Immune cell profiling reveals expanded stem cell–like memory T cells in anti-GAD65-associated neurological syndromes

Sumanta Barman; Saskia Räuber; Katharina Eisenhut; Daniela Esser; Martijn van Duijn; Madeleine Scharf; Marisol Herrera-Rivero; Paul Disse; Lara-Maria Preuth; Valeria Gulyaeva; Ilja Schwan; Eliza vom Stein; Marius Jonas; Duygu Pul; Michael Heming; Louisa Müller-Miny; Manuela Paunovic; Christine Strippel; Ebru Haholu; Jan Bartosch; Elisabeth Kaufmann; Justina Dargvainiene; Sabine Kahl; Marius Ringelstein; Eric Bindels; Heinz Wiendl; Nikolas H Stoecklein; Johannes Fischer; Norbert Goebels; Lars Komorowski; Michael Roden; Andrea Rossi; Monika Stoll; Albert J Becker; Motaz Hamed; Christian G Bien; Romana Höftberger; Jan Bauer; Sven G Meuth; Maarten J Titulaer; Frank Leypoldt; Gerd Meyer zu Hörste; Franziska Thaler; Nico Melzer; in cooperation with the Erasmus Medical Center-AIE Study Group; the German Network for Research on Autoimmune Encephalitis

doi:10.1073/pnas.2514753123

. 2026 Mar 25;123(13):e2514753123. doi: 10.1073/pnas.2514753123

Immune cell profiling reveals expanded stem cell–like memory T cells in anti-GAD65-associated neurological syndromes

Sumanta Barman ^a,¹, Saskia Räuber ^a,¹, Katharina Eisenhut ^b,^c,¹, Daniela Esser ^d, Martijn van Duijn ^e, Madeleine Scharf ^f, Marisol Herrera-Rivero ^g, Paul Disse ^a, Lara-Maria Preuth ^a, Valeria Gulyaeva ^a, Ilja Schwan ^a, Eliza vom Stein ^a, Marius Jonas ^a, Duygu Pul ^a, Michael Heming ^h, Louisa Müller-Miny ^h, Manuela Paunovic ^e, Christine Strippel ^h, Ebru Haholu ^b,^c,ⁱ, Jan Bartosch ^b,^c,ⁱ, Elisabeth Kaufmann ^j, Justina Dargvainiene ^d, Sabine Kahl ^k,^l,^m, Marius Ringelstein ^a,ⁿ, Eric Bindels ^o, Heinz Wiendl ^h, Nikolas H Stoecklein ^p, Johannes Fischer ^q, Norbert Goebels ^a, Lars Komorowski ^f, Michael Roden ^k,^l,^m, Andrea Rossi ^r, Monika Stoll ^s, Albert J Becker ^t, Motaz Hamed ^u, Christian G Bien ^v, Romana Höftberger ^w, Jan Bauer ^x, Sven G Meuth ^a, Maarten J Titulaer ^e, Frank Leypoldt ^d,^y, Gerd Meyer zu Hörste ^h, Franziska Thaler ^b,^c,², Nico Melzer ^a,^2,³; in cooperation with the Erasmus Medical Center-AIE Study Group⁴; the German Network for Research on Autoimmune Encephalitis⁴

^aDepartment of Neurology, Medical Faculty and University Hospital Düsseldorf, Heinrich Heine University, Düsseldorf 40225, Germany

^bInstitute of Clinical Neuroimmunology, University Hospital, Ludwig-Maximilians-Universität München, Munich 81377, Germany

^cBiomedical Center, Faculty of Medicine, Ludwig-Maximilians-Universität München, Martinsried 82152, Germany

^dInstitute of Clinical Chemistry, University Hospital Schleswig-Holstein, Kiel 24105/Lübeck 23538, Germany

^eDepartment of Neurology, Erasmus MC University Medical Center, Rotterdam 3015 GD, the Netherlands

^fInstitute for Experimental Immunology, affiliated to Euroimmun Medizinische Labordiagnostika Aktiengesellschaft, Lübeck 23560, Germany

^gInstitute of Epidemiology and Social Medicine, University of Münster, Münster 48149, Germany

^hDepartment of Neurology, University Hospital Münster, Münster 48149, Germany

ⁱGraduate School of Systemic Neurosciences, Ludwig-Maximilians-Universität München, Planegg 82152, Germany

^jDepartment of Neurology, University Hospital, Ludwig-Maximilians-Universität München, Munich 81377, Germany

^kDepartment of Endocrinology and Diabetology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University, Düsseldorf 40225, Germany

^lInstitute for Clinical Diabetology, German Diabetes Center, Leibniz Institute for Diabetes Research at Heinrich-Heine-University, Düsseldorf 40225, Germany

^mGerman Center for Diabetes Research, München-Neuherberg 85764, Germany

ⁿDepartment of Neurology, Center for Neurology and Neuropsychiatry, Landschaftsverband Rheinland-Klinikum, Heinrich Heine University, Düsseldorf 40629, Germany

^oDepartment of Hematology, Erasmus MC University Medical Center, Rotterdam 3015 GD, the Netherlands

^pDepartment of General, Visceral and Pediatric Surgery, Medical Faculty and University Hospital Düsseldorf, Heinrich Heine University, Düsseldorf 40225, Germany

^qInstitute for Transplantation Diagnostics and Cell Therapeutics, Medical Faculty and University Hospital Düsseldorf, Heinrich Heine University, Düsseldorf 40225, Germany

^rGenome Engineering and Model Development Lab, Institut für umweltmedizinische Forschung-Leibniz Research Institute for Environmental Medicine, Düsseldorf 40225, Germany

^sDepartment of Genetic Epidemiology, Institute of Human Genetics, University of Münster, Münster 48149, Germany

^tInstitute of Cellular Neurosciences II, University Hospital Bonn, Bonn 53127, Germany

^uClinic for Neurosurgery, University Hospital of Bonn, Bonn 53127, Germany

^vDepartment of Epileptology, Bethel Epilepsy Center, Medical School Ostwestfalen-Lippe, Bielefeld University, Bielefeld 33617, Germany

^wDivision of Neuropathology and Neurochemistry, Department of Neurology and Comprehensive Center for Clinical Neurosciences and Mental Health, Medical University of Vienna, Vienna 1090, Austria

^xDepartment of Neuroimmunology, Centre for Brain Research, Medical University of Vienna, Vienna 1090, Austria

^yDepartment of Neurology, University Hospital Schleswig-Holstein, Kiel 24105, Germany

To whom correspondence may be addressed. Email: Nico.Melzer@med.uni-duesseldorf.de.

Edited by Lawrence Steinman, Stanford University Arnold and Mabel Beckman Center for Molecular and Genetic Medicine, Stanford, CA; received June 8, 2025; accepted February 10, 2026

¹S.B., S.R., and K.E. contributed equally to this work.

²F.T. and N.M. contributed equally to this work.

⁴Complete lists of Erasmus Medical Center-AIE Study group and German Network for Research on Autoimmune Encephalitis can be found in the SI Appendix.

PMCID: PMC13038060 PMID: 41880578

Significance

Autoimmune neurological syndromes (AINS) with antibodies targeting glutamic acid decarboxylase 65 (anti-GAD65 AINS) show limited response to currently available immunotherapies and are associated with poor clinical outcomes. A better understanding of the mechanisms underlying anti-GAD65 AINS is critical for developing targeted therapeutic strategies. Here, we identified an expansion of stem cell–like memory T cells (TSCM) in the peripheral blood and cerebrospinal fluid (CSF) and confirmed their presence in brain tissue of anti-GAD65 AINS individuals. Unlike CSF T cell receptors, the B cell receptor (BCR) repertoire displayed only little to no clonal expansion. Our findings shed light on the pathophysiology of anti-GAD65 AINS and nominate TSCM as a potential therapeutic target and disease biomarker.

Keywords: anti-GAD65 autoimmune neurological syndromes, single cell transcriptomics, immune repertoire profiling

Abstract

The immunopathogenesis of autoimmune neurological syndromes (AINS) with antibodies against the 65 kDa isoform of glutamic acid decarboxylase (anti-GAD65 AINS) remains poorly understood. To elucidate underlying disease mechanisms and identify relevant cell populations, we performed single-cell RNA and immune repertoire sequencing of cerebrospinal fluid (CSF) and peripheral blood mononuclear cells (PBMCs) of eight anti-GAD65 AINS individuals compared to eight noninflammatory controls. In addition, PBMCs from 19 anti-GAD65 AINS individuals and 20 healthy controls were analyzed by multidimensional flow cytometry, and brain tissue specimens from four anti-GAD65 AINS individuals were examined histologically. We detected higher frequencies of stem cell–like memory T cells (TSCM) within the PBMCs and a marked enrichment and clonal expansion of activated CD4⁺ TSCM in the CSF of anti-GAD65 AINS individuals. Expanded T cells exhibited increased expression of proinflammatory genes. Histological analyses confirmed intraparenchymal CD8⁺ TSCM in three of four anti-GAD65 AINS individuals and rare meningeal/intraparenchymal CD4⁺ TSCM in one person. Although CSF B cell receptors (BCRs) displayed little to no clonal expansion, recombinant expression of 40 CSF BCRs revealed that 25% were GAD65-reactive with increased somatic hypermutations compared to non-GAD65-reactive BCRs. These findings further support the concept of an antigen-specific intrathecal immune response. In summary, we characterize the immune landscape of anti-GAD65 AINS at single-cell resolution and identify clonally expanded TSCM with cytotoxic properties as a hallmark of this disease.

Autoimmune neurological syndromes (AINS) with autoantibodies against the 65 kDa isoform of glutamic acid decarboxylase (GAD65) are immune-mediated disorders typically presenting with distinct clinical phenotypes: 1) limbic encephalitis (LE) with temporal lobe seizures (TLS) or epilepsy (TLE), 2) stiff person syndrome (SPS), 3) cerebellar ataxia (CA), or overlaps of the aforementioned (1–3). The disease course may be subacute or slowly progressing with limited treatment response often leading to lifelong neurological disability.

Autoantibodies are directed against GAD65, the rate-limiting enzyme of gamma-aminobutyric acid (GABA) synthesis. Given the cytoplasmic localization of the target antigen, the pathophysiological relevance of GAD65-antibodies remains controversial. While GAD65-antibodies are intrathecally produced in anti-GAD65 AINS individuals and were shown to inhibit GAD65 enzymatic activity in vitro, neither a direct interaction between antibody and target antigen nor an antibody-mediated disease pathology could be demonstrated in animal models (4, 5). Interestingly, recent neuropathological evidence from GAD-TLE individuals highlights a pronounced infiltration of cytotoxic CD8⁺ T cells and plasma cells into the brain parenchyma (6, 7). We recently reported a genome-wide association study (GWAS) of anti-GAD65 AINS, where the top genetic variant localized to a segment in the HLA class I region and several other variants with regulatory functions on gene expression mapped to CD4⁺ T cells and the cerebral cortex (8). These findings underline the complex immunopathogenesis of anti-GAD65 AINS and the need for systems biology approaches to unravel the concise pathophysiological mechanisms.

Here, we present an in-depth characterization of peripheral and intrathecal immune cell populations in individuals with anti-GAD65 AINS compared with noninflammatory and healthy controls (HC), using multidimensional flow cytometry, histology, and single-cell RNA sequencing combined with immune repertoire profiling. We identify a clonal expansion of activated CD4⁺ stem cell–like memory T cells (TSCM) with a cytotoxic phenotype in the cerebrospinal fluid (CSF) of anti-GAD65 AINS individuals, likely contributing to disease pathogenesis. TSCM were likewise enriched in peripheral blood (PB), and multiplex immune fluorescence analyses confirmed intraparenchymal CD8⁺ TSCM in three of four anti-GAD65 AINS individuals, while meningeal and intraparenchymal CD4⁺ TSCM were identified in one person. Furthermore, we detected GAD65-reactive B cells with an increased frequency of somatic hypermutations circulating in the CSF of these individuals, suggesting they had undergone antigen-driven affinity maturation.

