash;17879 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakahashi, Y. et al. Vaccination and infection as causative factors in Japanese patients with Rasmussen syndrome: molecular mimicry and HLA class I. \u003cem\u003eClin. Dev. Immunol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e (2\u0026ndash;4), 381\u0026ndash;387 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePasternack, N. et al. Endogenous retroviruses are dysregulated in ALS. \u003cem\u003eiScience\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e (7), 110147 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShah, A. H. et al. Human endogenous retrovirus K contributes to a stem cell niche in glioblastoma. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e ;\u003cb\u003e133\u003c/b\u003e(13). (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoucet-O'Hare, T. T. et al. SMARCB1 deletion in atypical teratoid rhabdoid tumors results in human endogenous retrovirus K (HML-2) expression. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (1), 12893 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMayer, J. et al. An almost-intact human endogenous retrovirus K on human chromosome 7. \u003cem\u003eNat. Genet.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (3), 257\u0026ndash;258 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDewannieux, M., Blaise, S. \u0026amp; Heidmann, T. Identification of a functional envelope protein from the HERV-K family of human endogenous retroviruses. \u003cem\u003eJ. Virol.\u003c/em\u003e \u003cb\u003e79\u003c/b\u003e (24), 15573\u0026ndash;15577 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDembny, P. et al. Human endogenous retrovirus HERV-K(HML-2) RNA causes neurodegeneration through Toll-like receptors. \u003cem\u003eJCI Insight\u003c/em\u003e ;\u003cb\u003e5\u003c/b\u003e(7). (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNevalainen, T., Autio, A. \u0026amp; Hurme, M. Human endogenous retroviruses of the HERV-K (HML-2) family are expressed in the brain of healthy individuals and modify the composition of the brain-infiltrating immune cells. \u003cem\u003eHeliyon\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (11), e21283 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUkita, M. et al. CXCL13-producing CD4\u0026thinsp;+\u0026thinsp;T cells accumulate in the early phase of tertiary lymphoid structures in ovarian cancer. \u003cem\u003eJCI Insight\u003c/em\u003e ;\u003cb\u003e7\u003c/b\u003e(12). (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePipi, E. et al. Tertiary Lymphoid Structures: Autoimmunity Goes Local. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 1952 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKobayashi, S. et al. A distinct human CD4\u0026thinsp;+\u0026thinsp;T cell subset that secretes CXCL13 in Rheumatoid synovium. \u003cem\u003eArthritis Rheum.\u003c/em\u003e \u003cb\u003e65\u003c/b\u003e (12), 3063\u0026ndash;3072 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, R. et al. Autoimmunity to munc-18 in Rasmussen's encephalitis. \u003cem\u003eNeuron\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e (2), 375\u0026ndash;383 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlvarez-Baron, E. et al. Autoantibodies to Munc18, cerebral plasma cells and B-lymphocytes in Rasmussen encephalitis. \u003cem\u003eEpilepsy Res.\u003c/em\u003e \u003cb\u003e80\u003c/b\u003e (1), 93\u0026ndash;97 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNibber, A. et al. Antibodies to AMPA receptors in Rasmussen's encephalitis. \u003cem\u003eEur. J. Paediatr. Neurol.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e (2), 222\u0026ndash;227 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThilo, B. et al. A case of Rasmussen encephalitis treated with rituximab. \u003cem\u003eNat. Rev. Neurol.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e (8), 458\u0026ndash;462 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTimarova, G., Lisa, I. \u0026amp; Kukumberg, P. Long-term effect of rituximab in a case with late-onset Rasmussen s encephalitis with anti-ganglioside IgGQ1b and anti-GAD antibodies positivity. Case Report. \u003cem\u003eNeuro Endocrinol. Lett.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e (3), 179\u0026ndash;183 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJagtap, S. A. et al. Rituximab in Rasmussen's encephalitis: A single center experience and review of the literature. \u003cem\u003eEpilepsy Behav. Rep.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 100540 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErdogan, S., Aliyeva, G., Kizil, T. E. O. \u0026amp; Sener, O. Rituximab-responsive late-onset Rasmussen's encephalitis. \u003cem\u003eSeizure\u003c/em\u003e \u003cb\u003e123\u003c/b\u003e, 57\u0026ndash;59 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJoller, N. et al. Treg cells expressing the coinhibitory molecule TIGIT selectively inhibit proinflammatory Th1 and Th17 cell responses. \u003cem\u003eImmunity\u003c/em\u003e \u003cb\u003e40\u003c/b\u003e (4), 569\u0026ndash;581 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSumida, T. S., Cheru, N. T. \u0026amp; Hafler, D. A. The regulation and differentiation of regulatory T cells and their dysfunction in autoimmune diseases. \u003cem\u003eNat. Rev. Immunol.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e (7), 503\u0026ndash;517 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohr, A., Atif, M., Balderas, R., Gorochov, G. \u0026amp; Miyara, M. The role of FOXP3(+) regulatory T cells in human autoimmune and inflammatory diseases. \u003cem\u003eClin. Exp. Immunol.\u003c/em\u003e \u003cb\u003e197\u003c/b\u003e (1), 24\u0026ndash;35 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSelck, C. \u0026amp; Dominguez-Villar, M. Antigen-Specific Regulatory T Cell Therapy in Autoimmune Diseases and Transplantation. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 661875 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchloder, J., Shahneh, F., Schneider, F. J. \u0026amp; Wieschendorf, B. Boosting regulatory T cell function for the treatment of autoimmune diseases - That's only half the battle! \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 973813 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWing, K., Yamaguchi, T. \u0026amp; Sakaguchi, S. Cell-autonomous and -non-autonomous roles of CTLA-4 in immune regulation. \u003cem\u003eTrends Immunol.