mRNA Profiling of Inflammatory Stress Responses after Aquaporin-4 Antibody and Human Complement Treatment Reveals Upregulation of NF-κB and IL6 Pathways

mRNA Profiling of Inflammatory Stress Responses after Aquaporin-4 Antibody and Human Complement Treatment Reveals Upregulation of NF-κB and IL6 Pathways | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article mRNA Profiling of Inflammatory Stress Responses after Aquaporin-4 Antibody and Human Complement Treatment Reveals Upregulation of NF-κB and IL6 Pathways Sarah Brandl, Qian Yu, Judith Hagenbuchner, Verena Endmayr, Romana Höftberger, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7064018/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted 8 You are reading this latest preprint version Abstract Neuromyelitis optica spectrum disorder (NMOSD) is a rare autoimmune disease affecting the central nervous system via autoantibodies that target the water channel aquaporin-4 (AQP4) on astrocytes. Binding to AQP4 initiates activation of innate immune components, especially the complement system. Both in vivo and in vitro models have been developed to investigate the molecular pathomechanisms of NMOSD. The goal of our study was to characterize the molecular response of four different human cell lines to a treatment with AQP4 antibody E5415A and human complement. We aimed to identify overlapping transcriptomic changes seen in the in vivo pathophysiology of NMOSD. Tested cell lines were AQP4-ECFP overexpressing U-87MG glioblastoma cells, U-87MG expressing only ECFP, HEK293 cells transiently transfected with AQP4-EmGFP, and human primary astrocytes. Complement-dependent cytotoxicity was induced after E5415A and active human complement treatment in AQP4-expressing cells, primarily by the classical complement pathway, but also with a contribution of the alternative pathway. Transcriptomic analysis revealed that both the in vitro U-87MG-AQP4-ECFP model and an in vivo rat model share genes primarily involved in nuclear factor K-light-chain-enhancer of activated B cells (NF-κB) and interleukin-6 (IL6) pathways. These findings were confirmed on the mRNA and protein levels in the in vitro model. As further validation, serum samples from AQP4 antibody seropositive and seronegative NMOSD patients were applied instead of E5415A on U-87MG-AQP4-ECFP cells and showed the same outcome. Additionally, NF-κB upregulation was shown by immunohistochemistry in medulla oblongata lesions of AQP4 antibody seropositive NMOSD patients. To conclude, our findings demonstrate IL6 and NF-κB pathways as major contributors to inflammation caused by complement activation in AQP4 antibody-positive NMOSD. We observed U-87MG-AQP4-ECFP cells to be a suitable model to study NMOSD pathomechanisms, as they show a gene expression profile towards NF-κB and IL6 pathway upregulation comparable with an in vivo model. Biological sciences/Cell biology Health sciences/Diseases Biological sciences/Immunology Health sciences/Neurology Biological sciences/Neuroscience NMOSD AQP4-IgG complement system in vitro model transcriptomics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Background Neuromyelitis optica spectrum disorder (NMOSD) is a rare neuroinflammatory disease that primarily affects the spinal cord and optic nerves 1 . The majority of patients are seropositive for autoantibodies targeting the water channel aquaporin-4 (AQP4) on astrocyte endfeet 2 . AQP4 has two major isoforms produced by alternative splicing, M1 and M23. At the astrocytic foot processes, the isoform AQP4M23 is more abundant than AQP4M1 and is more prone to form supramolecular aggregates called orthogonal arrays of particles, which are targeted by anti-AQP4 immunoglobulin G (AQP4-IgG) autoantibodies 3 . Upon binding, immune responses are predominantly mediated by complement-dependent cytotoxicity (CDC) through the classical pathway, resulting in astrocyte damage and blood-brain barrier (BBB) disruption 4 . Several studies have shown that AQP4-IgG titers in patients correlate with the intensity of complement activation, but not with disease severity 5 – 7 . Antibody-dependent cellular cytotoxicity (ADCC) has also been implicated, as it activates various leukocytes or natural killer cells via Fcγ receptors, damaging astrocytes and adjacent non-AQP4 expressing cells 8 . The most important diagnostic tool to identify AQP4-IgG serostatus of NMOSD patients is a cell-based assay using human embryonic kidney 293 (HEK293) cells with a transiently transfected AQP4 construct and fluorescent label 9 , 10 . Beyond diagnostics, these cells are also used to study NMOSD pathophysiological mechanisms at the cellular level. To date, pathophysiological mechanisms mediated by AQP4-IgG have been studied in vitro primarily on AQP4-overexpressing HEK293A cells 5 , 6 , 11 – 14 , Chinese hamster ovary (CHO) cells 15 – 21 or primary astrocytes with human 14 , 22 – 26 or rodent origin 4 , 8 , 15 , 27 – 32 . Recently, we have established a cellular model for CDC activation after AQP4-IgG exposure in HEK293 expressing AQP4 and found a strong AQP4-IgG titer dependent activation of CDC 6 . Here, we further investigate our findings by molecular profiling of cellular models of CDC activation stably transduced U-87 MG astrocytoma cells with AQP4M23-ECFP overexpression (U-87MG-AQP4-ECFP) and a non-AQP4 overexpressing counterpart (U-87MG-ECFP) AQP4M23-EmGFP in addition to transfected HEK293 cells (HEK293-AQP4-EmGFP) and human primary astrocytes (HA). We exposed these cell lines to human complement and the monoclonal mouse anti-AQP4 antibody E5415A, analyzing cytotoxicity and transcriptomic changes post-treatment 33 , 34 . Furthermore, we compared our transcriptomic findings to data from spatial transcriptomics from an NMOSD in vivo model to identify and validate overlapping differentially expressed genes. Methods Cells Human primary astrocytes (HA), isolated from the human cerebral cortex of a female donor, gestational age 22 weeks, were purchased by ScienCell (Lot. No. 33619, ScienCell Research Laboratories, San Diego, CA). For experiments, exclusively passages 3 to 5 were used. HA were cultured in 2 µg/cm 2 poly-L-lysine coated flasks (Sciencell Research Laboratories, San Diego, CA) in astrocyte medium AM (Innoprot, Derio, Spain), containing 2% fetal calf serum, 1% P/S, and 1% astrocyte growth supplement. Medium was exchanged every other day and cells were passaged when 90% dense. The glioblastoma cell line U-87 MG (ATCC; LGC Standards GmbH, Wesel, Germany) was genetically modified for a stable protein overexpression. Thereby, an overexpression of enhanced cyan fluorescent protein (ECFP) with (U-87MG-AQP4-ECFP) or without (U-87MG-ECFP) AQP4M23 in U-87 MG cells was obtained via viral infection as described elsewhere 35 , 36 . The vectors were pLIB-MCS-ECFP-iresNeo and pLIB-MCS-AQP4m23-ECFP-iresNeo. The plasmid maps are shown in Supplementary Fig. 1. Cells were selected using 10 µg/mL neomycin for 72 hours. HEK293A and U-87 MG cell lines were cultured in DMEM high glucose, 10% fetal calf serum, and 1% non-essential amino acids and passaged twice per week. Three days before complement and antibody treatment HEK293A (ATCC; LGC Standards GmbH, Wesel, Germany) cells were seeded. After 24 hours, they were transiently transfected with a pcDNA 6.2 AQP4M23-EmGFP (HEK293-AQP4-EmGFP) plasmid with polyethylenimine (Sigma, St. Louis, Missouri) in a ratio of 1:3.6 plasmid:transfection reagent. Then, cells were incubated for two days until treatment. Patient Serum Samples and Ethical Approval As a proof of concept, serum samples from three NMOSD patients with (titers 1:20,480, 1:1,280, and 1:5,120) and from three NMOSD patients without AQP4-IgG positivity were provided by Romana Höftberger from the Division of Neuropathology and Neurochemistry, Department of Neurology, Medical University of Vienna, Austria. The use of these serum samples from a biobank was approved by the ethical committee of the Medical University of Vienna (EK 1636/2019 and 1123/2015). All patients or their legal representatives gave written informed consent to participate in the study and all methods were performed in accordance with the relevant guidelines and regulation. All samples were tested for complement activation and cytotoxicity on U-87MG-AQP4-ECFP cells. The test was performed blinded. The complement treatment and following procedures were performed the same way as with E5415A, but instead of 10 µg/mL of the monoclonal antibody, 10% of end volume of serum was applied. Additionally, the sera were heat-inactivated for 30 minutes (min) at 56°C to inactivate complement before usage. AQP4 antibody and complement treatment and assessment of cytotoxicity Treatment with the mouse monoclonal AQP4 antibody E5415A (isolated from the hybridoma cell line AQP4 E5415A-1H6-68, Resource no. RCB4883, provided by the Riken BRC through the National BioResource Project of the MEXT/AMED, Japan) 30 and human complement was done as recently described 6 . Briefly, U-87MG-ECFP, U-87MG-AQP4-ECFP, HEK293-AQP4 -EmGFP cells or HA were grown in a 12- or 96-well plate until confluent. Then, they were treated with 40% active or heat-inactivated (45 min at 56°C) pooled human complement serum (Cedarlane, Burlington, Ontario, Canada) in X-VIVO 15 medium (Lonza, Basel, Switzerland) and with or without E5415A (in-house production from E5415A-1H6-68 hybridoma cells, #RCB4883, Riken, Tsukubashi, Ibaraki, Japan) (10 µg/mL), followed by incubation for 90 min at 37°C 33 . Afterwards, the cell supernatant was taken to assess cytotoxicity, and cells were harvested for RNA isolation, or immunofluorescence staining was performed. The CDC was assessed with a lactate dehydrogenase (LDH) assay (CytoTox 96® Nonradioactive Cytotoxicity Assay, Promega, Madison, WI). Thereby, of the amount of cytosolic LDH, which is released after cell damage, can be measured indirectly via an enzymatic assay, resulting in a reduction of tetrazolium salt into a red formazan. Cells were treated as stated above. As a positive control, additional wells with cells were treated with 1x lysis buffer in X-VIVO 15 medium for the last 45 min of the incubation. As a negative control served untreated cells in X-VIVO 15 medium. Cell supernatant was taken and incubated with the same volume of substrate for 30 min at room temperature, according to the manufacturer. The reaction was stopped with the provided stop solution, and the absorption was measured with a plate reader at 492 nm (DTX880; Beckman Coulter, Brea, CA). The results were first normalized to the respective IC treated samples to correct for the LDH background reactivity of human complement serum and then to the respective lysis buffer-treated samples to assess the percentage of cell lysis. Per treatment, three technical replicates were done for each of the three biological replicates. mRNA transcriptomic analysis Cells were treated with complement and E5415A or human serum as described above. After removing the supernatant to perform cytotoxicity assays, cells were washed with PBS and harvested with RLT buffer. Further steps were performed according to the protocol of the RNeasy Mini Kit (QIAGEN, Hilden, Germany). After quantity and quality determination by a NanoDrop™ System (Thermo Fisher Scientific, Waltham, MA, USA), total RNA samples were sent for sequencing by the Illumina PE150 technology to Novogene (Martinsried, Germany). Messenger RNA was purified from total RNA using poly-T oligo-attached magnetic beads. After fragmentation, the first strand cDNA was synthesized using random hexamer primers, followed by the second strand cDNA synthesis using either dUTP for directional library or dTTP for non-directional library 37 . The non-directional library was prepared by end repair, A-tailing, adapter ligation, size selection, amplification, and purification. For the directional library, it was ready after end repair, A-tailing, adapter ligation, size selection, USER enzyme digestion, amplification, and purification. The library was checked with Qubit and real-time PCR for quantification and bioanalyzer for size distribution detection. Quantified libraries were pooled and sequenced on Illumina platforms, according to effective library concentration and data amount. The raw FASTQ format data was initially processed using fastp software by Novogene. In this step, clean data (clean reads) were obtained by removing reads containing adapter, reads containing ploy-N and low-quality reads from raw data. At the same time, Q20, Q30, and GC content of the clean data were calculated. All the downstream analyses were based on clean data with high quality. Reference genome and gene model annotation files were downloaded from the genome website directly. Index of the reference genome was built using Hisat2 v2.0.5 and paired-end clean 1 reads were aligned to the reference genome using Hisat2 v2.0.5. Hisat2 was used by Novogene as the mapping tool for that Hisat2 can generate a database of splice junctions based on the gene model annotation file and thus a better mapping result than other non-splice mapping tools 38 . The program featureCounts v1.5.0-p3 was used to count the reads numbers mapped to each gene 39 . Then, fragments per kilobase million (FPKM) of each gene were calculated based on the length of the gene and reads count mapped to this gene. FPKM, expected number of Fragments Per Kilobase of transcript sequence per Million base pairs sequenced, considers the effect of sequencing depth and gene length for the reads count at the same time, and is currently the most commonly used method for estimating gene expression levels. The acquired data were deposited in the Gene Expression Omnibus database under dataset accession number GSE291954. Reverse transcription-quantitative polymerase chain reaction Selected differentially regulated genes identified by transcriptomic analysis were validated by reverse transcription-quantitative polymerase chain reaction (RT-qPCR). First, 1 µg of RNA was retrotranscribed into cDNA with the High Capacity cDNA Reverse Transcription Kit (Applied Biosciences, Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer. The product was diluted with 480 µl of DEPC water and stored at -20°C until needed. Then, TaqMan Gene Expression Assay probes (Applied Biosciences) and iTaq™ Universal Probes Supermix (Bio-Rad, Hercules, CA, USA) were mixed with 5 µl of cDNA each and run with a CFX96 RT-PCR machine (Bio-Rad) with CFX maestro software for 40 cycles (initial denaturation at 95°C for 30 seconds; second denaturation at 95°C for 5 seconds, annealing/extension at 60°C for 30 seconds). GAPDH served as a housekeeping gene, and delta-Ct values were calculated for each gene. All TaqMan Gene Expression Assay probes are listed in Supplementary Table 1. Immunocytochemistry For immunocytochemistry (ICC), cells were grown on 0.1% gelatin solution (Sigma-Aldrich, St. Louis, MO, USA) coated ibidi µ-Slide 18 Well Glass Bottom (ibidi, ibidi GmbH, Gräfelfing, Germany). When reaching confluence, cells were exposed to human complement with or without E5415A or human serum, as described above. Live cell staining was performed on cells for terminal complement complex (TCC), as well as C3/C3b/iC3b deposition on the cell surface. Therefore, 10% heat-inactivated fetal calf serum in PBS served as a washing and antibody buffer. Mouse anti-C5b-C9 neo (aE11) AlexaFluor594 (Novus Biologicals, Centennial, CO, USA) or mouse IgG1 anti-C3/C3b/iC3b (Cedarlane, Burlington, Ontario, Canada) was applied at 5 µg/mL for 1 hour at 4°C. As secondary antibody for C3/C3b/iC3b served goat-anti mouse IgG1 AlexaFluor594 (Invitrogen™, Thermo Fisher Scientific, Waltham, MA, USA), which was applied for 30 min at room temperature in the dark. Thereafter, cells were fixed with 4% paraformaldehyde for 10 min. After washing, nucleus staining with 4′,6-diamidino-2-phenylindole (DAPI) and mounting was performed in one step by applying Immunoselect Antifading Mounting Medium DAPI (Dianova, BIOZOL, Hamburg, Germany) to the cells. For intracellular ICC, cells were fixed with 4% paraformaldehyde and permeabilized with 1% Triton X-100 for 10 min each at room temperature before blocking in 1% bovine serum albumin and 5% normal goat serum in PBS. Then, 2 µg/mL rabbit monoclonal anti-p65 (D14E12; Cell Signaling) was added to the antibody buffer (1% bovine serum albumin and 1% normal goat serum in PBS). Cells were incubated at 4°C overnight. Goat-anti rabbit AlexaFluor647 (Invitrogen™, Thermo Fisher Scientific, Waltham, MA, USA) in antibody buffer was incubated on the cells for 1 hour at room temperature in the dark. Cells were mounted the same way as for the extracellular ICC. Microscopy was performed with the Zeiss LSM700 microscope (Zeiss Plan-Apochromat 63x/1.40 oil M27 objective). Fiji ImageJ (version 1.64) was used for background subtraction and adjustments for contrast and brightness. C3a enzyme-linked immunosorbent assay C3a levels in different cell lines were evaluated by enzyme-linked immunosorbent assay (ELISA) with the Human C3a ELISA Kit (Invitrogen™, Thermo Fisher Scientific, Waltham, MA, USA). Supernatants of AQP4 antibody and complement treatment experiments were collected and stored at -20°C. Before usage, the cell supernatants were thawed and centrifuged briefly (10 min at 2000 g). The ELISA was performed according to the manufacturer’s instructions with a sample dilution of 1:2000 and a 25 min incubation time of the Substrate solution before adding Stop Solution. The plate was measured in single values at 450 nm and 620 nm with a plate reader (DTX880; Beckman Coulter, Brea, CA). Human interleukin-6 enzyme-linked immunosorbent assay To assess interleukin-6 (IL6) levels in cell supernatants after complement and E5415A exposition, an ELISA was used. The Human IL6 DuoSet ELISA and DuoSet Ancillary Reagent Kit 2 (R&D Systems, Minneapolis, MN, USA) were performed according to the manufacturer. Dilutions of 1:4 in reagent diluent were used for supernatants of U-87MG-AQP4-ECFP and HEK293-AQP4-EmGFP cells, and 1:2 dilutions for supernatants of HA and U-87MG-ECFP cells. The optical density of the samples was measured in technical duplicates at 450 nm. NMOSD animal model and spatial transcriptomics Spatial transcriptomics data deposited at the Sequence Read Archive (SRA) database (NCBI/SRA accession numbers PRJNA1258753, PRJNA1262739, PRJNA1262139, PRJNA1262895, PRJNA1262368 and PRJNA1263155) were re-analyzed from the inflamed medulla oblongata of a female 7-weeks old Lewis rat that received daily intraperitoneal injections of 1 mg E5415A in PBS for two consecutive days, inducing experimental NMOSD. The animal was euthanized by CO₂ inhalation 24 hours after the final antibody injection, and the medulla oblongata was dissected for Visium spatial transcriptomics according to protocols from 10x Genomics (10x Genomics B.V, Leiden, The Netherlands). Data were processed using Space Ranger (10x Genomics) to generate gene expression profiles per gene and spot, overlaid on HE-stained tissue images. Loupe Browser 6.5.0 (10x Genomics) was used to define a perivascular lesion area and identify differentially upregulated genes compared to the remaining medulla oblongata of the same tissue section through the “categories,” “globally distinguishing,” and “significant feature comparison” functions. We did not exclude genes with low average counts. Log2-fold changes and p-values were calculated in Space Ranger, with p-values adjusted for multiple comparisons using the Benjamini–Hochberg method. The experiments were approved by the Ethic Commission of the Medical University Vienna and performed with the license of the Austrian Ministry for Science and Research (GZ: BMBWF-66.009/0107-V/3b/2018). The study is reported in accordance with ARRIVE guidelines ( https://arriveguidelines.org ). NF-kB activation in human brain tissue Neuropathological analysis was performed on two human autopsy cases of patients with AQP4-IgG seropositive NMOSD. The use of these tissue samples from a biobank was approved by the ethical committee of the Medical University of Vienna (EK 1636/2019 and 1123/2015). All patients or their legal representatives gave written informed consent to participate in the study. Formalin-fixed and paraffin-embedded tissue sections were stained for haematoxylin and eosin (H&E) and Luxol fast blue-Periodic acid Schiff (LFB/PAS). Immunohistochemistry was performed on the automated platform Autostainer Link 48 using EnVision™ FLEX + secondary system (Dako/Agilent) according to the manufacturer´s protocol using the following primary antibodies: rabbit polyclonal anti-GFAP (1:3,000; Dako/Agilent) or mouse monoclonal anti-GFAP (1:800; Millipore), rabbit monoclonal anti-p65 (1:800; D14E12; Cell Signaling), and mouse monoclonal anti-SMI31 (phosphorylated neurofilament; 1:25,000; Sternberger). Manual immunohistochemistry was performed in a humidified chamber for complement C9 neoantigen (C9neo; complement-mediated tissue injury; rabbit polyclonal; 1:2000, from Professor Paul Morgan, Cardiff, UK). Double labelling of GFAP and NF-κB (p65) was performed with the chromogenic reactions with Fast Blue (blue) and 3-amino-9-ethylcarbazole (AEC; BioSB (red)). Image acquisition was performed using a NanoZoomer 2.0-HT digital slide scanner C9600 (Hamamatsu Photonics, Hamamatsu, Japan). Bioinformatic and statistical analysis Statistical analyses (2-way ANOVA with Šídák's multiple comparisons test, principal component analysis) and drawing of figures (bar and scatter graphs, volcano plots, principal component plots, and heat maps) were prepared using GraphPad Prism 10.4 (GraphPad Software Inc., La Jolla, California, United States). All tests were done using a significance threshold of p < 0.05 with correction for multiple comparisons if applicable. Data are expressed as mean ± standard deviation. For all experiments, at least three biological replicates were analyzed. Differential expression analysis of two conditions/groups (three biological replicates per condition) was performed using the DESeq2 R package using the WebMev platform 40 – 42 . WebMev and DESeq2 provide statistical routines for determining differential expression in digital gene expression data using a model based on the negative binomial distribution. The resulting p-values were adjusted using Benjamini and Hochberg's approach to control the false discovery rate. Genes with an adjusted p-value ≤ 0.001 and log2-fold change ≥ 1 or ≤ -1 found by DESeq2 were assigned as differentially expressed. The overlap of differentially expressed genes between different treatments was analyzed using Venn diagrams and the WebMev platform. To functionally annotate the most significant genes, gene ontology (GO) analysis was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) using the official gene symbols of Homo Sapiens of genes with a statistical significance of p < 0.05 (adjusted using the Benjamini-Hochberg correction for multiple tests) 43 , 44 . The DAVID annotation chart for the Gene Ontology Term “biological processes” (GOTERM-BP) was used. Enrichment of genes in annotation terms was evaluated using the EASE Score, a one-tail Fisher’s exact test p-value, with p-values ≤ 0.05 indicating strong enrichment. Moreover, significantly altered genes were analyzed and visualized by STRING network analysis 45 . Results Transcriptomic changes in vitro after anti-AQP4 antibody and human complement treatment indicate an inflammatory stress response in human U-87MG-AQP4-ECFP cells We analyzed the transcriptomic changes after treatment of four established NMOSD cellular models (U-87MG-AQP4-ECFP, U-87MG-ECFP, HEK293-AQP4-EmGFP, and HA) with active or heat-inactivated human complement and with or without the monoclonal AQP4 antibody E5415A, as previously described 6 . Cytotoxicity assessed by LDH release quantification was significantly increased in AQP4-expressing cells, whereas AQP4-negative U-87MG-ECFP cells showed less cell death (Fig. 1 A). The highest level of cytotoxicity (mean 37.2%) was observed in U-87MG-AQP4-ECFP after E5415A + AC treatment. This effect was significantly different to all other treatments, and to E5415A + AC-treated U-87MG-ECFP cells (all p < 0.001). In contrast, U-87MG-ECFP cells, which lack AQP4 expression, exhibited a lower level of cytotoxicity (9.6% E5415A + AC, 11.2% AC), with a statistically significant difference between E5415A + AC to IC ± E5415A, and untreated cells (p < 0.05) (Fig. 1 A). To validate that the classical complement pathway mediated cytotoxicity, ICC for the terminal complement complex (TCC, C5b-C9) was performed. TCC deposition was exclusively observed in U-87MG-AQP4-ECFP cells treated with E5415A + AC, while no staining was detected in cells treated with E5415A + IC, complement-only treatment, or untreated controls (Fig. 2 A). The background level of CDC observed in controls (E5145A + IC, AC, and IC) was already seen in our previous study in HEK-AQP4-EmGFP cells and was associated with opsonization with C3b and associated effector functions 6 . Bound C3/C3b on AC-only-treated cells was observed for U-87MG-AQP4-ECFP cells as well, although to a smaller extent (Fig. 2 B). In contrast, C3/C3b opsonization on U-87MG-ECFP was seen in comparable amounts in E5415A + AC or AC-only-treated cells (Fig. 2 C). As no TCC deposition on the cellular surface of U-87MG-ECFP cells could be observed for any of the treatments (data not shown), this demonstrates the alternative pathway contribution in the U-87MG cell line. Moreover, C3a levels in both U-87MG-AQP4-ECFP and U-87MG-ECFP cells were not significantly different between E5415A + AC and AC treatments, indicating an additional AQP4 antibody-independent activation of the alternative complement pathway (Fig. 1 B). Transcriptomic analysis was performed by pairwise comparisons of E5415A + AC-treated U-87MG-AQP4-ECFP cells, U-87MG-AQP4-ECFP, and U-87MG-ECFP cells with the other treatments using Dseq2 analysis. The volcano plot shown in Fig. 3 A and the pie chart in Fig. 3 B revealed that amongst all detectable transcripts, 13,224 (99.1%) genes remained unchanged. In comparison, 96 genes were upregulated (0.7%), and 20 genes were downregulated (0.2%) when comparing U-87MG-AQP4-ECFP cells treated with E5415A + AC and E5415A + IC. From Fig. 3 C and D it is evident that 59 genes were significantly changed in U-87MG-AQP4-ECFP cells after E5415A + AC exposure compared to all other treatments. Functional annotation using DAVID revealed a strong enrichment of genes associated with inflammation, cell death and stress responses, and NF-κB signaling pathways (Fig. 3 E), which is comparable to previous findings 32 , 46 . Next, a similar analysis of HEK293-AQP4-EmGFP cells was performed, comparing E5415A + AC-treated cells with the other treatments, respectively. HEK293-AQP4-EmGFP cells also showed significant cytotoxicity after E5415A + AC (mean 32.7%) compared to all other treatments (all p < 0.001, Fig. 1 A). Background levels of complement-derived cell lysis were, as in previous findings, likely caused by a partially activated alternative pathway. C3a levels detected in differently treated HEK293-AQP4-EmGFP cells showed an AC-specific effect (mean of 32.39 µg/mL), but this effect was not higher when adding E5415A (mean of 38.82 µg/mL) (Fig. 1 B). Only 6 genes were significantly changed in E5415A + AC-treated HEK293-AQP4-EmGFP cells compared with the other treatments (Fig. 4 A), thereof 4 genes overlapped with the genes changed in U-87MG-AQP4-ECFP cells after E5415A + AC treatment (Fig. 4 B). The expression of all 59 differentially regulated genes from U-87MG-AQP4-ECFP cells after E5415A + AC treatment in HEK293-AQP4-EmGFP cells is shown in Fig. 4 C. Finally, we performed a similar analysis of primary HA. Similarly, HA cells displayed moderate cytotoxicity after treatment with E5145A + AC (mean 11.9%) and AC (12.6%), but with a large variation between replicates, and a significant difference to untreated cells (p < 0.001) (Fig. 1 A). Furthermore, HA treated with AC alone or in combination with E5415A showed similar mean C3a levels (22.44 µg/mL and 24.87 µg/mL) (Fig. 1 B). The lack of an E5145A + AC-specific effect could be explained by the lower level of AQP4 gene expression (mean gene counts 970.9, standard deviation 37.4) as compared to U-87MG-AQP4-ECFP (50,645.9±11,295.7) or HEK293-AQP4-EmGFP (324,388.8±108,263.6). Moreover, HA showed a higher level of background cytotoxicity, which is related to C3 cleavage and opsonization (Fig. 1 B and Fig. 2 D), and the differential expression of complement associated genes (Supplementary Fig. 2). While HA showed a higher proportion of differentially expressed genes (n = 808) compared to untreated cells, we could not detect a specific effect of E5145A + AC treatments (Fig. 3 D and E). To summarize, despite high expression of glial fibrillary acidic protein (GFAP) and other astrocyte-specific genes, the HA used in this study were not useful as an in vitro NMOSD model for our research question. The AQP4 levels were found to be insufficient, nevertheless, similar AQP4-unspecific effects could be observed, which were probably due to the activation of the alternative complement pathway and C3 and C5 receptors or other antibodies in the complement serum that bind to HA. Transcriptomic changes in U87MG-AQP4-ECFP cells with anti-AQP4 antibody and human complement treatment compared to an NMOSD in vivo model To validate our findings with a NMOSD in vivo model, we compared the transcriptomic changes of the treated cells with the results of a spatial transcriptomic analysis of medulla oblongata tissue from Lewis treated with AQP4 antibody E5415A (Supplementary Fig. 3A). 