Results

Demographics and Basic Clinical Data.

Eight anti-GAD65 AINS individuals were included in the sequencing cohort (Fig. 1A and SI Appendix, Table S1). The median age of anti-GAD65 AINS individuals was 65 (39 to 71) y. Seven persons were female, four presented with limbic encephalitis/temporal lobe epilepsy (LE/TLE), two with CA, one with SPS, and one showed an overlap of SPS and LE/TLE (Fig. 1B). None of the anti-GAD65 AINS individuals received immunotherapy at the time of sampling. One person had received steroids more than six mo prior, and another had received immunoglobulins more than six y prior to sampling. Given the long interval between treatment and sampling, the persons can be considered immunotherapy-naïve (9). Three individuals showed CSF pleocytosis at sampling. Blood–CSF-barrier dysfunction (BCSFBD) was only apparent in one person. Three individuals had CSF-specific oligoclonal bands (OCBs) (SI Appendix, Table S2).

UMAP and plots showing increased CD4+ memory T cells in CSF and plasmablasts in blood of anti-GAD65 autoimmune neurological syndrome cases. — CD4⁺ TSCM and MLC are expanded in the CSF and plasmablasts in the PB of anti-GAD65 AINS individuals. (A) Study design. Created in BioRender. Räuber, S. (2026) https://BioRender.com/4j7w6s2; (B) Clinical phenotypes of anti-GAD65 AINS individuals. Number of persons with each clinical phenotype in relation to the total number of persons in the sequencing, mFC, and histology cohorts are shown. (C) UMAP showing 21 color-coded cell clusters of 1,18,492 single cell transcriptomes integrated from CSF cells and PBMC from anti-GAD65 AINS individuals (n = 8) and IIH individuals (n = 8). (D) Dot plot illustrating selected marker genes of cell clusters from CSF cells and PBMC from anti-GAD65 AINS and IIH individuals. Color encodes mean expression, dot size visualizes the fraction of cells expressing the gene. (E–M) Violin plots with overlaying box plots illustrating the differentially abundant Sc-Seq cell clusters between anti-GAD65 AINS and IIH individuals (E–I), the differences in PB plasmablasts (J), CSF activated CD4⁺ TSCM (K), CSF MLC (L), and CSF Bc1 (M) between anti-GAD65 AINS with short disease duration (<1 y), anti-GAD65 AINS with long disease duration (>1 y), and IIH individuals, Boxes display the median as well as the 25th and 75th percentiles. The whiskers extend from the hinge to the largest and smallest values, respectively, but no further than 1.5 * IQR from the hinge. P-values were calculated by the Kruskal–Wallis test with the Dunn post hoc test (p-adjustment method: Benjamini–Hochberg). *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. Act.—activated; Av—average; Bc—B cells; BCR—B cell receptor; CA—cerebellar ataxia; CSF—cerebrospinal fluid; cDC—conventional dendritic cell; DC—dendritic cells; expr.—expression; FC—fold change; GAD—individuals with glutamic acid decarboxylase antibody-associated AINS; Granulo—granulocytes; HC—healthy controls; IIH—idiopathic intracranial hypertension; LE—limbic encephalitis; long—long disease duration (>1 y); MAIT—Mucosal-associated invariant T cells; Mega—megakaryocytes; Mono—monocytes; NK—natural killer cells; PB—peripheral blood; PBMC—Peripheral Blood Mononuclear Cell; pDC—plasmacytoid dendritic cells; Sc-Seq—single-cell sequencing; short—short disease duration (<1 y); SPS—stiff person syndrome; Tc—T cells; TCR—T cell receptor; TCM—central memory T cells; TEM—effector memory T cells; TLE—temporal lobe epilepsy; TSCM—stem cell–like memory T cells; UMAP—Uniform Manifold Approximation and Projection.

In addition, eight persons with Idiopathic Intracranial Hypertension (IIH) were included as noninflammatory controls (Fig. 1A and SI Appendix, Table S3). The median age of the IIH individuals was 38 [27 to 52] y and seven were female. Two persons had CSF pleocytosis and/or BCSFBD. None of these individuals received immunotherapy at the time of or prior to sampling. Basic CSF characteristics of the IIH cohort are displayed in SI Appendix, Table S4.

In addition, a flow cytometry cohort consisting of 19 anti-GAD65 AINS individuals and 20 HC was included in the study (Fig. 1A and SI Appendix, Table S1). The median age was comparable between groups [anti-GAD65 AINS 47 (21 to 71) y and HC 47 (21 to 68) y]. 17 anti-GAD65 AINS and 14 HC were female. The majority of anti-GAD65 AINS individuals presented with LE/TLE (12/19), followed by SPS (2/19), overlap of SPS and LE/TLE as well as CA (2/19, respectively), and overlap of SPS, LE/TLE, and CA (1/19) (Fig. 1B). None of the persons received immunotherapy at the time of sampling. Six individuals were treated with immunotherapy prior to sampling (SI Appendix, Table S1). CSF routine parameters are summarized in SI Appendix, Table S2. For histological analysis, brain tissue specimens from four persons with anti-GAD65 LE/TLE who underwent epilepsy surgery were included (SI Appendix, Table S1).

Expansion of TSCM in the PB and CSF of Anti-GAD65 AINS Individuals.

Applying single-cell RNA sequencing (scRNA-seq), we analyzed peripheral blood mononuclear cells (PBMCs) and CSF cells in an unbiased manner. We integrated scRNA-seq data from PBMCs and CSF of anti-GAD65 AINS and IIH individuals, yielding an object encompassing 118,492 total single-cell transcriptomes (PBMCs: 63,176, CSF: 55,316) with 4231.8 ± 691.5 (mean ± SEM) cells per sample and 1280.1 ± 1.65 (mean ± SEM) genes detected per cell. Hereafter, we refer to single-cell transcriptomes as “cells”. Then, we performed unsupervised clustering, which resulted in 21 cell clusters (Fig. 1C). Cluster annotation was based on marker gene expression using a combination of manual and automated approaches (Fig. 1D). When comparing anti-GAD65-AINS to IIH individuals, we found an increased proportion of plasmablasts in the PB of anti-GAD65 AINS individuals (Fig. 1E). In addition, the percentage of activated CD4⁺ TSCM - identified by the expression of typical TSCM markers (e.g., PTPRC (CD45), FAS (CD95), ITGAL (CD11a), CCR7 (CD197), SELL (CD62L), IL7R (CD127), CD27, CXCR3) (10–13)—was higher in anti-GAD65 AINS individuals. Given that PBMCs were only available from five anti-GAD65 AINS individuals, statistical significance could not be reached in the peripheral compartment (Fig. 1F). However, in the CSF, anti-GAD65 AINS individuals showed a marked expansion of the activated CD4⁺ TSCM cluster compared to the IIH group (Fig. 1G). Furthermore, higher percentages of microglia-like cells (MLC) and reduced frequencies of the B cell (Bc) cluster 1 were found in the CSF of anti-GAD65 AINS in comparison to IIH individuals (Fig. 1 H and I). We next subdivided the anti-GAD65 AINS cohort in persons with short disease duration (<1 y; n = 3) and long disease duration (>1 y; n = 5) at the time of sampling and compared the significant cell clusters between those two subgroups and IIH controls. PB plasmablasts were especially increased in anti-GAD65 AINS individuals with short disease duration (Fig. 1J) while CSF activated CD4⁺ TSCM were significantly higher in anti-GAD65 AINS individuals with long disease duration compared to the IIH group (Fig. 1K). No relevant differences were noted regarding CSF MLC and CSF Bc1 between groups (Fig. 1 L and M).

Given the involvement of memory T cells in autoimmune diseases such as multiple sclerosis (MS) and type 1 diabetes (T1D) (14–19), we next performed an in-depth characterization of TSCM in the PB using multidimensional flow cytometry (mFC) to validate our sequencing data. TSCM were defined by the coexpression of CD45RA, CCR7 (CD197), FAS (CD95), and CD27 (SI Appendix, Fig. S1). The percentage of TSCM was significantly higher in anti-GAD65 AINS individuals compared to HC, particularly among CD4⁺ TSCM (Fig. 2 A–C). Analysis of HLADR expression—as marker of T cell activation (20, 21)—revealed higher proportions of HLADR⁺ CD4⁺ T cells in anti-GAD65 AINS individuals in comparison to HC (Fig. 2D). Interestingly, CD4⁺ T cells of anti-GAD65 AINS individuals more frequently expressed CD69, a marker of activation and tissue homing (Fig. 2E) (22). Subdivision of anti-GAD65-AINS individuals based on disease duration revealed a persisting increase in TSCM, especially CD4⁺ TSCM, during the course of the disease (Fig. 2 F–H). Differences in CD8⁺ TSCM as well as total TSCM between anti-GAD65-AINS individuals with short disease duration and HC did not reach statistical significance (Fig. 2 F and H). Furthermore, histological analysis could confirm the presence of CD8⁺ TSCM in the brain parenchyma of three out of four and CD4⁺ TSCM in the meninges and brain parenchyma of one out of four anti-GAD65 LE/TLE individuals (Fig. 2I).

Charts and images showing increased memory T cells (TSCM) in blood and brain of anti-GAD65 patients, plus gene differences versus IIH controls. — TSCM in the PB and brain parenchyma of anti-GAD65-AINS individuals & differentially expressed genes involved in regulation of the immune response between anti-GAD65 AINS and IIH individuals. (A–E) Violin plots with overlaying box plots showing the differentially abundant cell populations assessed by mFC between anti-GAD65 AINS individuals and HC. Boxes display the median as well as the 25th and 75th percentiles. The whiskers extend from the hinge to the largest and smallest values, respectively, but no further than 1.5 * IQR from the hinge. P-values were calculated by the t test if normality could be assumed based on Shapiro–Wilk test, otherwise Wilcoxon signed-rank test was used. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. (F–H) Differences in PB TSCM between anti-GAD65 AINS individuals with short disease duration (<1 y), anti-GAD65 AINS individuals with long disease duration (>1 y), and HC. Boxes display the median as well as the 25th and 75th percentiles. The whiskers extend from the hinge to the largest and smallest values, respectively, but no further than 1.5 * IQR from the hinge. P-values were calculated by the Kruskal–Wallis test with the Dunn post hoc test (P-adjustment method: Benjamini–Hochberg). *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001; (I) Representative figures of multiplex immune fluorescence analyses of hippocampi from two persons with anti-GAD65 LE/TLE, respectively, depicting intraparenchymal CD27⁺CD45RA⁺CXCR3⁺CD8⁺ TSCM (*Upper* panel, short disease duration, 1 mo) and CD27⁺CD45RA⁺CXCR3⁺CD4⁺ TSCM (*Lower* panel, long disease duration, 6,3 y). (J–N) Volcano plots depicting the differentially expressed genes in the PB and CSF of anti-GAD65 AINS compared to IIH individuals. Pseudobulking was performed and the differential expression was tested with DESeq2. Adjusted P < 0.05 and log2FC ≥ 0.585 were used as cut-offs: J, K PBMC and CSF bulk analysis, respectively. L DGE analysis of CSF activated CD4⁺ TSCM. M: DGE analysis of CSF cDC cluster 1. N DGE analysis of CSF B cell cluster 1. Adj.—adjusted; Bc—B cells; CEBPD—CCAAT/enhancer-binding protein delta; CSF—cerebrospinal fluid; DC—dendritic cells; DGE—differential gene expression; EID2—EP300-interacting inhibitor of differentiation 2; FC—fold change; GAD—individuals with glutamic acid decarboxylase antibody-associated AINS; GNAS—Neuroendocrine secretory protein 55; HLA-C—Human Leukocyte Antigen-C; HC—healthy controls; IIH—idiopathic intracranial hypertension; KLF2—Krueppel-like factor 2; LE—limbic encephalitis; LINC01578—Long noncoding RNA LINC01578; long—long disease duration (>1 y); MAP2K2—Dual specificity mitogen-activated protein kinase kinase 2; mFC—multidimensional flow cytometry; PB—peripheral blood; PBMC—Peripheral Blood Mononuclear Cell; PCBP2—Poly(rC)-binding protein 2; short—short disease duration (<1 y); Tc—T cells; TCFL5—Transcription factor-like 5 protein; TGFB1—transforming growth factor-β; TPT1—Translationally-controlled tumor protein; TSCM—stem cell–like memory T cells.

