Комплексный метаанализ транскриптомов астроцитов человека показывает, что в ответ на подстрекательские и иммунные стимулы потенциальные признаки повторяются.

Introduction

Astrocytes are glial cells that play essential roles in the normal functioning of the central nervous system (CNS). Their diverse functions in maintaining brain homeostasis include regulating blood flow, providing metabolic and antioxidant support to neurons, secreting pro- and anti-inflammatory molecules, regulating ionic and neurotransmitter balance, and controlling immune cell activation ( Baxter et al., 2021 ; Liddelow and Barres, 2017 ; Mahmoud et al., 2019 ; Patani et al., 2023 ). In addition, as immunocompetent cells, astrocytes can express damage-associated molecular patterns (DAMP) and pathogen-associated molecular pattern (PAMP) receptors by undergoing “reactive transformation” ( Gong et al., 2020 ).

Reactive astrogliosis represents a defensive process in which astrocytes undergo a series of changes to minimize and repair damage. During such reactive states, astrocytes experience morphological, molecular, cellular, and functional changes, reflecting the heterogeneity of their responses to various factors. For example, reactive astrocytes can exert neuroprotective effects by promoting blood–brain barrier (BBB) repair, sequestering infections via glial scarring, and facilitating metabolic support for surviving neurons ( Lawrence et al., 2023 ) However, this reactivity can also generate detrimental effects if an imbalance occurs between their neuroprotective and neurotoxic effects, which is often related to the intensity and duration of the inflammatory response ( Colombo and Farina, 2016 ; Matusova et al., 2023 ; Sofroniew, 2025 ).

Neuroinflammatory responses are driven primarily by a variety of mediators, including cytokines (e.g., IL-1β, IL-6, and TNF-α), second messengers, and reactive oxygen species. Although multiple cell types contribute to the production of these mediators, including microglia, peripheral macrophages, and neurons, astrocytes stand out for their ability to functionally integrate into these processes. Specifically, their strategic localization, functional plasticity, and capacity to amplify inflammatory signals enable them to actively participate in the initiation, modulation, and resolution of neuroinflammation, positioning them as central regulators of the neuroimmune environment ( Ben Haim et al., 2015 ). In addition to signaling, these mediators are essential for neuronal metabolism, immune survival signaling, leukocyte trafficking and other neuroinflammatory processes. Consequently, their dysregulation is a critical factor in the progression of neuroinflammation, neurodegeneration, and demyelination within the CNS ( Cieri and Ramos, 2025 ; Naveed et al., 2025 ). For example, during chronic or pathological neuroinflammation induced by infection, injury, or insult, the integrity of the BBB is compromised by immune cell infiltration, edema, cell death, and increased permeability. As neuroinflammation is associated with severe pathological conditions and may trigger autoimmune responses or neurodegenerative disorders, there is a crucial need to investigate the reactivity and molecular mechanisms underlying the response of astrocytes to different stimuli ( Hawkins and Davis, 2005 ; Michael et al., 2026 ; Monahan et al., 2008 )

Reactive astrocyte states are typically characterized by phenotypic changes, such as hypertrophy and increased immune reactivity of glial fibrillary acidic protein (GFAP) ( Cieri et al., 2023 ). While initial frameworks in murine models proposed a dichotomous classification of astrocyte reactivity, distinguishing neurotoxic (A1) and neuroprotective (A2) states, emerging transcriptomic evidence reveals a highly diverse spectrum of activation states. This molecular diversity is tailored to the nature of the stimulus, regional heterogeneity and species-specific genomic architectures ( Cunningham et al., 2019 ; González-Giraldo et al., 2021 ; Liddelow and Barres, 2017 ). Consequently, current international consensus guidelines underscore that the binary A1/A2 nomenclature oversimplifies the multidimensional nature of reactivity ( Escartin et al., 2021 )

Despite the abundance of stimulus-specific transcriptomic data, a definitive, cross-stimulus ‘core’ signature of human astrocyte reactivity remains elusive. By integrating diverse RNA-seq datasets, this meta-analysis aims to uncover a candidate recurrent inflammatory gene signature that may be shared across reactive astrocyte states, providing a basis for future investigations. To this end, we employed Robust Rank Aggregation (RRA) to synthesize transcriptomic responses derived from 11 independent human astrocyte datasets. This approach aims to establish a robust gene signature of neuroinflammation that effectively minimizes the influence of individual study bias and stimulus-specific variability, ensuring the biological reproducibility of the transcriptional features.

Methodology

Research strategy and inclusion and exclusion criteria

The study selection process was conducted according to the PRISMA 2020 guidelines ( Page et al., 2021 ), as detailed in Figure 1 . A comprehensive systematic search was conducted across multiple databases, including PubMed, Scopus, and ScienceDirect, as well as genomic repositories such as the Gene Expression Omnibus (GEO) and the European Nucleotide Archive (ENA). Specific Boolean strings were tailored for each database (e.g., PubMed: ((Astrocytes [MeSH Terms] OR ‘human astrocytes’[Title/Abstract]) AND (‘Sequence Analysis, RNA’[Mesh] OR ‘RNA-seq’[Title/Abstract] OR ‘transcriptome’[Title/Abstract] OR ‘RNAseq’[Title/Abstract])) AND (Interleukins[MeSH Terms] OR ‘cytokine’[Title/Abstract] OR ‘neuroinflammation’[Title/Abstract] OR ‘reactive astrogliosis’[Title/Abstract]) NOT (‘Neoplasms’[Mesh] OR ‘Astrocytoma’[Mesh]). The complete search strategy, including specific terms per database, is provided in Supplementary Table 1 .

To be included in the study, the following criteria were used: (1) original article published between 2017 and 2023; (2) experimental design with paired samples, where each stimulated sample was matched to its corresponding control from the same biological source; (3) model of human astrocytes derived from pluripotent stem cells, purified fetal, or commercial cell lines; (4) stimulation with molecules or agents that promote the inflammatory process; (5) transcriptomic profiling via RNA-seq; and (6) availability of raw RNA-seq data in the GEO or EBI (European Bioinformatics Institute) databases. Studies were excluded if they met any of the following criteria: (1) book chapters and reviews; (2) experimental design without paired samples; (3) tumor or other nervous tissue cell models; (4) samples from astrocytes purified from patients diagnosed with a neurodegenerative disease or disorder; (5) studies in which astrocytes had been genetically manipulated prior to the inflammatory stimulus; and (6) transcriptomic profiles obtained from microarrays or single-cell, as illustrated in Figure 1 .

For screening and selection studies, the titles and abstracts were independently reviewed by two reviewers (AL-B & KR) to identify potentially relevant studies that met the previously described eligibility criteria. Any disagreements between reviewers were resolved through consensus or with the help of a third evaluator (AF-AP).

RNA-seq data extraction and preprocessing

After articles were selected based on the search strategy and criteria, raw sequencing read files (.fastq) were downloaded from the GEO and EBI databases for each study. All bioinformatics analyses were performed using the Galaxy platform (version 25.1.1. dev0) ( Galaxy Community, 2024 ). The complete computational workflow, including tool versions and parameters, is available as a Galaxy workflow file (.ga format) in the GitHub repository 1 . An initial quality analysis of the reads was conducted to determine the presence of adapters and overrepresented reads using the FASTQC tool v0.12.1 ( Andrews, 2010 ). Adapter trimming and low-quality reads were filtered out via Trimmomatic v0.39 ( Bolger et al., 2014 ). Subsequently, ribosomal RNAs (rRNAs) were removed with the SortMeRNA tool v4.3.6 using the rRNA, Rfam and SILVA databases ( Supplementary Table 2 ) ( Kopylova et al., 2012 ). Nonribosomal RNAs were aligned to the reference genome obtained from GENCODE (GRCh38 version 44) with the STAR tool v2.7.10b ( Dobin et al., 2013 ). Gene-level counts from reads mapped to the genome were quantified via the HTSeq-count package ( Anders et al., 2015 ). Differential expression analysis was performed with the Limma-Voom package ( Law et al., 2014 ). The trimmed mean of M values (TMM) normalization method was employed to correct for differences in overall expression. For each dataset, a design matrix based on the experimental condition was constructed, and treated samples were compared against their corresponding controls. The Benjamini–Hochberg method was employed to calculate adjusted p -values. This workflow was performed to minimize the degree of technical heterogeneity between studies.