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e (9), 428\u0026ndash;433 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVignali, D. A., Collison, L. W. \u0026amp; Workman, C. J. How regulatory T cells work. \u003cem\u003eNat. Rev. Immunol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (7), 523\u0026ndash;532 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDikiy, S. \u0026amp; Rudensky, A. Y. Principles of regulatory T cell function. \u003cem\u003eImmunity\u003c/em\u003e \u003cb\u003e56\u003c/b\u003e (2), 240\u0026ndash;255 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalliday, N. et al. CD86 Is a Selective CD28 Ligand Supporting FoxP3\u0026thinsp;+\u0026thinsp;Regulatory T Cell Homeostasis in the Presence of High Levels of CTLA-4. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 600000 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarsh, S. E. et al. Dissection of artifactual and confounding glial signatures by single-cell sequencing of mouse and human brain. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e (3), 306\u0026ndash;316 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChatila, Z. K. et al. RNA- and ATAC-sequencing Reveals a Unique CD83\u0026thinsp;+\u0026thinsp;Microglial Population Focally Depleted in Parkinson's Disease. bioRxiv. (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchetters, S. T. T., Gomez-Nicola, D., Garcia-Vallejo, J. J. \u0026amp; Van Kooyk, Y. Neuroinflammation: Microglia and T Cells Get Ready to Tango. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 1905 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChhatbar, C. \u0026amp; Prinz, M. The roles of microglia in viral encephalitis: from sensome to therapeutic targeting. \u003cem\u003eCell. Mol. Immunol.\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e (2), 250\u0026ndash;258 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacLachlan, B. J. et al. Molecular characterization of HLA class II binding to the LAG-3 T cell co-inhibitory receptor. \u003cem\u003eEur. J. Immunol.\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e (2), 331\u0026ndash;341 (2021).\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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Rasmussen encephalitis, T cells, microglia, single cell RNA sequencing, endogenous retrovirus","lastPublishedDoi":"10.21203/rs.3.rs-8584794/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8584794/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eRasmussen encephalitis (RE) is a rare neuroinflammatory disease characterized by intractable seizures and progressive brain atrophy that is usually confined to one cerebral hemisphere. Disease management involves anti-inflammatory, immune modulatory and anti-epileptic drugs, although surgical resection remains the only effective treatment option to achieve seizure freedom. The presence of clonally expanded resident memory T cells in brain tissue removed to control seizures suggests the involvement of an autoimmune response in the etiology of the disease.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eBlocks of fresh brain tissue were obtained from three RE surgery cases (ages 5, 8, and 26 years at the time of surgery) and immune cells were isolated. Single cell RNA sequencing was used to define the types of immune cells present in the affected brain tissue and potential crosstalk between them, along with multiplex immunofluorescence immunostaining of sections from the same specimens. We matched T cell receptor sequences to T cell phenotypes and used ViralTrack software to search for evidence of activation of latent viruses in the immune cells.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe immune cells isolated from the three RE cases comprised primarily activated microglia and resident memory CD8 T cells with fewer CD4 T cells, NK cells and monocyte-derived macrophages and dendritic cells. The majority of CD8 T cells expressed killer cell lectin-like receptors, and a virus responsive gene signature that included XCL1, TNFRSF9 and CRTAM, but also the exhaustion markers LAG3 and TIM3. Microglia expressed transcripts found in disease-associated microglia and transcripts associated with NLRP3 inflammasomes. We found no evidence for active latent viruses; however, we found endogenous HERV-K retrovirus sequences that were transcribed from multiple provirus insertion sites.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eOur analysis highlights the complexity of the immune landscape in brain areas affected by RE and supports a central role for clonally expanded antigen experienced resident memory CD8 T cells. From the RNA sequencing data, we conclude that there is extensive cross talk between T cells and activated microglia, and monocyte-derived macrophages and dendritic cells that may regulate T cell activity.\u003c/p\u003e","manuscriptTitle":"Immune landscape of the affected brain in Rasmussen encephalitis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-29 00:47:07","doi":"10.21203/rs.3.rs-8584794/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-16T08:32:25+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-11T20:15:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"236064080479505992966036729061672859882","date":"2026-01-26T07:18:54+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-24T16:41:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"85230081158399325745294413699133599444","date":"2026-01-23T20:30:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"113816475651978810350649502779357255838","date":"2026-01-23T15:55:40+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-23T15:10:17+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-16T09:31:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-13T11:42:23+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-13T11:37:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-12T18:39:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"dc58e069-98d5-4385-9334-aef319f9f7d4","owner":[],"postedDate":"January 29th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":61735490,"name":"Biological sciences/Immunology"},{"id":61735491,"name":"Health sciences/Neurology"},{"id":61735492,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-04-27T08:40:58+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-29 00:47:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8584794","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8584794","identity":"rs-8584794","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}