Differences in Gene Expression Involved in the Regulation of Inflammatory Immune Responses Between Anti-GAD65 AINS and IIH Individuals.

To identify changes in gene expression patterns in anti-GAD65 AINS individuals in comparison to the IIH group, we performed differential gene expression (DGE) analysis and created volcano plots of differentially expressed genes (Fig. 2 J–N). First, CSF and PBMC analyses were performed across all cell types, respectively (Fig. 2 J and K). Next, the analysis was repeated for all PBMC and CSF cell clusters separately (Fig. 2 L–N).

Among the top differentially expressed genes in the PBMC bulk analysis were EID2—which functions as a chemo-attractant receptor mediating cell migration (23)—TCFL5—which was previously linked to impaired germinal center formation, B cell differentiation, and B cell receptor (BCR) signaling (24)—MAP2K2—reported to modulate expression of MHCII, CD40, CD80 and to have anti-inflammatory effects (25, 26)—TPT1—which was found to be increased in twins with MS compared with unaffected twins (27)—and PCBP2—essential for CD4⁺ T cell activation and proliferation (28). Furthermore, one gene, which we found to be associated with anti-GAD65-AINS in a GWAS, was downregulated in the current anti-GAD65 AINS cohort (HLA-C) (8) (Fig. 2J). Regarding the CSF, the following genes were among the top differentially regulated ones: GNAS—a gene which was previously associated with tumor immune cell infiltration and Th2 polarization (29, 30)—KLF2—a transcription factor critical for immune cell differentiation, activation, proliferation, and migration as well as generation of Tregs (31–33)—CD69—a classical marker of lymphocyte activation and tissue retention and regulator of Treg differentiation (22)—and TGFB1—a cytokine critical for both immune tolerance and immunity (34) (Fig. 2K). Having a closer look at the different cell clusters, we found the expression of LINC01578—encoding a long noncoding RNA (lncRNA) (35)—to be lower in activated CD4⁺ TSCM of anti-GAD65 AINS in comparison to IIH individuals (Fig. 2L). Conventional dendritic cells (cDC) 1 and Bc1 of anti-GAD65 AINS persons had reduced CEBPD expression—known to be implicated in regulation of IL-10 expression (36)—compared to the IIH group. Bc1 of anti-GAD65 AINS individuals also showed decreased GNAS expression (Fig. 2 M and N).

In summary, DGE analysis identified differences in the expression of several genes involved in the regulation of the inflammatory immune response between anti-GAD65 AINS and IIH persons. These might present therapeutic targets warranting further investigation.

Numerous CSF BCRs Are GAD65-Reactive and Somatically Hypermutated, But Originate From Nonexpanded B Cells.

We next sought to characterize the intrathecal B cell response in anti-GAD65 AINS individuals. BCR sequence analysis identified little to no clonally expanded B cells in the CSF compartment of anti-GAD65 AINS individuals (SI Appendix, Fig. S2 A–C). To determine whether those BCRs were still autoantigen-specific, we randomly selected CSF BCRs of different Ig isotypes with concurrently available consensus heavy and light chain BCR variable region sequence information from a representative subgroup of individuals (n = 4, Fig. 3A). The corresponding heavy and light chain sequences were synthesized, cloned, expressed, and affinity purified as recombinant monoclonal antibodies (mAbs) (Fig. 3A). Remarkably, out of 40 mAbs, 10 (25%) were GAD65-reactive, as shown by indirect immunofluorescence staining of transfected HEK293T cells (Fig. 3 B–D and SI Appendix, Fig. S2D) and nonhuman primate brain tissue (SI Appendix, Fig. S2E). Notably, 3 out of 10 GAD65-reactive mAbs additionally showed Purkinje-cell reactivity on nonhuman primate brain tissue stainings (SI Appendix, Fig. S2E). None of the 40 mAbs displayed any other clear anti-CNS reactivity. GAD65-reactive BCRs that could be assigned to specific B cell subsets (3/10) were mapped to memory and naïve B cell clusters. While immunoglobulin isotype distribution did not differ between anti-GAD65 AINS and IIH individuals in the CSF (SI Appendix, Fig. S2F), GAD65-reactive mAbs were exclusively of the IgG1 isotype and significantly more somatically hypermutated than non-GAD65-reactive mAbs (Fig. 3 C and E), suggesting that they likely have undergone antigen-driven affinity maturation. In general, significantly more somatic hypermutations were found in the CSF BCRs of anti-GAD65 AINS compared to IIH individuals (Fig. 3F).

Producing 40 recombinant mAbs from CSF B cells of AINS patients, identifying 10 GAD65-reactive antibodies with high mutation rates. — Recombinant production of mAbs from CSF BCR data reveals numerous GAD65-reactive BCRs with increased somatic hypermutations. (A) Monoclonal antibody production workflow: 40 recombinant mAbs were produced from n = 4 anti-GAD65 AINS individuals by gene synthesis of CSF BCR variable region sequences of matching heavy and light chains as extracted from Cell Ranger-processed data, subsequent cloning, expression, and affinity purification. GAD65-reactivity testing by cell-based assay and indirect immunofluorescence stainings on primate brain tissue showed GAD65-reactivity in (B) 10 out of 40 produced mAbs (25%) and (C) 10 out of 20 produced IgG1 mAbs (50%). (D) Immunofluorescence stainings of the 10 GAD65-reactive recombinant human mAbs (green) on HEK cells overexpressing GAD65 or control-transfected cells (as indicated in row caption). Each column represents a tested mAb (as indicated by ID of individuals and clonotype number in column caption). 7A2 serves as positive and HK3 as negative control, both mAbs which were produced in the same expression vector. (E) Quantification of somatic hypermutations in produced GAD65-reactive and nonreactive mAbs and (F) all CSF BCRs from anti-GAD65 AINS as compared to IIH individuals. BCR—B cell receptor; CBA—cell-based assay; cl.—clonotype; GAD—individuals with glutamic acid decarboxylase antibody-associated AINS; GAD65+—glutamic acid decarboxylase 65-reactive; GAD65−—not glutamic acid decarboxylase 65-reactive; IFT—indirect immunofluorescence testing; IIH—idiopathic intracranial hypertension; CSF—cerebrospinal fluid; mAb—monoclonal antibody; neg. ctrl—negative control; pos. ctrl—positive control; VR—variable region.

In summary, anti-GAD65 AINS individuals harbor numerous GAD65-reactive and somatically hypermutated BCRs in the CSF compartment, which however originate from nonexpanded B cells.

Compartment-Specific TCR Clonal Expansion in Anti-GAD65 AINS.

To analyze the T cell receptor (TCR) repertoire, we used IgBlast-aligned V(D)J data from T cells, combined with transcriptomic data from T cell clusters. Analysis of TCR repertoires revealed distinct patterns of clonal expansion between anti-GAD65 AINS and IIH individuals across different compartments. In the CSF of anti-GAD65 AINS individuals, we observed significantly expanded T cell clones characterized by unique TCR sequences, suggesting an antigen-driven immune response (Fig. 4 A–F). These expanded CSF clones demonstrated a distinct transcriptional profile, with DGE analysis revealing significant upregulation of genes involved in cellular cytotoxicity, TCR signaling, T cell proliferation and survival, chemotaxis, as well as in inflammatory immune responses (Fig. 4 B and C). In contrast, the CSF compartment of the control group showed limited clonal expansion, indicating a more diverse and nontargeted immune repertoire (Fig. 4 D and E). Intriguingly, this pattern was reversed in the PBMC compartment, where control subjects exhibited higher frequencies of expanded clones compared to anti-GAD65 AINS individuals (Fig. 4 D and E). Analysis of the top 20 most abundant TCR clones in the CSF further confirmed this pattern, demonstrating elevated clonal frequencies in anti-GAD65 AINS compared to IIH individuals suggesting a focused, disease-specific immune response (SI Appendix, Fig. S3 A and B). Again, this pattern was inverted in the PBMC compartment, where control subjects displayed higher frequencies of expanded clones compared to anti-GAD65 AINS individuals (SI Appendix, Fig. S3 A and B). Furthermore, TCR clonal overlap was observed between PBMC and CSF compartments in both anti-GAD65 AINS and IIH groups. Notably, a higher degree of clonal sharing was evident in the majority of IIH samples (5 out of 6) compared to anti-GAD65 AINS samples (2 out of 4). Moreover, a limited number of TCR clones were found to be shared across different samples in both groups (SI Appendix, Fig. S3 C and D).

TCR clones in anti-GAD65 AINS vs. IIH, showing expanded CSF T cells, gene enrichment, and clonal networks between CSF and PBMC. — Comprehensive analysis of T cell repertoire in anti-GAD65 AINS individuals and controls reveals distinct clonal expansion patterns. (A) UMAP visualization of T cell repertoire composition of anti-GAD65 AINS and IIH individuals. (B) Pseudobulk DGE analysis displayed as a volcano plot comparing expanded versus nonexpanded T cell clones in CSF of anti-GAD65 AINS individuals. Adjusted P < 0.05 and log2FC ≥ 0.585 were used as cut-offs. Top 10 up-regulated and down-regulated genes are labeled. (C) Gene Ontology (GO) enrichment analysis of differentially expressed genes in expanded CSF T cell clones from anti-GAD65 AINS individuals. Dot size represents gene count; color intensity indicates statistical significance (adjusted P-value). (D) Clonal network visualization depicting TCR repertoire relationships between CSF and PBMC compartments in anti-GAD65 AINS and IIH individuals. Node size indicates clonal frequency; edges represent shared TCR sequences. (E) Stacked bar plot represents quantitative comparison of TCR clonal distribution in expanded clonal groups, comparing CSF and PBMC compartments in anti-GAD65 AINS and IIH individuals. Each stack shows the relative frequency of a distinct expanded clonal group. (F) Comparative clonal network analysis of CD4⁺ TSCM in CSF compartments between anti-GAD65 AINS and IIH individuals, highlighting disease-specific repertoire patterns. Node size indicates clonal frequency; edges represent shared TCR sequences. (G) Stacked bar plot represents quantitative comparison of CD4⁺ TSCM cell TCR clonal compositions in expanded clonal groups from CSF, contrasting anti-GAD65 AINS with IIH individuals. Each stack represents the relative frequency of a distinct expanded clonal group. Act.—activated; Ag—antigen; UMAP—Uniform Manifold Approximation and Projection; CSF—cerebrospinal fluid; GAD—individuals with glutamic acid decarboxylase antibody-associated AINS; IIH—idiopathic intracranial hypertension; FC—fold change; GO—Gene Ontology; Leuko—leukocyte; med.—mediated; NK—natural killer cells; pos.—positive; reg.—regulation; TCM—central memory T cells; TCR—T cell receptor; TEM—effector memory T cells; TSCM—stem cell–like memory T cells; PBMC—Peripheral Blood Mononuclear Cell.

In addition to the T cell clonal expansion analysis, which identified expansion of specific T cell clones in the CSF of anti-GAD65 AINS individuals, indicating proliferation in response to disease-relevant antigens, we went on and performed TCR diversity analysis to assess the overall composition and variability of the TCR repertoire. We found a significantly reduced diversity in the PBMC compartment of anti-GAD65 AINS compared to IIH individuals, consistent with a migration of pathogenic T cell clones to the CSF (SI Appendix, Fig. S3E). However, no significant differences in TCR diversity were observed in the CSF compartment between the two groups (SI Appendix, Fig. S3F).

Expansion of CD4⁺ TSCM in the CSF of Anti-GAD65 AINS Individuals.