Robust rank aggregation (RRA)

For the Robust Rank Aggregation (RRA) analysis, genes were ranked according to log fold change (logFC) values obtained using the Limma-Voom package. Datasets containing multiple treatment conditions contributed one ranked gene list for each treatment-versus-control comparison. The ranked gene lists were then aggregated using the “RobustRankAggreg” package (version 1.2.1) ( Kolde et al., 2012 ). This method employs an adjusted p-value to assess the likelihood that differentially expressed genes (DEGs) are consistently identified across datasets with highly ranked genes. Genes with an adjusted p -value < 0.05 and | log2FC| > 1 were considered significant DEGs in the RRA analysis. This method was selected based on the criteria that conditions of the experimental analysis varied. Differentially expressed genes were visualized using Morpheus tool 2 , which were stratified according to astrocyte source, including human induced pluripotent stem cell-derived astrocytes, fetal and primary astrocytes.

Functional enrichment and pathway analysis

Gene Ontology (GO) and pathway enrichment analysis were performed for the identified DEGs from the RRA analysis (as defined by adjusted p -value < 0.05 and | log2FC| > 1, see RRA section), using gProfiler tool 3 ( Kolberg et al., 2023 ). The enrichment calculations were performed using the default background implemented by g:Profiler for the selected organism. g:Profiler analysis was performed including GO biological process, Reactome and KEGG pathways. Statistical significance was assessed using the g:SCS multiple testing correction method, with a significance threshold of adjusted p < 0.05. Additionally, the REVIGO tool 4 was used to remove redundant GO terms, emphasize representative terms that group semantically similar terms, and prioritize statistically significant terms ( Supek et al., 2011 ). From this clustering, biological processes related to immune and inflammatory responses were selected to identify the genes directly involved in these responses.

Protein–protein interaction network

A protein–protein interaction (PPI) network was constructed via STRING (v.12.0) ( Szklarczyk et al., 2023 ) as a downstream exploratory analysis to prioritize highly connected genes within the candidate inflammatory and immune-associated signature. All possible interaction sources that STRING database offers were included; these were “Text mining”, “Experiments”, “Databases”, “Co-expression”, “Neighborhood”, “Gene Fusion” and “Co-occurrence”. A minimum STRING confidence score of 0.4 was used, retaining interactions with confidence scores ≥ 0.4. The PPIs reported in STRING were visualized via Cytoscape (v.3.10.2) ( Shannon et al., 2003 ) software. Key centrality measures including degree, betweenness centrality, eccentricity, average shortest path length, clustering coefficient and topological coefficient, were calculated. Network topology analysis was performed to identify candidate central nodes using betweenness centrality and degree distribution metrics. The degree method was employed to select nodes with the highest connectivity and betweenness centrality to quantify the number of times that a node lies on the shortest path connecting two other nodes; nodes with a high betweenness are considered more biologically informative that are responsible for transferring communication information ( Opsahl et al., 2010 ).

Additionally, to identify functional modules within the network, the Molecular Complex Detection (MCODE) plug-in (v.2.0.3) ( Bader and Hogue, 2003 ) was applied. This approach enabled to find functional subunits that may represent biological mechanisms. The parameters used for MCODE analysis, as default, were: degree cutoff = 2, node score cutoff = 0.2, k-core = 2 and max depth = 100. An enrichment analysis was performed using the STRING database. A 0.05 False Discovery Rate (FDR) threshold was applied for statistical significance and only terms with a minimum signal strength of 0.01 were included. Terms were considered only if at least two genes were present in the network. Enriched terms were grouped based on a similarity threshold of ≥0.8.

Results

Following the search strategy, 11 RNA-Seq datasets were selected and are detailed in Table 1 . In total, 153 samples were analysed, comprising 91 stimulated samples and 62 controls. The stimuli used included proinflammatory molecules (TNF-α, IL-6, IL-1β, palmitic acid, and FBS) as well as pathogens such as viruses and bacteria. The astrocytes used in this study were derived from diverse sources, including induced pluripotent stem cell (iPSC)-derived astrocytes as well as primary and fetal astrocytes, as detailed in Table 1 .

Read processing and mapping

After data cleaning and preprocessing, we observed that most samples were sequenced using Illumina HiSeq or Illumina NovaSeq platforms. Of the eleven datasets analyzed, seven employed a single-end sequencing strategy, while four used paired-end sequencing. On average, 32.7 million raw reads were obtained per sample, with an average GC content of 49% and read lengths ranging from 50 to 150 bp. After quality filtering (Q > 30), approximately 91% of the reads survived trimming with Trimmomatic. On average, only 1.87% of reads exceeded the E -value threshold in SortMeRNA, indicating a low level of ribosomal contamination, while 98.13% were classified as non-ribosomal. Mapping with STAR showed that, on average, 83.68% of the reads aligned uniquely to the reference genome, 9.18% mapped to multiple loci, 0.75% mapped to too many loci, and 6.4% remained unmapped (1.22% due to short length and 5.18% remained unmapped due to insufficient similarity to the reference or high biological noise. In total, 93.6% of the reads were successfully mapped. Gene quantification using HTSeq-count yielded an average of 6.17 million reads assigned to annotated genes per sample. These results ( Supplementary Table 3 ) indicate that the RNA-seq data were of high quality, with low ribosomal content, good mapping performance, and reliable gene expression quantification across datasets. Gene-level counts were analyzed using limma-voom for differential expression within each dataset.

Robust rank aggregation analysis

The RRA analysis was performed separately for up- and down-regulated genes using 22 ranked gene lists per regulation direction, generated from individual treatment-versus-control contrasts ( Supplementary Table 4 ). From RRA were identified a total of 130 DEGs, of which 125 were upregulated and 5 were downregulated ( Supplementary Table 5 ). The top 10 significantly upregulated genes were SOD2, CXCL8, WAKMAR2, ELF3, VCAM1, PARP9, BIRC3, NFKBIA , and novel lncRNA (ENSG00000288528), and the 5 downregulated genes were IL17D, GDF10, REEP1, PSMB7 and SKP2. The expression patterns of these genes were visualized using heatmaps generated in Morpheus, which were stratified according to astrocyte source, including human induced pluripotent stem cell-derived astrocytes, fetal and primary astrocytes (Supplementary Figure 1 ).

Functional enrichment and pathway analysis

The five downregulated genes identified ( IL17D, GDF10, REEP1, PSMB7 , and SKP2 ) were assessed for functional relevance; however, they did not show significant clustering within biological pathways related to inflammatory processes. In contrast, functional enrichment analysis of the 125 upregulated genes revealed significant activation of immune-related pathways, with prominent examples including “cytokine signaling in the immune system” and “interferon alpha/beta signaling” pathways. Additionally, pathways linked to viral infections, including SARS-CoV-1, influenza A, and COVID-19, were identified, indicating potential modulation of immune responses in these contexts. Other relevant pathways, including interleukin signaling, TNF, NF-kB, and chemokine receptor interactions, were also implicated, underscoring the critical role of immune pathway regulation in response to inflammatory stimuli ( Figure 2 ).

Following the identification of 241 enriched GO-BP terms underlying the astrocytic response, we employed the REVIGO tool to resolve high semantic redundancy and identify representative functional clusters. By applying semantic similarity clustering algorithms, the list was condensed to 26 key biological processes ( Figure 3 ). The 26 GO-BP terms encompass a diverse range of processes, including the regulation of immune system processes, cytokine production, response to stimuli, viral processes, and cell death. This analysis identified two main clusters: immune response and inflammatory response.

The biological processes associated with each cluster and their corresponding genes were then analyzed. Subsequently, genes shared between both clusters were identified, and a score was calculated based on the frequency with which each gene was associated with the different processes. Based on these scores, only genes with a relevant representation in one or both clusters were selected. This approach reduced the initial gene set to 33 genes, which were subsequently used for the construction of the protein–protein interaction (PPI) network.