322 genes were significantly upregulated in the inflammatory experimental NMOSD lesion (Fig. 5 A). As can be seen in a Venn diagram, 15 of these genes were shared with the 59 differentially expressed genes from U-87MG-AQP4-ECFP E5415A + AC-treated cells (Fig. 5 B; ATF3, CEBPD, CXCL1, CXCL2, GEM, IL11, IL6, IRAK2, IRF1, JUNB, NFKB2, NFKBIA, NFKBIZ, NR4A2, and PTX3). Hence, this gene set represents genes that are also upregulated in the in vivo AQP4-IgG positive NMOSD situation. These genes were associated with inflammation (CXCL1, CXCL2, IL11, IL6, IRAK2, NFKB2, NFKBIZ, PTX3), cell death and stress responses (ATF3, CEBPD, CXCL1, CXCL2, IL6, IRAK2, IRF1, JUNB, NFKB2, NFKBIA, NFKBIZ, NR4A2, PTX3), and NF-κB signaling pathways (IRAK2, NFKB2, NFKBIA, NFKBIZ) identified by functional annotation with DAVID (Fig. 3 E). The summative expression of these genes in tissue from rats treated with E5415A for two days, one day and a untreated control is shown in Supplementary Fig. 3A-C, and the expression of the individual genes is shown in Supplementary Fig. 3D STRING pathway analysis (Supplementary Fig. 4) revealed potential connections among these genes, except GEM (a guanidine triphosphate-binding protein that might be participating in receptor-mediated signal transduction at the plasma membrane), which showed no associations with other genes and no association with the GO terms mentioned above and was therefore excluded from further analysis 47 . The transcriptomic analysis comparing the expression changes of these genes in treated cells compared with the rat E5415A MO lesion is visualized in the heat map in Fig. 5 C. As an internal control, we added AQP4 to our gene list as the target protein of AQP4-IgG. The difference in expression is expressed with a log2-fold change of E5145A + AC-treated cells compared to the other treatments. E5145A + AC-treated U-87MG-AQP4-ECFP cells showed an upregulation in selected genes compared with all other treatments, which was comparable to the changes seen in the rat E5415A MO lesion. In HA, these gene expression changes were further accentuated, however, they were only upregulated after E5145A + AC compared to untreated cells. This pattern was not observed in HEK293-AQP4-EmGFP cells treated with E5415A + AC compared to the other treatments. To validate these findings, we performed RT-qPCR on cDNA synthesized from RNA used for transcriptomic analysis of all four cell lines. The results were visualized in a heat map showing the standardized expression (log2-fold change of E5415A + AC-treated cells compared to the other treatments, Fig. 5 D). Overall, gene expression changes were comparable to the transcriptomic analysis. For better visualization of total gene expression levels, results are additionally shown as ΔCt values (Supplementary Fig. 5). In U-87MG-AQP4-ECFP cells, highly significant gene expression differences after E5415A + AC treatment compared to other treatments were observed. Expression levels of AQP4 were significantly downregulated, NFKB2 was only upregulated compared to IC or untreated cells, and all other tested genes were significantly upregulated compared to all other treatments. In contrast, HA and U87MG-ECFP cells showed less significant differences in expression between E5415A + AC and the different treatments, only compared to untreated cells, there was a significant upregulation of most genes. In HEK293-AQP4-EmGFP cells, NR4A2 and CXCL1 were upregulated significantly after E5415A + AC treatment compared to all other treatments, whereas some of the other genes were only upregulated compared to untreated cells. No significant changes in any treatment of HEK293-AQP4-EmGFP cells were observed in AQP4, CEBPB, IL6, IL11, IRAK2, and NFKB2. Validation of transcriptomic changes of cells with anti-AQP4 antibody and human complement treatment compared to an NMOSD in vivo model on protein level In general, most of the differentially expressed genes were involved in NF-κB signaling, stress and cell death, and inflammation. For the analysis on protein level, two key proteins were selected: i) of the differentially expressed genes in the gene set are directly or indirectly related to NF-κB signaling. To efficiently test NF-κB activation, we decided to investigate RelA (p65) translocation upon E5415A and complement treatment in U-87MG-AQP4-ECFP cells since they showed the most pronounced difference on mRNA level regarding their NF-κB activation. For the ICC of RelA, U-87MG-AQP4-ECFP were treated with E5415A in combination with active or heat-inactivated human complement. Untreated cells served as a negative control, and TNF-α-treated cells as a positive control. Thereafter, intracellular ICC was performed. Representative confocal microscopy images are shown in Fig. 6 A. As expected, no translocation into the nucleus was observed in untreated cells, whereas the majority of nuclei in TNF-α-treated cells showed translocated p65. In E5415A + AC-treated cells, approximately 70% (visual estimation) showed NF-κB activation by RelA nuclear translocation. Treatments with E5415A + IC or complement only showed RelA translocation in some cells but in a smaller proportion of approximately 10%. ii) IL6 is one of the key players in NMOSD and was significantly upregulated on mRNA level in U-87MG-AQP4-ECFP cells treated with E5415A + AC. For the assessment of IL6 production after treatment, a human IL6 ELISA was used. U-87MG-AQP4-ECFP, HA, and HEK293-AQP4-EmGFP cells were treated as before; the cell supernatant was collected after the end of incubation and analyzed for IL6 levels. Results are shown in Fig. 6 B. Similar to the gene expression results, treatment with E5415A + AC resulted in the significantly increased secretion of IL6 only in U-87MG-AQP4-ECFP cells compared to the other treatments (all p < 0.001). Importantly, IL6 was released into the cell supernatant in higher amounts by U-87MG-AQP4-ECFP cells as well as in complement-only treated cells compared to the other cell lines. Whereas HA showed no IL6 production in untreated cells, they had low IL6 levels in treated cells, which did not show significant changes between E5415A + AC-treated cells and the other conditions. No IL6 secretion was found in HEK293-AQP4-EmGFP cells. Confirmatory translational analysis of NMOSD patient samples Since U-87MG-AQP4-ECFP proved to be the most applicable cell line for AQP4-IgG seropositive NMOSD modeling on the cellular level we next investigated the effects of human AQP4-IgG from NMOSD patients in this model. Therefore, we applied six NMOSD patient sera to the cells; three of them were AQP4-IgG seropositive (NMOSD#1, #4, and #5), and three were seronegative (NMOSD#2, #3, and #6). The binding of AQP4-IgG and AQP4-IgM was confirmed by cell-based assays (Supplementary Figs. 6 and 7). Cells were treated with 10% human serum sample and active or heat-inactivated human complement. Cell supernatants were analyzed for LDH to assess the degree of cytotoxicity and for IL6 production. Furthermore, cells were either lysed for RNA isolation and analysis or stained for TCC deposition or p65 translocation. Results of gene expression levels are shown as a heat map with log2-fold change to complement only controls (Fig. 7 A). Two-fold changes in gene expression were only found in the AQP4-IgG seropositive samples NMOSD#1 (AQP4-IgG titer 1:20,480), #4 (1:1,280), and #5 (1:5,120) in combination with AC, but not IC. In general, the most pronounced changes were found for NMOSD#1 and AC treatment. Next, a 2-way ANOVA with Šídák's multiple comparisons test was performed comparing each serum treatment combined with AC vs. IC. The results are shown in Supplementary Fig. 8. For AQP4, only NMOSD#1 showed significant differences (adj. p-value < 0.01). All gene expressions were significantly upregulated between NMOSD#1 with AC vs. IC treatment with a p < 0.001, except for JUNB and NFKB2 (both not significant). NMOSD#2 serum treated cells showed no significance in any of the tested genes between AC or IC addition. Similarly, NMOSD#3 + AC-treated cells showed no significant changes, except for a downregulation of IL11 (p = 0.03) and JUNB (p = 0.01). NMOSD#4 + AC treated cells showed gene upregulation in CXCL1, CXCL2, IL6, IRAK2, NFKBIA, NFKBIZ, NR4A2, and PTX3 (all p < 0.001), as well as in IRF1 (p = 0.007). Likewise, CXCL1, CXCL2, IL6, NFKBIZ, and NR4A2 were highly upregulated in NMOSD#5 + AC treated cells (p < 0.001). IL11 (p = 0.001), NFKBIA, and PTX3 (both p = 0.003) were upregulated as well. NMOSD#6 serum with AC treatment led to the upregulation of three genes, CXCL2 (p = 0.002), IL6 (p = 0.008), and NR4A2 (p < 0.001). The results from the cytotoxicity assay are shown in Fig. 7 B. U-87MG-AQP4-ECFP treated with the AQP4-IgG positive serum samples NMOSD#1, #4, and #5, and AC showed the highest cytotoxicity, but also the AQP4-IgG seronegative sample NMOSD#6 + AC was cytotoxic. Again, the greatest effect was seen for NMOSD#1. To assess the complement-dependent cell lysis further, we additionally performed ICC for TCC of treated cells (Fig. 8 A). Thereby, only cells treated with AQP4-IgG seropositive NMOSD patient sera showed TCC deposition on the cellular surface, and only when combined with AC. NMOSD#6 serum-treated cells here did not show any TCC deposition. Furthermore, the intensity of TCC was higher in NMOSD#1-treated cells, in line with the LDH assay results. IL6 levels were significantly increased after treatment of serum samples NMOSD#1, #4, and #5 in combination with AC, but not after treatment with AQP4-IgG seronegative samples (Fig. 7 C). Again, NMOSD#1 and AC-treated cells produced the highest levels of IL6. To assess NF-κB activation in NMOSD patient sera and human complement-treated U-87MG-AQP4-ECFP cells, ICC for RelA (p65) was performed (Fig. 8 B). Although NF-κB-related genes were highly upregulated in AQP4-IgG seropositive NMOSD patient sera on mRNA level, nuclear translocation of p65 was not as distinctive. NMOSD#1 + AC-treated cells showed the highest proportion of p65 in the nuclei at approximately 70% (visual estimation), whereas NMOSD#4 and NMOSD#5 showed only weak p65 translocation in approximately 10% of nuclei. However, there was also weak translocation of p65 into the nucleus observed in some of the sera plus IC-treated cells, but none in the AQP4-IgG seronegative NMOSD patient sera with AC (NMOSD#2, #3, and #6). Finally, we checked whether the NF-κB pathway is also activated in human NMOSD neuropathology. Therefore, we investigated the medulla oblongata from two NMOSD patients, who presented different lesion types: one patient showed acute astrocyte loss (Supplementary Fig. 9A and B), extensive complement deposition C9neo (Supplementary Fig. 9C) and acute axonal injury in the lesion (Supplementary Fig. 9D); the other patient revealed pronounced tissue vacuolization along with selective loss of AQP4 immunoreactivity and some axonal spheroids, but without complement deposition (data not shown). Both lesions were located in the raphe and affected axons crossing the midline, derived from the inferior olivary nucleus. Some neurons in the inferior olivary nucleus showed strong nuclear translocation of p65 (Supplementary Fig. 9E, arrows), while astrocytes in the olivary nucleus (Supplementary Fig. 9E, arrowheads) and lesion rim (Supplementary Fig. 9F, arrowheads) were p65 negative. Discussion It has been shown both in vitro and in vivo that the combination of antibodies against AQP4 and human complement induces CDC in astrocytes 4,6,8,11–17,19−32,34,48–53 . We treated the four human cellular models with E5415A, a monoclonal antibody against AQP4, in combination with human complement. Then, we performed mRNA-seq and compared transcriptomic changes to those of an NMOSD in vivo rat model. To ensure that mRNA changes after E5415A and active complement treatment were associated with CDC, we tested all four cell lines for cytotoxicity. To confirm that cell lysis resulted from the classic complement pathway, we checked for TCC and C3/C3b deposition. First, we used a cellular model previously established in our lab by Lerch et al. using transfected HEK293 cells 6 . Moreover, we used a stable AQP4-ECFP expressing U-87MG cell line, with an ECFP only overexpressing equivalent to compare AQP4-specific effects. CDC was observed in HEK293-AQP4-EmGFP cells and in U-87MG-AQP4-ECFP cells, confirming previous findings 3 , 15 , 54 . Importantly, U-87MG-AQP4-ECFP showed the highest toxicity after incubation, followed by HEK293-AQP4-EmGFP cells, which might be explained by AQP4 expression (stable in U-87MG cells and transient in HEK293 cells). However, partial activation of the alternative complement pathway was observed as well in AC-only-treated cells. Levels of C3a were elevated in AC-treated cells in the cell supernatant of, and C3/C3b deposition was found on U-87MG-AQP4-ECFP cells in our study, and in the previous study of Lerch et al. 6 . Typically, the alternative pathway is additionally eventually activated, although the classical pathway is the primary cause of CDC, leading to complement amplification accompanied with inflammation 55 . As we wanted to compare the cellular effects of these immortalized cell lines with those of physiologically relevant cells, we additionally used HA. However, the HA chosen for our study turned out to be inappropriate for our research question. First, cell culture with primary cells is limited to a few passages, as they rapidly alter their morphology, growth behavior, and gene expression, including AQP4 levels. Second, cytotoxicity was induced in these cells with no significant differences between the different treatments, and they showed high variation between replicates. Although the gene expression levels of our target genes were altered compared to untreated cells, differences between treatments could only be observed in some genes. Furthermore, no TCC deposition was detected on HA after complement treatment (data not shown), but C3/C3b on cells treated with AC with or without E5415A. Additionally, increased C3a levels in HA indicate the activation of the alternative complement pathway. Since gene counts of general astrocyte markers, such as GFAP and S100B, were high in HA samples, and AQP4 was moderately expressed across treatments as well, we wondered which developmental state the cells had at the point of isolation. Upon request at ScienCell, we got the information that the Lot we had purchased (#33619) was derived from the cerebral cortex of a female donor with a gestational age of 22 weeks. At this stage of development, AQP4 expression is only rarely found in the neocortex, and astrocyte differentiation and BBB encirclement occur postnatally 56 – 58 . On the one hand, the overexpression of AQP4 in immortalized cell lines led to a 100-fold higher expression, which might explain the majority of why changes were more pronounced in the immortalized cell lines. On the other hand, a small proportion might have been caused by the immaturity of our HA as well. This limitation should be addressed in future studies by differentiating primary astrocytes to promote maturation. However, the question remains if a sufficient AQP4 expression at the astrocyte endfeet could be achieved without co-culturing the astrocytes with brain microvascular endothelial cells. The treatment with E5415A and AC led to CDC in AQP4-expressing cells and profound transcriptomic changes on the mRNA level. As expected, AQP4 expression levels were downregulated after treatment with E5415A and AC in U-87MG-AQP4-ECFP and HEK293-AQP4-EmGFP cells, indicating a specific cell loss and confirming previous findings 31 , 59 . Moreover, most changes of E5415A and AC treatment compared to all other treatments led to most genes being changed in U-87MG-AQP4-ECFP cells (compared to all other treatments in AQP4-positive and -negative cells; n = 59). In contrast, only 6 genes changed compared to differently treated cells in HEK293-AQP4-EmGFP cells, and none in HA. Interestingly, most genes changed in HA compared to the untreated control (n = 808), but the antibody addition did not alter any gene expression compared to AC-only treated cells. This was surprising, since when compared to IC with or without E5415A-treated cells, the number of changed genes deviated a lot. This could be due to the high variability in HA replicates. HEK293-AQP4-EmGFP cells showed an upregulation of NR4A1-3 and downregulation of TXNIP when treated with E5415A and AC, which were overlapping genes that had been altered after the same treatment in U-87MG-AQP4-ECFP as well (Fig. 4 C). NR4A1-3 encode for nuclear receptor 4A family proteins and are immediate early genes that are upregulated during inflammation and cellular stress 60 , 61 . In astrocytes, NR4A2 is induced via pro-inflammatory cytokines and binds NF-κB component p65 on the target inflammatory gene promoter and results in transcriptional repression 62 . Upregulation of NR4A2 suggests a cellular response to the NF-κB activation overshoot the cells experience after treatment. TXNIP, an antioxidant and key protein in regular stress response that usually promotes apoptosis, was downregulated in both AQP4 overexpressing cell lines, indicating an anti-apoptotic gene regulation 63 . CGA upregulation and ARRDC3 downregulation were detected in HEK293-AQP4-EmGFP cells exclusively. Differently expressed genes were then compared to spatial transcriptomic results of an in vivo rat model (medulla oblongata inflammatory NMOSD lesion) after peripheral injection of E5415A antibody. Most of the overlapping differently expressed genes were directly or indirectly associated with stress responses, inflammation, and NF-κB signaling (interactions are represented schematically in Fig. 9 ). Moreover, activation of NF-κB signaling was also observed in human NMOSD pathological tissue. Although NF-κB is also activated due to the glioblastoma origin of U-87 MG cells and activating transcription factor-3 (ATF3) and CCAAT/enhancer-binding protein beta (CEBPB) are transcription factors that are associated with malignant glioblastoma, they are likewise activated under general cellular stress and inflammation 64 – 66 . These genes were elevated in treated HA compared to untreated cells as well. ATF3 and CEBPB work downstream of NF-κB to amplify or modulate immune responses 67 – 69 . Our findings are in line with a previous study of Walker-Caulfield and colleagues, which investigated primary astrocyte-enriched mouse cultures after stimulation with AQP4 antibody-positive serum samples from NMOSD patients for their transcriptomic changes 46 . This study found an upregulation of mouse equivalents of genes detected in our study (CXCL1, CXCL2, IL6, Cebpb, Nfkbia, Nfkbiz). The authors highlighted a “NMOSD granulocytic footprint”, which they indicated is activated not only downstream via CDC but also as an early event in the onset of NMOSD pathology, creating a pro-granulocytic inflammatory environment. Furthermore, this study demonstrated the efficiency of bortezomib, a small-molecule proteasome inhibitor, which is currently approved for myeloma treatment, to successfully inhibit NF-κB signaling in astrocytes 46 . Similar results of a predominantly NF-kB and IL6-driven response to AQP4 antibodies and complement were seen by two other studies in primary rat astrocytes: Du et al. found an IL6 upregulation after NMOSD patient serum application, which was assigned to Janus kinase/signal transducer and activator of transcription 3-dependent inflammatory response, as they were able to decrease IL6 levels with a Janus kinase1/2 specific inhibitor AZD1480 59 . Additionally, Wang et al. were able to prevent IL6 level increase after NMOSD patient AQP4 antibody exposure by blocking NF-κB with the inhibitor S3633, indicating the contribution of NF-κB for elevated IL6 in NMOSD 29 . IL6 has often been highlighted as a leading inflammatory cytokine in NMOSD, given its abundance in the blood and cerebrospinal fluid of affected patients. 