Detailed characterization of CD4⁺ TSCM in the CSF revealed distinct clonal expansion patterns between anti-GAD65 AINS and IIH individuals. Analysis of the TCR repertoire demonstrated a marked enrichment of expanded clones within the CSF CD4⁺ TSCM cell population of anti-GAD65 AINS compared to IIH individuals (Fig. 4 F and G). This pattern was particularly evident in the clonal network analysis, where anti-GAD65 AINS individuals exhibited larger, interconnected clusters of clonally expanded TSCM (frequency ≥2), suggesting the presence of antigen-driven clonal selection (Fig. 4F). Quantitative assessment of clonal distributions further supported this observation. Anti-GAD65 AINS individuals showed a higher frequency of expanded clonal groups within their CSF TSCM compartment (Fig. 4G). The disease-specific nature of this expansion was further emphasized by individual clonotype network analysis, which revealed distinct architectural patterns in anti-GAD65 AINS individuals characterized by larger nodes representing highly expanded clones (SI Appendix, Fig. S4 A and B). Analysis of the top 20 most abundant TCR clones in the CSF TSCM population demonstrated higher clonal frequencies in anti-GAD65 AINS compared to IIH individuals, indicating a more focused and potentially disease-relevant immune response within this memory T cell subset (SI Appendix, Fig. S4C).

Discussion

Our study characterizes the immune landscape of the CSF and PB of anti-GAD65 AINS individuals at single-cell resolution. We report several important aspects: 1) The fraction of TSCM, particularly CD4⁺ TSCM, was increased in the PB and in the CSF of anti-GAD65 AINS individuals throughout the disease course, while plasmablasts were only increased in the PB, particularly in persons with short disease duration. 2) CSF CD4⁺ TSCM were clonally expanded in anti-GAD65 AINS individuals, while peripheral TCR diversity was reduced. 3) Clonally expanded T cells in the CSF of anti-GAD65 AINS individuals showed a proinflammatory gene signature. 4) CD8⁺ TSCM and to a lesser extent CD4⁺ TSCM could be detected in the brain parenchyma of anti-GAD65 AINS individuals. 5) CSF BCR repertoire analysis revealed a relevant proportion of GAD65-reactive BCRs that belong to the IgG1 subtype and were derived from nonclonally expanded cells.

We propose that the marked clonal expansion of activated CD4⁺ TSCM in the CSF, together with CD8⁺ TSCM in the brain parenchyma and the occasional meningeal/intraparenchymal CD4⁺ TSCM, plays a central role in driving the disease pathology of anti-GAD65 AINS. Memory T cells are an important part of the adaptive immune system, providing long-term protection against pathogens. When they re-encounter an antigen, they proliferate rapidly and mediate cellular cytotoxicity (37). Memory T cells have been implicated in several autoimmune disorders, including MS and T1D (14–19). In T1D, CD8⁺ T cells target and destroy insulin-producing β-cells in the pancreas. As GAD65 can be found in the brain but also in islets of Langerhans (18, 38), a considerable number of persons with T1D have GAD65-antibodies in the serum and many individuals with anti-GAD65 AINS do present with concomitant T1D (4, 7, 39). Previous studies identified an autoimmune stem-like CD8⁺ T cell population driving T1D (18, 19). Those TSCM self-renew continuously and give rise to cytotoxic immune mediators destroying β-cells in the pancreas (18). Apart from T1D and MS, stem-like T cells were found to maintain and promote the progression of other autoimmune diseases, e.g., systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA). Frequencies of CD4⁺ TSCM were found to be significantly increased in the PB of SLE and RA individuals when compared to HC (40, 41). Similar to these reports, we detected higher proportions of CD4⁺ as well as CD8⁺ TSCM in the PB and clonal expansion of CD4⁺ TSCM in the CSF of anti-GAD65 AINS individuals. Histological analyses revealed the presence of multiple CD8⁺ TSCM in brain parenchyma and occasional meningeal/intraparenchymal CD4⁺ TSCM. Furthermore, sc-RNA-seq and mFC analyses demonstrated pronounced activation within the CD4⁺ T cell compartment. The significantly higher expression of CD69 on CD4⁺ T cells in persons with anti-GAD65 AINS compared to controls further supports ongoing activation and may reflect a phenotype primed for CNS migration (22). The preferential detection of CD4⁺ TSCM in the CSF, alongside the predominance of CD8⁺ TSCM in the brain parenchyma, likely reflects fundamental differences in antigen presentation and tissue retention mechanisms within the CNS. This observation is in agreement with previously found parenchymal CD69⁺CD103⁺CD8⁺ T resident memory cells in anti-GAD65 AINS individuals (7). CD4⁺ T cells require MHC class II-restricted antigen presentation, which is predominantly provided by professional antigen-presenting cells localized to the CSF, meninges, and perivascular spaces. In contrast, neurons generally do not express MHC class II, even under inflammatory conditions. CD8⁺ T cells, however, can directly engage neurons through MHC class I molecules, which are upregulated on stressed or inflamed neuronal populations. This differential MHC restriction provides distinct retention signals (42–45). Thus, CD4⁺ TSCM in the CSF and meninges may serve as a self-renewing reservoir continuously supporting CD8⁺ effector responses, while CD8⁺ TSCM establish residency within the brain parenchyma at sites of pathology. In line with these findings, previous studies reported intraparenchymal accumulation of cytotoxic CD8⁺ T cells in anti-GAD65-AINS individuals (6, 7). Taken together, our data support a dual-compartment in which a CD4⁺ TSCM reservoir in the CSF and meninges cooperates with a parenchymal CD8⁺ TSCM population to perpetuate the chronic autoimmune response in anti-GAD65 AINS. The persistent increase in TSCM frequencies observed throughout the disease course further aligns with this model, supporting the concept of sustained, T cell–mediated pathology in anti-GAD65-AINS (6, 7, 46).

Our DGE analysis further revealed reduced expression of HLA-C in the PBMCs of anti-GAD65 AINS compared to IIH individuals. HLA-C is a gene locus on chromosome 6 encoding HLA-C alleles, which are involved in the regulation of the innate and adaptive immune response (47). They can interact with NK cells (via their killer-cell immunoglobulin-like (KIR) receptors) and with CD8⁺ T cells as well as indirectly with CD4⁺ T cells (47). A recently published GWAS analysis of persons with anti-GAD65 AINS has identified HLA-C as one protein-coding gene mapping to the anti-GAD65 AINS GWAS loci (8). Likewise, previous studies linked alterations in the HLA-C locus to other autoimmune diseases like psoriasis, Crohn’s disease, alopecia areata, and rheumatic diseases (48–51). Thus, HLA-C might be an interesting locus in the pathophysiology of anti-GAD65 AINS suggesting a need for more comprehensive research in that field.

Our analysis of TCR repertoires in anti-GAD65 AINS revealed distinct patterns of clonal expansion and compartmentalization between the CSF and PB. The observation of significantly expanded T cell clones in the CSF of anti-GAD65 AINS individuals is consistent with findings from other autoimmune encephalitis studies, such as in anti-GABA_A receptor encephalitis, where Bracher et al. demonstrated that a CD8⁺ T cell clone was strongly expanded in the CSF (52). The reciprocal relationship between peripheral TCR diversity and CSF clonal expansion supports the notion of a focused autoimmune response, in which pathogenic T cells migrate from the periphery to the CNS, resulting in an accumulation of disease-relevant T cells within the CNS and leaving a less diverse T cell repertoire in the peripheral circulation. This pattern mirrors observations in MS, where Jacobsen et al. reported compartmentalized oligoclonal expansion of memory CD8⁺ T cells in the CSF (15). Moreover, Sousa et al. have also found that highly expanded T cell clones were enriched in the CSF compartment of persons with MS compared to PB (53). The inverse relationship between CSF clonal expansion and peripheral diversity may represent a hallmark of CNS-directed autoimmunity, as similarly described in other inflammatory immune-mediated diseases of the CNS (54). The distinct clonal network patterns in anti-GAD65 AINS individuals, characterized by larger interconnected clusters of expanded TSCM and the high clonal frequencies observed in the top 20 most abundant TCR clones of anti-GAD65 AINS persons’ CSF TSCM further support the notion of a focused, disease-relevant immune response. This is consistent with the concept of epitope-spreading in autoimmune diseases, where the initial autoimmune response against a specific antigen (in this case, GAD65) leads to the activation and expansion of T cells recognizing additional epitopes (55). Future research should focus on characterizing the antigen specificity of these expanded clones and exploring their potential as therapeutic targets or biomarkers for disease progression and treatment response. Notably, our finding of distinct transcriptional profiles in expanded CSF clones, particularly the upregulation of T cell proliferation and inflammatory response genes, is consistent with observations in other neuroinflammatory conditions like MS, which was reported by Pappalardo et al. (56). Moreover, a study by Gate et al. on Alzheimer’s disease found that clonally expanded CD8⁺ T cells in the CSF showed increased expression of cytotoxic effector genes, including NKG7 and GZMH (57). This finding suggests that the upregulation of activation- and inflammatory response-related genes in expanded CSF clones may be a common feature across various neurodegenerative and autoimmune conditions affecting the CNS.

In contrast to the clonal expansion patterns observed in the TCR repertoire, we did not identify relevant clonally expanded B cells in the CSF of anti-GAD65 AINS individuals. However, recombinant production and reactivity testing of mAbs derived from randomly chosen CSF BCRs with concurrently available heavy and light chains revealed a remarkably high frequency of GAD65-reactive mAbs. Somatic hypermutations were increased in GAD65-reactive as compared to nonreactive mAbs, suggesting that they likely have undergone antigen-driven affinity maturation, as described previously (58). In contrast to a previous study (58) that identified two (out of seven) persons with GAD65-reactive BCRs derived from antibody secreting cells (ASCs) in the CSF, we did not find GAD65-reactive BCRs in CSF ASCs of the four persons included in our BCR analysis, indicating that only few anti-GAD65 AINS individuals harbor them. As Tröscher et al. (7) have found plasma cell infiltrations into the brain parenchyma in persons with GAD-TLE at early disease stages, another point to consider is that GAD65-reactive plasma cells could have already evaded the CSF by the time of sampling in our study. Similarly, the increase in plasmablasts in the PB was mainly found in persons with a disease duration below one y in our anti-GAD65 AINS sequencing cohort, suggesting that the B cellular immune response in anti-GAD65 AINS individuals might be primarily established in early disease stages.

Taken together, our findings, demonstrating enrichment and clonal expansion of TSCM in anti-GAD65 AINS, point toward a predominantly T cell–driven disease pathophysiology. Additionally, we observed that anti-GAD65 AINS individuals harbor numerous GAD65-reactive and somatically hypermutated, albeit nonexpanded, B cells in the CSF, suggesting a B cell–assisted autoimmune process. This may explain recently reported treatment responses in two SPS individuals to anti-CD19 CAR T cell therapy (59, 60). Nevertheless, the marked enrichment and expansion of TSCM suggests that therapeutic strategies targeting the T cell compartment could be more effective in anti-GAD65 AINS. Approaches such as anti-CD3 therapy, already approved for the treatment of T1D, could represent promising candidates for future clinical trials. Furthermore, characterizing the antigen specificity of these TSCM populations will be crucial to understand their role as potential drivers of disease pathology. Such insights could enable the development of highly targeted interventions aimed at selectively eliminating or modulating antigen-specific TSCM, thereby limiting excessive immune responses while preserving overall immune competence.

Our study is limited by the small sample size and the differences in age and sex between anti-GAD65 AINS and IIH individuals. Different disease durations within the anti-GAD65 AINS cohort further limit our analyses. In addition, different kits were used for library preparation.

However, our study exceeds previous ones by providing an in-depth characterization of peripheral and intrathecal immune cell populations in mainly treatment-naive anti-GAD65 AINS individuals compared to noninflammatory controls, yielding insights into disease pathogenesis and identifying potential treatment targets.