Protein–protein interaction network

Because GO enrichment and REVIGO analyses consistently highlighted immune- and inflammation-related biological processes, we performed a targeted downstream analysis to identify a candidate immune-inflammatory core. Specifically, the 130 DEGs were filtered for genes annotated to immune and inflammatory response processes, resulting in a subnetwork of 33 nodes. These genes were then used to construct and visualize the PPI network via the STRING database and Cytoscape software. The resulting network ( Figure 4 ) comprised 33 nodes and 200 interaction edges with an average node degree of 12.1, a cluster coefficient of 0.711, a network diameter of 3, and a p -value enrichment of less than 1.0e-16. To ensure the mathematical robustness of the inflammatory core, key nodes were defined using a multi-parametric topological filter. Nodes were prioritized if they ranked within top 15% of the distribution for both degree and betweenness centrality metrics. Based on these criteria, five primary nodes were identified: CXCL10, DDX58, IFIH1, IL-1B, and TLR3. These nodes exhibited the highest connectivity and centrality, suggesting their role as candidate hub nodes within the astrocyte-mediated inflammatory response. Comprehensive network parameters for each node, including eccentricity, average shortest path length, and topological coefficients, are detailed in Supplementary Table 6 .

Additionally, we used the MCODE plug-in to investigate whether associated proteins might form a highly connected molecular module. The MCODE clustering analysis identified three connected modules, ranging from 16 to 5 nodes. Module 1 was the most interconnected cluster (MCODE score = 13.87) with 104 edges and 16 nodes, identifying TLR3 as the seed node ( Figure 5A ). Enrichment analysis performed using STRING database revealed that the biological processes of this cluster are primarily associated with the activation of antiviral innate immune responses. Module 2 (MCODE score = 4.57) consisted of 16 edges and 8 nodes, with B2M as seed protein ( Figure 5B ). Enrichment analysis demonstrated biological process primarily involved in antigen processing and presentation via MHC class I, suggesting an activation of adaptive immune responses. Finally, the Module 3 (MCODE score = 3.5) showed fewer nodes and consisted of 5 nodes and 7 edges, with BIRC2 as seed ( Figure 5C ). Based on the enrichment analysis, the biological process identified focused on the activation of cytokine signaling and NF-κB pathway.

Discussion

Astrocytes, the most abundant glial cells in the CNS, play a critical role in maintaining physiological homeostasis and responding to injuries or infections. While they remain in a homeostatic state under normal conditions, sensing pathological cues triggers a ‘reactive transformation’ characterized by finely tuned molecular and functional shifts ( Kim et al., 2026 ). In inflammatory settings, these immunocompetent cells are essential for promoting both innate and adaptive immune responses. Specifically, they induce the expression of major histocompatibility complex class II (MHC-II) molecules and costimulatory signals, aiding peripheral immune cell recruitment ( Le Thuc and García-Cáceres, 2024 ; Xie and Yang, 2015 ). Various inflammatory stimuli have been studied to clarify how astrocytes regulate these immune functions. For example, stimulation with LPS and Poly(I:C) triggers a significantly stronger innate immune response and effector molecule production than cytokine stimulation alone. Furthermore, pattern recognition receptor (PRR) activation by distinct viral or bacterial ligands modulates specific chemokine profiles according to pathogen type, underscoring how astrocytes tailor their responses to specific stimulus ( McKimmie and Graham, 2010 ).

By employing Robust Rank Aggregation (RRA), we identified a candidate recurrent immune and inflammatory signature of 130 genes across 11 highly heterogeneous datasets. Within this signature, 125 genes were upregulated in response to a broad range of biological and biochemical proinflammatory stimuli ( Supplementary Table 5 ). A large portion of these genes are associated with immune responses: 47 are specifically involved in immune regulation and 20 are linked to the induction of interferon alpha, beta, or gamma. These results align with previous studies that emphasized the role of astrocytes in the innate immune response, where they exhibit neuroprotective functions. To explore the biological significance of these 125 DEGs, we performed KEGG and REACTOME pathway enrichment analyses. These analyses revealed enrichment in key inflammatory pathways, including TNF (14 genes), NF-κB (15 genes), and NOD-like receptor (15 genes) signaling.

Previous studies have described enrichment in pathways related to immune system activation, antiviral innate immunity, major histocompatibility complex, cytokine/interferon signaling, and the inflammatory response through TNF signaling, NOD-like receptor, and NF-κB signaling in astrocytes exposed to similar conditions ( Fields et al., 2022 ; Kim et al., 2022 ; Kong et al., 2022 ; LaRocca et al., 2021 ; Li et al., 2021 ; Rojas-Cruz et al., 2023 ; Santos et al., 2017 ; Vadodaria et al., 2021 ). Consistent with these observations, these pathways were also significantly enriched among the differentially expressed genes identified in our analysis. However, it is important to note that pathway enrichment analyses derived from transcriptomic data reflect coordinated changes in gene expression rather than direct measurements of protein abundance or signaling activity. Therefore, these results should be interpreted as indicators of transcriptional regulation of biological processes rather than definitive evidence of pathway activation.

TNF-α is a principal cytokine mediator regulating innate and adaptive immunity. Studies suggest that TNF-α can drive astrogliosis and play an integral role in the pathogenesis of various neurological disorders ( Gao et al., 2024 ; Kim et al., 2022 ). NOD-like receptors (NLRs) are a specialized group of intracellular proteins that play pivotal roles in sensing molecules associated with intracellular infection and stress triggering signaling cascades that promote pathogen clearance. These receptors serve as scaffolding proteins that ease the assembly of diverse signaling complexes and modulate the activation of inflammatory caspases ( Jorgačevski and Potokar, 2023 ). For instance, NOD1 and NOD2 in primary murine astrocytes recognize motifs of bacterial peptidoglycan and mediate IL-6 secretion in response to specific ligands ( Giovannoni and Quintana, 2020 ; Sterka et al., 2006 ).

Notably, the NF-κB pathway was enriched with 15 genes, including VCAM1, TNFAIP3, BIRC3, NFKBIA, CXCL2, ICAM1, CXCL5, CXCL1 , IL-1β , CXCL10, CSF1, NFKB1, BIRC2, CXCL3 , and TRAF1 . In the CNS, this pathway plays diverse roles depending on the physiological context, serving as a key regulator of the immune response, cell proliferation, and apoptosis ( Wang et al., 2022 ). The NF-κB complex consists of five subunits that assemble in various combinations, and its activity is modulated by post-translational modifications such as O-GlcNAcylation and phosphorylation ( Dong et al., 2023 ; Ridder and Schwaninger, 2009 ). NF-κB translocation affects cells differently depending on the location and duration of the signaling event. By promoting either pro-survival or pro-apoptotic states, it can ultimately exert neuroprotective or neurotoxic effects ( Pozniak et al., 2014 ).

Functionally, NF-κB-regulated genes are primarily involved in immunoregulatory, inflammatory, and proliferative processes. Among the enriched genes, we found TNFAIP3 which encodes a deubiquitinating enzyme that negatively regulates the TNF-induced NF-κB proinflammatory signaling pathway ( Catrysse et al., 2014 ). Interestingly, NF-κB not only drives the expression of proinflammatory genes but also induces the transcription of its own inhibitors, such as TNFAIP3. This protein, deubiquitinates RIP1 and IKKγ, and inhibits the NF-κB-induced responses. This process forms a negative feedback loop that limits excessive inflammatory responses ( Ruland, 2011 ).

Specifically, in astrocytes, NF-κB plays crucial roles in synaptic glutamate clearance, metabolic regulation and structural plasticity ( Li et al., 2024 ; Pozniak et al., 2014 ). For example, NF-κB activation by TNF-α upregulates the production of nerve growth factor (NGF) and glia-derived neurotrophic factor (GDNF), highlighting its neuroprotective potential ( Pozniak et al., 2014 ). However, under pathological conditions, chronic NF-κB signaling is associated with the transcription the transcription of downstream target genes, including TNF-α, IL-1β, IL-6 and iNOS, which promotes the expression and release of pro-inflammatory factors ( Xu et al., 2021 ). Prolonged NF-κB activation can impair neuronal function and alter dendritic arborization morphology, contributing to adverse neurological outcomes ( Kaltschmidt et al., 2005 ; Pozniak et al., 2014 ). In other mammalian species such as rodents, a chronic activation of astrocytic NF-κB promotes a loss of mitochondrial-associated proteins and the rise of inflammatory-related proteins ( Jong Huat et al., 2024 ).