70 . Its major impact on the disease, as well as its efficient effects on relapse prevention when blocked, mark IL6 as one of the major cytokines that should be inducible in any NMOSD model 70 . Importantly, we observed an upregulation of IL6 at both mRNA and protein levels in U-87MG-AQP4-ECFP cells after treatment with E5415A or AQP4 antibody-positive NMOSD patient serum samples, but not in HEK293-AQP4-EmGFP cells. Regarding the other changed genes in our set, NF-κB pathways are highlighted by direct or indirect involvement even more: C-X-C motif chemokine ligand 1 and 2 (CXCL1/2) are pro-inflammatory chemokines, which recruit neutrophils to sites of inflammation and are directly regulated by NF-κB. IL6 and IL11 are cytokines involved in acute and chronic inflammatory processes, with IL6 being a well-established downstream target of NF-κB 71 . Interleukin-1 receptor-associated kinase-like-2 (IRAK2) plays a role in Toll-like receptor (TLR) and IL1 receptor signaling, which activates NF-κB 71 , 72 . Interferon regulatory factor-1 (IRF1) cooperates with NF-κB in regulating genes involved in immune defense 73 . NF-κB subunit-2/p100 (NFKB2) and NF-κB inhibitor alpha (IκBα; NFKBIA) encode components of the NF-κB pathway itself. NFKB2 contributes to the non-canonical NF-κB pathway 74 . NFKBIA is an inhibitor that regulates NF-κB activity through feedback in the canonical pathway 74 . Upon activation by pro-inflammatory stimuli such as tumor necrosis factor-alpha (TNF-α), IL1β, or pathogen-associated molecular patterns, IκBα is phosphorylated and degraded. This frees the NF-κB dimer, comprising p50/ NF-κB1 and p65/RelA, allowing it to translocate into the nucleus 74 . NFKBIA is downregulated in peripheral blood mononuclear cells (PBMCs) from NMOSD patients during NMOSD relapses, perhaps reflecting the underlying inflammatory pathway in NMOSD during a disease flare 75 . NF-κB inhibitor zeta (IκBζ; NFKBIZ) functions as a co-activator for specific NF-κB target genes, particularly during inflammatory responses 74 . RelA, when unshed from IκBα plays a pivotal role by binding to κB sites in the promoters of target genes and driving the transcription of pro-inflammatory mediators, such as CXCL1, IL6, and pentraxin-3 (PTX3). When NF-κB signaling was tested via RelA translocation in this study, it was more elevated in our U-87MG-AQP4-ECFP cells treated with E5415A + AC combination treatment. Nevertheless, it needs to be mentioned that this effect was only as pronounced when adding the monoclonal antibody or anti-AQP4 IgG seropositive human serum with a high titer (1:20,480). PTX3, an acute-phase protein, is involved in innate immunity and complement component interaction (C1q, factor H, ficolins, mannan-binding lectin) 76 . PTX3 has also been shown to be associated with inflammatory responses in NMOSD patients 77 . NR4A2, as mentioned before, plays an anti-inflammatory role by suppressing RelA binding to the target inflammatory gene promoter 62 . Therefore, NR4A2 would be a possible candidate for NF-κB blockade by upregulating its expression. Indeed, we have seen E5415A + AC-specific upregulation of NR4A2 in AQP4-overexpressing cells. Its NF-κB downregulating effect, however, was not observed in this study. Transcription factor jun-B (JUNB), part of the AP-pathogenicity and alongside NF-κB in cytokine expression, is essential for IL23-dependent Th17 pathogenicity and is itself induced by IL6 78,79 . Nishiyama et al. observed the release of pro-inflammatory Th17 cytokines in AQP4-IgG seropositive NMOSD patient PBMCs following incubation with AQP4-immunocomplexes and complement. Notably, upregulation of IL17A and IL23 was detected only in treatment-naïve PBMCs that had not undergone B cell depletion, whereas Rituximab-treated PBMCs exhibited enhanced IL6 production. These findings underscore the necessity of IL6-activated B and T cells for downstream Th17 cytokine production 80 . This was highlighted before by Agasing et al. in transcriptomic analysis of NMOSD patient PBMCs as well, as they showed the co-upregulation of IL6 and interferon type I, thereby interferon type I being essential for B cell activation, and IL6 for further Th17 differentiation 81 . Our study has some limitations. First, cell lysis was also observed to a lower degree in samples with complement only or E5415A with IC as well. TCC staining showed that this cell lysis was not due to terminal complement complex formation. A possible explanation for this could be unspecific binding or alternative pathway activation by C3 opsonization after the addition of human complement. Chen and colleagues saw C3 upregulation in mouse astrocytes after NMO-IgG exposure that led to microglial interactions via C3a receptors 82 . We detected high gene counts of C3 in U-87MG cell lines independent of treatment. In contrast, gene counts were lower in E5415A + AC-treated HEK293-AQP4-EmGFP cells compared to the other treatments, and higher in E5415A + AC-treated HA compared to the remaining samples. C1R and C1S components were elevated in these cells as well, but gene counts were moderate to low in all four cell lines and did not correlate notably with different treatment (Supplementary Fig. 2). Second, besides complement components being directly involved in the classical complement pathway, there are additional complement regulatory proteins such as CD46 (membrane cofactor protein), CD55 (decay accelerating factor), and CD59 (protectin), which either inhibit the C3 convertase, or the TCC formation 83 . A study by Saadoun and Papadopoulus revealed that complement regulatory proteins are expressed in human astrocytes, but not in NMOSD lesions and after co-culture of astrocytes with endothelial cells 26 . These complement inhibitors, CFH (complement factor H) and CLU (clusterin), were expressed at different levels in the cell lines used in our study (Supplementary Fig. 3). The presence of these complement inhibitors might explain why relatively high amounts of human complement were necessary to induce CDC. Third, findings were rather unspecific in HA, which is explained by their low AQP4 expression. These primary HA were obtained from human cerebral cortex (purchased via ScienCell) and cultured for up to 5 passages. Finally, the results from human serum samples were AQP4 antibody titer dependent. Most genes were significantly changed after treatment with human AQP4 antibodies and AC. Cytotoxicity also correlated with AQP4 antibody levels. Surprisingly, serum from the AQP4-antibody negative NMOSD patient #6 induced some CDC, probably due to the presence of other factors in the serum of this patient. Finally, in the human neuropathological sections investigated in our study strong nuclear translocation of NF-κB was only seen in some neurons, while astrocytes were negative. This could be explained by the selective loss of astrocytes in the lesions and the activation of NF-κB signaling pathways by the alternative complement pathway in neurons, similar as seen in other cells in this study. Conclusions Our study demonstrates that AQP4-expressing U-87MG-AQP4-ECFP cells exhibit significant CDC upon exposure to the monoclonal AQP4 antibody E5415A in combination with active human complement. Transcriptomic analysis revealed an upregulation of genes directly or indirectly linked to IL6 and NF-κB after this treatment. Similar changes were also seen in an in vivo rat model of NMOSD: This suggests that U-87MG-AQP4-ECFP cells effectively mimic CDC-related astrocytic responses in AQP4 antibody seropositive NMOSD and enable to testing of new treatment strategies, especially regarding the NF-κB pathway, on a cellular level. Abbreviations AC active complement ADCC antibody-dependent cellular cytotoxicity AQP4 aquaporin-4 ATF3 activating transcription factor-3 BBB blood-brain barrier CDC complement-dependent cytotoxicity CEBPB CCAAT/enhancer-binding protein beta CFH complement factor H CHO Chinese hamster ovary CXCL C-X-C motif chemokine ligand DAPI 4′,6-diamidino-2-phenylindole DAVID Database for Annotation, Visualization, and Integrated Discovery DEG differentially expressed gene ECFP enhanced cyan fluorescent protein ELISA enzyme-linked immunosorbent assay EmGFP emerald green fluorescent protein FPKM fragments per kilobase million GFAP glial fibrillary acidic protein GO gene ontology HEK293 human embryonic kidney 293 IL interleukin IC inactive complement ICC immunocytochemistry IgG immunoglobulin-G IRAK2 interleukin-1 receptor-associated kinase-like-2 IRF1 interferon regulatory factor-1 JUNB transcription factor jun-B LDH lactate dehydrogenase min minutes NF-κB nuclear factor K-light-chain-enhancer of activated B cells NFKB2 NF-κB subunit-2 NFKBIA NF-κB inhibitor alpha NFKBIZ NF-κB inhibitor zeta NR4A Nuclear receptor subfamily 4 group A protein PBMCs peripheral blood mononuclear cells PC principal component PTX3 pentraxin-3 RT-qPCR reverse transcription-quantitative polymerase chain reaction TCC terminal complement complex TNF-α tumor necrosis factor-alpha Declarations Ethics approval and consent to participate Serum samples from six NMOSD patients and neuropathological tissue sections from two NMOSD patients were provided by the biobank of the Department of Neuropathology, Medical University of Vienna, Austria. The use of these samples from a biobank for research studies was approved by the ethical committee of the Medical University of Vienna (EK 1636/2019 and 1123/2015). All patients or their legal representatives gave written informed consent to participate in the study and all methods were performed in accordance with the relevant guidelines and regulation. Consent for publication Not applicable. Competing Interests The authors declare that this study received funding from Roche Austria GmbH (to MR). The funder was not involved in the study design, analysis, or critical revision of the article for important intellectual content. RH reports speaker honoraria from UCB and BMS. The Medical Universities of Innsbruck (Austria; employer of MR) and Vienna (Austria; employer of RH) receive payments for antibody assays and for antibody validation experiments organized by Euroimmun (Lübeck, Germany). QY, JH, VE and MB declare no competing interests. Funding This study was financially supported by the intramural funding program of the Medical University of Innsbruck Ph.D. Research Training Groups, Project 2022-1-2 “CONNECT” (to SB, JH and MR) and a restricted research grant from Roche Austria GmbH supporting this study (to MR). It was also supported by the Austrian Science Fund (FWF grant 10.55776/PAT6054424 ) to MB, the China Scholarship Council (CSC 202306170046) to QY, and the Austrian Research Promotion Agency (FFG, project number FO999920011) to RH and VE. Author Contribution SB analyzed and interpreted the data, wrote the manuscript, and performed all experiments. MR designed the study, supervised the work, analyzed and interpreted data, and participated in preparing the manuscript. JH produced the stable cell lines and contributed to the manuscript. QY and MB provided the spatial transcriptomics analysis of the experimental rat NMO model and helped with the analysis and interpretation of data. VE and RH provided serum samples of NMOSD patients and performed the immunohistochemistry of human NMOSD neuropathological sections. All authors reviewed the manuscript critically for important intellectual content and approved the final version of the manuscript. Data Availability The datasets used and/or analyzed during the current study are available at GEO (GSE291954) or shown in the manuscript and supplementary file. The submission is still private, to review the reviewer should go to https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE291954 and enter the token urmxuskqlnmvtkz into the box. References Jarius, S. et al. Neuromyelitis optica. Nat. Rev. Dis. Primers . 6 , 85. 10.1038/s41572-020-0214-9 (2020). Wingerchuk, D. M. et al. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology 85 , 177–189. 10.1212/wnl.0000000000001729 (2015). Crane, J. M. et al. Binding affinity and specificity of neuromyelitis optica autoantibodies to aquaporin-4 M1/M23 isoforms and orthogonal arrays. J. Biol. Chem. 286 , 16516–16524. 10.1074/jbc.M111.227298 (2011). Hinson, S. R. et al. Molecular outcomes of neuromyelitis optica (NMO)-IgG binding to aquaporin-4 in astrocytes. Proceedings of the National Academy of Sciences 109, 1245–1250, doi: (2012). 10.1073/pnas.1109980108 Jitprapaikulsan, J. et al. Clinical utility of AQP4-IgG titers and measures of complement-mediated cell killing in NMOSD. Neurology(R) Neuroimmunol. neuroinflammation . 7 10.1212/nxi.0000000000000727 (2020). Lerch, M. et al. More Efficient Complement Activation by Anti-Aquaporin-4 Compared With Anti-Myelin Oligodendrocyte Glycoprotein Antibodies. Neurology(R) neuroimmunology & neuroinflammation 10, (2023). 10.1212/nxi.0000000000200059 Yandamuri, S. S. et al. MOGAD patient autoantibodies induce complement, phagocytosis, and cellular cytotoxicity. JCI Insight . 8 10.1172/jci.insight.165373 (2023). Duan, T., Smith, A. J. & Verkman, A. S. Complement-independent bystander injury in AQP4-IgG seropositive neuromyelitis optica produced by antibody-dependent cellular cytotoxicity. Acta Neuropathol. Commun. 7 , 112. 10.1186/s40478-019-0766-7 (2019). Waters, P. J. et al. Serologic diagnosis of NMO: a multicenter comparison of aquaporin-4-IgG assays. Neurology 78 , 665–671. 10.1212/WNL.0b013e318248dec1 (2012). discussion 669. Waters, P. et al. Multicentre comparison of a diagnostic assay: aquaporin-4 antibodies in neuromyelitis optica. J. Neurol. Neurosurg. Psychiatry . 87 , 1005–1015. 10.1136/jnnp-2015-312601 (2016). Ishikura, T. et al. Anti-AQP4 autoantibodies promote ATP release from astrocytes and induce mechanical pain in rats. J. Neuroinflammation . 18 , 181. 10.1186/s12974-021-02232-w (2021). Lennon, V. A., Kryzer, T. J., Pittock, S. J., Verkman, A. S. & Hinson, S. R. IgG marker of optic-spinal multiple sclerosis binds to the aquaporin-4 water channel. J. Exp. Med. 202 , 473–477. 10.1084/jem.20050304 (2005). Kaneko, K. et al. Different Complement Activation Patterns Following C5 Cleavage in MOGAD and AQP4-IgG + NMOSD. Neurol. Neuroimmunol. Neuroinflammation . 11 , e200293. 10.1212/NXI.0000000000200293 (2024). Yamashita, K. et al. Cerebrospinal fluid mitochondrial DNA in neuromyelitis optica spectrum disorder. J. Neuroinflammation . 15 , 125. 10.1186/s12974-018-1162-0 (2018). Soltys, J. et al. Membrane assembly of aquaporin-4 autoantibodies regulates classical complement activation in neuromyelitis optica. J. Clin. Invest. 129 , 2000–2013. 10.1172/jci122942 (2019). Ratelade, J. et al. Neuromyelitis optica IgG and natural killer cells produce NMO lesions in mice without myelin loss. Acta Neuropathol. 123 , 861–872. 10.1007/s00401-012-0986-4 (2012). Zhu, W. et al. Human C5-specific single-chain variable fragment ameliorates brain injury in a model of NMOSD. Neurol. Neuroimmunol. Neuroinflamm . 6 , e561. 10.1212/nxi.0000000000000561 (2019). Tradtrantip, L., Yeaman, M. R. & Verkman, A. S. Cytoprotective IgG antibodies in sera from a subset of patients with AQP4-IgG seropositive neuromyelitis optica spectrum disorder. Sci. Rep. 11 , 21962. 10.1038/s41598-021-01294-3 (2021). Phuan, P. W., Ratelade, J., Rossi, A., Tradtrantip, L. & Verkman, A. S. Complement-dependent cytotoxicity in neuromyelitis optica requires aquaporin-4 protein assembly in orthogonal arrays. J. Biol. Chem. 287 , 13829–13839. 10.1074/jbc.M112.344325 (2012). Tradtrantip, L., Ratelade, J., Zhang, H. & Verkman, A. S. Enzymatic deglycosylation converts pathogenic neuromyelitis optica anti-aquaporin-4 immunoglobulin G into therapeutic antibody. Ann. Neurol. 73 , 77–85. 10.1002/ana.23741 (2013). Mader, S., Brimberg, L., Soltys, J. N., Bennett, J. L. & Diamond, B. Mutations of Recombinant Aquaporin-4 Antibody in the Fc Domain Can Impair Complement-Dependent Cellular Cytotoxicity and Transplacental Transport. Front. Immunol. 9 , 1599. 10.3389/fimmu.2018.01599 (2018). Vincent, T. et al. Functional Consequences of Neuromyelitis Optica-IgG Astrocyte Interactions on Blood-Brain Barrier Permeability and Granulocyte Recruitment1. J. Immunol. 181 , 5730–5737. 10.4049/jimmunol.181.8.5730 (2008). Haruki, H. et al. NMO sera down-regulate AQP4 in human astrocyte and induce cytotoxicity independent of complement. J. Neurol. Sci. 331 , 136–144. 10.1016/j.jns.2013.05.035 (2013). Nishiyama, S. et al. Complement-dependent and -independent aquaporin 4-antibody-mediated cytotoxicity in human astrocytes: Pathogenetic implications in neuromyelitis optica. Biochem. Biophys. Rep. 7 , 45–51. 10.1016/j.bbrep.2016.05.012 (2016). Piatek, P. et al. C5a-Preactivated Neutrophils Are Critical for Autoimmune-Induced Astrocyte Dysregulation in Neuromyelitis Optica Spectrum Disorder. Front. Immunol. 9 , 1694. 10.3389/fimmu.2018.01694 (2018). Saadoun, S. & Papadopoulos, M. C. Role of membrane complement regulators in neuromyelitis optica. Multiple Scler. J. 21 , 1644–1654. 10.1177/1352458515571446 (2015). Sabater, L. et al. Cytotoxic effect of neuromyelitis optica antibody (NMO-IgG) to astrocytes: an in vitro study. J. Neuroimmunol. 215 , 31–35. 10.1016/j.jneuroim.2009.07.014 (2009). Alexopoulos, H. et al. Anti-aquaporin-4 autoantibodies in systemic lupus erythematosus persist for years and induce astrocytic cytotoxicity but not CNS disease. J. Neuroimmunol. 289 , 8–11. 10.1016/j.jneuroim.2015.10.007 (2015). Wang, Y. et al. NMO-IgG Induce Interleukin-6 Release via Activation of the NF-κB Signaling Pathway in Astrocytes. Neuroscience 496 , 96–104. 10.1016/j.neuroscience.2022.05.038 (2022). Huang, P. et al. The binding property of a monoclonal antibody against the extracellular domains of aquaporin-4 directs aquaporin-4 toward endocytosis. Biochem. Biophys. Rep. 7 , 77–83. 10.1016/j.bbrep.2016.05.017 (2016). Hinson, S. R. et al. Aquaporin-4-binding autoantibodies in patients with neuromyelitis optica impair glutamate transport by down-regulating EAAT2. J. Exp. Med. 205 , 2473–2481. 10.1084/jem.20081241 (2008). Howe, C. L. et al. Neuromyelitis optica IgG stimulates an immunological response in rat astrocyte cultures. Glia 62 , 692–708. 10.1002/glia.22635 (2014). Kurosawa, K. et al. Severely exacerbated neuromyelitis optica rat model with extensive astrocytopathy by high affinity anti-aquaporin-4 monoclonal antibody. Acta Neuropathol. Commun. 3 , 82. 10.1186/s40478-015-0259-2 (2015). Hillebrand, S. et al. Circulating AQP4-specific auto-antibodies alone can induce neuromyelitis optica spectrum disorder in the rat. Acta Neuropathol. 137 , 467–485. 10.1007/s00401-018-1950-8 (2019). Hagenbuchner, J. et al. The tubulin inhibitor MG-2477 induces autophagy-regulated cell death, ROS accumulation and activation of FOXO3 in neuroblastoma. Oncotarget 8 (2017). Nothdurfter, D. et al. 3D bioprinted, vascularized neuroblastoma tumor environment in fluidic chip devices for precision medicine drug testing. Biofabrication 14 , 035002. 10.1088/1758-5090/ac5fb7 (2022). Parkhomchuk, D. et al. Transcriptome analysis by strand-specific sequencing of complementary DNA. Nucleic Acids Res. 37 , e123. 10.1093/nar/gkp596 (2009). Mortazavi, A., Williams, B. A., McCue, K., Schaeffer, L. & Wold, B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat. Methods . 5 , 621–628. 10.1038/nmeth.1226 (2008). Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30 , 923–930. 10.1093/bioinformatics/btt656 (2014). Wang, Y. E. et al. A Cloud Platform for Analyzing and Visualizing Cancer Genomic Data. Cancer Res. 77 , e11–e14. 10.1158/0008-5472.Can-17-0802 (2017). Anders, S. & Huber, W. Differential expression analysis for sequence count data. Genome Biol 11, R106, (2010). 10.1186/gb-2010-11-10-r106 Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15 , 550. 10.1186/s13059-014-0550-8 (2014). Sherman, B. T. et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 50 , W216–w221. 10.1093/nar/gkac194 (2022). Huang da, W., Sherman, B. T. & Lempicki, R. A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 4 , 44–57. 10.1038/nprot.2008.211 (2009). Szklarczyk, D. et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 51 , D638–d646. 10.1093/nar/gkac1000 (2023). Walker-Caulfield, M. E. et al. NFκB signaling drives pro-granulocytic astroglial responses to neuromyelitis optica patient IgG. J. Neuroinflammation . 12 , 185. 10.1186/s12974-015-0403-8 (2015). Maguire, J. et al. Gem: an induced, immediate early protein belonging to the Ras family. Science 265 , 241–244. 10.1126/science.7912851 (1994). Kinoshita, M. et al. Astrocytic necrosis is induced by anti-aquaporin-4 antibody-positive serum. Neuroreport 20 , 508–512. 10.1097/wnr.0b013e32832776f4 (2009). Stathopoulos, P. & Dalakas, M. C. The role of complement and complement therapeutics in neuromyelitis optica spectrum disorders. Expert Rev. Clin. Immunol. 18 , 933–945. 10.1080/1744666X.2022.2105205 (2022). Bennett, J. L. et al. Intrathecal pathogenic anti-aquaporin-4 antibodies in early neuromyelitis optica. Ann. Neurol. 66 , 617–629. 10.1002/ana.21802 (2009). Bradl, M. et al. Neuromyelitis optica: pathogenicity of patient immunoglobulin in vivo. Ann. Neurol. 66 , 630–643. 10.1002/ana.21837 (2009). Saadoun, S. et al. Intra-cerebral injection of neuromyelitis optica immunoglobulin G and human complement produces neuromyelitis optica lesions in mice. Brain 133 , 349–361. 10.1093/brain/awp309 (2010). Tradtrantip, L., Yeaman, M. R. & Verkman, A. S. Cytoprotective IgG antibodies in sera from a subset of patients with AQP4-IgG seropositive neuromyelitis optica spectrum disorder. Sci. Rep. 11 , 21962. 10.1038/s41598-021-01294-3 (2021). Rossi, A., Moritz, T. J., Ratelade, J. & Verkman, A. S. Super-resolution imaging of aquaporin-4 orthogonal arrays of particles in cell membranes. J. Cell. Sci. 125 , 4405–4412. 10.1242/jcs.109603 (2012). Ho, B. H. T., Spicer, B. A. & Dunstone, M. A. Action of the Terminal Complement Pathway on Cell Membranes. J. Membr. Biol. 10.1007/s00232-025-00343-6 (2025). Gömöri, E. et al. Fetal development of membrane water channel proteins aquaporin-1 and aquaporin-4 in the human brain. Int. J. Dev. Neurosci. 24 , 295–305. 10.1016/j.ijdevneu.2006.05.003 (2006). Holst, C. B., Brøchner, C. B., Vitting-Seerup, K. & Møllgård, K. Astrogliogenesis in human fetal brain: complex spatiotemporal immunoreactivity patterns of GFAP, S100, AQP4 and YKL-40. J. Anat. 235 , 590–615. 10.1111/joa.12948 (2019). Goasdoué, K., Miller, S. M., Colditz, P. B., Björkman, S. T. & Review The blood-brain barrier; protecting the developing fetal brain. Placenta 54 , 111–116. https://doi.org/10.1016/j.placenta.2016.12.005 (2017). Du, L. et al. Effect of NMO-IgG on the interleukin-6 cascade in astrocytes via activation of the JAK/STAT3 signaling pathway. Life Sci. 258 , 118217. https://doi.org/10.1016/j.lfs.2020.118217 (2020). Chen, L., Fan, F., Wu, L. & Zhao, Y. The nuclear receptor 4A family members: mediators in human disease and autophagy. Cell. Mol. Biol. Lett. 25 10.1186/s11658-020-00241-w (2020). Safe, S. et al. Nuclear receptor 4A (NR4A) family - orphans no more. Saijo, K. et al. A Nurr1/CoREST pathway in microglia and astrocytes protects dopaminergic neurons from inflammation-induced death. Choi, E. H., Park, S. J. & TXNIP A key protein in the cellular stress response pathway and a potential therapeutic target. Exp. Mol. Med. 55 , 1348–1356. 10.1038/s12276-023-01019-8 (2023). Ji, J. et al. TRIM22 activates NF-κB signaling in glioblastoma by accelerating the degradation of IκBα. Cell. Death Differ. 28 , 367–381. 10.1038/s41418-020-00606-w (2021). Nagai, S. et al. Aberrant nuclear factor-kappaB activity and its participation in the growth of human malignant astrocytoma. J. Neurosurg. 96 , 909–917. 10.3171/jns.2002.96.5.0909 (2002). Li, W. et al. Sphingosine kinase 1 promotes growth of glioblastoma by increasing inflammation mediated by the NF-κB /IL-6/STAT3 and JNK/PTX3 pathways. Acta Pharm. Sinica B . 12 , 4390–4406. https://doi.org/10.1016/j.apsb.2022.09.012 (2022). Ko, C. Y., Chang, W. C. & Wang, J. M. Biological roles of CCAAT/Enhancer-binding protein delta during inflammation. J. Biomed. Sci. 22 , 6. 10.1186/s12929-014-0110-2 (2015). Nyunt, T. et al. Mitochondrial oxidative stress-induced transcript variants of ATF3 mediate lipotoxic brain microvascular injury. Free Radic. Biol. Med. 143 , 25–46. https://doi.org/10.1016/j.freeradbiomed.2019.07.024 (2019). Holland, S. D. & Ramer, M. S. Microglial activating transcription factor 3 upregulation: An indirect target to attenuate inflammation in the nervous system. Front. Mol. Neurosci. 16 10.3389/fnmol.2023.1150296 (2023). Fujihara, K. et al. Interleukin-6 in neuromyelitis optica spectrum disorder pathophysiology. Neurol. Neuroimmunol. Neuroinflamm . 7 10.1212/nxi.0000000000000841 (2020). Hirano, T. IL-6 in inflammation, autoimmunity and cancer. Int. Immunol. 33 , 127–148. 10.1093/intimm/dxaa078 (2021). Pereira, M. & Gazzinelli, R. T. Regulation of innate immune signaling by IRAK proteins. Front. Immunol. 14 10.3389/fimmu.2023.1133354 (2023). Feng, H., Zhang, Y. B., Gui, J. F., Lemon, S. M. & Yamane, D. Interferon regulatory factor 1 (IRF1) and anti-pathogen innate immune responses. PLoS Pathog. 17 , e1009220. 10.1371/journal.ppat.1009220 (2021). Guo, Q. et al. NF-κB in biology and targeted therapy: new insights and translational implications. Signal. Transduct. Target. Therapy . 9 10.1038/s41392-024-01757-9 (2024). Yamamura, R. et al. Transcriptome signature in the blood of neuromyelitis optica spectrum disorder under steroid tapering. Front. Immunol. 16 , 1508977. 10.3389/fimmu.2025.1508977 (2025). Inforzato, A. et al. PTX3 as a paradigm for the interaction of pentraxins with the complement system. Semin Immunol. 25 , 79–85. 10.1016/j.smim.2013.05.002 (2013). Wang, H. et al. Increased plasma levels of pentraxin 3 in patients with multiple sclerosis and neuromyelitis optica. Mult Scler. 19 , 926–931. 10.1177/1352458512457845 (2013). Ren, F. J., Cai, X. Y., Yao, Y. & Fang, G. Y. JunB: a paradigm for Jun family in immune response and cancer. Front. Cell. Infect. Microbiol. 13 , 1222265. 10.3389/fcimb.2023.1222265 (2023). Hasan, Z. et al. JunB is essential for IL-23-dependent pathogenicity of Th17 cells. Nat. Commun. 8 , 15628. 10.1038/ncomms15628 (2017). Nishiyama, S. et al. Anti-aquaporin-4 immune complex stimulates complement-dependent Th17 cytokine release in neuromyelitis optica spectrum disorders. Sci. Rep. 14 , 3146. 10.1038/s41598-024-53661-5 (2024). Agasing, A. M. et al. Transcriptomics and proteomics reveal a cooperation between interferon and T-helper 17 cells in neuromyelitis optica. Nat. Commun. 11 , 2856. 10.1038/s41467-020-16625-7 (2020). Chen, T. et al. Astrocyte-microglia interaction drives evolving neuromyelitis optica lesion. J. Clin. Investig. 130 , 4025–4038. 10.1172/JCI134816 (2020). Zipfel, P. F. & Skerka, C. Complement regulators and inhibitory proteins. Nat. Rev. Immunol. 9 , 729–740. 