In conclusion, our findings contribute to the growing body of evidence supporting the role of antigen-specific B and T cell responses in autoimmune encephalitis. The compartment-specific changes in the TCR repertoire and clonal expansion patterns emphasize the importance of studying both CSF and PB to improve the understanding of the pathogenesis of CNS autoimmune disorders. Our findings highlight the potential role of TSCM in the pathogenesis of anti-GAD65 AINS and suggest that these cells may serve as a reservoir for disease-specific T cells in the CNS. Future research should focus on characterizing the antigen specificity of expanded TSCM clones and assessing their potential as therapeutic targets or disease biomarkers.

Materials and Methods

Experimental Model and Study Participant Details.

Anti-GAD65 AINS individuals were prospectively included in the GAD sequencing-cohort at four specialized clinical centers (Münster, Rotterdam, Kiel, and Munich) according to the following inclusion criteria:

1.
Suspected anti-GAD65 AINS presenting with one of the typical clinical phenotypes or confirmed anti-GAD65 AINS.
2.
Performance of a lumbar puncture (LP) during clinical routine workup.
3.
No ongoing or previous immunomodulatory therapy (treatment-naïve) except from IVIG, steroids, immunoadsorption, or plasmapheresis at least 6 mo prior to sampling.
4.
Written informed consent to participate in the study.

Individuals were excluded if one or more of the following exclusion criteria applied:

1.
Questionable diagnosis.
2.
Signs of acute infection, assessed by routine blood and CSF analysis.
3.
Known comorbid systemic autoimmune disease (e.g., SLE) or chronic infections (e.g., HIV, chronic hepatitis), except for T1D.
4.
Pregnancy or breastfeeding.
5.
Age <18 y at time of inclusion.
6.
Mental illness impairing the ability to give informed consent.
7.
Artificial blood contamination of the CSF resulting in >200 erythrocytes/μL.

The study design is illustrated in Fig. 1A. Detailed clinical data of the anti-GAD65 AINS and IIH cohorts are summarized in the supporting information file (SI Appendix, Table S1–S4 and supplementary method details). Samples were processed according to a standardized protocol at all centers minimizing the risk for systematic technical bias. The study was performed in accordance with the declaration of Helsinki and was approved by the local ethics committees: 1) the “Ethikkommission der Ärztekammer Westfalen-Lippe (ÄKWL) und der Westfälischen-Wilhelms-Universität” (Ethics Committee of the Board of Physicians of the Region Westfalen-Lippe and of the Westfälische Wilhelms-University; reference numbers 2015-522-f-S, 2021-800-f-S, and 2015-088-f-S) in Münster, 2) the “Erasmus MC Medical Ethical Research Board” (reference numbers: MEC-2015-397, MEC-2020-0418, and MEC-2020-0650) in Rotterdam, 3) the “Ethikkommission” of the Medical Faculty Kiel (reference numbers: D498/19, B337/13, B300/19, and D578/18), 4) by the “Ethikkommission” of the Medical Faculty Munich (projects: 18-419, 163-16, 22-0150, 173-14), 5) by the Ethics Committee of the Board of Physicians of the Region Nordrhein and of the Medical Faculty at Heinrich Heine University Düsseldorf (reference number: 5951R), 6) by the ethical committees of the Medical University of Vienna (EK 1206/2013, 1123/2015, and 1636/2019), and 7) the Medical University of Bonn (229/00, 360/12).

Method Details

Method details on collection of PB and CSF samples, sample preparation for scRNA-seq and mFC, single cell RNA-sequencing, histological analyses, BCR cloning, recombinant mAb production, and anti-GAD65 reactivity screening are provided in the SI Appendix, Supplementary Method Details.

Quantification and Statistical Analysis.

Bioinformatic Analyses of Single Cell RNA-Sequencing Data.

The 10× Genomics’ Cell Ranger (v6.1.2) multipipeline with the transcriptome reference (refdata-gex-GRCh38-2020-A) and V(D)J reference (refdata-cellranger-vdj-GRCh38-alts-ensembl-7.1.0) was used for data processing. Filtered feature-barcode matrix files were then analyzed with the R package “Seurat (v5.0.1)” (61). Further details regarding quality control, cluster annotation, cell abundance analysis, cluster-specific DGE, as well as BCR and TCR bioinformatic analysis are provided in the supporting information file (supplementary method details).

Supplementary Material

Appendix 01 (PDF)

pnas.2514753123.sapp.pdf^{(3.8MB, pdf)}

Dataset S01 (DOCX)

pnas.2514753123.sd01.docx^{(4.5MB, docx)}

Acknowledgments

This project was supported by the German Research Foundation (ERARE18-202 UltraAIE) and the Netherlands Medical Research Foundation (ZonMW) (ERARE-JTC2018 202: UltraAIE) under the frame of E-Rare-3, the ERA-Net for Research on Rare Diseases; the German Federal Ministry of Education and Research (Comprehensive, Orchestrated, National Network to Explain, Categorize and Treat autoimmune encephalitis and allied diseases within the German NEtwork for Research on AuToimmune Encephalitis – CONNECT GENERATE and CONNECT GENERATE 2.0; 01GM1908A and 01GM2208A); Dioraphte (2001 0403); EpilepsieNL (19-18); the Research Committee of the Faculty of Medicine of the Heinrich Heine University Düsseldorf (Grant number 2022-04); the “Else Kröner-Fresenius Stiftung” (Grant number 2023_EKMS.05); the Fritz Thyssen Stiftung (Grant number 10.23.1.015MN); LMU excellent (Grant number: AOST 867603-4); the Friedrich-Baur-Stiftung (Grant number: 54/23), and the Austrian Science Fund (FWF, Grant number: P 34864-B). We would like to thank all participating persons and their relatives and caregivers for their invaluable contribution to this study. We acknowledge High-Performance Computing (HPC) support from the Centre for Information and Media Technology (ZIM) at the Heinrich Heine University Düsseldorf. Furthermore, we would like to thank the Core Facility Flow Cytometry at the Medical Faculty of the Heinrich Heine University Düsseldorf, Germany, for assistance in sample analysis. This work was part of Saskia Räuber’s and Paul Disse’s PhD research.

Author contributions

M.J.T., F.L., G.M.z.H., F.T., and N.M. designed research; S.B., S.R., K.E., P.D., L.-M.P., V.G., I.S., E.v.S., M.J., D.P., M. Heming, L.M.-M., M.P., C.S., E.H., J. Bartosch, J.D., J. Bauer, M.J.T., F.L., G.M.z.H., F.T., and N.M. performed research; S.B., S.R., K.E., D.E., M.v.D., M. Scharf, M. Heming, and J. Bauer contributed new reagents/analytic tools; S.B., S.R., K.E., D.E., M. Scharf, and J. Bauer analyzed data; S.R. and K.E. funding acquisition; M.J.T., F.L., G.M.z.H., F.T., and N.M. supervision, Funding acquisition; and S.B., S.R., K.E., D.E., M.v.D., M. Scharf, M.H.-R., P.D., L.P., V.G., I.S., E.v.S., M. Heming, L.M.-M., J. Bartosch, E.K., S.K., M. Ringelstein, E.B., H.W., N.H.S., J.F., N.G., L.K., M. Roden, A.R., M. Stoll, A.J.B., M. Hamed, C.G.B., R.H., J. Bauer, S.G.M., M.J.T., F.L., G.M.z.H., F.T., and N.M. wrote the paper.

Competing interests

S.B. is affiliated with Novartis, Basel, Switzerland. M.S., R.M., and L.K. are employees of EUROIMMUN. S.R. received travel Grants from Merck Healthcare Germany GmbH, Alexion Pharmaceuticals, Jazz Pharmaceuticals, and Bristol Myers Squibb. She served on a scientific advisory board from Merck Healthcare Germany GmbH and received honoraria for lecturing from Roche and Merck Healthcare Germany GmbH. Her research was supported by Novartis, Sanofi-Aventis Deutschland GmbH, “Stiftung zur Förderung junger Neurowissenschaftler”, and “Else Kröner-Fresenius-Stiftung”. K.E. was supported by the “Friedrich-Baur-Stiftung”. E.K. has received personal honoraria for lectures or advice from UCB, UNEEG, Medtronic, Desitin and Precisis and has participated in trials sponsored by Medtronic, UCB, Ergomed, and Precisis. Her research is supported by the Munich Clinican Scientist Program (MCSP); M.R. received lecture fees and/or served on advisory boards for AstraZeneca, Echosens, Eli Lilly, Madrigal, Merck-MSD, Novo Nordisk, Synlab and Target RWE and performed investigator-initiated research with support from Boehringer Ingelheim and Novo Nordisk to the German Diabetes Center (DDZ). H.W. received speaker honoraria from Alexion, Biogen, Bristol Myers Squibb, Genzyme, Merck, Neurodiem, Novartis, Ology, Roche, TEVA, and WebMD Global. He received honoraria for consulting services from Abbvie, Actelion, Argenx, B.D., Bristol Myers Squibb, EMD Serono, Fondazione Cariplo, Gossamer Bio, Idorsia, Immunic, Immunovant, INmune Bio_Syneos Health, Janssen, Merck, NexGen, Novartis, Roche, Sanofi, Swiss MS Society, UCB and Worldwide Clinical Trials. His research is supported by the German Myasthenia Gravis Society. J.B.’s studies are funded by the Austrian Science Fund (FWF). NHS declares financial ties to Menarini Silicon Biosystems (CellSearch Assay) in the form of third-party funding for research support, as well as to Illumina (NGS products) in the form of lecture fees. S.G.M. receives honoraria for lecturing, and travel expenses for attending meetings from Academy 2, Argenx, Alexion, Almirall, Amicus Therapeutics Germany, Bayer Health Care, Biogen, BioNtech, BMS, Celgene, Datamed, Demecan, Desitin, Diamed, Diaplan, DIU Dresden, DPmed, Gen Medicine and Healthcare products, Genzyme, Hexal AG, IGES, Impulze GmbH, Janssen Cilag, KW Medipoint, MedDay Pharmaceuticals, Merck Serono, MICE, Mylan, Neuraxpharm, Neuropoint, Novartis, Novo Nordisk, ONO Pharma, Oxford PharmaGenesis, QuintilesIMS, Roche, Sanofi-Aventis, Springer Medizin Verlag, STADA, Chugai Pharma, Teva, UCB, Viatris, Wings for Life international and Xcenda, his research is funded by the German Ministry for Education and Research (BMBF), Bundesinstitut für Risikobewertung (BfR), Deutsche Forschungsgemeinschaft (DFG), Else Kröner Fresenius Foundation, Gemeinsamer Bundesausschuss (G-BA), German Academic Exchange Service, Hertie Foundation, Interdisciplinary Center for Clinical Studies (IZKF) Muenster, German Foundation Neurology and Alexion, Almirall, Amicus Therapeutics Germany, Biogen, Diamed, DGM e.v., Fresenius Medical Care, Genzyme, Gesellschaft von Freunden und Förderern der Heinrich-Heine-Universität Düsseldorf e.V., HERZ Burgdorf, Merck Serono, Novartis, ONO Pharma, Roche, and Teva. M.J.T. has received funds for serving on a scientific advisory board of AmGen, ArgenX, Arialys and UCB, received funds from Dioraphte (2001 0403). EpilepsieNL (19-18), received research funds for consultation at Guidepoint Global LLC, an unrestricted research Grant from CSL Behring, and an unrestricted research Grant from Euroimmun, was supported by an Interlaken Leadership Award and an E-RARE3 Grant [Netherlands Medical Research Foundation (ZonMW), ERARE-JTC2018 202: UltraAIE]. F.L. is also supported by E-Rare Joint Transnational research support (ERA-Net, LE3064/2-1), Stiftung Pathobiochemie of the German Society for Laboratory Medicine and HORIZON MSCA 2022 Doctoral Network 101119457 - IgG4-TREAT and discloses speaker honoraria from Grifols, Teva, Biogen, Bayer, Roche, Novartis, Fresenius, travel funding from Merck, Grifols and Bayer and serving on advisory boards for Roche, Biogen and Alexion. G.M.z.H. received research support from Biogen and Merck Germany. He received honoraria from Alexion and LFB Pharma and participated in Data Safety Monitoring and/or Advisory Boards of LFB Pharma, Roche and Immunovant. F.T. received Grant support from Novartis Pharma GmbH and speaker honoraria from Alexion Pharmaceuticals. N.M. received honoraria for lecturing and travel expenses for attending meetings from Biogen Idec, GlaxoSmithKline, Teva, Novartis Pharma, Bayer Healthcare, Genzyme, Alexion Pharmaceuticals, Fresenius Medical Care, Diamed, UCB Pharma, AngeliniPharma, BIAL and Sanofi-Aventis, received royalties for consulting from UCB Pharma, Alexion Pharmaceuticals and Sanofi, and received financial research support from Euroimmun, Fresenius Medical Care, Diamed, Alexion Pharmaceuticals, Novartis Pharma, and Sanofi. The remaining authors declare no conflict of interest.