Previous studies have consistently identified NF-κB as a major regulator of the pro-inflammatory phenotype adopted by reactive astrocytes ( Hasel et al., 2021 ; Jong Huat et al., 2024 ; Leng et al., 2022 ). In addition to its canonical function as a transcription factor, NF-κB act as a critical signal transduction mediator, linking extracellular inflammatory cues to the activation of downstream gene expression programs that shape astrocyte responses ( Kaltschmidt et al., 2005 ; Pozniak et al., 2014 ).

Based on the 33 nodes of interest related to immune response and inflammation, a PPI network was constructed. The top five hub nodes with the highest degree of connectivity were CXCL10, DDX58, IFIH1, IL-1β, and TLR3. A remarkable finding of this meta-analysis is the consistent upregulation of the cytosolic sensors DDX58 (RIG-I) and IFIH1 (MDA5) across diverse non-viral inflammatory contexts, including stress induced by palmitic acid and systemic pro-inflammatory cytokine exposure. This could suggest that astrocytes adopt a phenotypic state of antiviral mimicry, engaging in ‘sterile’ inflammatory responses. Mechanistically, this state is likely orchestrated by the sensing of endogenous retroelements or mitochondrial DNA/RNA release under stress ( Dhir et al., 2018 ; Russ and Iordanskiy, 2023 ), as seen in other neurological contexts ( Blank and Prinz, 2017 ) Specifically, palmitic acid-induced mitochondrial impairment triggers metabolic and immune dysfunction, ultimately leading to antiviral mimicry. ( Rojas-Cruz et al., 2023 ).

Both DDX58 and IFIH1, members of the RIG-I-like receptor family, play critical roles in the immune response by mediating the transcriptional induction of type I interferons. Although expression of these interferons is low under homeostatic conditions, it is significantly upregulated upon innate immune activation. Specifically, DDX58 acts as a double-stranded RNA receptor that, upon activation, triggers innate immune responses through mitochondrial antiviral signaling and the induction of interferon synthesis ( Rehwinkel and Gack, 2020 ; Yoneyama et al., 2004 ). This interferon pathway, which was enriched in our analysis, is critical for promoting cell death in infected cells, activating adaptive immunity, and supporting hematopoietic stem cell renewal and proliferation ( Wilkins and Gale, 2010 ). In neurotropic viral models generated from human astrocytes, DDX58 expression is constitutive and increases following infection, highlighting its role as an intracellular viral sensor ( Furr and Marriott, 2012 ; Potokar et al., 2023 ). Additionally, the activation of RIG-I signaling in brain regions of patients with mild cognitive impairment may be involved in the accumulation of amyloid precursor proteins and in the exacerbated production of cytokines. This supports the hypothesis that this receptor is involved in the early events related to progression to Alzheimer’s disease ( de Rivero Vaccari et al., 2014 ). In nonneuronal cells, DDX58 activation enhances autophagy, aiding in the clearance of lipid bodies that contribute to inflammation and subsequent apoptosis ( Frietze et al., 2022 ). Moreover, IFIH1, which encodes the pattern recognition receptor MDA5, plays a complementary role by activating the TLR3 pathway via NF-κB, generating a distinct interferon signature. This activity is consistent with studies showing that astrocytes respond to the synthetic double-stranded RNA Poly(I:C) ( De Miranda et al., 2009 ).

Like IFIH1, DDX58 is involved in canonical PRR signaling pathways, which interact by activating Toll-like receptor (TLR) pathways. In astrocytes, the expression of functional Toll-like receptors has been well documented, and their activation leads to the upregulation of genes encoding inflammatory mediators ( Krasowska-Zoladek et al., 2007 ). Therefore, these receptors are considered key regulators of the innate immune response and are implicated in the progression of neuropathologies such as Alzheimer’s disease, Parkinson’s disease and amyotrophic lateral sclerosis ( Kinsella et al., 2018 ). Members of the TLR family activate two primary signaling pathways, both of which depend on myeloid differentiation primary response gene 88 (MyD88), with the notable exception of TLR3, which utilizes an alternative signaling mechanism.

TLR3 has been described as a receptor expressed by human astrocytes under basal culture conditions. In addition to being activated by double-stranded RNA (dsRNA), TLR3 can also recognize endogenous mRNAs ( Jack et al., 2005 ). Once activated, TLR3, along with the recruitment of other molecules such as TRIF (interferon inducer), initiates a signaling cascade that leads to the production of IFN-β and the activation of MAP kinases and NF-κB. This cascade promotes inflammation-related pathways, including those involved in glioma tumor progression ( Krasowska-Zoladek et al., 2007 ; Li et al., 2017 ). In addition, TLR3 activation has been found to be related to cell death progression, with sex-dependent differences, where female cells induce apoptosis and male cells induce necrotic apoptosis ( Chavez-Valdez et al., 2019 ).

Although TLR3 is expressed in various cell types, with its activation eliciting distinct cellular responses in each context, its activation in cultured astrocytes leads to prolonged inflammation ( Seo et al., 2015 ). This inflammation results in the upregulation of several genes encoding cytokines, such as IL-1β, IL-1α, IL-6, and TNF-α, as well as chemokines, such as CXCL10, growth factors, enzymes, other inflammatory mediators and TLR receptors ( Borysiewicz et al., 2009 ; Chavez-Valdez et al., 2019 ; Jack et al., 2005 ; Krasowska-Zoladek et al., 2007 ).

Our meta-analysis identified CXCL10 as a top-ranked hub node with high connectivity. Previous studies suggest it acts as a primary gatekeeper for the CXCR3-mediated recruitment of Th1 and NK cells during the astrocyte-driven neuroinflammatory response ( Elemam et al., 2022 ; McKimmie and Graham, 2010 ). The CXCL10-CXCR3 axis functions as a regulatory interface between chemotactic recruitment and cellular reprogramming, ensuring that infiltrated Th1 and NK cells undergo metabolic and transcriptional preconditioning to withstand the neuroinflammatory environment ( Bantug et al., 2018 ; Bufi et al., 2025 ; Corrado and Pearce, 2022 ; Koch et al., 2009 ). Transcriptionally, CXCL10 is regulated by external stimuli, including IFN-γ, TNF-α, and LPS, through NF-κB-dependent pathways ( Vazirinejad et al., 2014 ). Its expression has been observed in neurons, glial cells, and stromal cells, where it can exert either protective or detrimental effects, depending on the etiology and context of neuroinflammation ( McKimmie and Michlmayr, 2014 ).

Specifically, CXCL10 secretion from astrocytes has been shown to mediate T-cell recruitment to the brain under conditions such as multiple sclerosis and CNS viral infections ( Cheng and Chen, 2014 ; Phares et al., 2013 ; Sørensen et al., 2002 ). Furthermore, during neuroinflammation induced by SCI, CXCL10 may play a role in disrupting the integrity of the blood–brain barrier, thereby fostering an environment that promotes the accumulation of peripheral immune cells and the amplification of immune responses. In this context, the release of CXCL10 by astrocytes may adversely affect oligodendrocytes and inhibit neuronal regeneration by attracting inflammatory cells, although the underlying molecular mechanisms remain to be fully elucidated ( Qiao et al., 2022 ).

The critical and nonredundant roles of CXCL10 in neuroinflammation are controversial and have been examined across various models. For example, studies involving viral encephalitis, such as those involving multiple virus infections, such as mouse hepatitis virus (MHV) and West Nile virus (WNV), suggest that CXCL10 plays a beneficial role by enhancing the clearance of infected cells. Conversely, CXCL10-induced inflammation has been shown to be detrimental in conditions such as malaria encephalitis and trypanosomiasis ( Amin et al., 2009 ; Campanella et al., 2008 ; Christensen et al., 2009 ). While numerous studies have linked CXCL10 expression to viral responses in the CNS, CXCL10 binding to the CXCR3 receptor promotes Th1 responses, generating positive feedback loops and facilitating the recruitment of microglia. Additionally, other studies have indicated that increased CXCL10 expression acts as an inflammatory mediator during the progression of neurodegenerative diseases and is correlated with the activation of ERK1/2 signaling pathways that mediate glial–neuronal interactions ( Koper et al., 2018 ; Liu et al., 2014 ; Sui et al., 2006 ). Indeed, the overexpression of CXCL10 has been associated with neurotoxic processes attributed to increased intracellular calcium influx into mitochondria, which promotes membrane permeabilization and the subsequent release of cytochrome c, ultimately initiating caspase activation and leading to apoptosis ( Sui et al., 2006 ).