10.1038/nri2620 (2009). Additional Declarations Competing interest reported. The authors declare that this study received funding from Roche Austria GmbH (to MR). The funder was not involved in the study design, analysis, or critical revision of the article for important intellectual content. RH reports speaker honoraria from UCB and BMS. The Medical Universities of Innsbruck (Austria; employer of MR) and Vienna (Austria; employer of RH) receive payments for antibody assays and for antibody validation experiments organized by Euroimmun (Lübeck, Germany). QY, JH, VE and MB declare no competing interests. Supplementary Files Supplementaryfile.pdf Cite Share Download PDF Status: Published Journal Publication published 08 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 11 Sep, 2025 Reviews received at journal 09 Sep, 2025 Reviewers agreed at journal 20 Aug, 2025 Reviewers invited by journal 20 Aug, 2025 Editor assigned by journal 20 Aug, 2025 Editor invited by journal 18 Aug, 2025 Submission checks completed at journal 11 Aug, 2025 First submitted to journal 11 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7064018","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":513606364,"identity":"4b22ba34-cc25-47a9-a383-174b69232292","order_by":0,"name":"Sarah Brandl","email":"","orcid":"","institution":"Medical University of Innsbruck","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Brandl","suffix":""},{"id":513606365,"identity":"04f3d3ea-8acd-413a-a870-df8cc56e647a","order_by":1,"name":"Qian Yu","email":"","orcid":"","institution":"Medical University of Vienna","correspondingAuthor":false,"prefix":"","firstName":"Qian","middleName":"","lastName":"Yu","suffix":""},{"id":513606366,"identity":"dfe86f7c-4c19-457d-a7ea-434bcb319ccc","order_by":2,"name":"Judith Hagenbuchner","email":"","orcid":"","institution":"Medical University of Innsbruck","correspondingAuthor":false,"prefix":"","firstName":"Judith","middleName":"","lastName":"Hagenbuchner","suffix":""},{"id":513606367,"identity":"e270cf7c-31a6-461e-afa1-7399c8f010cd","order_by":3,"name":"Verena Endmayr","email":"","orcid":"","institution":"Medical University of Vienna","correspondingAuthor":false,"prefix":"","firstName":"Verena","middleName":"","lastName":"Endmayr","suffix":""},{"id":513606368,"identity":"b9c085bc-4a06-4f2b-b335-6a2cb372a690","order_by":4,"name":"Romana Höftberger","email":"","orcid":"","institution":"Medical University of Vienna","correspondingAuthor":false,"prefix":"","firstName":"Romana","middleName":"","lastName":"Höftberger","suffix":""},{"id":513606369,"identity":"2e3397b2-cdad-4a11-8d32-8c072d68bdd0","order_by":5,"name":"Monika Bradl","email":"","orcid":"","institution":"Medical University of Vienna","correspondingAuthor":false,"prefix":"","firstName":"Monika","middleName":"","lastName":"Bradl","suffix":""},{"id":513606370,"identity":"7499c750-1b54-49a1-bb18-83890b6f1684","order_by":6,"name":"Markus Reindl","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/klEQVRIiWNgGAWjYHACZjgl8aBCAsQ0IEFLwhkkLTwEtQCBRGIbA2Etuu1nHxvz/GHI52fnPXgjcZ4FUKR5A8OPGgZ5exxazM6kGyfztjFYzmzmS7ZI3CYBFDlWwNhzjMGwB5eWA2nMh3kbGAwMDvOYSQC11G+7kWPAABRhxKnl/DPmw0CHGdiDtcwB2nL/jQHj3wYGe5xabqQxJ/OwAW1hBmlpAGq5wWPADLQlEbeWZ8yGc9skDCQO8xhbJBwD+SWt4LDMMYnkngO4HJbGLPHmj40Bf/8ZwxsfauoYzI4f3vjwTY2NbXsDDmsgQAKVewBDZBSMglEwCkYBSQAAuHxOf8c7sigAAAAASUVORK5CYII=","orcid":"","institution":"Medical University of Innsbruck","correspondingAuthor":true,"prefix":"","firstName":"Markus","middleName":"","lastName":"Reindl","suffix":""}],"badges":[],"createdAt":"2025-07-07 09:53:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7064018/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7064018/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-27335-9","type":"published","date":"2025-12-08T15:59:17+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":91527332,"identity":"3637b376-6da3-4783-ab67-d43f578b9191","added_by":"auto","created_at":"2025-09-17 11:17:06","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":319223,"visible":true,"origin":"","legend":"\u003cp\u003eCytotoxicity assay after human complement and E5415A treatment. The cell lines U-87MG-AQP4-ECFP, U-87MG-ECFP, HEK293-AQP4-EmGFP, and HA were treated with or without the monoclonal AQP4 antibody E5415A (10 µg/mL) in combination with active or heat-inactivated human complement. (\u003cstrong\u003eA\u003c/strong\u003e) Cytotoxicity was assessed with the CytoTox 96® Non-Radioactive Cytotoxicity Assay. Values were normalized to cells treated with lysis buffer, which served as a positive control (=100% cell lysis), and human complement. (\u003cstrong\u003eB\u003c/strong\u003e) Levels of C3a were determined by ELISA. Values were normalized to the C3a levels present in the human complement serum. Bar charts (means with standard deviation and individual values) were created with GraphPad Prism 10.4. Groups (for (A) n=9, for (B) n=3) were compared with a 2-way ANOVA with Šídák's multiple comparisons test, all treatments per cell line against E5415A+AC. ns = not significant, * p\u0026lt;0.05, ** p\u0026lt;0.01, *** p\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/596030acb235c4e042dab75d.jpeg"},{"id":91526429,"identity":"a2a16ea0-b72c-4be7-b5e8-6725c6f0dfcc","added_by":"auto","created_at":"2025-09-17 11:09:05","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":536602,"visible":true,"origin":"","legend":"\u003cp\u003eComplement component deposition on the surface of different cell lines after application of the monoclonal AQP4 antibody E5415A (10 µg/mL) and active or heat-inactivated human complement. (\u003cstrong\u003eA\u003c/strong\u003e) Terminal complement complex (TCC) formation on U-87MG-AQP4-ECFP cells. After the treatments, cells were washed with PBS with 10% heat-inactivated FCS, and mouse anti-human C5b-C9 neo (aE11) Alexa Fluor 594 (Novus Biologicals) was applied with 5 µg/mL to visualize TCC deposition on the cell surface. (\u003cstrong\u003eB-D\u003c/strong\u003e) Complement component C3/C3b/iC3b deposition. After the treatments, cells were washed with PBS with 10% heat-inactivated FCS, and mouse anti-human/mouse C3/C3b/iC3b purified (6C9) (Cedarlane) was applied with 5 µg/mL, followed by 2 µg/mL goat-anti mouse IgG1 Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 594 (Invitrogen™, Thermo Fisher Scientific) to visualize opsonization by C3/C3b/iC3b on the cell surface of (\u003cstrong\u003eB\u003c/strong\u003e) U-87MG-AQP4-ECFP cells, (\u003cstrong\u003eC\u003c/strong\u003e) U-87MG-ECFP cells and (\u003cstrong\u003eD\u003c/strong\u003e) primary human astrocytes. Scale bar = 20 µm. AF = Alexa Fluor, AQP4 = aquaporin-4, DAPI = 4′,6-diamidino-2-phenylindole, ECFP = enhanced cyan fluorescent protein.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/3adc457dd2f2dcc56f2a7a14.jpeg"},{"id":91526436,"identity":"74e5b717-91b7-49d8-af9c-4b804580a6ea","added_by":"auto","created_at":"2025-09-17 11:09:06","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":486458,"visible":true,"origin":"","legend":"\u003cp\u003eRNA-seq analysis reveals that treatment with the monoclonal AQP4 antibody E5145A and human complement induces the expression of inflammatory and reactive genes in U-87MG cells expressing AQP4.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Volcano plot of differentially expressed genes between U-87MG-AQP4-ECFP cells treated with E5145+AC compared to E5145+IC; green indicates upregulated genes and red indicates downregulated genes. Groups (n=3) were compared using DSeq2 analysis, and changes with an adjusted p-value \u0026lt;0.001 and a two-fold change were considered statistically significant. (\u003cstrong\u003eB\u003c/strong\u003e) Pie chart of the results shown in (A) summarizing the upregulated, downregulated, and unchanged genes. (\u003cstrong\u003eC\u003c/strong\u003e) Pairwise comparisons using DSeq2 analysis of E5145+AC-treated U-87MG-AQP4-ECFP cells with U-87MG-AQP4-ECFP and U-87MG-ECFP cells after other treatments indicated 59 specifically changed genes. (\u003cstrong\u003eD\u003c/strong\u003e) A heat map (created with GraphPad Prism 10.4.) showing the mean standardized expression (z-scores, according to decreasing principal component 1 scores) of these 59 differential genes. Each row shows the relative expression level for a single sample, and each column shows the expression level of single genes. (\u003cstrong\u003eE\u003c/strong\u003e) A bubble plot (created with GraphPad Prism 10.4) showing the functional annotation by DAVID of the 59 specifically changed genes showing the most significant associations indicated three functional clusters. P-values are visualized by a heatmap and different sizes. AC = active complement, AQP4 = aquaporin-4, ECFP = enhanced cyan fluorescent protein, EmGFP = emerald green fluorescent protein, IC = inactive complement.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/8696593c75d196665962202b.jpeg"},{"id":91526449,"identity":"4b6d7804-4c7d-40f6-9336-7b1dad3624b9","added_by":"auto","created_at":"2025-09-17 11:09:07","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":480283,"visible":true,"origin":"","legend":"\u003cp\u003eRNA-seq analyses revealed that treatment with the monoclonal AQP4 antibody E5145A and human complement does not induce the expression of a specific response in HEK293 cells expressing AQP4 and human astrocytes.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Pairwise comparisons using DSeq2 analysis of E5145+AC-treated HEK293A-AQP4-EmGFP cells with other treatments indicated only 6 specifically changed genes. (\u003cstrong\u003eB\u003c/strong\u003e) A heat map (created with GraphPad Prism 10.4.) showing the mean standardized expression (z-scores, according to decreasing principal component 1 scores) of the 59 differential genes from U-87MG-AQP4-ECFP cells in HEK293A-AQP4-EmGFP cells. (\u003cstrong\u003eC\u003c/strong\u003e) From the 59 differential genes from U-87MG-AQP4-ECFP cells only 4 genes were also differentially expressed in E5145+AC-treated HEK293A-AQP4-EmGFP cells. (\u003cstrong\u003eD\u003c/strong\u003e) Pairwise comparisons using DSeq2 analysis of E5145+AC-treated HA with other treatments indicated no specifically changed genes. AC = active complement, AQP4 = aquaporin-4, ECFP = enhanced cyan fluorescent protein, EmGFP = emerald green fluorescent protein, HA = human astrocytes, IC = inactive complement.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/5f8211683634dcd06c131285.jpeg"},{"id":91526435,"identity":"9d83305a-e3cf-45e2-b9f4-29edc5253914","added_by":"auto","created_at":"2025-09-17 11:09:06","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":564945,"visible":true,"origin":"","legend":"\u003cp\u003eRNA-seq validation by qRT-PCR in cell lines treated with AQP4 antibody E5415A and complement.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Volcano plot of differentially regulated genes identified by spatial transcriptomics in a medulla oblongata blood perivascular lesion from rats treated with E5415A (rat E5415A MO). The area was compared to a control area using DSeq analysis and changes with an adjusted p-value \u0026lt;0.05 and a two-fold change were considered as statistically significant. (\u003cstrong\u003eB\u003c/strong\u003e) Pie chart showing the overlap of significantly altered genes between U87MG-AQP4-ECFP and the rat E5415A MO. (\u003cstrong\u003eC\u003c/strong\u003e) A heat map showing the standardized expression analyzed by RNA-Seq (log2-fold change to E5415A+AC) of the 14 differentially expressed genes overlapping between U87MG-AQP4-ECFP and the rat E5415A MO lesion. Additionally, AQP4 was selected as an internal control. Each row shows the mean expression level for a single gene, and each column shows the mean expression level of 3 replicates respectively. (\u003cstrong\u003eD\u003c/strong\u003e) A heat map showing the standardized expression (log2-fold change to E5415A+AC) analyzed by RT-qPCR as validation of the 15 genes from RNA-seq. Gene expression was first normalized to the housekeeping gene GAPDH (ΔCt), followed by normalization to E5415A+AC treatment (ΔΔCt). The data represent six biological replicates from two independent experiments. AC = active complement, AQP4 = aquaporin-4, ECFP = enhanced cyan fluorescent protein, EmGFP = emerald green fluorescent protein, HA = human astrocytes, IC = inactive complement, MO = medulla oblongata.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/a18f4f87ccdaa9a078457ac9.jpeg"},{"id":91526456,"identity":"4d4987bf-8ba0-458f-9bc3-8b828fb024b0","added_by":"auto","created_at":"2025-09-17 11:09:08","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":388643,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of RNA-seq compared to in vivo transcriptomics on protein level. (\u003cstrong\u003eA\u003c/strong\u003e) U-87MG-AQP4-ECFP were treated with the monoclonal AQP4 antibody E5415A (10 µg/mL) in combination with active or heat-inactivated human complement. Thereafter, immunocytochemistry was performed for NF-κB component RelA (p65). As positive control served cells that had been treated with TNF-α (10ng/mL). Cells were fixed, permeabilized, and blocked. Then, 2 µg/mL rabbit anti-p65 (D14E12) (cell signaling) served as primary antibody, followed by goat-anti rabbit AlexaFluor647 (Thermo Fisher). The white arrows in the composite images indicate translocation of p65 into the nucleus. Scale bar = 20 µm. (\u003cstrong\u003eB\u003c/strong\u003e) U-87MG-AQP4-ECFP, HA, and HEK293-AQP4-EmGFP cells were treated as in (A) and the supernatant was assessed for IL6 levels by human IL6 ELISA (R\u0026amp;D Systems). The bar graph (means, standard deviations and individual values) was created with GraphPad Prism 10.4. Groups (n=3) were compared 2-way ANOVA with Dunnett’s multiple comparisons test. *** p\u0026lt;0.001. AF = Alexa Fluor, DAPI = 4′,6-diamidino-2-phenylindole, ECFP = enhanced cyan fluorescent protein, EmGFP = emerald green fluorescent protein, HA = human astrocytes, NF-κB = nuclear factor kappa B, IC = inactive complement.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/5894eaf06a74fe255470313f.jpeg"},{"id":91526430,"identity":"7908948f-ea6b-42ad-b3ec-0faea1f1b2e3","added_by":"auto","created_at":"2025-09-17 11:09:05","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":546395,"visible":true,"origin":"","legend":"\u003cp\u003eEffects of treatment with AQP4-IgG positive and negative NMOSD serum samples on U-87MG-AQP4-ECFP. U-87MG-AQP4-ECFP cells were treated with human NMOSD patient sera #1-6 in combination and active or heat-inactivated human complement. After the treatment, cells were harvested for RNA isolation (\u003cstrong\u003eA\u003c/strong\u003e) and the cell supernatant was used for cytotoxicity (\u003cstrong\u003eB\u003c/strong\u003e) and IL6 (\u003cstrong\u003eC\u003c/strong\u003e) assessment. (\u003cstrong\u003eA\u003c/strong\u003e) Log2-fold changes of 2\u003csup\u003e-ΔΔCt\u003c/sup\u003e values of human sera plus complement (AC/IC). ΔCt values were calculated with the Ct values of the respective genes minus the Ct values of housekeeping gene GAPDH. To assess serum-specific effects (ΔΔCt), ΔCt values of complement only samples were subtracted from those with serum plus complement. (\u003cstrong\u003eB\u003c/strong\u003e) Cytotoxicity was assessed with the CytoTox 96® Non-Radioactive Cytotoxicity Assay (Promega). Afterwards, each mean value of active or inactive complement only was subtracted from the respective values with serum. These values were normalized to those of lysis buffer-treated cells (set to 100%), to obtain serum-specific impact on cell lysis. (\u003cstrong\u003eC\u003c/strong\u003e) IL6 levels of the cell supernatants after treatment were measured by human IL6 ELISA (R\u0026amp;D Systems). Afterwards, each value median of active or inactive complement only was subtracted from the respective values with serum. Negative values were set to zero. (\u003cstrong\u003eA\u003c/strong\u003e) Heat map was created with GraphPad Prism 10.4. (\u003cstrong\u003eB\u003c/strong\u003e, \u003cstrong\u003eC\u003c/strong\u003e) Bar charts (means, standard deviation and individual values) were created with GraphPad Prism 10.4. Groups (n=3) were compared by 2-way ANOVA with Šídák’s multiple comparisons test * p\u0026lt;0.05, ** p\u0026lt;0.01, *** p\u0026lt;0.001. AC = active complement, AQP4 = aquaporin-4, ECFP = enhanced cyan fluorescent protein, IC = inactive complement.\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/35c86b97f735c234122dc761.jpeg"},{"id":91527331,"identity":"9c0ca866-e4bc-4331-8b76-1370f6a5a085","added_by":"auto","created_at":"2025-09-17 11:17:06","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1121890,"visible":true,"origin":"","legend":"\u003cp\u003eImmunocytochemistry of U-87MG-AQP4-ECFP cells after treatment with NMOSD patient sera and human complement.\u003cstrong\u003e \u003c/strong\u003eU-87MG-AQP4-ECFP were treated with 10% NMOSD patient sera (with or without AQP4-IgG) in combination with active or heat-inactivated human complement. Then, cells were washed with PBS with 10% heat-inactivated fetal calf serum and immunocytochemistry was performed. (\u003cstrong\u003eA\u003c/strong\u003e) Terminal complement complex (TCC) staining. The cellular surface was stained for TCC deposition by applying mouse anti-C5b-C9 neo (aE11) Alexa Fluor 594 (Novus Biologicals) with 5 µg/mL. (\u003cstrong\u003eB\u003c/strong\u003e) NF-κB component p65 staining. Translocation of p65 from the cytosol into the nucleus was visualized by intracellular immunocytochemistry. Cells were fixed, permeabilized, and blocked. Then, 2 µg/mL rabbit anti-human p65 (D14E12) (Cell Signaling) served as the primary antibody, followed by goat anti-rabbit AlexaFluor647 (Thermo Fisher). White arrows indicate p65 translocation into the nucleus. Scale bar = 20 µm. AF = Alexa Fluor, DAPI = 4′,6-diamidino-2-phenylindole, TCC = terminal complement complex.\u003c/p\u003e","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/af3e3b73193d49423aaca7c9.jpeg"},{"id":91526433,"identity":"97f91847-b34d-4d4f-b096-c14f6bf9e749","added_by":"auto","created_at":"2025-09-17 11:09:05","extension":"jpeg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":329569,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic representation of differently expressed genes after AQP4-IgG and active human complement treatment. Differently expressed genes observed in this study are highlighted in red. Grey elements were not investigated in the study. Binding of AQP4-IgG to AQP4 on U-87MG-AQP4-ECFP, primary human astrocytes, or on rat astrocytes enables complement C1q initiation of the classical complement cascade, resulting in the terminal complement complex (TCC). Moreover, the alternative complement pathway and cellular stress are induced. Differently expressed genes are primarily involved in interleukin (IL)-6 and nuclear factor kappa B (NF-κB) activation. AQP4 = aquaporin-4, ATF-3 = activating transcription factor-3, CEBPB = CCAAT/enhancer-binding protein beta, ERK = extracellular-signal regulated kinase, IRAK2 = interleukin-1 receptor-associated kinase 2, IRF1 = interferon regulatory factor-1, Jak = Janus kinase, JNK = c-jun N-terminal kinases, JUNB = transcription factor jun-B, MAPK = mitogen-activated protein kinase, NFKB2 = NF-κB subunit 2, NFKBIA = NF-κB inhibitor alpha, NFKBIZ = NF-κB inhibitor zeta, NR4A2 = nuclear receptor 4A2, P = phosphorylation, PI3K = phosphoinositide 3-kinase, STAT3 = signal transducers and activators of transcription-3, Th = T helper, TLR = toll like receptor, TNFR = tumor necrosis factor receptor, PTX-3 = pentraxin-3, Ub = ubiquitin. Created in BioRender. Brandl, S. (2025) https://BioRender.com/1x2unnn.\u003c/p\u003e","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/75f592a3be595a58d1a08a35.jpeg"},{"id":98244079,"identity":"f4a192e3-e426-4bbe-ba3c-1463a53019c4","added_by":"auto","created_at":"2025-12-15 16:12:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6089565,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/db2cd7a5-f5af-4af9-9a76-462e4509e859.pdf"},{"id":91526445,"identity":"b3cc8760-55f8-48d7-9731-526a3c3f31c9","added_by":"auto","created_at":"2025-09-17 11:09:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":19641869,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryfile.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7064018/v1/347795ff1bbc5b7800b4724b.pdf"}],"financialInterests":"Competing interest reported. The authors declare that this study received funding from Roche Austria GmbH (to MR). The funder was not involved in the study design, analysis, or critical revision of the article for important intellectual content. RH reports speaker honoraria from UCB and BMS. The Medical Universities of Innsbruck (Austria; employer of MR) and Vienna (Austria; employer of RH) receive payments for antibody assays and for antibody validation experiments organized by Euroimmun (Lübeck, Germany). QY, JH, VE and MB declare no competing interests.","formattedTitle":"mRNA Profiling of Inflammatory Stress Responses after Aquaporin-4 Antibody and Human Complement Treatment Reveals Upregulation of NF-κB and IL6 Pathways","fulltext":[{"header":"Background","content":"\u003cp\u003eNeuromyelitis optica spectrum disorder (NMOSD) is a rare neuroinflammatory disease that primarily affects the spinal cord and optic nerves \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. The majority of patients are seropositive for autoantibodies targeting the water channel aquaporin-4 (AQP4) on astrocyte endfeet \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. AQP4 has two major isoforms produced by alternative splicing, M1 and M23. At the astrocytic foot processes, the isoform AQP4M23 is more abundant than AQP4M1 and is more prone to form supramolecular aggregates called orthogonal arrays of particles, which are targeted by anti-AQP4 immunoglobulin G (AQP4-IgG) autoantibodies \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Upon binding, immune responses are predominantly mediated by complement-dependent cytotoxicity (CDC) through the classical pathway, resulting in astrocyte damage and blood-brain barrier (BBB) disruption \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Several studies have shown that AQP4-IgG titers in patients correlate with the intensity of complement activation, but not with disease severity \u003csup\u003e\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Antibody-dependent cellular cytotoxicity (ADCC) has also been implicated, as it activates various leukocytes or natural killer cells via Fcγ receptors, damaging astrocytes and adjacent non-AQP4 expressing cells \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe most important diagnostic tool to identify AQP4-IgG serostatus of NMOSD patients is a cell-based assay using human embryonic kidney 293 (HEK293) cells with a transiently transfected AQP4 construct and fluorescent label \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Beyond diagnostics, these cells are also used to study NMOSD pathophysiological mechanisms at the cellular level. To date, pathophysiological mechanisms mediated by AQP4-IgG have been studied \u003cem\u003ein vitro\u003c/em\u003e primarily on AQP4-overexpressing HEK293A cells \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan additionalcitationids=\"CR12 CR13\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e, Chinese hamster ovary (CHO) cells \u003csup\u003e\u003cspan additionalcitationids=\"CR16 CR17 CR18 CR19 CR20\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e or primary astrocytes with human \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan additionalcitationids=\"CR23 CR24 CR25\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e or rodent origin \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan additionalcitationids=\"CR28 CR29 CR30 CR31\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eRecently, we have established a cellular model for CDC activation after AQP4-IgG exposure in HEK293 expressing AQP4 and found a strong AQP4-IgG titer dependent activation of CDC \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Here, we further investigate our findings by molecular profiling of cellular models of CDC activation stably transduced U-87 MG astrocytoma cells with AQP4M23-ECFP overexpression (U-87MG-AQP4-ECFP) and a non-AQP4 overexpressing counterpart (U-87MG-ECFP) AQP4M23-EmGFP in addition to transfected HEK293 cells (HEK293-AQP4-EmGFP) and human primary astrocytes (HA). We exposed these cell lines to human complement and the monoclonal mouse anti-AQP4 antibody E5415A, analyzing cytotoxicity and transcriptomic changes post-treatment \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Furthermore, we compared our transcriptomic findings to data from spatial transcriptomics from an NMOSD \u003cem\u003ein vivo\u003c/em\u003e model to identify and validate overlapping differentially expressed genes.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eCells\u003c/h2\u003e\u003cp\u003eHuman primary astrocytes (HA), isolated from the human cerebral cortex of a female donor, gestational age 22 weeks, were purchased by ScienCell (Lot. No. 33619, ScienCell Research Laboratories, San Diego, CA). For experiments, exclusively passages 3 to 5 were used. HA were cultured in 2 \u0026micro;g/cm\u003csup\u003e2\u003c/sup\u003e poly-L-lysine coated flasks (Sciencell Research Laboratories, San Diego, CA) in astrocyte medium AM (Innoprot, Derio, Spain), containing 2% fetal calf serum, 1% P/S, and 1% astrocyte growth supplement. Medium was exchanged every other day and cells were passaged when 90% dense.\u003c/p\u003e\u003cp\u003eThe glioblastoma cell line U-87 MG (ATCC; LGC Standards GmbH, Wesel, Germany) was genetically modified for a stable protein overexpression. Thereby, an overexpression of enhanced cyan fluorescent protein (ECFP) with (U-87MG-AQP4-ECFP) or without (U-87MG-ECFP) AQP4M23 in U-87 MG cells was obtained via viral infection as described elsewhere \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. The vectors were pLIB-MCS-ECFP-iresNeo and pLIB-MCS-AQP4m23-ECFP-iresNeo. The plasmid maps are shown in Supplementary Fig.\u0026nbsp;1. Cells were selected using 10 \u0026micro;g/mL neomycin for 72 hours. HEK293A and U-87 MG cell lines were cultured in DMEM high glucose, 10% fetal calf serum, and 1% non-essential amino acids and passaged twice per week.\u003c/p\u003e\u003cp\u003eThree days before complement and antibody treatment HEK293A (ATCC; LGC Standards GmbH, Wesel, Germany) cells were seeded. After 24 hours, they were transiently transfected with a pcDNA 6.2 AQP4M23-EmGFP (HEK293-AQP4-EmGFP) plasmid with polyethylenimine (Sigma, St. Louis, Missouri) in a ratio of 1:3.6 plasmid:transfection reagent. Then, cells were incubated for two days until treatment.