Footnotes

Preprint server: A previous version of this manuscript has been deposited on BioRxiv under a CC-BY-NC 4.0 international license, https://doi.org/10.1101/2025.05.18.654720.

This article is a PNAS Direct Submission.

Contributor Information

Nico Melzer, Email: nico.melzer@med.uni-duesseldorf.de.

Collaborators: Juna M. de Vries, Mariska M. P. Nagtzaam, Suzanne C. Franken, Yvette S. Crijnen, Juliette Brenner, Robin W. van Steenhoven, Jeroen Kerstens, Marienke A. A. M. de Bruijn, Anna E. M. Bastiaansen, Remco M. Hoogenboezem, Sharon Veenbergen, Peter A. E. Sillevis Smitt, Marwa Al-Dubai, Luise Appeltshauser, Ilya Ayzenberg, Carolin Baade-Büttner, Andreas van Baalen, Sebastian Baatz, Oliver Bähr, Bettina Balint, Iason Bartzokis, Sebastian Bauer, Annette Baumgartner, Tobias Baumgartner, Antonios Bayas, Stefanie Becker, Sonka Benesch, Robert Berger, Birgit Berger, Martin Berghoff, Sarah Bernsen, Achim Berthele, Christian Bien, Corinna Bien, Julia Bierwith, Andreas Binder, Stefan Bittner, Daniel Bittner, Franz Blaes, Astrid Blaschek, Amelie Bohn, Moritz Böhringer, Marie Braun, Sergio Castro-Gomez, Justina Dargvainiene, Timo Deba, Julia Maren Decker, Johanna-Maria Dietmaier, Andre Dik, Julian Dominik, Kathrin Doppler, Mona Dreesmann, Lena Edelhoff, Laura Ehrhardt, Sven Ehrlich, Katharina Eisenhut, Alexander Emmer, Dominique Endres, Marina Entscheva-Storr, Daniela Esser, Thorleif Etgen, Jürgen Hartmut Faiss, Kim Kristin Falk, Timo Faustmann, Walid Fazeli, Alexander Finke, Carsten Finke, Felix Fischbach, Dirk Fitzner, Marina Flotats-Bastardas, Mathias Fousse, Tobias Freilinger, Manuel Friese, Hannah Fuhrer, Armin Johannes Gaebler, Marco Gallus, Marcel Gebhard, Christian Geis, Clemens Gödel, Anna Gorsler, Armin Grau, Oliver Grauer, Britta Greshake, Catharina roß, Thomas Grüter, Aiden Haghikia, Robert Handreka, Iris Hannibal, Niels Hansen, Sandra Hansmann, Jens Harmel, Antonia Harms, Yetzenia Dubraska Haro Alizo, Alkomiet Hasan, Martin Häusler, Ida Sybille Haussleiter, Joachim Havla, Chung Ha-Yeun, Wolfgang Heide, Valentin Held, Kerstin Hellwig, Marina Herwerth, Philip Hillebrand, Frank Hoffmann, Christian Hofmann, Ulrich Hofstadt-van Oy, Dominica Hudasch (ehem. Ratuszny), Yannik Hülsmann, Martinv Hümmert, Peter Huppke, Hagen Huttner, Fatme Seval Ismail, Martina Jansen, Mareike Jansen, Marius Jonas, Aleksandra Juranek, Daniel Kamp, Michael Karenfort, Annika Kather, Max Kaufmann, Christoph Kellinghaus, Constanze Kerin (geb. Mönig), Ruth Kerkhoff, Rolf Kern, Jaqueline Klausewitz, Michael Kluge, Susanne Knake, Benjamin Knier, Ellen Knierim, Frank Kohlert, Inga Koneczny, Felix Konen, Peter Körtvélyessy, Stjepana Kovac, Andrea Kraft, Markus Krämer, Verena Kraus, Christos Krogias, Gregor Kuhlenbäumer, Olga Kukhlenko, Tania Kümpfel, Albrecht Kunze, Hanna Lapp, Christoph Lehrich, Martin Lesser, Jan Lewerenz, Frank Leypoldt, Andreas Linsa, Daniel Lüdecke, Jan Lünemann, Marie Madlener, Michael Malter, Nils Margraf, Carlos Martinez Quesada, Monika Meister, Nico Melzer, Kristin Stefanie Melzer, Til Menge, Sven Meuth, Gerd Meyer zu Hörste, Fabian Möller, Marie-Luise Mono, Sigrid Mues, Hiltrud Muhle, Anna-Katharina Mundorf, Marc Nikolaus, Jost Obrocki, Friedemann Paul, Loana Penner, Lena Kristina Pfeffer, Thomas Pfefferkorn, Steffen Pfeuffer, Alexandra Philipsen, Johannes Piepgras, Julika Pitsch, Felix von Podewils, Mosche Pompsch, Josef Priller, Anne-Katrin Pröbstel, Harald Prüß, Duygu Pul, Daniel Rapp, Johanna Maria Helena Rau, Saskia Räuber, Markus Rauchenzauner, Robert Rehmann, Ina Reichen, Gernot Reimann, Momsen Reincke, Raphael Reinecke, Jonathan Repple, Nele Retzlaff, Sigrid Reuter, Marius Ringelstein, Henrik Rohner, Felix Rosenow, Kevin Rostásy, Theodor Rüber, Stephan Rüegg, Yannic Saathoff, Jens Schaumberg, Hanna Schellhorn, Ruth Schilling, Jens Schmidt, Melissa Schmitz, Ina-Isabelle Schmütz, Hauke Schneider, Patrick Schramm, Stephan Schreiber, Stefanie Schreiber, Gesa Schreyer, Ina Schröder, Simon Schuster, Philip Schwenkenbecher, Günter Seidel, Frank Seifert, Thomas Seifert-Held, Makbule Senel, Kai Siebenbrodt, Olga Simova, Claudia Sommer, Juliane Spiegler, Oliver Stammel, Slobodan Stankovic, Andreas Steinbrecher, Johann Steiner, Henning Stolze, Muriel Stoppe, Karin Storm van’s Gravesande, Christine Strippel, Dietrich Sturm, Klarissa Hanja Stürner, Kurt-Wolfram Sühs, Steffen Syrbe, Simone Taube, Franziska Thaler, Florian Then Bergh, Katrin Thies, Anja Tietz, Corinna Trebst, George Trendelenburg, Regina Trollmann, Thanos Tsaktanis, Hayrettin Tumani, Mehtap Türedi, Christian Urbanek, Rem Vaizian, Niklas Vogel, Max Vogtmann, Matthias von Mering, Katharina von Zedtwitz, Judith Wagner, Jan Wagner, Elias Wagner, Barbara Wagner, Klaus-Peter Wandinger, Mirko Wegschneider, Judith Weiser, Robert Weissert, Njiku Melchior Wellmer, Brigitte Wildemann, Jonathan Wickel, Karsten Witt, Katharina Wurdack, Yavor Yalachkov, and Lara Zieger

Data, Materials, and Software Availability

Data underlying this study are registered with the ABCD-J data catalog at https://data.abcd-j.de/dataset/d1b56ad8-f8e9-5d4b-aa76-fe1c155ad201/1.0. The sequencing raw data and processed data will be made available in the European Genome Phenome Archive upon reasonable request and after material transfer agreement to regulate data protection of potentially reidentifiable genetic data. Sequencing data from parts of the IIH cohort have been published before (62, 63) and have been deposited in GEO repository with the accession code GSE138266. Regarding the computational analyses, the official tutorial of the packages listed were followed and a custom in-house pipeline was used. The codes used in this study have been made available on GitHub (https://github.com/sumanta-barman/immune-cell-profiling-anti-gad65).