Finally, IL-1β is recognized as one of the primary proinflammatory cytokines and plays a pivotal role in both the immune system and the central nervous system ( Allan et al., 2005 ). Its presence has been associated with the maintenance of neuroinflammation, which contributes to the pathogenesis of Parkinson’s disease ( Koprich et al., 2008 ). Pharmacological studies have validated its role in the progression of neuroretinal inflammatory responses, which lead to neural tissue damage ( Dabouz et al., 2020 ). In Alzheimer’s disease patients, IL-1β has been identified alongside IL-1α, indicating the presence of active and persistent inflammation at the tissue level ( Italiani et al., 2018 ). Under physiological conditions, particularly during aging, increased expression levels of TNF-α, IL-1β, and IL-6 have been observed in rat astrocytes via immunofluorescence techniques, a phenomenon not observed in neurons or microglia. These findings suggest that the astrocyte-driven loss of tissue homeostasis, characterized by a cytokine storm in the aging brain, may promote the development of neurodegenerative diseases ( Willis et al., 2020 ).

The genes and pathways identified in this analysis provide a basis for generating hypotheses regarding astrocyte-associated inflammatory programs under pathological conditions. Rather than established therapeutic targets, mediators such as IL-1β, CXCL10, and NF-κB-related genes represent recurrent components of transcriptional inflammatory signatures across heterogeneous datasets. Further experimental studies are required to determine their functional relevance and potential implications in astrocyte-mediated neuroinflammation.

Overall, the findings of this study highlight the central role of astrocytes as immunocompetent mediators in response to inflammatory stimuli. The differential expression of genes involved in pathways such as TNF, NF-κB, and NOD-like receptors might reflects an activation of the innate immune system, characterized by the upregulation of proinflammatory cytokines and chemokines such as IL-1β and CXCL10, as well as viral RNA sensors like DDX58 and IFIH1. These results reinforce the active role of astrocytes in detecting damage- or pathogen-associated signals and in modulating the neuroimmune environment, positioning them as key contributors to brain inflammatory responses.

Several important limitations should be considered when interpreting these findings. Although the robust rank aggregation approach facilitates the identification of genes consistently regulated across independent datasets, the resulting signature may still be influenced by the characteristics of the studies included in the meta-analysis. In our analysis, the datasets comprised different astrocyte sources: iPSC-derived, fetal, and primary cells. While primary astrocytes exhibited a polarized response across a wide range of stimuli, fetal and iPSC-derived astrocytes showed higher sample-to-sample and stimulus-specific heterogeneity in their expression profiles ( Supplementary Figure 1 ). Consequently, these differences in astrocyte origin, developmental maturity, experimental conditions, and dataset composition may shape the observed transcriptional patterns. Additionally, the inclusion of multiple treatment-control contrasts from the same datasets introduces a potential bias, as certain studies or stimulus categories may be overrepresented in the final RRA signature. For instance, because several viral and pathogen related datasets were included, interferon and antiviral signaling programs likely exerted an influence on the transcriptional profile.

The candidate inflammatory signature reported here should be interpreted within the context of the analyzed datasets and may not fully capture the diversity of astrocyte responses across all biological conditions. In addition, it is important to note that the candidate hub nodes identified based on network connectivity should not be interpreted as drivers of astrocyte inflammatory responses and have not been independently experimentally validated.

Edited by: Daniel Ortuño-Sahagún , University of Guadalajara, Mexico

Reviewed by: Erdem Tüzün , Istanbul University, Türkiye

Amanda Herrero González , University of León, Spain

https://github.com/karemra/Meta-analysis-astrocytes.git .

https://software.broadinstitute.org/morpheus/ .

http://biit.cs.ut.ee/gprofiler/ .

http://revigo.irb.hr .

Data availability statement

The original contributions presented in the study are included in the article/ Supplementary material , further inquiries can be directed to the corresponding author.

Author contributions

AL-B: Data curation, Methodology, Visualization, Conceptualization, Formal analysis, Writing – original draft, Writing – review & editing. KR-A: Visualization, Data curation, Methodology, Conceptualization, Writing – original draft, Writing – review & editing, Formal analysis. AA-P: Methodology, Conceptualization, Formal analysis, Writing – original draft, Supervision, Writing – review & editing. JG: Writing – review & editing, Methodology, Supervision, Conceptualization, Funding acquisition.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fncel.2026.1870142/full#supplementary-material

References

Allan S. M. Tyrrell P. J. Rothwell N. J. ( 2005 ). Interleukin-1 and neuronal injury. Nat. Rev. Immunol. 5 629 – 640 . 10.1038/nri1664 16034365

Amin D. N. Rottenberg M. E. Thomsen A. R. Mumba D. Fenger C. Kristensson K. et al . ( 2009 ). Expression and role of CXCL10 during the encephalitic stage of experimental and clinical African trypanosomiasis. J. Infect. Dis. 200 1556 – 1565 . 10.1086/644597 19827943

Anders S. Pyl P. T. Huber W. ( 2015 ). HTSeq–a Python framework to work with high-throughput sequencing data. Bioinformatics 31 166 – 169 . 10.1093/bioinformatics/btu638 25260700 PMC4287950

Andrews S. ( 2010 ). FastQC A Quality Control tool for High Throughput Sequence Data. United Kingdom : Babraham Institute .

Bader G. D. Hogue C. W. ( 2003 ). An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinform. 4 : 2 . 10.1186/1471-2105-4-2 12525261 PMC149346

Bantug G. R. Galluzzi L. Kroemer G. Hess C. ( 2018 ). The spectrum of T cell metabolism in health and disease. Nat. Rev. Immunol. 18 19 – 34 . 10.1038/nri.2017.99 28944771

Baxter P. S. Dando O. Emelianova K. He X. McKay S. Hardingham G. E. et al . ( 2021 ). Microglial identity and inflammatory responses are controlled by the combined effects of neurons and astrocytes. Cell Rep. 34 : 108882 . 10.1016/j.celrep.2021.108882 33761343 PMC7994374

Ben Haim L. Carrillo-de Sauvage M. A. Ceyzériat K. Escartin C. ( 2015 ). Elusive roles for reactive astrocytes in neurodegenerative diseases. Front. Cell Neurosci. 9 : 278 . 10.3389/fncel.2015.00278 26283915 PMC4522610

Blank T. Prinz M. ( 2017 ). Type I interferon pathway in CNS homeostasis and neurological disorders. Glia 65 1397 – 1406 . 10.1002/glia.23154 28519900

Bolger A. M. Lohse M. Usadel B. ( 2014 ). Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30 2114 – 2120 . 10.1093/bioinformatics/btu170 24695404 PMC4103590

Borysiewicz E. Fil D. Konat G. W. ( 2009 ). Rho proteins are negative regulators of TLR2. TLR3, and TLR4 signaling in astrocytes. J. Neurosci. Res. 87 1565 – 1572 . 10.1002/jnr.21968 19115402

Bufi A. A. Di Stefano J. Papait A. Silini A. R. Parolini O. Ponsaerts P. ( 2025 ). The central role of CXCL10-CXCR3 signaling in neuroinflammation and neuropathology. Cytokine Growth Factor Rev. 84 20 – 34 . 10.1016/j.cytogfr.2025.05.003 40473519

Campanella G. S. Tager A. M. El Khoury J. K. Thomas S. Y. Abrazinski T. A. Manice L. A. et al . ( 2008 ). Chemokine receptor CXCR3 and its ligands CXCL9 and CXCL10 are required for the development of murine cerebral malaria. Proc. Natl. Acad. Sci. U. S. A. 105 4814 – 4819 . 10.1073/pnas.0801544105 18347328 PMC2290783

Catrysse L. Vereecke L. Beyaert R. van Loo G. ( 2014 ). A20 in inflammation and autoimmunity. Trends Immunol. 35 22 – 31 . 10.1016/j.it.2013.10.005 24246475