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePatient Serum Samples and Ethical Approval\u003c/h3\u003e\n\u003cp\u003e As a proof of concept, serum samples from three NMOSD patients with (titers 1:20,480, 1:1,280, and 1:5,120) and from three NMOSD patients without AQP4-IgG positivity were provided by Romana H\u0026ouml;ftberger from the Division of Neuropathology and Neurochemistry, Department of Neurology, Medical University of Vienna, Austria. The use of these serum samples from a biobank was approved by the ethical committee of the Medical University of Vienna (EK 1636/2019 and 1123/2015). All patients or their legal representatives gave written informed consent to participate in the study and all methods were performed in accordance with the relevant guidelines and regulation.\u003c/p\u003e\u003cp\u003eAll samples were tested for complement activation and cytotoxicity on U-87MG-AQP4-ECFP cells. The test was performed blinded. The complement treatment and following procedures were performed the same way as with E5415A, but instead of 10 \u0026micro;g/mL of the monoclonal antibody, 10% of end volume of serum was applied. Additionally, the sera were heat-inactivated for 30 minutes (min) at 56\u0026deg;C to inactivate complement before usage.\u003c/p\u003e\n\u003ch3\u003eAQP4 antibody and complement treatment and assessment of cytotoxicity\u003c/h3\u003e\n\u003cp\u003eTreatment with the mouse monoclonal AQP4 antibody E5415A (isolated from the hybridoma cell line AQP4 E5415A-1H6-68, Resource no. RCB4883, provided by the Riken BRC through the National BioResource Project of the MEXT/AMED, Japan) \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e and human complement was done as recently described \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Briefly, U-87MG-ECFP, U-87MG-AQP4-ECFP, HEK293-AQP4 -EmGFP cells or HA were grown in a 12- or 96-well plate until confluent. Then, they were treated with 40% active or heat-inactivated (45 min at 56\u0026deg;C) pooled human complement serum (Cedarlane, Burlington, Ontario, Canada) in X-VIVO 15 medium (Lonza, Basel, Switzerland) and with or without E5415A (in-house production from E5415A-1H6-68 hybridoma cells, #RCB4883, Riken, Tsukubashi, Ibaraki, Japan) (10 \u0026micro;g/mL), followed by incubation for 90 min at 37\u0026deg;C \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Afterwards, the cell supernatant was taken to assess cytotoxicity, and cells were harvested for RNA isolation, or immunofluorescence staining was performed.\u003c/p\u003e\u003cp\u003eThe CDC was assessed with a lactate dehydrogenase (LDH) assay (CytoTox 96\u0026reg; Nonradioactive Cytotoxicity Assay, Promega, Madison, WI). Thereby, of the amount of cytosolic LDH, which is released after cell damage, can be measured indirectly via an enzymatic assay, resulting in a reduction of tetrazolium salt into a red formazan. Cells were treated as stated above. As a positive control, additional wells with cells were treated with 1x lysis buffer in X-VIVO 15 medium for the last 45 min of the incubation. As a negative control served untreated cells in X-VIVO 15 medium. Cell supernatant was taken and incubated with the same volume of substrate for 30 min at room temperature, according to the manufacturer. The reaction was stopped with the provided stop solution, and the absorption was measured with a plate reader at 492 nm (DTX880; Beckman Coulter, Brea, CA). The results were first normalized to the respective IC treated samples to correct for the LDH background reactivity of human complement serum and then to the respective lysis buffer-treated samples to assess the percentage of cell lysis. Per treatment, three technical replicates were done for each of the three biological replicates.\u003c/p\u003e\n\u003ch3\u003emRNA transcriptomic analysis\u003c/h3\u003e\n\u003cp\u003eCells were treated with complement and E5415A or human serum as described above. After removing the supernatant to perform cytotoxicity assays, cells were washed with PBS and harvested with RLT buffer. Further steps were performed according to the protocol of the RNeasy Mini Kit (QIAGEN, Hilden, Germany). After quantity and quality determination by a NanoDrop\u0026trade; System (Thermo Fisher Scientific, Waltham, MA, USA), total RNA samples were sent for sequencing by the Illumina PE150 technology to Novogene (Martinsried, Germany).\u003c/p\u003e\u003cp\u003eMessenger RNA was purified from total RNA using poly-T oligo-attached magnetic beads. After fragmentation, the first strand cDNA was synthesized using random hexamer primers, followed by the second strand cDNA synthesis using either dUTP for directional library or dTTP for non-directional library \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. The non-directional library was prepared by end repair, A-tailing, adapter ligation, size selection, amplification, and purification. For the directional library, it was ready after end repair, A-tailing, adapter ligation, size selection, USER enzyme digestion, amplification, and purification. The library was checked with Qubit and real-time PCR for quantification and bioanalyzer for size distribution detection. Quantified libraries were pooled and sequenced on Illumina platforms, according to effective library concentration and data amount.\u003c/p\u003e\u003cp\u003eThe raw FASTQ format data was initially processed using fastp software by Novogene. In this step, clean data (clean reads) were obtained by removing reads containing adapter, reads containing ploy-N and low-quality reads from raw data. At the same time, Q20, Q30, and GC content of the clean data were calculated. All the downstream analyses were based on clean data with high quality.\u003c/p\u003e\u003cp\u003eReference genome and gene model annotation files were downloaded from the genome website directly. Index of the reference genome was built using Hisat2 v2.0.5 and paired-end clean 1 reads were aligned to the reference genome using Hisat2 v2.0.5. Hisat2 was used by Novogene as the mapping tool for that Hisat2 can generate a database of splice junctions based on the gene model annotation file and thus a better mapping result than other non-splice mapping tools \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe program featureCounts v1.5.0-p3 was used to count the reads numbers mapped to each gene \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Then, fragments per kilobase million (FPKM) of each gene were calculated based on the length of the gene and reads count mapped to this gene. FPKM, expected number of Fragments Per Kilobase of transcript sequence per Million base pairs sequenced, considers the effect of sequencing depth and gene length for the reads count at the same time, and is currently the most commonly used method for estimating gene expression levels.\u003c/p\u003e\u003cp\u003eThe acquired data were deposited in the Gene Expression Omnibus database under dataset accession number GSE291954.\u003c/p\u003e\n\u003ch3\u003eReverse transcription-quantitative polymerase chain reaction\u003c/h3\u003e\n\u003cp\u003eSelected differentially regulated genes identified by transcriptomic analysis were validated by reverse transcription-quantitative polymerase chain reaction (RT-qPCR). First, 1 \u0026micro;g of RNA was retrotranscribed into cDNA with the High Capacity cDNA Reverse Transcription Kit (Applied Biosciences, Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer. The product was diluted with 480 \u0026micro;l of DEPC water and stored at -20\u0026deg;C until needed. Then, TaqMan Gene Expression Assay probes (Applied Biosciences) and iTaq\u0026trade; Universal Probes Supermix (Bio-Rad, Hercules, CA, USA) were mixed with 5 \u0026micro;l of cDNA each and run with a CFX96 RT-PCR machine (Bio-Rad) with CFX maestro software for 40 cycles (initial denaturation at 95\u0026deg;C for 30 seconds; second denaturation at 95\u0026deg;C for 5 seconds, annealing/extension at 60\u0026deg;C for 30 seconds). GAPDH served as a housekeeping gene, and delta-Ct values were calculated for each gene. All TaqMan Gene Expression Assay probes are listed in Supplementary Table\u0026nbsp;1.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eImmunocytochemistry\u003c/h2\u003e\u003cp\u003eFor immunocytochemistry (ICC), cells were grown on 0.1% gelatin solution (Sigma-Aldrich, St. Louis, MO, USA) coated ibidi \u0026micro;-Slide 18 Well Glass Bottom (ibidi, ibidi GmbH, Gr\u0026auml;felfing, Germany). When reaching confluence, cells were exposed to human complement with or without E5415A or human serum, as described above. Live cell staining was performed on cells for terminal complement complex (TCC), as well as C3/C3b/iC3b deposition on the cell surface. Therefore, 10% heat-inactivated fetal calf serum in PBS served as a washing and antibody buffer. Mouse anti-C5b-C9 neo (aE11) AlexaFluor594 (Novus Biologicals, Centennial, CO, USA) or mouse IgG1 anti-C3/C3b/iC3b (Cedarlane, Burlington, Ontario, Canada) was applied at 5 \u0026micro;g/mL for 1 hour at 4\u0026deg;C. As secondary antibody for C3/C3b/iC3b served goat-anti mouse IgG1 AlexaFluor594 (Invitrogen\u0026trade;, Thermo Fisher Scientific, Waltham, MA, USA), which was applied for 30 min at room temperature in the dark. Thereafter, cells were fixed with 4% paraformaldehyde for 10 min. After washing, nucleus staining with 4\u0026prime;,6-diamidino-2-phenylindole (DAPI) and mounting was performed in one step by applying Immunoselect Antifading Mounting Medium DAPI (Dianova, BIOZOL, Hamburg, Germany) to the cells. For intracellular ICC, cells were fixed with 4% paraformaldehyde and permeabilized with 1% Triton X-100 for 10 min each at room temperature before blocking in 1% bovine serum albumin and 5% normal goat serum in PBS. Then, 2 \u0026micro;g/mL rabbit monoclonal anti-p65 (D14E12; Cell Signaling) was added to the antibody buffer (1% bovine serum albumin and 1% normal goat serum in PBS). Cells were incubated at 4\u0026deg;C overnight. Goat-anti rabbit AlexaFluor647 (Invitrogen\u0026trade;, Thermo Fisher Scientific, Waltham, MA, USA) in antibody buffer was incubated on the cells for 1 hour at room temperature in the dark. Cells were mounted the same way as for the extracellular ICC. Microscopy was performed with the Zeiss LSM700 microscope (Zeiss Plan-Apochromat 63x/1.40 oil M27 objective). Fiji ImageJ (version 1.64) was used for background subtraction and adjustments for contrast and brightness.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eC3a enzyme-linked immunosorbent assay\u003c/h3\u003e\n\u003cp\u003eC3a levels in different cell lines were evaluated by enzyme-linked immunosorbent assay (ELISA) with the Human C3a ELISA Kit (Invitrogen\u0026trade;, Thermo Fisher Scientific, Waltham, MA, USA). Supernatants of AQP4 antibody and complement treatment experiments were collected and stored at -20\u0026deg;C. Before usage, the cell supernatants were thawed and centrifuged briefly (10 min at 2000 g). The ELISA was performed according to the manufacturer\u0026rsquo;s instructions with a sample dilution of 1:2000 and a 25 min incubation time of the Substrate solution before adding Stop Solution. The plate was measured in single values at 450 nm and 620 nm with a plate reader (DTX880; Beckman Coulter, Brea, CA).\u003c/p\u003e\n\u003ch3\u003eHuman interleukin-6 enzyme-linked immunosorbent assay\u003c/h3\u003e\n\u003cp\u003eTo assess interleukin-6 (IL6) levels in cell supernatants after complement and E5415A exposition, an ELISA was used. The Human IL6 DuoSet ELISA and DuoSet Ancillary Reagent Kit 2 (R\u0026amp;D Systems, Minneapolis, MN, USA) were performed according to the manufacturer. Dilutions of 1:4 in reagent diluent were used for supernatants of U-87MG-AQP4-ECFP and HEK293-AQP4-EmGFP cells, and 1:2 dilutions for supernatants of HA and U-87MG-ECFP cells. The optical density of the samples was measured in technical duplicates at 450 nm.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eNMOSD animal model and spatial transcriptomics\u003c/h2\u003e\u003cp\u003eSpatial transcriptomics data deposited at the Sequence Read Archive (SRA) database (NCBI/SRA accession numbers PRJNA1258753, PRJNA1262739, PRJNA1262139, PRJNA1262895, PRJNA1262368 and PRJNA1263155) were re-analyzed from the inflamed medulla oblongata of a female 7-weeks old Lewis rat that received daily intraperitoneal injections of 1 mg E5415A in PBS for two consecutive days, inducing experimental NMOSD. The animal was euthanized by CO₂ inhalation 24 hours after the final antibody injection, and the medulla oblongata was dissected for Visium spatial transcriptomics according to protocols from 10x Genomics (10x Genomics B.V, Leiden, The Netherlands). Data were processed using Space Ranger (10x Genomics) to generate gene expression profiles per gene and spot, overlaid on HE-stained tissue images. Loupe Browser 6.5.0 (10x Genomics) was used to define a perivascular lesion area and identify differentially upregulated genes compared to the remaining medulla oblongata of the same tissue section through the \u0026ldquo;categories,\u0026rdquo; \u0026ldquo;globally distinguishing,\u0026rdquo; and \u0026ldquo;significant feature comparison\u0026rdquo; functions. We did not exclude genes with low average counts. Log2-fold changes and p-values were calculated in Space Ranger, with p-values adjusted for multiple comparisons using the Benjamini\u0026ndash;Hochberg method. The experiments were approved by the Ethic Commission of the Medical University Vienna and performed with the license of the Austrian Ministry for Science and Research (GZ: BMBWF-66.009/0107-V/3b/2018). The study is reported in accordance with ARRIVE guidelines (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://arriveguidelines.org\u003c/span\u003e\u003cspan address=\"https://arriveguidelines.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eNF-kB activation in human brain tissue\u003c/h2\u003e\u003cp\u003eNeuropathological analysis was performed on two human autopsy cases of patients with AQP4-IgG seropositive NMOSD. The use of these tissue samples from a biobank was approved by the ethical committee of the Medical University of Vienna (EK 1636/2019 and 1123/2015). All patients or their legal representatives gave written informed consent to participate in the study.\u003c/p\u003e\u003cp\u003eFormalin-fixed and paraffin-embedded tissue sections were stained for haematoxylin and eosin (H\u0026amp;E) and Luxol fast blue-Periodic acid Schiff (LFB/PAS). Immunohistochemistry was performed on the automated platform Autostainer Link 48 using EnVision\u0026trade; FLEX\u0026thinsp;+\u0026thinsp;secondary system (Dako/Agilent) according to the manufacturer\u0026acute;s protocol using the following primary antibodies: rabbit polyclonal anti-GFAP (1:3,000; Dako/Agilent) or mouse monoclonal anti-GFAP (1:800; Millipore), rabbit monoclonal anti-p65 (1:800; D14E12; Cell Signaling), and mouse monoclonal anti-SMI31 (phosphorylated neurofilament; 1:25,000; Sternberger). Manual immunohistochemistry was performed in a humidified chamber for complement C9 neoantigen (C9neo; complement-mediated tissue injury; rabbit polyclonal; 1:2000, from Professor Paul Morgan, Cardiff, UK). Double labelling of GFAP and NF-κB (p65) was performed with the chromogenic reactions with Fast Blue (blue) and 3-amino-9-ethylcarbazole (AEC; BioSB (red)). Image acquisition was performed using a NanoZoomer 2.0-HT digital slide scanner C9600 (Hamamatsu Photonics, Hamamatsu, Japan).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eBioinformatic and statistical analysis\u003c/h2\u003e\u003cp\u003eStatistical analyses (2-way ANOVA with Š\u0026iacute;d\u0026aacute;k's multiple comparisons test, principal component analysis) and drawing of figures (bar and scatter graphs, volcano plots, principal component plots, and heat maps) were prepared using GraphPad Prism 10.4 (GraphPad Software Inc., La Jolla, California, United States). All tests were done using a significance threshold of p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 with correction for multiple comparisons if applicable. Data are expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation. For all experiments, at least three biological replicates were analyzed.\u003c/p\u003e\u003cp\u003eDifferential expression analysis of two conditions/groups (three biological replicates per condition) was performed using the DESeq2 R package using the WebMev platform \u003csup\u003e\u003cspan additionalcitationids=\"CR41\" citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. WebMev and DESeq2 provide statistical routines for determining differential expression in digital gene expression data using a model based on the negative binomial distribution. The resulting p-values were adjusted using Benjamini and Hochberg's approach to control the false discovery rate. Genes with an adjusted p-value\u0026thinsp;\u0026le;\u0026thinsp;0.001 and log2-fold change\u0026thinsp;\u0026ge;\u0026thinsp;1 or \u0026le; -1 found by DESeq2 were assigned as differentially expressed. The overlap of differentially expressed genes between different treatments was analyzed using Venn diagrams and the WebMev platform.\u003c/p\u003e\u003cp\u003eTo functionally annotate the most significant genes, gene ontology (GO) analysis was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) using the official gene symbols of Homo Sapiens of genes with a statistical significance of p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 (adjusted using the Benjamini-Hochberg correction for multiple tests) \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. The DAVID annotation chart for the Gene Ontology Term \u0026ldquo;biological processes\u0026rdquo; (GOTERM-BP) was used. Enrichment of genes in annotation terms was evaluated using the EASE Score, a one-tail Fisher\u0026rsquo;s exact test p-value, with p-values\u0026thinsp;\u0026le;\u0026thinsp;0.05 indicating strong enrichment. Moreover, significantly altered genes were analyzed and visualized by STRING network analysis \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eTranscriptomic changes\u003c/b\u003e \u003cb\u003ein vitro\u003c/b\u003e \u003cb\u003eafter anti-AQP4 antibody and human complement treatment indicate an inflammatory stress response in human U-87MG-AQP4-ECFP cells\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe analyzed the transcriptomic changes after treatment of four established NMOSD cellular models (U-87MG-AQP4-ECFP, U-87MG-ECFP, HEK293-AQP4-EmGFP, and HA) with active or heat-inactivated human complement and with or without the monoclonal AQP4 antibody E5415A, as previously described \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Cytotoxicity assessed by LDH release quantification was significantly increased in AQP4-expressing cells, whereas AQP4-negative U-87MG-ECFP cells showed less cell death (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). The highest level of cytotoxicity (mean 37.2%) was observed in U-87MG-AQP4-ECFP after E5415A\u0026thinsp;+\u0026thinsp;AC treatment. This effect was significantly different to all other treatments, and to E5415A\u0026thinsp;+\u0026thinsp;AC-treated U-87MG-ECFP cells (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In contrast, U-87MG-ECFP cells, which lack AQP4 expression, exhibited a lower level of cytotoxicity (9.6% E5415A\u0026thinsp;+\u0026thinsp;AC, 11.2% AC), with a statistically significant difference between E5415A\u0026thinsp;+\u0026thinsp;AC to IC\u0026thinsp;\u0026plusmn;\u0026thinsp;E5415A, and untreated cells (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo validate that the classical complement pathway mediated cytotoxicity, ICC for the terminal complement complex (TCC, C5b-C9) was performed. TCC deposition was exclusively observed in U-87MG-AQP4-ECFP cells treated with E5415A\u0026thinsp;+\u0026thinsp;AC, while no staining was detected in cells treated with E5415A\u0026thinsp;+\u0026thinsp;IC, complement-only treatment, or untreated controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The background level of CDC observed in controls (E5145A\u0026thinsp;+\u0026thinsp;IC, AC, and IC) was already seen in our previous study in HEK-AQP4-EmGFP cells and was associated with opsonization with C3b and associated effector functions \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Bound C3/C3b on AC-only-treated cells was observed for U-87MG-AQP4-ECFP cells as well, although to a smaller extent (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). In contrast, C3/C3b opsonization on U-87MG-ECFP was seen in comparable amounts in E5415A\u0026thinsp;+\u0026thinsp;AC or AC-only-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). As no TCC deposition on the cellular surface of U-87MG-ECFP cells could be observed for any of the treatments (data not shown), this demonstrates the alternative pathway contribution in the U-87MG cell line. Moreover, C3a levels in both U-87MG-AQP4-ECFP and U-87MG-ECFP cells were not significantly different between E5415A\u0026thinsp;+\u0026thinsp;AC and AC treatments, indicating an additional AQP4 antibody-independent activation of the alternative complement pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTranscriptomic analysis was performed by pairwise comparisons of E5415A\u0026thinsp;+\u0026thinsp;AC-treated U-87MG-AQP4-ECFP cells, U-87MG-AQP4-ECFP, and U-87MG-ECFP cells with the other treatments using Dseq2 analysis. The volcano plot shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and the pie chart in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB revealed that amongst all detectable transcripts, 13,224 (99.1%) genes remained unchanged. In comparison, 96 genes were upregulated (0.7%), and 20 genes were downregulated (0.2%) when comparing U-87MG-AQP4-ECFP cells treated with E5415A\u0026thinsp;+\u0026thinsp;AC and E5415A\u0026thinsp;+\u0026thinsp;IC. From Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC and D it is evident that 59 genes were significantly changed in U-87MG-AQP4-ECFP cells after E5415A\u0026thinsp;+\u0026thinsp;AC exposure compared to all other treatments. Functional annotation using DAVID revealed a strong enrichment of genes associated with inflammation, cell death and stress responses, and NF-κB signaling pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE), which is comparable to previous findings \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNext, a similar analysis of HEK293-AQP4-EmGFP cells was performed, comparing E5415A\u0026thinsp;+\u0026thinsp;AC-treated cells with the other treatments, respectively. HEK293-AQP4-EmGFP cells also showed significant cytotoxicity after E5415A\u0026thinsp;+\u0026thinsp;AC (mean 32.7%) compared to all other treatments (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Background levels of complement-derived cell lysis were, as in previous findings, likely caused by a partially activated alternative pathway. C3a levels detected in differently treated HEK293-AQP4-EmGFP cells showed an AC-specific effect (mean of 32.39 \u0026micro;g/mL), but this effect was not higher when adding E5415A (mean of 38.82 \u0026micro;g/mL) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Only 6 genes were significantly changed in E5415A\u0026thinsp;+\u0026thinsp;AC-treated HEK293-AQP4-EmGFP cells compared with the other treatments (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), thereof 4 genes overlapped with the genes changed in U-87MG-AQP4-ECFP cells after E5415A\u0026thinsp;+\u0026thinsp;AC treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). The expression of all 59 differentially regulated genes from U-87MG-AQP4-ECFP cells after E5415A\u0026thinsp;+\u0026thinsp;AC treatment in HEK293-AQP4-EmGFP cells is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFinally, we performed a similar analysis of primary HA. Similarly, HA cells displayed moderate cytotoxicity after treatment with E5145A\u0026thinsp;+\u0026thinsp;AC (mean 11.9%) and AC (12.6%), but with a large variation between replicates, and a significant difference to untreated cells (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Furthermore, HA treated with AC alone or in combination with E5415A showed similar mean C3a levels (22.44 \u0026micro;g/mL and 24.87 \u0026micro;g/mL) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). The lack of an E5145A\u0026thinsp;+\u0026thinsp;AC-specific effect could be explained by the lower level of AQP4 gene expression (mean gene counts 970.9, standard deviation 37.4) as compared to U-87MG-AQP4-ECFP (50,645.9\u0026plusmn;11,295.7) or HEK293-AQP4-EmGFP (324,388.8\u0026plusmn;108,263.6). Moreover, HA showed a higher level of background cytotoxicity, which is related to C3 cleavage and opsonization (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), and the differential expression of complement associated genes (Supplementary Fig.\u0026nbsp;2). While HA showed a higher proportion of differentially expressed genes (n\u0026thinsp;=\u0026thinsp;808) compared to untreated cells, we could not detect a specific effect of E5145A\u0026thinsp;+\u0026thinsp;AC treatments (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD and E). To summarize, despite high expression of glial fibrillary acidic protein (GFAP) and other astrocyte-specific genes, the HA used in this study were not useful as an \u003cem\u003ein vitro\u003c/em\u003e NMOSD model for our research question. The AQP4 levels were found to be insufficient, nevertheless, similar AQP4-unspecific effects could be observed, which were probably due to the activation of the alternative complement pathway and C3 and C5 receptors or other antibodies in the complement serum that bind to HA.