Supporting Information

References

1.Saiz A., et al. , Spectrum of neurological syndromes associated with glutamic acid decarboxylase antibodies: Diagnostic clues for this association. Brain 131, 2553–2563 (2008). [DOI] [PubMed] [Google Scholar]
2.Dalakas M. C., Stiff-person syndrome and related disorders-diagnosis, mechanisms and therapies. Nat. Rev. Neurol. 20, 587–601 (2024). [DOI] [PubMed] [Google Scholar]
3.Dalakas M. C., Stiff-person syndrome and GAD antibody-spectrum disorders: GABAergic neuronal excitability, immunopathogenesis and update on antibody therapies. Neurotherapeutics 19, 832–847 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Graus F., Saiz A., Dalmau J., GAD antibodies in neurological disorders-insights and challenges. Nat. Rev. Neurol. 16, 353–365 (2020). [DOI] [PubMed] [Google Scholar]
5.Hackert J. K., et al. , Anti-GAD65 containing cerebrospinal fluid does not alter GABAergic transmission. Front. Cell. Neurosci. 10, 130 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Dik A., et al. , Impact of T cells on neurodegeneration in anti-GAD65 limbic encephalitis. Ann. Clin. Transl. Neurol. 8, 2289–2301 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Tröscher A. R., et al. , Temporal lobe epilepsy with GAD antibodies: Neurons killed by T cells not by complement membrane attack complex. Brain 146, 1436–1452 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Strippel C., et al. , A genome-wide association study in autoimmune neurological syndromes with anti-GAD65 autoantibodies. Brain 146, 977–990 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Gross C. C., et al. , Classification of neurological diseases using multi-dimensional CSF analysis. Brain 144, 2625–2634 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Bangs D. J., et al. , CXCR3 regulates stem and proliferative CD8+ T cells during chronic infection by promoting interactions with DCs in splenic bridging channels. Cell Rep. 38, 110266 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Zhao Y., et al. , Human CD8 T-stem cell memory subsets phenotypic and functional characterization are defined by expression of CD122 or CXCR3. Eur. J. Immunol. 51, 1732–1747 (2021). [DOI] [PubMed] [Google Scholar]
12.Hong H., et al. , The distribution of human stem cell-like memory T cell in lung cancer. J. Immunother. 39, 233–240 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Gattinoni L., Speiser D. E., Lichterfeld M., Bonini C., T memory stem cells in health and disease. Nat. Med. 23, 18–27 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Jilek S., et al. , CSF enrichment of highly differentiated CD8+ T cells in early multiple sclerosis. Clin. Immunol. 123, 105–113 (2007). [DOI] [PubMed] [Google Scholar]
15.Jacobsen M., et al. , Oligoclonal expansion of memory CD8+ T cells in cerebrospinal fluid from multiple sclerosis patients. Brain 125, 538–550 (2002). [DOI] [PubMed] [Google Scholar]
16.Fransen N. L., et al. , Tissue-resident memory T cells invade the brain parenchyma in multiple sclerosis white matter lesions. Brain 143, 1714–1730 (2020). [DOI] [PubMed] [Google Scholar]
17.Kaskow B. J., Baecher-Allan C., Effector T cells in multiple sclerosis. Cold Spring Harb. Perspect. Med. 8, a029025 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Gearty S. V., et al. , An autoimmune stem-like CD8 T cell population drives type 1 diabetes. Nature 602, 156–161 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Abdelsamed H. A., et al. , Beta cell-specific CD8+ T cells maintain stem cell memory-associated epigenetic programs during type 1 diabetes. Nat. Immunol. 21, 578–587 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Saraiva D. P., et al. , Expression of HLA-DR in cytotoxic T lymphocytes: A validated predictive biomarker and a potential therapeutic strategy in breast cancer. Cancers (Basel) 13, 3841 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Reddy M., Eirikis E., Davis C., Davis H. M., Prabhakar U., Comparative analysis of lymphocyte activation marker expression and cytokine secretion profile in stimulated human peripheral blood mononuclear cell cultures: An in vitro model to monitor cellular immune function. J. Immunol. Methods 293, 127–142 (2004). [DOI] [PubMed] [Google Scholar]
22.Cibrián D., Sánchez-Madrid F., CD69: From activation marker to metabolic gatekeeper. Eur. J. Immunol. 47, 946–953 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Barington L., et al. , EBI2 in splenic and local immune responses and in autoimmunity. J. Leukoc. Biol. 104, 313–322 (2018). [DOI] [PubMed] [Google Scholar]
24.Sánchez Gómez I., “The role of the transcription factor TCFL5 in the physiopathology of B lymphocytes,” PhD thesis, Universidad Autónoma de Madrid, Madrid, Spain (2021).
25.Li S., et al. , Circular RNA MAP2K2-modified immunosuppressive dendritic cells for preventing alloimmune rejection in organ transplantation. Bioeng. Transl. Med. 9, e10615 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Gong K.-Q., et al. , MAP2K2 delays recovery in murine models of acute lung injury and associates with acute respiratory distress syndrome outcome. Am. J. Respir. Cell Mol. Biol. 66, 555–563 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Ingelfinger F., et al. , Twin study reveals non-heritable immune perturbations in multiple sclerosis. Nature 603, 152–158 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Martinelli M., et al. , The poly(C)-binding protein Pcbp2 is essential for CD4+ T cell activation and proliferation. iScience 26, 105860 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Lee J., et al. , Cyclic AMP concentrations in dendritic cells induce and regulate Th2 immunity and allergic asthma. Proc. Natl. Acad. Sci. U.S.A. 112, 1529–1534 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Hernandez-Barco Y. G., et al. , 84 The role of the GNAS oncogene in the immune microenvironment of intraductal papillary mucinous neoplasm. Am. J. Gastroenterol. 114, S50–S51 (2019). [Google Scholar]
31.Wittner J., Schuh W., Krüppel-like factor 2: A central regulator of B cell differentiation and plasma cell homing. Front. Immunol. 14, 1172641 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Hart G. T., Wang X., Hogquist K. A., Jameson S. C., Krüppel-like factor 2 (KLF2) regulates B-cell reactivity, subset differentiation, and trafficking molecule expression. Proc. Natl. Acad. Sci. U.S.A. 108, 716–721 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Pabbisetty S. K., et al. , KLF2 is a rate-limiting transcription factor that can be targeted to enhance regulatory T-cell production. Proc. Natl. Acad. Sci. U.S.A. 111, 9579–9584 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Wang J., Zhao X., Wan Y. Y., Intricacies of TGF-β signaling in Treg and Th17 cell biology. Cell. Mol. Immunol. 20, 1002–1022 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Liu J., et al. , Long noncoding RNA LINC01578 drives colon cancer metastasis through a positive feedback loop with the NF-κB/YY1 axis. Mol. Oncol. 14, 3211–3233 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Ko C.-Y., Chang W.-C., Wang J.-M., Biological roles of CCAAT/enhancer-binding protein delta during inflammation. J. Biomed. Sci. 22, 6 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Martin M. D., Badovinac V. P., Defining memory CD8 T cell. Front. Immunol. 9, 2692 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Herold K. C., et al. , The immunology of type 1 diabetes. Nat. Rev. Immunol. 24, 435–451 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Muñoz-Lopetegi A., et al. , Neurologic syndromes related to anti-GAD65: Clinical and serologic response to treatment. Neurol. Neuroimmunol. Neuroinflamm. 7, e696 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Cianciotti B. C., et al. , CD4+ memory stem T cells recognizing citrullinated epitopes are expanded in patients with rheumatoid arthritis and sensitive to tumor necrosis factor blockade. Arthritis Rheumatol. 72, 565–575 (2020). [DOI] [PubMed] [Google Scholar]
41.Lee Y. J., et al. , Role of stem cell-like memory T cells in Systemic Lupus Erythematosus. Arthritis Rheumatol. 70, 1459–1469 (2018). [DOI] [PubMed] [Google Scholar]
42.Redwine J. M., Buchmeier M. J., Evans C. F., In vivo expression of major histocompatibility complex molecules on oligodendrocytes and neurons during viral infection. Am. J. Pathol. 159, 1219–1224 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Strazielle N., Creidy R., Malcus C., Boucraut J., Ghersi-Egea J.-F., T-lymphocytes traffic into the brain across the blood-CSF barrier: Evidence using a reconstituted choroid plexus epithelium. PLoS ONE 11, e0150945 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Aguirre K., Miller S., MHC class II-positive perivascular microglial cells mediate resistance to Cryptococcus neoformans brain infection. Glia 39, 184–188 (2002). [DOI] [PubMed] [Google Scholar]
45.Melzer N., Budde T., Stork O., Meuth S. G., Limbic encephalitis: Potential impact of adaptive autoimmune inflammation on neuronal circuits of the amygdala. Front. Neurol. 6, 171 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Räuber S., et al. , Lymphocyte signatures correspond to clinical phenotypes in autoimmune limbic encephalitis. Brain Commun. 7, fcaf156 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Velastegui E., Vera E., Vanden Berghe W., Muñoz M. S., Orellana-Manzano A., HLA-C: Evolution, epigenetics, and pathological implications in the major histocompatibility complex. Front. Genet. 14, 1206034 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Siegel R. J., Bridges S. L., Ahmed S., HLA-C: An accomplice in rheumatic diseases. ACR Open Rheumatol. 1, 571–579 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Chen M., et al. , Hypermethylation of HLA-C may be an epigenetic marker in psoriasis. J. Dermatol. Sci. 83, 10–16 (2016). [DOI] [PubMed] [Google Scholar]
50.Haida Y., et al. , Association analysis of the HLA-C gene in Japanese alopecia areata. Immunogenetics 65, 553–557 (2013). [DOI] [PubMed] [Google Scholar]
51.Kulkarni S., et al. , Genetic interplay between HLA-C and MIR148A in HIV control and Crohn disease. Proc. Natl. Acad. Sci. U.S.A. 110, 20705–20710 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Bracher A., et al. , An expanded parenchymal CD8+ T cell clone in GABAA receptor encephalitis. Ann. Clin. Transl. Neurol. 7, 239–244 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Alves A. D. P., et al. , Intrathecal T-cell clonal expansions in patients with multiple sclerosis. Ann. Clin. Transl. Neurol. 3, 422–433 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Alves Sousa A. de P., et al. , Comprehensive analysis of TCR-β repertoire in patients with neurological immune-mediated disorders. Sci. Rep. 9, 344 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Sercarz E. E., Driver clones and determinant spreading. J. Autoimmun. 14, 275–277 (2000). [DOI] [PubMed] [Google Scholar]
56.Pappalardo J. L., et al. , Transcriptomic and clonal characterization of T cells in the human central nervous system. Sci. Immunol. 5, eabb8786 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Gate D., et al. , Clonally expanded CD8 T cells patrol the cerebrospinal fluid in Alzheimer’s disease. Nature 577, 399–404 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Biljecki M., et al. , Antibodies against glutamic acid decarboxylase 65 are locally produced in the CSF and arise during affinity maturation. Neurol. Neuroimmunol. Neuroinflamm. 10, e200090 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Faissner S., et al. , Successful use of anti-CD19 CAR T cells in severe treatment-refractory stiff-person syndrome. Proc. Natl. Acad. Sci. U.S.A. 121, e2403227121 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Ayzenberg I., et al. , Anti-CD19 CAR T-cell therapy in advanced stiff-person syndrome and concomitant myasthenia gravis. Neurol. Neuroimmunol. Neuroinflamm. 12, e200479 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Hao Y., et al. , Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat. Biotechnol. 42, 293–304 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Heming M., et al. , Neurological manifestations of COVID-19 feature T cell exhaustion and dedifferentiated monocytes in cerebrospinal fluid. Immunity 54, 164–175 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Esser D., et al. , Activated Αβ T− and Reduced Mucosa-associated Invariant T Cells in LGI1− and CASPR2-encephalitis. Brain 148, 3170–3183 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix 01 (PDF)

pnas.2514753123.sapp.pdf^{(3.8MB, pdf)}

Dataset S01 (DOCX)

pnas.2514753123.sd01.docx^{(4.5MB, docx)}

Data Availability Statement

[r1] 1.Saiz A., et al. , Spectrum of neurological syndromes associated with glutamic acid decarboxylase antibodies: Diagnostic clues for this association. Brain 131, 2553–2563 (2008). [DOI] [PubMed] [Google Scholar]

[r2] 2.Dalakas M. C., Stiff-person syndrome and related disorders-diagnosis, mechanisms and therapies. Nat. Rev. Neurol. 20, 587–601 (2024). [DOI] [PubMed] [Google Scholar]

[r3] 3.Dalakas M. C., Stiff-person syndrome and GAD antibody-spectrum disorders: GABAergic neuronal excitability, immunopathogenesis and update on antibody therapies. Neurotherapeutics 19, 832–847 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r4] 4.Graus F., Saiz A., Dalmau J., GAD antibodies in neurological disorders-insights and challenges. Nat. Rev. Neurol. 16, 353–365 (2020). [DOI] [PubMed] [Google Scholar]

[r5] 5.Hackert J. K., et al. , Anti-GAD65 containing cerebrospinal fluid does not alter GABAergic transmission. Front. Cell. Neurosci. 10, 130 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r6] 6.Dik A., et al. , Impact of T cells on neurodegeneration in anti-GAD65 limbic encephalitis. Ann. Clin. Transl. Neurol. 8, 2289–2301 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r7] 7.Tröscher A. R., et al. , Temporal lobe epilepsy with GAD antibodies: Neurons killed by T cells not by complement membrane attack complex. Brain 146, 1436–1452 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r8] 8.Strippel C., et al. , A genome-wide association study in autoimmune neurological syndromes with anti-GAD65 autoantibodies. Brain 146, 977–990 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r9] 9.Gross C. C., et al. , Classification of neurological diseases using multi-dimensional CSF analysis. Brain 144, 2625–2634 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r10] 10.Bangs D. J., et al. , CXCR3 regulates stem and proliferative CD8+ T cells during chronic infection by promoting interactions with DCs in splenic bridging channels. Cell Rep. 38, 110266 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r11] 11.Zhao Y., et al. , Human CD8 T-stem cell memory subsets phenotypic and functional characterization are defined by expression of CD122 or CXCR3. Eur. J. Immunol. 51, 1732–1747 (2021). [DOI] [PubMed] [Google Scholar]

[r12] 12.Hong H., et al. , The distribution of human stem cell-like memory T cell in lung cancer. J. Immunother. 39, 233–240 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r13] 13.Gattinoni L., Speiser D. E., Lichterfeld M., Bonini C., T memory stem cells in health and disease. Nat. Med. 23, 18–27 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r14] 14.Jilek S., et al. , CSF enrichment of highly differentiated CD8+ T cells in early multiple sclerosis. Clin. Immunol. 123, 105–113 (2007). [DOI] [PubMed] [Google Scholar]

[r15] 15.Jacobsen M., et al. , Oligoclonal expansion of memory CD8+ T cells in cerebrospinal fluid from multiple sclerosis patients. Brain 125, 538–550 (2002). [DOI] [PubMed] [Google Scholar]