Chavez-Valdez R. Mottahedin A. Stridh L. Yellowhair T. R. Jantzie L. L. Northington F. J. et al . ( 2019 ). Evidence for sexual dimorphism in the response to TLR3 activation in the developing neonatal mouse brain: A pilot study. Front. Physiol. 10 : 306 . 10.3389/fphys.2019.00306 30971945 PMC6443881

Cheng W. Chen G. ( 2014 ). Chemokines and chemokine receptors in multiple sclerosis. Med. Inflamm. 2014 : 659206 . 10.1155/2014/659206 24639600 PMC3930130

Christensen J. E. Simonsen S. Fenger C. Sørensen M. R. Moos T. Christensen J. P. et al . ( 2009 ). Fulminant lymphocytic choriomeningitis virus-induced inflammation of the CNS involves a cytokine-chemokine-cytokine-chemokine cascade. J. Immunol. 182 1079 – 1087 . 10.4049/jimmunol.182.2.1079 19124751

Cieri M. B. Ramos A. J. ( 2025 ). Astrocytes, reactive astrogliosis, and glial scar formation in traumatic brain injury. Neural Regenerat. Res. 20 973 – 989 . 10.4103/NRR.NRR-D-23-02091 38989932 PMC11438322

Cieri M. B. Villarreal A. Gomez-Cuautle D. D. Mailing I. Ramos A. J. ( 2023 ). Progression of reactive gliosis and astroglial phenotypic changes following stab wound-induced traumatic brain injury in mice. J. Neurochem. 167 183 – 203 . 10.1111/jnc.15941 37592830

Colombo E. Farina C. ( 2016 ). Astrocytes: Key regulators of neuroinflammation. Trends Immunol. 37 608 – 620 . 10.1016/j.it.2016.06.006 27443914

Corrado M. Pearce E. L. ( 2022 ). Targeting memory T cell metabolism to improve immunity. J. Clin. Invest. 132 : e148546 . 10.1172/JCI148546 34981777 PMC8718135

Cunningham C. Dunne A. Lopez-Rodriguez A. B. ( 2019 ). Astrocytes: Heterogeneous and dynamic phenotypes in neurodegeneration and innate immunity. Neuroscientist 25 455 – 474 . 10.1177/1073858418809941 30451065 PMC6525076

Dabouz R. Cheng C. W. H. Abram P. Omri S. Cagnone G. Sawmy K. V. et al . ( 2020 ). An allosteric interleukin-1 receptor modulator mitigates inflammation and photoreceptor toxicity in a model of retinal degeneration. J. Neuroinflamm. 17 : 359 . 10.1186/s12974-020-02032-8 33246504 PMC7694438

De Miranda J. Yaddanapudi K. Hornig M. Lipkin W. I. ( 2009 ). Astrocytes recognize intracellular polyinosinic-polycytidylic acid via MDA-5. FASEB J. 23 1064 – 1071 . 10.1096/fj.08-121434 19036857 PMC2660646

de Rivero Vaccari J. P. Dietrich W. D. Keane R. W. ( 2014 ). Activation and regulation of cellular inflammasomes: Gaps in our knowledge for central nervous system injury. J. Cereb. Blood Flow Metab. 34 369 – 375 . 10.1038/jcbfm.2013.227 24398940 PMC3948131

Dhir A. Dhir S. Borowski L. S. Jimenez L. Teitell M. Rötig A. et al . ( 2018 ). Mitochondrial double-stranded RNA triggers antiviral signalling in humans. Nature 560 238 – 242 . 10.1038/s41586-018-0363-0 30046113 PMC6570621

Dobin A. Davis C. A. Schlesinger F. Drenkow J. Zaleski C. Jha S. et al . ( 2013 ). STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29 15 – 21 . 10.1093/bioinformatics/bts635 23104886 PMC3530905

Dong X. Shu L. Zhang J. Yang X. Cheng X. Zhao X. et al . ( 2023 ). Ogt-mediated O-GlcNAcylation inhibits astrocytes activation through modulating NF-κB signaling pathway. J. Neuroinflamm. 20 : 146 . 10.1186/s12974-023-02824-8 37349834 PMC10286367

Elemam N. M. Talaat I. M. Maghazachi A. A. ( 2022 ). CXCL10 chemokine: A critical player in RNA and DNA viral infections. Viruses 14 : 2445 . 10.3390/v14112445 36366543 PMC9696077

Escartin C. Galea E. Lakatos A. O’Callaghan J. P. Petzold G. C. Serrano-Pozo A. et al . ( 2021 ). Reactive astrocyte nomenclature, definitions, and future directions. Nat. Neurosci. 24 312 – 325 . 10.1038/s41593-020-00783-4 33589835 PMC8007081

Fields J. A. Swinton M. K. Montilla-Perez P. Ricciardelli E. Telese F. ( 2022 ). The cannabinoid receptor agonist, WIN-55212-2, suppresses the activation of proinflammatory genes induced by interleukin 1 beta in human astrocytes. Cannabis Cannabinoid Res. 7 78 – 92 . 10.1089/can.2020.0128 33998879 PMC8864424

Frietze K. K. Brown A. M. Das D. Franks R. G. Cunningham J. L. Hayward M. et al . ( 2022 ). Lipotoxicity reduces DDX58/Rig-1 expression and activity leading to impaired autophagy and cell death. Autophagy 18 142 – 160 . 10.1080/15548627.2021.1920818 33966599 PMC8865291

Furr S. R. Marriott I. ( 2012 ). Viral CNS infections: Role of glial pattern recognition receptors in neuroinflammation. Front. Microbiol. 3 : 201 . 10.3389/fmicb.2012.00201 22723794 PMC3379540

Galaxy Community ( 2024 ). The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update. Nucleic Acids Res. 52 W83 – W94 . 10.1093/nar/gkae410 38769056 PMC11223835

Gao M. Song Y. Liu Y. Miao Y. Guo Y. Chai H. ( 2024 ). TNF-α/TNFR1 activated astrocytes exacerbate depression-like behavior in CUMS mice. Cell Death Discov. 10 : 220 . 10.1038/s41420-024-01987-4 38710713 PMC11074147

Giovannoni F. Quintana F. J. ( 2020 ). The role of astrocytes in CNS inflammation. Trends Immunol. 41 805 – 819 . 10.1016/j.it.2020.07.007 32800705 PMC8284746

Gong T. Liu L. Jiang W. Zhou R. ( 2020 ). DAMP-sensing receptors in sterile inflammation and inflammatory diseases. Nat. Rev. Immunol. 20 95 – 112 . 10.1038/s41577-019-0215-7 31558839

González-Giraldo Y. Forero D. A. Barreto G. E. Aristizábal-Pachón A. ( 2021 ). Common genes and pathways involved in the response to stressful stimuli by astrocytes: A meta-analysis of genome-wide expression studies. Genomics 113 669 – 680 . 10.1016/j.ygeno.2021.01.008 33485956

Hasel P. Rose I. V. L. Sadick J. S. Kim R. D. Liddelow S. A. ( 2021 ). Neuroinflammatory astrocyte subtypes in the mouse brain. Nat. Neurosci. 24 1475 – 1487 . 10.1038/s41593-021-00905-6 34413515

Hawkins B. T. Davis T. P. ( 2005 ). The blood-brain barrier/neurovascular unit in health and disease. Pharmacol. Rev. 57 173 – 185 . 10.1124/pr.57.2.4 15914466

Italiani P. Puxeddu I. Napoletano S. Scala E. Melillo D. Manocchio S. et al . ( 2018 ). Circulating levels of IL-1 family cytokines and receptors in Alzheimer’s disease: New markers of disease progression? J. Neuroinflamm. 15 : 342 . 10.1186/s12974-018-1376-1 30541566 PMC6292179

Jack C. S. Arbour N. Manusow J. Montgrain V. Blain M. McCrea E. et al . ( 2005 ). TLR signaling tailors innate immune responses in human microglia and astrocytes. J. Immunol. 175 4320 – 4330 . 10.4049/jimmunol.175.7.4320 16177072

Jong Huat T. Camats-Perna J. Newcombe E. A. Onraet T. Campbell D. Sucic J. T. et al . ( 2024 ). The impact of astrocytic NF-κB on healthy and Alzheimer’s disease brains. Sci. Rep. 14 : 14305 . 10.1038/s41598-024-65248-1 38906984 PMC11192733