\u003c/p\u003e\u003cp\u003e\u003cb\u003eTranscriptomic changes in U87MG-AQP4-ECFP cells with anti-AQP4 antibody and human complement treatment compared to an NMOSD\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e \u003cb\u003emodel\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo validate our findings with a NMOSD \u003cem\u003ein vivo\u003c/em\u003e model, we compared the transcriptomic changes of the treated cells with the results of a spatial transcriptomic analysis of medulla oblongata tissue from Lewis treated with AQP4 antibody E5415A (Supplementary Fig.\u0026nbsp;3A). 322 genes were significantly upregulated in the inflammatory experimental NMOSD lesion (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). As can be seen in a Venn diagram, 15 of these genes were shared with the 59 differentially expressed genes from U-87MG-AQP4-ECFP E5415A\u0026thinsp;+\u0026thinsp;AC-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB; ATF3, CEBPD, CXCL1, CXCL2, GEM, IL11, IL6, IRAK2, IRF1, JUNB, NFKB2, NFKBIA, NFKBIZ, NR4A2, and PTX3). Hence, this gene set represents genes that are also upregulated in the \u003cem\u003ein vivo\u003c/em\u003e AQP4-IgG positive NMOSD situation. These genes were associated with inflammation (CXCL1, CXCL2, IL11, IL6, IRAK2, NFKB2, NFKBIZ, PTX3), cell death and stress responses (ATF3, CEBPD, CXCL1, CXCL2, IL6, IRAK2, IRF1, JUNB, NFKB2, NFKBIA, NFKBIZ, NR4A2, PTX3), and NF-κB signaling pathways (IRAK2, NFKB2, NFKBIA, NFKBIZ) identified by functional annotation with DAVID (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). The summative expression of these genes in tissue from rats treated with E5415A for two days, one day and a untreated control is shown in Supplementary Fig.\u0026nbsp;3A-C, and the expression of the individual genes is shown in Supplementary Fig.\u0026nbsp;3D STRING pathway analysis (Supplementary Fig.\u0026nbsp;4) revealed potential connections among these genes, except GEM (a guanidine triphosphate-binding protein that might be participating in receptor-mediated signal transduction at the plasma membrane), which showed no associations with other genes and no association with the GO terms mentioned above and was therefore excluded from further analysis \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe transcriptomic analysis comparing the expression changes of these genes in treated cells compared with the rat E5415A MO lesion is visualized in the heat map in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC. As an internal control, we added AQP4 to our gene list as the target protein of AQP4-IgG. The difference in expression is expressed with a log2-fold change of E5145A\u0026thinsp;+\u0026thinsp;AC-treated cells compared to the other treatments. E5145A\u0026thinsp;+\u0026thinsp;AC-treated U-87MG-AQP4-ECFP cells showed an upregulation in selected genes compared with all other treatments, which was comparable to the changes seen in the rat E5415A MO lesion. In HA, these gene expression changes were further accentuated, however, they were only upregulated after E5145A\u0026thinsp;+\u0026thinsp;AC compared to untreated cells. This pattern was not observed in HEK293-AQP4-EmGFP cells treated with E5415A\u0026thinsp;+\u0026thinsp;AC compared to the other treatments.\u003c/p\u003e\u003cp\u003eTo validate these findings, we performed RT-qPCR on cDNA synthesized from RNA used for transcriptomic analysis of all four cell lines. The results were visualized in a heat map showing the standardized expression (log2-fold change of E5415A\u0026thinsp;+\u0026thinsp;AC-treated cells compared to the other treatments, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Overall, gene expression changes were comparable to the transcriptomic analysis. For better visualization of total gene expression levels, results are additionally shown as ΔCt values (Supplementary Fig.\u0026nbsp;5). In U-87MG-AQP4-ECFP cells, highly significant gene expression differences after E5415A\u0026thinsp;+\u0026thinsp;AC treatment compared to other treatments were observed. Expression levels of AQP4 were significantly downregulated, NFKB2 was only upregulated compared to IC or untreated cells, and all other tested genes were significantly upregulated compared to all other treatments. In contrast, HA and U87MG-ECFP cells showed less significant differences in expression between E5415A\u0026thinsp;+\u0026thinsp;AC and the different treatments, only compared to untreated cells, there was a significant upregulation of most genes. In HEK293-AQP4-EmGFP cells, NR4A2 and CXCL1 were upregulated significantly after E5415A\u0026thinsp;+\u0026thinsp;AC treatment compared to all other treatments, whereas some of the other genes were only upregulated compared to untreated cells. No significant changes in any treatment of HEK293-AQP4-EmGFP cells were observed in AQP4, CEBPB, IL6, IL11, IRAK2, and NFKB2.\u003c/p\u003e\u003cp\u003e\u003cb\u003eValidation of transcriptomic changes of cells with anti-AQP4 antibody and human complement treatment compared to an NMOSD\u003c/b\u003e \u003cb\u003ein vivo\u003c/b\u003e \u003cb\u003emodel on protein level\u003c/b\u003e\u003c/p\u003e\u003cp\u003eIn general, most of the differentially expressed genes were involved in NF-κB signaling, stress and cell death, and inflammation. For the analysis on protein level, two key proteins were selected: i) of the differentially expressed genes in the gene set are directly or indirectly related to NF-κB signaling. To efficiently test NF-κB activation, we decided to investigate RelA (p65) translocation upon E5415A and complement treatment in U-87MG-AQP4-ECFP cells since they showed the most pronounced difference on mRNA level regarding their NF-κB activation. For the ICC of RelA, U-87MG-AQP4-ECFP were treated with E5415A in combination with active or heat-inactivated human complement. Untreated cells served as a negative control, and TNF-α-treated cells as a positive control. Thereafter, intracellular ICC was performed. Representative confocal microscopy images are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA. As expected, no translocation into the nucleus was observed in untreated cells, whereas the majority of nuclei in TNF-α-treated cells showed translocated p65. In E5415A\u0026thinsp;+\u0026thinsp;AC-treated cells, approximately 70% (visual estimation) showed NF-κB activation by RelA nuclear translocation. Treatments with E5415A\u0026thinsp;+\u0026thinsp;IC or complement only showed RelA translocation in some cells but in a smaller proportion of approximately 10%.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eii) IL6 is one of the key players in NMOSD and was significantly upregulated on mRNA level in U-87MG-AQP4-ECFP cells treated with E5415A\u0026thinsp;+\u0026thinsp;AC. For the assessment of IL6 production after treatment, a human IL6 ELISA was used. U-87MG-AQP4-ECFP, HA, and HEK293-AQP4-EmGFP cells were treated as before; the cell supernatant was collected after the end of incubation and analyzed for IL6 levels. Results are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB. Similar to the gene expression results, treatment with E5415A\u0026thinsp;+\u0026thinsp;AC resulted in the significantly increased secretion of IL6 only in U-87MG-AQP4-ECFP cells compared to the other treatments (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Importantly, IL6 was released into the cell supernatant in higher amounts by U-87MG-AQP4-ECFP cells as well as in complement-only treated cells compared to the other cell lines. Whereas HA showed no IL6 production in untreated cells, they had low IL6 levels in treated cells, which did not show significant changes between E5415A\u0026thinsp;+\u0026thinsp;AC-treated cells and the other conditions. No IL6 secretion was found in HEK293-AQP4-EmGFP cells.\u003c/p\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eConfirmatory translational analysis of NMOSD patient samples\u003c/h2\u003e\u003cp\u003eSince U-87MG-AQP4-ECFP proved to be the most applicable cell line for AQP4-IgG seropositive NMOSD modeling on the cellular level we next investigated the effects of human AQP4-IgG from NMOSD patients in this model. Therefore, we applied six NMOSD patient sera to the cells; three of them were AQP4-IgG seropositive (NMOSD#1, #4, and #5), and three were seronegative (NMOSD#2, #3, and #6). The binding of AQP4-IgG and AQP4-IgM was confirmed by cell-based assays (Supplementary Figs.\u0026nbsp;6 and 7). Cells were treated with 10% human serum sample and active or heat-inactivated human complement. Cell supernatants were analyzed for LDH to assess the degree of cytotoxicity and for IL6 production. Furthermore, cells were either lysed for RNA isolation and analysis or stained for TCC deposition or p65 translocation.\u003c/p\u003e\u003cp\u003eResults of gene expression levels are shown as a heat map with log2-fold change to complement only controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Two-fold changes in gene expression were only found in the AQP4-IgG seropositive samples NMOSD#1 (AQP4-IgG titer 1:20,480), #4 (1:1,280), and #5 (1:5,120) in combination with AC, but not IC. In general, the most pronounced changes were found for NMOSD#1 and AC treatment. Next, a 2-way ANOVA with Š\u0026iacute;d\u0026aacute;k's multiple comparisons test was performed comparing each serum treatment combined with AC vs. IC. The results are shown in Supplementary Fig.\u0026nbsp;8. For AQP4, only NMOSD#1 showed significant differences (adj. p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.01). All gene expressions were significantly upregulated between NMOSD#1 with AC vs. IC treatment with a p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, except for JUNB and NFKB2 (both not significant). NMOSD#2 serum treated cells showed no significance in any of the tested genes between AC or IC addition. Similarly, NMOSD#3\u0026thinsp;+\u0026thinsp;AC-treated cells showed no significant changes, except for a downregulation of IL11 (p\u0026thinsp;=\u0026thinsp;0.03) and JUNB (p\u0026thinsp;=\u0026thinsp;0.01). NMOSD#4\u0026thinsp;+\u0026thinsp;AC treated cells showed gene upregulation in CXCL1, CXCL2, IL6, IRAK2, NFKBIA, NFKBIZ, NR4A2, and PTX3 (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), as well as in IRF1 (p\u0026thinsp;=\u0026thinsp;0.007). Likewise, CXCL1, CXCL2, IL6, NFKBIZ, and NR4A2 were highly upregulated in NMOSD#5\u0026thinsp;+\u0026thinsp;AC treated cells (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). IL11 (p\u0026thinsp;=\u0026thinsp;0.001), NFKBIA, and PTX3 (both p\u0026thinsp;=\u0026thinsp;0.003) were upregulated as well. NMOSD#6 serum with AC treatment led to the upregulation of three genes, CXCL2 (p\u0026thinsp;=\u0026thinsp;0.002), IL6 (p\u0026thinsp;=\u0026thinsp;0.008), and NR4A2 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe results from the cytotoxicity assay are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB. U-87MG-AQP4-ECFP treated with the AQP4-IgG positive serum samples NMOSD#1, #4, and #5, and AC showed the highest cytotoxicity, but also the AQP4-IgG seronegative sample NMOSD#6\u0026thinsp;+\u0026thinsp;AC was cytotoxic. Again, the greatest effect was seen for NMOSD#1. To assess the complement-dependent cell lysis further, we additionally performed ICC for TCC of treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). Thereby, only cells treated with AQP4-IgG seropositive NMOSD patient sera showed TCC deposition on the cellular surface, and only when combined with AC. NMOSD#6 serum-treated cells here did not show any TCC deposition. Furthermore, the intensity of TCC was higher in NMOSD#1-treated cells, in line with the LDH assay results.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIL6 levels were significantly increased after treatment of serum samples NMOSD#1, #4, and #5 in combination with AC, but not after treatment with AQP4-IgG seronegative samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC). Again, NMOSD#1 and AC-treated cells produced the highest levels of IL6.\u003c/p\u003e\u003cp\u003eTo assess NF-κB activation in NMOSD patient sera and human complement-treated U-87MG-AQP4-ECFP cells, ICC for RelA (p65) was performed (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB). Although NF-κB-related genes were highly upregulated in AQP4-IgG seropositive NMOSD patient sera on mRNA level, nuclear translocation of p65 was not as distinctive. NMOSD#1\u0026thinsp;+\u0026thinsp;AC-treated cells showed the highest proportion of p65 in the nuclei at approximately 70% (visual estimation), whereas NMOSD#4 and NMOSD#5 showed only weak p65 translocation in approximately 10% of nuclei. However, there was also weak translocation of p65 into the nucleus observed in some of the sera plus IC-treated cells, but none in the AQP4-IgG seronegative NMOSD patient sera with AC (NMOSD#2, #3, and #6).\u003c/p\u003e\u003cp\u003eFinally, we checked whether the NF-κB pathway is also activated in human NMOSD neuropathology. Therefore, we investigated the medulla oblongata from two NMOSD patients, who presented different lesion types: one patient showed acute astrocyte loss (Supplementary Fig.\u0026nbsp;9A and B), extensive complement deposition C9neo (Supplementary Fig.\u0026nbsp;9C) and acute axonal injury in the lesion (Supplementary Fig.\u0026nbsp;9D); the other patient revealed pronounced tissue vacuolization along with selective loss of AQP4 immunoreactivity and some axonal spheroids, but without complement deposition (data not shown). Both lesions were located in the raphe and affected axons crossing the midline, derived from the inferior olivary nucleus. Some neurons in the inferior olivary nucleus showed strong nuclear translocation of p65 (Supplementary Fig.\u0026nbsp;9E, arrows), while astrocytes in the olivary nucleus (Supplementary Fig.\u0026nbsp;9E, arrowheads) and lesion rim (Supplementary Fig.\u0026nbsp;9F, arrowheads) were p65 negative.\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIt has been shown both \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e that the combination of antibodies against AQP4 and human complement induces CDC in astrocytes \u003csup\u003e4,6,8,11\u0026ndash;17,19\u0026minus;32,34,48\u0026ndash;53\u003c/sup\u003e. We treated the four human cellular models with E5415A, a monoclonal antibody against AQP4, in combination with human complement. Then, we performed mRNA-seq and compared transcriptomic changes to those of an NMOSD \u003cem\u003ein vivo\u003c/em\u003e rat model.\u003c/p\u003e\u003cp\u003eTo ensure that mRNA changes after E5415A and active complement treatment were associated with CDC, we tested all four cell lines for cytotoxicity. To confirm that cell lysis resulted from the classic complement pathway, we checked for TCC and C3/C3b deposition. First, we used a cellular model previously established in our lab by Lerch et al. using transfected HEK293 cells \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Moreover, we used a stable AQP4-ECFP expressing U-87MG cell line, with an ECFP only overexpressing equivalent to compare AQP4-specific effects. CDC was observed in HEK293-AQP4-EmGFP cells and in U-87MG-AQP4-ECFP cells, confirming previous findings \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Importantly, U-87MG-AQP4-ECFP showed the highest toxicity after incubation, followed by HEK293-AQP4-EmGFP cells, which might be explained by AQP4 expression (stable in U-87MG cells and transient in HEK293 cells). However, partial activation of the alternative complement pathway was observed as well in AC-only-treated cells. Levels of C3a were elevated in AC-treated cells in the cell supernatant of, and C3/C3b deposition was found on U-87MG-AQP4-ECFP cells in our study, and in the previous study of Lerch et al. \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Typically, the alternative pathway is additionally eventually activated, although the classical pathway is the primary cause of CDC, leading to complement amplification accompanied with inflammation \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAs we wanted to compare the cellular effects of these immortalized cell lines with those of physiologically relevant cells, we additionally used HA. However, the HA chosen for our study turned out to be inappropriate for our research question. First, cell culture with primary cells is limited to a few passages, as they rapidly alter their morphology, growth behavior, and gene expression, including AQP4 levels. Second, cytotoxicity was induced in these cells with no significant differences between the different treatments, and they showed high variation between replicates. Although the gene expression levels of our target genes were altered compared to untreated cells, differences between treatments could only be observed in some genes. Furthermore, no TCC deposition was detected on HA after complement treatment (data not shown), but C3/C3b on cells treated with AC with or without E5415A. Additionally, increased C3a levels in HA indicate the activation of the alternative complement pathway. Since gene counts of general astrocyte markers, such as GFAP and S100B, were high in HA samples, and AQP4 was moderately expressed across treatments as well, we wondered which developmental state the cells had at the point of isolation. Upon request at ScienCell, we got the information that the Lot we had purchased (#33619) was derived from the cerebral cortex of a female donor with a gestational age of 22 weeks. At this stage of development, AQP4 expression is only rarely found in the neocortex, and astrocyte differentiation and BBB encirclement occur postnatally \u003csup\u003e\u003cspan additionalcitationids=\"CR57\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. On the one hand, the overexpression of AQP4 in immortalized cell lines led to a 100-fold higher expression, which might explain the majority of why changes were more pronounced in the immortalized cell lines. On the other hand, a small proportion might have been caused by the immaturity of our HA as well. This limitation should be addressed in future studies by differentiating primary astrocytes to promote maturation. However, the question remains if a sufficient AQP4 expression at the astrocyte endfeet could be achieved without co-culturing the astrocytes with brain microvascular endothelial cells.\u003c/p\u003e\u003cp\u003eThe treatment with E5415A and AC led to CDC in AQP4-expressing cells and profound transcriptomic changes on the mRNA level. As expected, AQP4 expression levels were downregulated after treatment with E5415A and AC in U-87MG-AQP4-ECFP and HEK293-AQP4-EmGFP cells, indicating a specific cell loss and confirming previous findings \u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. Moreover, most changes of E5415A and AC treatment compared to all other treatments led to most genes being changed in U-87MG-AQP4-ECFP cells (compared to all other treatments in AQP4-positive and -negative cells; n\u0026thinsp;=\u0026thinsp;59). In contrast, only 6 genes changed compared to differently treated cells in HEK293-AQP4-EmGFP cells, and none in HA. Interestingly, most genes changed in HA compared to the untreated control (n\u0026thinsp;=\u0026thinsp;808), but the antibody addition did not alter any gene expression compared to AC-only treated cells. This was surprising, since when compared to IC with or without E5415A-treated cells, the number of changed genes deviated a lot. This could be due to the high variability in HA replicates.\u003c/p\u003e\u003cp\u003eHEK293-AQP4-EmGFP cells showed an upregulation of NR4A1-3 and downregulation of TXNIP when treated with E5415A and AC, which were overlapping genes that had been altered after the same treatment in U-87MG-AQP4-ECFP as well (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). NR4A1-3 encode for nuclear receptor 4A family proteins and are immediate early genes that are upregulated during inflammation and cellular stress \u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. In astrocytes, NR4A2 is induced via pro-inflammatory cytokines and binds NF-κB component p65 on the target inflammatory gene promoter and results in transcriptional repression \u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. Upregulation of NR4A2 suggests a cellular response to the NF-κB activation overshoot the cells experience after treatment. TXNIP, an antioxidant and key protein in regular stress response that usually promotes apoptosis, was downregulated in both AQP4 overexpressing cell lines, indicating an anti-apoptotic gene regulation \u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. CGA upregulation and ARRDC3 downregulation were detected in HEK293-AQP4-EmGFP cells exclusively.\u003c/p\u003e\u003cp\u003eDifferently expressed genes were then compared to spatial transcriptomic results of an \u003cem\u003ein vivo\u003c/em\u003e rat model (medulla oblongata inflammatory NMOSD lesion) after peripheral injection of E5415A antibody. Most of the overlapping differently expressed genes were directly or indirectly associated with stress responses, inflammation, and NF-κB signaling (interactions are represented schematically in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Moreover, activation of NF-κB signaling was also observed in human NMOSD pathological tissue. Although NF-κB is also activated due to the glioblastoma origin of U-87 MG cells and activating transcription factor-3 (ATF3) and CCAAT/enhancer-binding protein beta (CEBPB) are transcription factors that are associated with malignant glioblastoma, they are likewise activated under general cellular stress and inflammation \u003csup\u003e\u003cspan additionalcitationids=\"CR65\" citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. These genes were elevated in treated HA compared to untreated cells as well. ATF3 and CEBPB work downstream of NF-κB to amplify or modulate immune responses \u003csup\u003e\u003cspan additionalcitationids=\"CR68\" citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. Our findings are in line with a previous study of Walker-Caulfield and colleagues, which investigated primary astrocyte-enriched mouse cultures after stimulation with AQP4 antibody-positive serum samples from NMOSD patients for their transcriptomic changes \u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. This study found an upregulation of mouse equivalents of genes detected in our study (CXCL1, CXCL2, IL6, Cebpb, Nfkbia, Nfkbiz). The authors highlighted a \u0026ldquo;NMOSD granulocytic footprint\u0026rdquo;, which they indicated is activated not only downstream via CDC but also as an early event in the onset of NMOSD pathology, creating a pro-granulocytic inflammatory environment. Furthermore, this study demonstrated the efficiency of bortezomib, a small-molecule proteasome inhibitor, which is currently approved for myeloma treatment, to successfully inhibit NF-κB signaling in astrocytes \u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. Similar results of a predominantly NF-kB and IL6-driven response to AQP4 antibodies and complement were seen by two other studies in primary rat astrocytes: Du et al. found an IL6 upregulation after NMOSD patient serum application, which was assigned to Janus kinase/signal transducer and activator of transcription 3-dependent inflammatory response, as they were able to decrease IL6 levels with a Janus kinase1/2 specific inhibitor AZD1480 \u003csup\u003e59\u003c/sup\u003e. Additionally, Wang et al. were able to prevent IL6 level increase after NMOSD patient AQP4 antibody exposure by blocking NF-κB with the inhibitor S3633, indicating the contribution of NF-κB for elevated IL6 in NMOSD \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. IL6 has often been highlighted as a leading inflammatory cytokine in NMOSD, given its abundance in the blood and cerebrospinal fluid of affected patients. \u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e. Its major impact on the disease, as well as its efficient effects on relapse prevention when blocked, mark IL6 as one of the major cytokines that should be inducible in any NMOSD model \u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e. Importantly, we observed an upregulation of IL6 at both mRNA and protein levels in U-87MG-AQP4-ECFP cells after treatment with E5415A or AQP4 antibody-positive NMOSD patient serum samples, but not in HEK293-AQP4-EmGFP cells.