[r16] 16.Fransen N. L., et al. , Tissue-resident memory T cells invade the brain parenchyma in multiple sclerosis white matter lesions. Brain 143, 1714–1730 (2020). [DOI] [PubMed] [Google Scholar]

[r17] 17.Kaskow B. J., Baecher-Allan C., Effector T cells in multiple sclerosis. Cold Spring Harb. Perspect. Med. 8, a029025 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r18] 18.Gearty S. V., et al. , An autoimmune stem-like CD8 T cell population drives type 1 diabetes. Nature 602, 156–161 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r19] 19.Abdelsamed H. A., et al. , Beta cell-specific CD8+ T cells maintain stem cell memory-associated epigenetic programs during type 1 diabetes. Nat. Immunol. 21, 578–587 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r20] 20.Saraiva D. P., et al. , Expression of HLA-DR in cytotoxic T lymphocytes: A validated predictive biomarker and a potential therapeutic strategy in breast cancer. Cancers (Basel) 13, 3841 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r21] 21.Reddy M., Eirikis E., Davis C., Davis H. M., Prabhakar U., Comparative analysis of lymphocyte activation marker expression and cytokine secretion profile in stimulated human peripheral blood mononuclear cell cultures: An in vitro model to monitor cellular immune function. J. Immunol. Methods 293, 127–142 (2004). [DOI] [PubMed] [Google Scholar]

[r22] 22.Cibrián D., Sánchez-Madrid F., CD69: From activation marker to metabolic gatekeeper. Eur. J. Immunol. 47, 946–953 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r23] 23.Barington L., et al. , EBI2 in splenic and local immune responses and in autoimmunity. J. Leukoc. Biol. 104, 313–322 (2018). [DOI] [PubMed] [Google Scholar]

[r24] 24.Sánchez Gómez I., “The role of the transcription factor TCFL5 in the physiopathology of B lymphocytes,” PhD thesis, Universidad Autónoma de Madrid, Madrid, Spain (2021).

[r25] 25.Li S., et al. , Circular RNA MAP2K2-modified immunosuppressive dendritic cells for preventing alloimmune rejection in organ transplantation. Bioeng. Transl. Med. 9, e10615 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r26] 26.Gong K.-Q., et al. , MAP2K2 delays recovery in murine models of acute lung injury and associates with acute respiratory distress syndrome outcome. Am. J. Respir. Cell Mol. Biol. 66, 555–563 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r27] 27.Ingelfinger F., et al. , Twin study reveals non-heritable immune perturbations in multiple sclerosis. Nature 603, 152–158 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r28] 28.Martinelli M., et al. , The poly(C)-binding protein Pcbp2 is essential for CD4+ T cell activation and proliferation. iScience 26, 105860 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r29] 29.Lee J., et al. , Cyclic AMP concentrations in dendritic cells induce and regulate Th2 immunity and allergic asthma. Proc. Natl. Acad. Sci. U.S.A. 112, 1529–1534 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r30] 30.Hernandez-Barco Y. G., et al. , 84 The role of the GNAS oncogene in the immune microenvironment of intraductal papillary mucinous neoplasm. Am. J. Gastroenterol. 114, S50–S51 (2019). [Google Scholar]

[r31] 31.Wittner J., Schuh W., Krüppel-like factor 2: A central regulator of B cell differentiation and plasma cell homing. Front. Immunol. 14, 1172641 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r32] 32.Hart G. T., Wang X., Hogquist K. A., Jameson S. C., Krüppel-like factor 2 (KLF2) regulates B-cell reactivity, subset differentiation, and trafficking molecule expression. Proc. Natl. Acad. Sci. U.S.A. 108, 716–721 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r33] 33.Pabbisetty S. K., et al. , KLF2 is a rate-limiting transcription factor that can be targeted to enhance regulatory T-cell production. Proc. Natl. Acad. Sci. U.S.A. 111, 9579–9584 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r34] 34.Wang J., Zhao X., Wan Y. Y., Intricacies of TGF-β signaling in Treg and Th17 cell biology. Cell. Mol. Immunol. 20, 1002–1022 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r35] 35.Liu J., et al. , Long noncoding RNA LINC01578 drives colon cancer metastasis through a positive feedback loop with the NF-κB/YY1 axis. Mol. Oncol. 14, 3211–3233 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r36] 36.Ko C.-Y., Chang W.-C., Wang J.-M., Biological roles of CCAAT/enhancer-binding protein delta during inflammation. J. Biomed. Sci. 22, 6 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r37] 37.Martin M. D., Badovinac V. P., Defining memory CD8 T cell. Front. Immunol. 9, 2692 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r38] 38.Herold K. C., et al. , The immunology of type 1 diabetes. Nat. Rev. Immunol. 24, 435–451 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r39] 39.Muñoz-Lopetegi A., et al. , Neurologic syndromes related to anti-GAD65: Clinical and serologic response to treatment. Neurol. Neuroimmunol. Neuroinflamm. 7, e696 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r40] 40.Cianciotti B. C., et al. , CD4+ memory stem T cells recognizing citrullinated epitopes are expanded in patients with rheumatoid arthritis and sensitive to tumor necrosis factor blockade. Arthritis Rheumatol. 72, 565–575 (2020). [DOI] [PubMed] [Google Scholar]

[r41] 41.Lee Y. J., et al. , Role of stem cell-like memory T cells in Systemic Lupus Erythematosus. Arthritis Rheumatol. 70, 1459–1469 (2018). [DOI] [PubMed] [Google Scholar]

[r42] 42.Redwine J. M., Buchmeier M. J., Evans C. F., In vivo expression of major histocompatibility complex molecules on oligodendrocytes and neurons during viral infection. Am. J. Pathol. 159, 1219–1224 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r43] 43.Strazielle N., Creidy R., Malcus C., Boucraut J., Ghersi-Egea J.-F., T-lymphocytes traffic into the brain across the blood-CSF barrier: Evidence using a reconstituted choroid plexus epithelium. PLoS ONE 11, e0150945 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r44] 44.Aguirre K., Miller S., MHC class II-positive perivascular microglial cells mediate resistance to Cryptococcus neoformans brain infection. Glia 39, 184–188 (2002). [DOI] [PubMed] [Google Scholar]

[r45] 45.Melzer N., Budde T., Stork O., Meuth S. G., Limbic encephalitis: Potential impact of adaptive autoimmune inflammation on neuronal circuits of the amygdala. Front. Neurol. 6, 171 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r46] 46.Räuber S., et al. , Lymphocyte signatures correspond to clinical phenotypes in autoimmune limbic encephalitis. Brain Commun. 7, fcaf156 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r47] 47.Velastegui E., Vera E., Vanden Berghe W., Muñoz M. S., Orellana-Manzano A., HLA-C: Evolution, epigenetics, and pathological implications in the major histocompatibility complex. Front. Genet. 14, 1206034 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r48] 48.Siegel R. J., Bridges S. L., Ahmed S., HLA-C: An accomplice in rheumatic diseases. ACR Open Rheumatol. 1, 571–579 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r49] 49.Chen M., et al. , Hypermethylation of HLA-C may be an epigenetic marker in psoriasis. J. Dermatol. Sci. 83, 10–16 (2016). [DOI] [PubMed] [Google Scholar]

[r50] 50.Haida Y., et al. , Association analysis of the HLA-C gene in Japanese alopecia areata. Immunogenetics 65, 553–557 (2013). [DOI] [PubMed] [Google Scholar]

[r51] 51.Kulkarni S., et al. , Genetic interplay between HLA-C and MIR148A in HIV control and Crohn disease. Proc. Natl. Acad. Sci. U.S.A. 110, 20705–20710 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r52] 52.Bracher A., et al. , An expanded parenchymal CD8+ T cell clone in GABAA receptor encephalitis. Ann. Clin. Transl. Neurol. 7, 239–244 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r53] 53.Alves A. D. P., et al. , Intrathecal T-cell clonal expansions in patients with multiple sclerosis. Ann. Clin. Transl. Neurol. 3, 422–433 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r54] 54.Alves Sousa A. de P., et al. , Comprehensive analysis of TCR-β repertoire in patients with neurological immune-mediated disorders. Sci. Rep. 9, 344 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r55] 55.Sercarz E. E., Driver clones and determinant spreading. J. Autoimmun. 14, 275–277 (2000). [DOI] [PubMed] [Google Scholar]

[r56] 56.Pappalardo J. L., et al. , Transcriptomic and clonal characterization of T cells in the human central nervous system. Sci. Immunol. 5, eabb8786 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r57] 57.Gate D., et al. , Clonally expanded CD8 T cells patrol the cerebrospinal fluid in Alzheimer’s disease. Nature 577, 399–404 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r58] 58.Biljecki M., et al. , Antibodies against glutamic acid decarboxylase 65 are locally produced in the CSF and arise during affinity maturation. Neurol. Neuroimmunol. Neuroinflamm. 10, e200090 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r59] 59.Faissner S., et al. , Successful use of anti-CD19 CAR T cells in severe treatment-refractory stiff-person syndrome. Proc. Natl. Acad. Sci. U.S.A. 121, e2403227121 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r60] 60.Ayzenberg I., et al. , Anti-CD19 CAR T-cell therapy in advanced stiff-person syndrome and concomitant myasthenia gravis. Neurol. Neuroimmunol. Neuroinflamm. 12, e200479 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r61] 61.Hao Y., et al. , Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat. Biotechnol. 42, 293–304 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r62] 62.Heming M., et al. , Neurological manifestations of COVID-19 feature T cell exhaustion and dedifferentiated monocytes in cerebrospinal fluid. Immunity 54, 164–175 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[r63] 63.Esser D., et al. , Activated Αβ T− and Reduced Mucosa-associated Invariant T Cells in LGI1− and CASPR2-encephalitis. Brain 148, 3170–3183 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Immune cell profiling reveals expanded stem cell–like memory T cells in anti-GAD65-associated neurological syndromes

Sumanta Barman

Saskia Räuber

Katharina Eisenhut

Daniela Esser

Martijn van Duijn

Madeleine Scharf

Marisol Herrera-Rivero

Paul Disse

Lara-Maria Preuth

Valeria Gulyaeva

Ilja Schwan

Eliza vom Stein

Marius Jonas

Duygu Pul

Michael Heming

Louisa Müller-Miny

Manuela Paunovic

Christine Strippel

Ebru Haholu

Jan Bartosch

Elisabeth Kaufmann

Justina Dargvainiene

Sabine Kahl

Marius Ringelstein

Eric Bindels

Heinz Wiendl

Nikolas H Stoecklein

Johannes Fischer

Norbert Goebels

Lars Komorowski

Michael Roden

Andrea Rossi

Monika Stoll

Albert J Becker

Motaz Hamed

Christian G Bien

Romana Höftberger

Jan Bauer

Sven G Meuth

Maarten J Titulaer

Frank Leypoldt

Gerd Meyer zu Hörste

Franziska Thaler

Nico Melzer

Significance

Abstract

Results

Demographics and Basic Clinical Data.

Fig. 1.

Expansion of TSCM in the PB and CSF of Anti-GAD65 AINS Individuals.

Fig. 2.

Differences in Gene Expression Involved in the Regulation of Inflammatory Immune Responses Between Anti-GAD65 AINS and IIH Individuals.

Numerous CSF BCRs Are GAD65-Reactive and Somatically Hypermutated, But Originate From Nonexpanded B Cells.

Fig. 3.

Compartment-Specific TCR Clonal Expansion in Anti-GAD65 AINS.

Fig. 4.

Expansion of CD4+ TSCM in the CSF of Anti-GAD65 AINS Individuals.

Discussion

Materials and Methods

Experimental Model and Study Participant Details.

Method Details

Quantification and Statistical Analysis.

Bioinformatic Analyses of Single Cell RNA-Sequencing Data.

Supplementary Material

Acknowledgments

Author contributions

Competing interests

Footnotes

Contributor Information

Data, Materials, and Software Availability

Supporting Information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Expansion of CD4⁺ TSCM in the CSF of Anti-GAD65 AINS Individuals.