Jorgačevski J. Potokar M. ( 2023 ). Immune functions of astrocytes in viral neuroinfections. Int. J. Mol. Sci. 24 3514 . 10.3390/ijms24043514 36834929 PMC9960577

Kaltschmidt B. Widera D. Kaltschmidt C. ( 2005 ). Signaling via NF-kappaB in the nervous system. Biochim. Biophys. Acta 1745 287 – 299 . 10.1016/j.bbamcr.2005.05.009 15993497

Kim H. Y. Kim S. Akaydin A. N. Kim S. Hyeon S. J. Lee J. et al . ( 2026 ). The rise of astrocytes: Are they guardians or troublemakers of the brain disorder? Exp. Mol. Med. 58 301 – 318 . 10.1038/s12276-025-01627-6 41639424 PMC12992555

Kim H. Leng K. Park J. Sorets A. G. Kim S. Shostak A. et al . ( 2022 ). Reactive astrocytes transduce inflammation in a blood-brain barrier model through a TNF-STAT3 signaling axis and secretion of alpha 1-antichymotrypsin. Nat. Commun. 13 : 6581 . 10.1038/s41467-022-34412-4 36323693 PMC9630454

Kinsella S. Fichtner M. Watters O. König H. G. Prehn J. H. M. ( 2018 ). Increased A20-E3 ubiquitin ligase interactions in bid-deficient glia attenuate TLR3- and TLR4-induced inflammation. J. Neuroinflamm. 15 : 130 . 10.1186/s12974-018-1143-3 29720226 PMC5930864

Koch M. A. Tucker-Heard G. Perdue N. R. Killebrew J. R. Urdahl K. B. Campbell D. J. ( 2009 ). The transcription factor T-bet controls regulatory T cell homeostasis and function during type 1 inflammation. Nat. Immunol. 10 595 – 602 . 10.1038/ni.1731 19412181 PMC2712126

Kolberg L. Raudvere U. Kuzmin I. Adler P. Vilo J. Peterson H. ( 2023 ). g:Profiler-interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update). Nucleic Acids Res. 51 W207 – W212 . 10.1093/nar/gkad347 37144459 PMC10320099

Kolde R. Laur S. Adler P. Vilo J. ( 2012 ). Robust rank aggregation for gene list integration and meta-analysis. Bioinformatics 28 573 – 580 . 10.1093/bioinformatics/btr709 22247279 PMC3278763

Kong W. Montano M. Corley M. J. Helmy E. Kobayashi H. Kinisu M. et al . ( 2022 ). Neuropilin-1 mediates SARS-CoV-2 infection of astrocytes in brain organoids, inducing inflammation leading to dysfunction and death of neurons. mBio 13 : e02308-22 . 10.1128/mbio.02308-22 36314791 PMC9765283

Koper O. M. Kamińska J. Sawicki K. Kemona H. ( 2018 ). CXCL9, CXCL10, CXCL11, and their receptor (CXCR3) in neuroinflammation and neurodegeneration. Adv. Clin. Exp. Med. 27 849 – 856 . 10.17219/acem/68846 29893515

Koprich J. B. Reske-Nielsen C. Mithal P. Isacson O. ( 2008 ). Neuroinflammation mediated by IL-1β increases susceptibility of dopamine neurons to degeneration in an animal model of Parkinson’s disease. J. Neuroinflamm. 5 : 8 . 10.1186/1742-2094-5-8 18304357 PMC2292163

Kopylova E. Noé L. Touzet H. ( 2012 ). SortMeRNA: Fast and accurate filtering of ribosomal RNAs in metatranscriptomic data. Bioinformatics 28 3211 – 3217 . 10.1093/bioinformatics/bts611 23071270

Krasowska-Zoladek A. Banaszewska M. Kraszpulski M. Konat G. W. ( 2007 ). Kinetics of inflammatory response of astrocytes induced by TLR 3 and TLR4 ligation. J. Neurosci. Res. 85 205 – 212 . 10.1002/jnr.21088 17061254

LaRocca T. J. Cavalier A. N. Roberts C. M. Lemieux M. R. Ramesh P. Garcia M. A. et al . ( 2021 ). Amyloid beta acts synergistically as a pro-inflammatory cytokine. Neurobiol. Dis. 159 : 105493 . 10.1016/j.nbd.2021.105493 34464705 PMC8502211

Law C. W. Chen Y. Shi W. Smyth G. K. ( 2014 ). voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 15 : R29 . 10.1186/gb-2014-15-2-r29 24485249 PMC4053721

Lawrence J. M. Schardien K. Wigdahl B. Nonnemacher M. R. ( 2023 ). Roles of neuropathology-associated reactive astrocytes: A systematic review. Acta Neuropathol. Commun. 11 : 42 . 10.1186/s40478-023-01526-9 36915214 PMC10009953

Le Thuc O. García-Cáceres C. ( 2024 ). Obesity-induced inflammation: Connecting the periphery to the brain. Nat. Metab. 6 1237 – 1252 . 10.1038/s42255-024-01079-8 38997442

Leng K. Rose I. V. L. Kim H. Xia W. Romero-Fernandez W. Rooney B. et al . ( 2022 ). CRISPRi screens in human iPSC-derived astrocytes elucidate regulators of distinct inflammatory reactive states. Nat. Neurosci. 25 1528 – 1542 . 10.1038/s41593-022-01180-9 36303069 PMC9633461

Li H. Li Q. Guo T. He W. Dong C. Wang Y. ( 2017 ). LncRNA CRNDE triggers inflammation through the TLR3-NF-κB-Cytokine signaling pathway. Tumour Biol. 39 : 1010428317703821 . 10.1177/1010428317703821 28621230

Li J. Pan L. Pembroke W. G. Rexach J. E. Godoy M. I. Condro M. C. et al . ( 2021 ). Conservation and divergence of vulnerability and responses to stressors between human and mouse astrocytes. Nat. Commun. 12 : 3958 . 10.1038/s41467-021-24232-3 34172753 PMC8233314

Li K. Ling H. Wang X. Xie Q. Gu C. Luo W. et al . ( 2024 ). The role of NF-κB signaling pathway in reactive astrocytes among neurodegeneration after methamphetamine exposure by integrated bioinformatics. Prog. Neuropsychopharmacol. Biol. Psychiatry 129 : 110909 . 10.1016/j.pnpbp.2023.110909 38061485

Liddelow S. A. Barres B. A. ( 2017 ). Reactive astrocytes: Production, function, and therapeutic potential. Immunity 46 957 – 967 . 10.1016/j.immuni.2017.06.006 28636962

Liu C. Cui G. Zhu M. Kang X. Guo H. ( 2014 ). Neuroinflammation in Alzheimer’s disease: Chemokines produced by astrocytes and chemokine receptors. Intern. J. Clin. Exp. Pathol. 7 8342 – 8355 . 25674199 PMC4314046

Mahmoud S. Gharagozloo M. Simard C. Gris D. ( 2019 ). Astrocytes maintain glutamate homeostasis in the CNS by controlling the balance between glutamate uptake and release. Cells 8 : 184 . 10.3390/cells8020184 30791579 PMC6406900

Matusova Z. Hol E. M. Pekny M. Kubista M. Valihrach L. ( 2023 ). Reactive astrogliosis in the era of single-cell transcriptomics. Front. Cell Neurosci. 17 : 1173200 . 10.3389/fncel.2023.1173200 37153637 PMC10157076

McKimmie C. S. Graham G. J. ( 2010 ). Astrocytes modulate the chemokine network in a pathogen-specific manner. Biochem. Biophys. Res. Commun. 394 1006 – 1011 . 10.1016/j.bbrc.2010.03.111 20331977

McKimmie C. Michlmayr D. ( 2014 ). Role of CXCL10 in central nervous system inflammation. IJICMR 6 1 – 18 . 10.2147/IJICMR.S35953

Michael B. D. Griffiths M. J. Granerod J. Brown D. Keir G. Wnêk M. et al . ( 2026 ). The Interleukin-1 balance during encephalitis is associated with clinical severity, blood-brain barrier permeability, neuroimaging changes, and disease outcome. J. Infect. Dis. 213 1651 – 1660 . 10.1093/infdis/jiv771 26712949 PMC4837908