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eRegarding the other changed genes in our set, NF-κB pathways are highlighted by direct or indirect involvement even more: C-X-C motif chemokine ligand 1 and 2 (CXCL1/2) are pro-inflammatory chemokines, which recruit neutrophils to sites of inflammation and are directly regulated by NF-κB. IL6 and IL11 are cytokines involved in acute and chronic inflammatory processes, with IL6 being a well-established downstream target of NF-κB \u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e. Interleukin-1 receptor-associated kinase-like-2 (IRAK2) plays a role in Toll-like receptor (TLR) and IL1 receptor signaling, which activates NF-κB \u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e,\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e. Interferon regulatory factor-1 (IRF1) cooperates with NF-κB in regulating genes involved in immune defense \u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e. NF-κB subunit-2/p100 (NFKB2) and NF-κB inhibitor alpha (IκBα; NFKBIA) encode components of the NF-κB pathway itself. NFKB2 contributes to the non-canonical NF-κB pathway \u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e. NFKBIA is an inhibitor that regulates NF-κB activity through feedback in the canonical pathway \u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e. Upon activation by pro-inflammatory stimuli such as tumor necrosis factor-alpha (TNF-α), IL1β, or pathogen-associated molecular patterns, IκBα is phosphorylated and degraded. This frees the NF-κB dimer, comprising p50/ NF-κB1 and p65/RelA, allowing it to translocate into the nucleus \u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e. NFKBIA is downregulated in peripheral blood mononuclear cells (PBMCs) from NMOSD patients during NMOSD relapses, perhaps reflecting the underlying inflammatory pathway in NMOSD during a disease flare \u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e. NF-κB inhibitor zeta (IκBζ; NFKBIZ) functions as a co-activator for specific NF-κB target genes, particularly during inflammatory responses \u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e. RelA, when unshed from IκBα plays a pivotal role by binding to κB sites in the promoters of target genes and driving the transcription of pro-inflammatory mediators, such as CXCL1, IL6, and pentraxin-3 (PTX3). When NF-κB signaling was tested via RelA translocation in this study, it was more elevated in our U-87MG-AQP4-ECFP cells treated with E5415A\u0026thinsp;+\u0026thinsp;AC combination treatment. Nevertheless, it needs to be mentioned that this effect was only as pronounced when adding the monoclonal antibody or anti-AQP4 IgG seropositive human serum with a high titer (1:20,480). PTX3, an acute-phase protein, is involved in innate immunity and complement component interaction (C1q, factor H, ficolins, mannan-binding lectin) \u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. PTX3 has also been shown to be associated with inflammatory responses in NMOSD patients \u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. NR4A2, as mentioned before, plays an anti-inflammatory role by suppressing RelA binding to the target inflammatory gene promoter \u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. Therefore, NR4A2 would be a possible candidate for NF-κB blockade by upregulating its expression. Indeed, we have seen E5415A\u0026thinsp;+\u0026thinsp;AC-specific upregulation of NR4A2 in AQP4-overexpressing cells. Its NF-κB downregulating effect, however, was not observed in this study.\u003c/p\u003e\u003cp\u003eTranscription factor jun-B (JUNB), part of the AP-pathogenicity and alongside NF-κB in cytokine expression, is essential for IL23-dependent Th17 pathogenicity and is itself induced by IL6 \u003csup\u003e78,79\u003c/sup\u003e. Nishiyama et al. observed the release of pro-inflammatory Th17 cytokines in AQP4-IgG seropositive NMOSD patient PBMCs following incubation with AQP4-immunocomplexes and complement. Notably, upregulation of IL17A and IL23 was detected only in treatment-na\u0026iuml;ve PBMCs that had not undergone B cell depletion, whereas Rituximab-treated PBMCs exhibited enhanced IL6 production. These findings underscore the necessity of IL6-activated B and T cells for downstream Th17 cytokine production \u003csup\u003e\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e. This was highlighted before by Agasing et al. in transcriptomic analysis of NMOSD patient PBMCs as well, as they showed the co-upregulation of IL6 and interferon type I, thereby interferon type I being essential for B cell activation, and IL6 for further Th17 differentiation \u003csup\u003e\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eOur study has some limitations. First, cell lysis was also observed to a lower degree in samples with complement only or E5415A with IC as well. TCC staining showed that this cell lysis was not due to terminal complement complex formation. A possible explanation for this could be unspecific binding or alternative pathway activation by C3 opsonization after the addition of human complement. Chen and colleagues saw C3 upregulation in mouse astrocytes after NMO-IgG exposure that led to microglial interactions via C3a receptors \u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e. We detected high gene counts of C3 in U-87MG cell lines independent of treatment. In contrast, gene counts were lower in E5415A\u0026thinsp;+\u0026thinsp;AC-treated HEK293-AQP4-EmGFP cells compared to the other treatments, and higher in E5415A\u0026thinsp;+\u0026thinsp;AC-treated HA compared to the remaining samples. C1R and C1S components were elevated in these cells as well, but gene counts were moderate to low in all four cell lines and did not correlate notably with different treatment (Supplementary Fig.\u0026nbsp;2). Second, besides complement components being directly involved in the classical complement pathway, there are additional complement regulatory proteins such as CD46 (membrane cofactor protein), CD55 (decay accelerating factor), and CD59 (protectin), which either inhibit the C3 convertase, or the TCC formation \u003csup\u003e\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e. A study by Saadoun and Papadopoulus revealed that complement regulatory proteins are expressed in human astrocytes, but not in NMOSD lesions and after co-culture of astrocytes with endothelial cells \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. These complement inhibitors, CFH (complement factor H) and CLU (clusterin), were expressed at different levels in the cell lines used in our study (Supplementary Fig.\u0026nbsp;3). The presence of these complement inhibitors might explain why relatively high amounts of human complement were necessary to induce CDC. Third, findings were rather unspecific in HA, which is explained by their low AQP4 expression. These primary HA were obtained from human cerebral cortex (purchased via ScienCell) and cultured for up to 5 passages. Finally, the results from human serum samples were AQP4 antibody titer dependent. Most genes were significantly changed after treatment with human AQP4 antibodies and AC. Cytotoxicity also correlated with AQP4 antibody levels. Surprisingly, serum from the AQP4-antibody negative NMOSD patient #6 induced some CDC, probably due to the presence of other factors in the serum of this patient. Finally, in the human neuropathological sections investigated in our study strong nuclear translocation of NF-κB was only seen in some neurons, while astrocytes were negative. This could be explained by the selective loss of astrocytes in the lesions and the activation of NF-κB signaling pathways by the alternative complement pathway in neurons, similar as seen in other cells in this study.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOur study demonstrates that AQP4-expressing U-87MG-AQP4-ECFP cells exhibit significant CDC upon exposure to the monoclonal AQP4 antibody E5415A in combination with active human complement. Transcriptomic analysis revealed an upregulation of genes directly or indirectly linked to IL6 and NF-κB after this treatment. Similar changes were also seen in an \u003cem\u003ein vivo\u003c/em\u003e rat model of NMOSD: This suggests that U-87MG-AQP4-ECFP cells effectively mimic CDC-related astrocytic responses in AQP4 antibody seropositive NMOSD and enable to testing of new treatment strategies, especially regarding the NF-κB pathway, on a cellular level.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eAC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eactive complement\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eADCC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eantibody-dependent cellular cytotoxicity\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eAQP4\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eaquaporin-4\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eATF3\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eactivating transcription factor-3\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eBBB\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eblood-brain barrier\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCDC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ecomplement-dependent cytotoxicity\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCEBPB\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eCCAAT/enhancer-binding protein beta\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCFH\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ecomplement factor H\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCHO\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eChinese hamster ovary\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCXCL\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eC-X-C motif chemokine ligand\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDAPI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003e4\u0026prime;,6-diamidino-2-phenylindole\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDAVID\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eDatabase for Annotation, Visualization, and Integrated Discovery\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDEG\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003edifferentially expressed gene\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eECFP\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eenhanced cyan fluorescent protein\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eELISA\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eenzyme-linked immunosorbent assay\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eEmGFP\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eemerald green fluorescent protein\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eFPKM\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003efragments per kilobase million\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGFAP\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eglial fibrillary acidic protein\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGO\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003egene ontology\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eHEK293\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ehuman embryonic kidney 293\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIL\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003einterleukin\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003einactive complement\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eICC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eimmunocytochemistry\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIgG\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eimmunoglobulin-G\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIRAK2\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003einterleukin-1 receptor-associated kinase-like-2\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIRF1\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003einterferon regulatory factor-1\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eJUNB\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003etranscription factor jun-B\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLDH\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003elactate dehydrogenase\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003emin\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eminutes\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNF-κB\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003enuclear factor K-light-chain-enhancer of activated B cells\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNFKB2\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNF-κB subunit-2\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNFKBIA\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNF-κB inhibitor alpha\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNFKBIZ\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNF-κB inhibitor zeta\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNR4A\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNuclear receptor subfamily 4 group A protein\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePBMCs\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eperipheral blood mononuclear cells\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eprincipal component\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePTX3\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003epentraxin-3\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eRT-qPCR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ereverse transcription-quantitative polymerase chain reaction\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTCC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eterminal complement complex\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTNF-α\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003etumor necrosis factor-alpha\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003cp\u003eSerum samples from six NMOSD patients and neuropathological tissue sections from two NMOSD patients were provided by the biobank of the Department of Neuropathology, Medical University of Vienna, Austria. The use of these samples from a biobank for research studies was approved by the ethical committee of the Medical University of Vienna (EK 1636/2019 and 1123/2015). All patients or their legal representatives gave written informed consent to participate in the study and all methods were performed in accordance with the relevant guidelines and regulation.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eCompeting Interests\u003c/h2\u003e\u003cp\u003eThe authors declare that this study received funding from Roche Austria GmbH (to MR). The funder was not involved in the study design, analysis, or critical revision of the article for important intellectual content. RH reports speaker honoraria from UCB and BMS. The Medical Universities of Innsbruck (Austria; employer of MR) and Vienna (Austria; employer of RH) receive payments for antibody assays and for antibody validation experiments organized by Euroimmun (L\u0026uuml;beck, Germany). QY, JH, VE and MB declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis study was financially supported by the intramural funding program of the Medical University of Innsbruck Ph.D. Research Training Groups, Project 2022-1-2 \u0026ldquo;CONNECT\u0026rdquo; (to SB, JH and MR) and a restricted research grant from Roche Austria GmbH supporting this study (to MR). It was also supported by the Austrian Science Fund (FWF grant \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.55776/PAT6054424\u003c/span\u003e\u003cspan address=\"10.55776/PAT6054424\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to MB, the China Scholarship Council (CSC 202306170046) to QY, and the Austrian Research Promotion Agency (FFG, project number FO999920011) to RH and VE.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eSB analyzed and interpreted the data, wrote the manuscript, and performed all experiments. MR designed the study, supervised the work, analyzed and interpreted data, and participated in preparing the manuscript. JH produced the stable cell lines and contributed to the manuscript. QY and MB provided the spatial transcriptomics analysis of the experimental rat NMO model and helped with the analysis and interpretation of data. VE and RH provided serum samples of NMOSD patients and performed the immunohistochemistry of human NMOSD neuropathological sections. All authors reviewed the manuscript critically for important intellectual content and approved the final version of the manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets used and/or analyzed during the current study are available at GEO (GSE291954) or shown in the manuscript and supplementary file. The submission is still private, to review the reviewer should go to https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE291954 and enter the token urmxuskqlnmvtkz into the box.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eJarius, S. et al. Neuromyelitis optica. \u003cem\u003eNat. Rev. Dis. Primers\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, 85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41572-020-0214-9\u003c/span\u003e\u003cspan address=\"10.1038/s41572-020-0214-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWingerchuk, D. M. et al. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e85\u003c/b\u003e, 177\u0026ndash;189. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/wnl.0000000000001729\u003c/span\u003e\u003cspan address=\"10.1212/wnl.0000000000001729\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCrane, J. M. et al. Binding affinity and specificity of neuromyelitis optica autoantibodies to aquaporin-4 M1/M23 isoforms and orthogonal arrays. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cb\u003e286\u003c/b\u003e, 16516\u0026ndash;16524. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1074/jbc.M111.227298\u003c/span\u003e\u003cspan address=\"10.1074/jbc.M111.227298\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHinson, S. R. et al. Molecular outcomes of neuromyelitis optica (NMO)-IgG binding to aquaporin-4 in astrocytes. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e 109, 1245\u0026ndash;1250, doi: (2012). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.1109980108\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1109980108\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJitprapaikulsan, J. et al. Clinical utility of AQP4-IgG titers and measures of complement-mediated cell killing in NMOSD. \u003cem\u003eNeurology(R) Neuroimmunol. neuroinflammation\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/nxi.0000000000000727\u003c/span\u003e\u003cspan address=\"10.1212/nxi.0000000000000727\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLerch, M. et al. More Efficient Complement Activation by Anti-Aquaporin-4 Compared With Anti-Myelin Oligodendrocyte Glycoprotein Antibodies. \u003cem\u003eNeurology(R) neuroimmunology \u0026amp; neuroinflammation\u003c/em\u003e 10, (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/nxi.0000000000200059\u003c/span\u003e\u003cspan address=\"10.1212/nxi.0000000000200059\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYandamuri, S. S. et al. MOGAD patient autoantibodies induce complement, phagocytosis, and cellular cytotoxicity. \u003cem\u003eJCI Insight\u003c/em\u003e. \u003cb\u003e8\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1172/jci.insight.165373\u003c/span\u003e\u003cspan address=\"10.1172/jci.insight.165373\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDuan, T., Smith, A. J. \u0026amp; Verkman, A. S. Complement-independent bystander injury in AQP4-IgG seropositive neuromyelitis optica produced by antibody-dependent cellular cytotoxicity. \u003cem\u003eActa Neuropathol. Commun.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 112. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s40478-019-0766-7\u003c/span\u003e\u003cspan address=\"10.1186/s40478-019-0766-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWaters, P. J. et al. Serologic diagnosis of NMO: a multicenter comparison of aquaporin-4-IgG assays. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e78\u003c/b\u003e, 665\u0026ndash;671. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/WNL.0b013e318248dec1\u003c/span\u003e\u003cspan address=\"10.1212/WNL.0b013e318248dec1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012). discussion 669.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWaters, P. et al. Multicentre comparison of a diagnostic assay: aquaporin-4 antibodies in neuromyelitis optica. \u003cem\u003eJ. Neurol. Neurosurg. Psychiatry\u003c/em\u003e. \u003cb\u003e87\u003c/b\u003e, 1005\u0026ndash;1015. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/jnnp-2015-312601\u003c/span\u003e\u003cspan address=\"10.1136/jnnp-2015-312601\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIshikura, T. et al. Anti-AQP4 autoantibodies promote ATP release from astrocytes and induce mechanical pain in rats. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e, 181. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12974-021-02232-w\u003c/span\u003e\u003cspan address=\"10.1186/s12974-021-02232-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLennon, V. A., Kryzer, T. J., Pittock, S. J., Verkman, A. S. \u0026amp; Hinson, S. R. IgG marker of optic-spinal multiple sclerosis binds to the aquaporin-4 water channel. \u003cem\u003eJ. Exp. Med.\u003c/em\u003e \u003cb\u003e202\u003c/b\u003e, 473\u0026ndash;477. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1084/jem.20050304\u003c/span\u003e\u003cspan address=\"10.1084/jem.20050304\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKaneko, K. et al. Different Complement Activation Patterns Following C5 Cleavage in MOGAD and AQP4-IgG\u0026thinsp;+\u0026thinsp;NMOSD. \u003cem\u003eNeurol. Neuroimmunol. Neuroinflammation\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, e200293. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/NXI.0000000000200293\u003c/span\u003e\u003cspan address=\"10.1212/NXI.0000000000200293\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYamashita, K. et al. Cerebrospinal fluid mitochondrial DNA in neuromyelitis optica spectrum disorder. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e15\u003c/b\u003e, 125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12974-018-1162-0\u003c/span\u003e\u003cspan address=\"10.1186/s12974-018-1162-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoltys, J. et al. Membrane assembly of aquaporin-4 autoantibodies regulates classical complement activation in neuromyelitis optica. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e \u003cb\u003e129\u003c/b\u003e, 2000\u0026ndash;2013. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1172/jci122942\u003c/span\u003e\u003cspan address=\"10.1172/jci122942\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRatelade, J. et al. Neuromyelitis optica IgG and natural killer cells produce NMO lesions in mice without myelin loss. \u003cem\u003eActa Neuropathol.\u003c/em\u003e \u003cb\u003e123\u003c/b\u003e, 861\u0026ndash;872. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00401-012-0986-4\u003c/span\u003e\u003cspan address=\"10.1007/s00401-012-0986-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhu, W. et al. Human C5-specific single-chain variable fragment ameliorates brain injury in a model of NMOSD. \u003cem\u003eNeurol. Neuroimmunol. Neuroinflamm\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e561. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/nxi.0000000000000561\u003c/span\u003e\u003cspan address=\"10.1212/nxi.0000000000000561\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTradtrantip, L., Yeaman, M. R. \u0026amp; Verkman, A. S. Cytoprotective IgG antibodies in sera from a subset of patients with AQP4-IgG seropositive neuromyelitis optica spectrum disorder. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 21962. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-021-01294-3\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-01294-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePhuan, P. W., Ratelade, J., Rossi, A., Tradtrantip, L. \u0026amp; Verkman, A. S. Complement-dependent cytotoxicity in neuromyelitis optica requires aquaporin-4 protein assembly in orthogonal arrays. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cb\u003e287\u003c/b\u003e, 13829\u0026ndash;13839. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1074/jbc.M112.344325\u003c/span\u003e\u003cspan address=\"10.1074/jbc.M112.344325\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTradtrantip, L., Ratelade, J., Zhang, H. \u0026amp; Verkman, A. S. Enzymatic deglycosylation converts pathogenic neuromyelitis optica anti-aquaporin-4 immunoglobulin G into therapeutic antibody. \u003cem\u003eAnn. Neurol.\u003c/em\u003e \u003cb\u003e73\u003c/b\u003e, 77\u0026ndash;85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/ana.23741\u003c/span\u003e\u003cspan address=\"10.1002/ana.23741\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMader, S., Brimberg, L., Soltys, J. N., Bennett, J. L. \u0026amp; Diamond, B. Mutations of Recombinant Aquaporin-4 Antibody in the Fc Domain Can Impair Complement-Dependent Cellular Cytotoxicity and Transplacental Transport. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 1599. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2018.01599\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2018.01599\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVincent, T. et al. Functional Consequences of Neuromyelitis Optica-IgG Astrocyte Interactions on Blood-Brain Barrier Permeability and Granulocyte Recruitment1. \u003cem\u003eJ. Immunol.\u003c/em\u003e \u003cb\u003e181\u003c/b\u003e, 5730\u0026ndash;5737. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4049/jimmunol.181.8.5730\u003c/span\u003e\u003cspan address=\"10.4049/jimmunol.181.8.5730\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaruki, H. et al. NMO sera down-regulate AQP4 in human astrocyte and induce cytotoxicity independent of complement. \u003cem\u003eJ. Neurol. Sci.\u003c/em\u003e \u003cb\u003e331\u003c/b\u003e, 136\u0026ndash;144. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jns.2013.05.035\u003c/span\u003e\u003cspan address=\"10.1016/j.jns.2013.05.035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNishiyama, S. et al. Complement-dependent and -independent aquaporin 4-antibody-mediated cytotoxicity in human astrocytes: Pathogenetic implications in neuromyelitis optica. \u003cem\u003eBiochem. Biophys. Rep.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 45\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.bbrep.2016.05.012\u003c/span\u003e\u003cspan address=\"10.1016/j.bbrep.2016.05.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePiatek, P. et al. C5a-Preactivated Neutrophils Are Critical for Autoimmune-Induced Astrocyte Dysregulation in Neuromyelitis Optica Spectrum Disorder. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 1694. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2018.01694\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2018.01694\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSaadoun, S. \u0026amp; Papadopoulos, M. C. Role of membrane complement regulators in neuromyelitis optica. \u003cem\u003eMultiple Scler. J.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 1644\u0026ndash;1654. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/1352458515571446\u003c/span\u003e\u003cspan address=\"10.1177/1352458515571446\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSabater, L. et al. Cytotoxic effect of neuromyelitis optica antibody (NMO-IgG) to astrocytes: an in vitro study. \u003cem\u003eJ. Neuroimmunol.\u003c/em\u003e \u003cb\u003e215\u003c/b\u003e, 31\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jneuroim.2009.07.014\u003c/span\u003e\u003cspan address=\"10.1016/j.jneuroim.2009.07.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlexopoulos, H. et al. Anti-aquaporin-4 autoantibodies in systemic lupus erythematosus persist for years and induce astrocytic cytotoxicity but not CNS disease. \u003cem\u003eJ. Neuroimmunol.\u003c/em\u003e \u003cb\u003e289\u003c/b\u003e, 8\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jneuroim.2015.10.007\u003c/span\u003e\u003cspan address=\"10.1016/j.jneuroim.2015.10.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, Y. et al. NMO-IgG Induce Interleukin-6 Release via Activation of the NF-κB Signaling Pathway in Astrocytes. \u003cem\u003eNeuroscience\u003c/em\u003e \u003cb\u003e496\u003c/b\u003e, 96\u0026ndash;104. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuroscience.2022.05.038\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroscience.2022.05.038\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang, P. et al. The binding property of a monoclonal antibody against the extracellular domains of aquaporin-4 directs aquaporin-4 toward endocytosis. \u003cem\u003eBiochem. Biophys. Rep.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 77\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.bbrep.2016.05.017\u003c/span\u003e\u003cspan address=\"10.1016/j.bbrep.2016.05.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHinson, S. R. et al. Aquaporin-4-binding autoantibodies in patients with neuromyelitis optica impair glutamate transport by down-regulating EAAT2. \u003cem\u003eJ. Exp. Med.\u003c/em\u003e \u003cb\u003e205\u003c/b\u003e, 2473\u0026ndash;2481. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1084/jem.20081241\u003c/span\u003e\u003cspan address=\"10.1084/jem.20081241\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHowe, C. L. et al. Neuromyelitis optica IgG stimulates an immunological response in rat astrocyte cultures. \u003cem\u003eGlia\u003c/em\u003e \u003cb\u003e62\u003c/b\u003e, 692\u0026ndash;708. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/glia.22635\u003c/span\u003e\u003cspan address=\"10.1002/glia.22635\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKurosawa, K. et al. Severely exacerbated neuromyelitis optica rat model with extensive astrocytopathy by high affinity anti-aquaporin-4 monoclonal antibody. \u003cem\u003eActa Neuropathol. Commun.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s40478-015-0259-2\u003c/span\u003e\u003cspan address=\"10.1186/s40478-015-0259-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHillebrand, S. et al. Circulating AQP4-specific auto-antibodies alone can induce neuromyelitis optica spectrum disorder in the rat. \u003cem\u003eActa Neuropathol.\u003c/em\u003e \u003cb\u003e137\u003c/b\u003e, 467\u0026ndash;485. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00401-018-1950-8\u003c/span\u003e\u003cspan address=\"10.1007/s00401-018-1950-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHagenbuchner, J. et al. The tubulin inhibitor MG-2477 induces autophagy-regulated cell death, ROS accumulation and activation of FOXO3 in neuroblastoma. \u003cem\u003eOncotarget\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNothdurfter, D. et al. 3D bioprinted, vascularized neuroblastoma tumor environment in fluidic chip devices for precision medicine drug testing. \u003cem\u003eBiofabrication\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 035002. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1088/1758-5090/ac5fb7\u003c/span\u003e\u003cspan address=\"10.1088/1758-5090/ac5fb7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eParkhomchuk, D. et al. Transcriptome analysis by strand-specific sequencing of complementary DNA. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, e123. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkp596\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkp596\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMortazavi, A., Williams, B. A., McCue, K., Schaeffer, L. \u0026amp; Wold, B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. \u003cem\u003eNat. Methods\u003c/em\u003e. \u003cb\u003e5\u003c/b\u003e, 621\u0026ndash;628. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nmeth.1226\u003c/span\u003e\u003cspan address=\"10.1038/nmeth.1226\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiao, Y., Smyth, G. K. \u0026amp; Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 923\u0026ndash;930. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/btt656\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btt656\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, Y. E. et al. A Cloud Platform for Analyzing and Visualizing Cancer Genomic Data. \u003cem\u003eCancer Res.\u003c/em\u003e \u003cb\u003e77\u003c/b\u003e, e11\u0026ndash;e14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/0008-5472.Can-17-0802\u003c/span\u003e\u003cspan address=\"10.1158/0008-5472.Can-17-0802\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAnders, S. \u0026amp; Huber, W. Differential expression analysis for sequence count data. \u003cem\u003eGenome Biol\u003c/em\u003e 11, R106, (2010). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/gb-2010-11-10-r106\u003c/span\u003e\u003cspan address=\"10.1186/gb-2010-11-10-r106\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 550. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13059-014-0550-8\u003c/span\u003e\u003cspan address=\"10.1186/s13059-014-0550-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSherman, B. T. et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e50\u003c/b\u003e, W216\u0026ndash;w221. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkac194\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkac194\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang da, W., Sherman, B. T. \u0026amp; Lempicki, R. A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. \u003cem\u003eNat. Protoc.\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, 44\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nprot.2008.211\u003c/span\u003e\u003cspan address=\"10.1038/nprot.2008.211\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSzklarczyk, D. et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e, D638\u0026ndash;d646. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkac1000\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkac1000\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWalker-Caulfield, M. E. et al. NFκB signaling drives pro-granulocytic astroglial responses to neuromyelitis optica patient IgG. \u003cem\u003eJ. Neuroinflammation\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e, 185. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12974-015-0403-8\u003c/span\u003e\u003cspan address=\"10.1186/s12974-015-0403-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaguire, J. et al. Gem: an induced, immediate early protein belonging to the Ras family. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e265\u003c/b\u003e, 241\u0026ndash;244. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.7912851\u003c/span\u003e\u003cspan address=\"10.1126/science.7912851\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1994).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKinoshita, M. et al. Astrocytic necrosis is induced by anti-aquaporin-4 antibody-positive serum. \u003cem\u003eNeuroreport\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 508\u0026ndash;512. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/wnr.0b013e32832776f4\u003c/span\u003e\u003cspan address=\"10.1097/wnr.0b013e32832776f4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStathopoulos, P. \u0026amp; Dalakas, M. C. The role of complement and complement therapeutics in neuromyelitis optica spectrum disorders. \u003cem\u003eExpert Rev. Clin. Immunol.\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e, 933\u0026ndash;945. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/1744666X.2022.2105205\u003c/span\u003e\u003cspan address=\"10.1080/1744666X.2022.2105205\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBennett, J. L. et al. Intrathecal pathogenic anti-aquaporin-4 antibodies in early neuromyelitis optica. \u003cem\u003eAnn. Neurol.\u003c/em\u003e \u003cb\u003e66\u003c/b\u003e, 617\u0026ndash;629. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/ana.21802\u003c/span\u003e\u003cspan address=\"10.1002/ana.21802\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBradl, M. et al. Neuromyelitis optica: pathogenicity of patient immunoglobulin in vivo. \u003cem\u003eAnn. Neurol.\u003c/em\u003e \u003cb\u003e66\u003c/b\u003e, 630\u0026ndash;643. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/ana.21837\u003c/span\u003e\u003cspan address=\"10.1002/ana.21837\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSaadoun, S. et al. Intra-cerebral injection of neuromyelitis optica immunoglobulin G and human complement produces neuromyelitis optica lesions in mice. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e133\u003c/b\u003e, 349\u0026ndash;361. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/brain/awp309\u003c/span\u003e\u003cspan address=\"10.1093/brain/awp309\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTradtrantip, L., Yeaman, M. R. \u0026amp; Verkman, A. S. Cytoprotective IgG antibodies in sera from a subset of patients with AQP4-IgG seropositive neuromyelitis optica spectrum disorder. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 21962. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-021-01294-3\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-01294-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRossi, A., Moritz, T. J., Ratelade, J. \u0026amp; Verkman, A. S. Super-resolution imaging of aquaporin-4 orthogonal arrays of particles in cell membranes. \u003cem\u003eJ. Cell. Sci.\u003c/em\u003e \u003cb\u003e125\u003c/b\u003e, 4405\u0026ndash;4412. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1242/jcs.109603\u003c/span\u003e\u003cspan address=\"10.1242/jcs.109603\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHo, B. H. T., Spicer, B. A. \u0026amp; Dunstone, M. A. Action of the Terminal Complement Pathway on Cell Membranes. \u003cem\u003eJ. Membr. Biol.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00232-025-00343-6\u003c/span\u003e\u003cspan address=\"10.1007/s00232-025-00343-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eG\u0026ouml;m\u0026ouml;ri, E. et al. Fetal development of membrane water channel proteins aquaporin-1 and aquaporin-4 in the human brain. \u003cem\u003eInt. J. Dev. Neurosci.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 295\u0026ndash;305. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ijdevneu.2006.05.003\u003c/span\u003e\u003cspan address=\"10.1016/j.ijdevneu.2006.05.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHolst, C. B., Br\u0026oslash;chner, C. B., Vitting-Seerup, K. \u0026amp; M\u0026oslash;llg\u0026aring;rd, K. Astrogliogenesis in human fetal brain: complex spatiotemporal immunoreactivity patterns of GFAP, S100, AQP4 and YKL-40. \u003cem\u003eJ. Anat.\u003c/em\u003e \u003cb\u003e235\u003c/b\u003e, 590\u0026ndash;615. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/joa.12948\u003c/span\u003e\u003cspan address=\"10.1111/joa.12948\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoasdou\u0026eacute;, K., Miller, S. M., Colditz, P. B., Bj\u0026ouml;rkman, S. T. \u0026amp; Review The blood-brain barrier; protecting the developing fetal brain. \u003cem\u003ePlacenta\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e, 111\u0026ndash;116. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.placenta.2016.12.005\u003c/span\u003e\u003cspan address=\"10.1016/j.placenta.2016.12.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDu, L. et al. Effect of NMO-IgG on the interleukin-6 cascade in astrocytes via activation of the JAK/STAT3 signaling pathway. \u003cem\u003eLife Sci.\u003c/em\u003e \u003cb\u003e258\u003c/b\u003e, 118217. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lfs.2020.118217\u003c/span\u003e\u003cspan address=\"10.1016/j.lfs.2020.118217\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen, L., Fan, F., Wu, L. \u0026amp; Zhao, Y. The nuclear receptor 4A family members: mediators in human disease and autophagy. \u003cem\u003eCell. Mol. Biol. Lett.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s11658-020-00241-w\u003c/span\u003e\u003cspan address=\"10.1186/s11658-020-00241-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSafe, S. et al. Nuclear receptor 4A (NR4A) family - orphans no more.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSaijo, K. et al. A Nurr1/CoREST pathway in microglia and astrocytes protects dopaminergic neurons from inflammation-induced death.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoi, E. H., Park, S. J. \u0026amp; TXNIP A key protein in the cellular stress response pathway and a potential therapeutic target. \u003cem\u003eExp. Mol. Med.\u003c/em\u003e \u003cb\u003e55\u003c/b\u003e, 1348\u0026ndash;1356. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s12276-023-01019-8\u003c/span\u003e\u003cspan address=\"10.1038/s12276-023-01019-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJi, J. et al. TRIM22 activates NF-κB signaling in glioblastoma by accelerating the degradation of IκBα. \u003cem\u003eCell. Death Differ.\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e, 367\u0026ndash;381. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41418-020-00606-w\u003c/span\u003e\u003cspan address=\"10.1038/s41418-020-00606-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNagai, S. et al. Aberrant nuclear factor-kappaB activity and its participation in the growth of human malignant astrocytoma. \u003cem\u003eJ. Neurosurg.\u003c/em\u003e \u003cb\u003e96\u003c/b\u003e, 909\u0026ndash;917. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3171/jns.2002.96.5.0909\u003c/span\u003e\u003cspan address=\"10.3171/jns.2002.96.5.0909\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2002).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, W. et al. Sphingosine kinase 1 promotes growth of glioblastoma by increasing inflammation mediated by the NF-κB /IL-6/STAT3 and JNK/PTX3 pathways. \u003cem\u003eActa Pharm. Sinica B\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e, 4390\u0026ndash;4406. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.apsb.2022.09.012\u003c/span\u003e\u003cspan address=\"10.1016/j.apsb.2022.09.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKo, C. Y., Chang, W. C. \u0026amp; Wang, J. M. Biological roles of CCAAT/Enhancer-binding protein delta during inflammation. \u003cem\u003eJ. Biomed. Sci.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e, 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12929-014-0110-2\u003c/span\u003e\u003cspan address=\"10.1186/s12929-014-0110-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNyunt, T. et al. Mitochondrial oxidative stress-induced transcript variants of ATF3 mediate lipotoxic brain microvascular injury. \u003cem\u003eFree Radic. Biol. Med.\u003c/em\u003e \u003cb\u003e143\u003c/b\u003e, 25\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.freeradbiomed.2019.07.024\u003c/span\u003e\u003cspan address=\"10.1016/j.freeradbiomed.2019.07.024\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHolland, S. D. \u0026amp; Ramer, M. S. Microglial activating transcription factor 3 upregulation: An indirect target to attenuate inflammation in the nervous system. \u003cem\u003eFront. Mol. Neurosci.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fnmol.2023.1150296\u003c/span\u003e\u003cspan address=\"10.3389/fnmol.2023.1150296\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFujihara, K. et al. Interleukin-6 in neuromyelitis optica spectrum disorder pathophysiology. \u003cem\u003eNeurol. Neuroimmunol. Neuroinflamm\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1212/nxi.0000000000000841\u003c/span\u003e\u003cspan address=\"10.1212/nxi.0000000000000841\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHirano, T. IL-6 in inflammation, autoimmunity and cancer. \u003cem\u003eInt. Immunol.\u003c/em\u003e \u003cb\u003e33\u003c/b\u003e, 127\u0026ndash;148. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/intimm/dxaa078\u003c/span\u003e\u003cspan address=\"10.1093/intimm/dxaa078\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePereira, M. \u0026amp; Gazzinelli, R. T. Regulation of innate immune signaling by IRAK proteins. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2023.1133354\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2023.1133354\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFeng, H., Zhang, Y. B., Gui, J. F., Lemon, S. M. \u0026amp; Yamane, D. Interferon regulatory factor 1 (IRF1) and anti-pathogen innate immune responses. \u003cem\u003ePLoS Pathog.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, e1009220. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.ppat.1009220\u003c/span\u003e\u003cspan address=\"10.1371/journal.ppat.1009220\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuo, Q. et al. NF-κB in biology and targeted therapy: new insights and translational implications. \u003cem\u003eSignal. Transduct. Target. Therapy\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41392-024-01757-9\u003c/span\u003e\u003cspan address=\"10.1038/s41392-024-01757-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYamamura, R. et al. Transcriptome signature in the blood of neuromyelitis optica spectrum disorder under steroid tapering. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, 1508977. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2025.1508977\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2025.1508977\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eInforzato, A. et al. PTX3 as a paradigm for the interaction of pentraxins with the complement system. \u003cem\u003eSemin Immunol.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 79\u0026ndash;85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.smim.2013.05.002\u003c/span\u003e\u003cspan address=\"10.1016/j.smim.2013.05.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, H. et al. Increased plasma levels of pentraxin 3 in patients with multiple sclerosis and neuromyelitis optica. \u003cem\u003eMult Scler.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 926\u0026ndash;931. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/1352458512457845\u003c/span\u003e\u003cspan address=\"10.1177/1352458512457845\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRen, F. J., Cai, X. Y., Yao, Y. \u0026amp; Fang, G. Y. JunB: a paradigm for Jun family in immune response and cancer. \u003cem\u003eFront. Cell. Infect. Microbiol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 1222265. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fcimb.2023.1222265\u003c/span\u003e\u003cspan address=\"10.3389/fcimb.2023.1222265\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHasan, Z. et al. JunB is essential for IL-23-dependent pathogenicity of Th17 cells. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 15628. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ncomms15628\u003c/span\u003e\u003cspan address=\"10.1038/ncomms15628\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNishiyama, S. et al. Anti-aquaporin-4 immune complex stimulates complement-dependent Th17 cytokine release in neuromyelitis optica spectrum disorders. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 3146. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-024-53661-5\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-53661-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAgasing, A. M. et al. Transcriptomics and proteomics reveal a cooperation between interferon and T-helper 17 cells in neuromyelitis optica. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 2856. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-020-16625-7\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-16625-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen, T. et al. Astrocyte-microglia interaction drives evolving neuromyelitis optica lesion. \u003cem\u003eJ. Clin. Investig.\u003c/em\u003e \u003cb\u003e130\u003c/b\u003e, 4025\u0026ndash;4038. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1172/JCI134816\u003c/span\u003e\u003cspan address=\"10.1172/JCI134816\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZipfel, P. F. \u0026amp; Skerka, C. Complement regulators and inhibitory proteins. \u003cem\u003eNat. Rev. Immunol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 729\u0026ndash;740. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nri2620\u003c/span\u003e\u003cspan address=\"10.1038/nri2620\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"NMOSD, AQP4-IgG, complement system, in vitro model, transcriptomics","lastPublishedDoi":"10.21203/rs.3.rs-7064018/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7064018/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eNeuromyelitis optica spectrum disorder (NMOSD) is a rare autoimmune disease affecting the central nervous system via autoantibodies that target the water channel aquaporin-4 (AQP4) on astrocytes. Binding to AQP4 initiates activation of innate immune components, especially the complement system. Both \u003cem\u003ein vivo\u003c/em\u003e and \u003cem\u003ein vitro\u003c/em\u003e models have been developed to investigate the molecular pathomechanisms of NMOSD. The goal of our study was to characterize the molecular response of four different human cell lines to a treatment with AQP4 antibody E5415A and human complement. We aimed to identify overlapping transcriptomic changes seen in the \u003cem\u003ein vivo\u003c/em\u003e pathophysiology of NMOSD. Tested cell lines were AQP4-ECFP overexpressing U-87MG glioblastoma cells, U-87MG expressing only ECFP, HEK293 cells transiently transfected with AQP4-EmGFP, and human primary astrocytes. Complement-dependent cytotoxicity was induced after E5415A and active human complement treatment in AQP4-expressing cells, primarily by the classical complement pathway, but also with a contribution of the alternative pathway. Transcriptomic analysis revealed that both the \u003cem\u003ein vitro\u003c/em\u003e U-87MG-AQP4-ECFP model and an \u003cem\u003ein vivo\u003c/em\u003e rat model share genes primarily involved in nuclear factor K-light-chain-enhancer of activated B cells (NF-κB) and interleukin-6 (IL6) pathways. These findings were confirmed on the mRNA and protein levels in the \u003cem\u003ein vitro\u003c/em\u003e model. As further validation, serum samples from AQP4 antibody seropositive and seronegative NMOSD patients were applied instead of E5415A on U-87MG-AQP4-ECFP cells and showed the same outcome. Additionally, NF-κB upregulation was shown by immunohistochemistry in medulla oblongata lesions of AQP4 antibody seropositive NMOSD patients. To conclude, our findings demonstrate IL6 and NF-κB pathways as major contributors to inflammation caused by complement activation in AQP4 antibody-positive NMOSD. We observed U-87MG-AQP4-ECFP cells to be a suitable model to study NMOSD pathomechanisms, as they show a gene expression profile towards NF-κB and IL6 pathway upregulation comparable with an \u003cem\u003ein vivo\u003c/em\u003e model.\u003c/p\u003e","manuscriptTitle":"mRNA Profiling of Inflammatory Stress Responses after Aquaporin-4 Antibody and Human Complement Treatment Reveals Upregulation of NF-κB and IL6 Pathways","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-17 11:08:57","doi":"10.21203/rs.3.rs-7064018/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-11T10:54:59+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-10T02:32:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"277829306274926981831909883052084917806","date":"2025-08-20T13:52:46+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-20T13:48:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-20T13:13:51+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-08-18T09:44:02+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-11T14:29:04+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-08-11T14:23:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9fbc834d-f571-4761-b8fc-653e7771de0e","owner":[],"postedDate":"September 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":54563725,"name":"Biological sciences/Cell biology"},{"id":54563726,"name":"Health sciences/Diseases"},{"id":54563727,"name":"Biological sciences/Immunology"},{"id":54563728,"name":"Health sciences/Neurology"},{"id":54563729,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2025-12-15T16:05:24+00:00","versionOfRecord":{"articleIdentity":"rs-7064018","link":"https://doi.org/10.1038/s41598-025-27335-9","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-12-08 15:59:17","publishedOnDateReadable":"December 8th, 2025"},"versionCreatedAt":"2025-09-17 11:08:57","video":"","vorDoi":"10.1038/s41598-025-27335-9","vorDoiUrl":"https://doi.org/10.1038/s41598-025-27335-9","workflowStages":[]},"version":"v1","identity":"rs-7064018","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7064018","identity":"rs-7064018","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}