Monahan J. Warren M. Carvey P. M. ( 2008 ). Neuroinflammation and peripheral immune infiltration in Parkinson’s disease: An autoimmune hypothesis. Cell Trans. 17 363 – 372 . 10.3727/096368908784423328 18522239

Naveed M. Smedlund K. Zhou Q. G. Cai W. Hill J. W. ( 2025 ). Astrocyte involvement in metabolic regulation and disease. Trends Endocrinol. Metab. 36 219 – 234 . 10.1016/j.tem.2024.08.001 39214743 PMC11868460

Opsahl T. Agneessens F. Skvoretz J. ( 2010 ). Node centrality in weighted networks: Generalizing degree and shortest paths. Soc. Netw. 32 245 – 251 . 10.1016/j.socnet.2010.03.006

Page M. J. McKenzie J. E. Bossuyt P. M. Boutron I. Hoffmann T. C. Mulrow C. D. et al . ( 2021 ). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 372 : n71 . 10.1136/bmj.n71 33782057 PMC8005924

Patani R. Hardingham G. E. Liddelow S. A. ( 2023 ). Functional roles of reactive astrocytes in neuroinflammation and neurodegeneration. Nat. Rev. Neurol. 19 395 – 409 . 10.1038/s41582-023-00822-1 37308616

Phares T. W. Stohlman S. A. Hinton D. R. Bergmann C. C. ( 2013 ). Astrocyte-derived CXCL10 drives accumulation of antibody-secreting cells in the central nervous system during viral encephalomyelitis. J. Virol. 87 3382 – 3392 . 10.1128/JVI.03307-12 23302888 PMC3592145

Potokar M. Zorec R. Jorgačevski J. ( 2023 ). astrocytes are a key target for neurotropic viral infection. Cells 12 : 2307 . 10.3390/cells12182307 37759529 PMC10528686

Pozniak P. D. White M. K. Khalili K. ( 2014 ). TNF-α/NF-κB signaling in the CNS: Possible connection to EPHB2. J. Neuroimmune Pharmacol. 9 133 – 141 . 10.1007/s11481-013-9517-x 24277482 PMC3959244

Qiao X. Zhang W. Zhao W. ( 2022 ). Role of CXCL10 in spinal cord injury. Int. J. Med. Sci. 19 2058 – 2070 . 10.7150/ijms.76694 36483597 PMC9724238

Rehwinkel J. Gack M. U. ( 2020 ). RIG-I-like receptors: Their regulation and roles in RNA sensing. Nat Rev Immunol . 20 537 – 551 . 10.1038/s41577-020-0288-3 32203325 PMC7094958

Ridder D. A. Schwaninger M. ( 2009 ). NF-kappaB signaling in cerebral ischemia. Neuroscience 158 995 – 1006 . 10.1016/j.neuroscience.2008.07.007 18675321

Rojas-Cruz A. F. Martín-Jiménez C. A. González J. González-Giraldo Y. Pinzón A. M. Barreto G. E. et al . ( 2023 ). Palmitic acid upregulates Type I interferon-mediated antiviral response and cholesterol biosynthesis in human astrocytes. Mol. Neurobiol. 60 4842 – 4854 . 10.1007/s12035-023-03366-z 37184765 PMC10293381

Ruland J. ( 2011 ). Return to homeostasis: Downregulation of NF-κB responses. Nat. Immunol. 12 709 – 714 . 10.1038/ni.2055 21772279

Russ E. Iordanskiy S. ( 2023 ). Endogenous retroviruses as modulators of innate immunity. Pathogens 12 : 162 . 10.3390/pathogens12020162 36839434 PMC9963469

Santos R. Vadodaria K. C. Jaeger B. N. Mei A. Lefcochilos-Fogelquist S. Mendes A. P. D. et al . ( 2017 ). Differentiation of inflammation-responsive astrocytes from glial progenitors generated from human induced pluripotent stem cells. Stem Cell Rep. 8 1757 – 1769 . 10.1016/j.stemcr.2017.05.011 28591655 PMC5470172

Seo J. W. Yang E. J. Kim S. H. Choi I. H. ( 2015 ). An inhibitory alternative splice isoform of Toll-like receptor 3 is induced by type I interferons in human astrocyte cell lines. BMB Rep. 48 696 – 701 . 10.5483/bmbrep.2015.48.12.106 26077030 PMC4791326

Shannon P. Markiel A. Ozier O. Baliga N. S. Wang J. T. Ramage D. et al . ( 2003 ). Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13 2498 – 2504 . 10.1101/gr.1239303 14597658 PMC403769

Sofroniew M. V. ( 2025 ). Astrogliosis. Cold Spring Harb. Perspect. Biol. 7 : a020420 . 10.1101/cshperspect.a020420 25380660 PMC4315924

Sørensen T. L. Trebst C. Kivisäkk P. Klaege K. L. Majmudar A. Ravid R. et al . ( 2002 ). Multiple sclerosis: A study of CXCL10 and CXCR3 co-localization in the inflamed central nervous system. J. Neuroimmunol. 127 59 – 68 . 10.1016/s0165-5728(02)00097-8 12044976

Sterka D. Rati D. M. Marriott I. ( 2006 ). Functional expression of NOD2, a novel pattern recognition receptor for bacterial motifs, in primary murine astrocytes. Glia 53 322 – 330 . 10.1002/glia.20286 16265673

Sui Y. Stehno-Bittel L. Li S. Loganathan R. Dhillon N. K. Pinson D. et al . ( 2006 ). CXCL10-induced cell death in neurons: Role of calcium dysregulation. Eur. J. Neurosci. 23 957 – 964 . 10.1111/j.1460-9568.2006.04631.x 16519660

Supek F. Bošnjak M. Škunca N. Šmuc T. ( 2011 ). REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One 6 : e21800 . 10.1371/journal.pone.0021800 21789182 PMC3138752

Szklarczyk D. Kirsch R. Koutrouli M. Nastou K. Mehryary F. Hachilif R. et al . ( 2023 ). 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 36370105 PMC9825434

Vadodaria K. C. Mendes A. P. D. Mei A. Racha V. Erikson G. Shokhirev M. N. et al . ( 2021 ). Altered neuronal support and inflammatory response in bipolar disorder patient-derived astrocytes. Stem Cell Rep. 16 825 – 835 . 10.1016/j.stemcr.2021.02.004 33667413 PMC8072028

Vazirinejad R. Ahmadi Z. Kazemi Arababadi M. Hassanshahi G. Kennedy D. ( 2014 ). The biological functions, structure and sources of CXCL10 and its outstanding part in the pathophysiology of multiple sclerosis. Neuroimmunomodulation 21 322 – 330 . 10.1159/000357780 24642726

Wang Y. Zhang J. Chang H. Wang H. Xu W. Cong H. et al . ( 2022 ). 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 35659638

Wilkins C. Gale M. ( 2010 ). Recognition of viruses by cytoplasmic sensors. Curr. Opin. Immunol. 22 41 – 47 . 10.1016/j.coi.2009.12.003 20061127 PMC3172156

Willis C. M. Sutter P. Rouillard M. Crocker S. J. ( 2020 ). The effects of IL-1β on astrocytes are conveyed by extracellular vesicles and influenced by age , Neurochem. Res. 45 694 – 707 . 10.1007/s11064-019-02937-8 31900795

Xie L. Yang S.-H. ( 2015 ). Interaction of astrocytes and T cells in physiological and pathological conditions. Brain Res. 1623 63 – 73 . 10.1016/j.brainres.2015.03.026 25813828 PMC4569510

Xu D. Kong T. Shao Z. Liu M. Zhang R. Zhang S. et al . ( 2021 ). Orexin-A alleviates astrocytic apoptosis and inflammation via inhibiting OX1R-mediated NF-κB and MAPK signaling pathways in cerebral ischemia/reperfusion injury. Biochim. Biophys. Acta Mol. Basis Dis. 1867 : 166230 . 10.1016/j.bbadis.2021.166230 34358627

Yoneyama M. Kikuchi M. Natsukawa T. Shinobu N. Imaizumi T. Miyagishi M. et al . ( 2004 ). The RNA helicase RIG-I has an essential function in double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 5 730 – 737 . 10.1038/ni1087 15208624