Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 130 of the 131 references it cites were found in a registry or an index.

DNA damage contributes to neurotoxic inflammation in Aicardi-Goutières syndrome astrocytes

Open access
Giordano, Anna Maria Sole, Luciani, Marco, Gatto, Francesca, Abou Alezz, Monah, Beghè, Chiara, Della Volpe, Lucrezia, Migliara, Alessandro, Valsoni, Sara, Genua, Marco, Dzieciatkowska, Monika, Frati, Giacomo, Tahraoui-Bories, Julie, Giliani, Silvia Clara, Orcesi, Simona, Fazzi, Elisa, Ostuni, Renato, D’Alessandro, Angelo, Di Micco, Raffaella, Merelli, Ivan, Lombardo, Angelo, Reijns, Martin A.M., Gromak, Natalia, Gritti, Angela, Kajaste-Rudnitski, Anna · 2022 · Journal of Experimental Medicine

Abstract

Aberrant induction of type I IFN is a hallmark of the inherited encephalopathy Aicardi-Goutières syndrome (AGS), but the mechanisms triggering disease in the human central nervous system (CNS) remain elusive. Here, we generated human models of AGS using genetically modified and patient-derived pluripotent stem cells harboring TREX1 or RNASEH2B loss-of-function alleles. Genome-wide transcriptomic analysis reveals that spontaneous proinflammatory activation in AGS astrocytes initiates signaling cascades impacting multiple CNS cell subsets analyzed at the single-cell level. We identify accumulating DNA damage, with elevated R-loop and micronuclei formation, as a driver of STING- and NLRP3-related inflammatory responses leading to the secretion of neurotoxic mediators. Importantly, pharmacological inhibition of proapoptotic or inflammatory cascades in AGS astrocytes prevents neurotoxicity without apparent impact on their increased type I IFN responses. Together, our work identifies DNA damage as a major driver of neurotoxic inflammation in AGS astrocytes, suggests a role for AGS gene products in R-loop homeostasis, and identifies common denominators of disease that can be targeted to prevent astrocyte-mediated neurotoxicity in AGS.

Overview

Integrity signals

Registry

Crossref · 10.1084/jem.20211121

Cites

130 of 130 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Figures

Graphical abstract
Graphical abstract CC BY · PubMed Central
Figure 1. KO iPSCs efficiently differentiate into proinflammatory astrocytes. (A) Human genomic DNA sequencing. Green arrowhead, sgRNA cut site; purple, nucleotide substitutions; cyan, start codon; red, premature stop codon generated by nonhomologous end joining–mediated out-of-frame repair. Nucleotides deletions are dashed. Reported are the amino acids encoded by the nucleotides upstream of the sgRNA cut site. (B) WB detection of TREX1 and RNASEH2B protein levels in WT and edited clones. (Bottom panel also shown in Fig. S1 A.) (C) RNase H2 enzymatic activity was analyzed in WT and KO iPSCs as cleavage of an in vitro synthetized and labeled RNA/DNA molecule measured by fluorescence. (Mean ± SEM; n = 4 independent differentiations; one-tailed Mann–Whitney U test; *, P < 0.05.) (D) Scheme of proinflammatory astrocytes differentiation protocol created with BioRender.com. (E) GFAP astrocyte marker was detected by gene expression in full KO clones and WT at passage 1 after the iPSC-derived astrocytes enrichment step and expressed as fold vs. day 0, normalized to the HPRT1 housekeeping gene. (Mean ± SEM; n = 5 independent differentiations.) (F) Percentages of S100b- and EAAT2-expressing double-positive cells were measured by IF at passage 1 and quantified. One representative image per genotype is shown. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent differentiations.) (G) Expression and ImageJ quantification of the astrocyte markers EAAT2 and GFAP by WB at passage 1. One representative gel is shown. (Mean ± SEM; n = 3 independent differentiations.) Source data are available for this figure: SourceDataF1.
Figure 1. KO iPSCs efficiently differentiate into proinflammatory astrocytes. (A) Human genomic DNA sequencing. Green arrowhead, sgRNA cut site; purple, nucleotide substitutions; cyan, start codon; red, premature stop codon generated by nonhomologous end joining–mediated out-of-frame repair. Nucleotides deletions are dashed. Reported are the amino acids encoded by the nucleotides upstream of the sgRNA cut site. (B) WB detection of TREX1 and RNASEH2B protein levels in WT and edited clones. (Bottom panel also shown in Fig. S1 A.) (C) RNase H2 enzymatic activity was analyzed in WT and KO iPSCs as cleavage of an in vitro synthetized and labeled RNA/DNA molecule measured by fluorescence. (Mean ± SEM; n = 4 independent differentiations; one-tailed Mann–Whitney U test; *, P < 0.05.) (D) Scheme of proinflammatory astrocytes differentiation protocol created with BioRender.com. (E) GFAP astrocyte marker was detected by gene expression in full KO clones and WT at passage 1 after the iPSC-derived astrocytes enrichment step and expressed as fold vs. day 0, normalized to the HPRT1 housekeeping gene. (Mean ± SEM; n = 5 independent differentiations.) (F) Percentages of S100b- and EAAT2-expressing double-positive cells were measured by IF at passage 1 and quantified. One representative image per genotype is shown. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent differentiations.) (G) Expression and ImageJ quantification of the astrocyte markers EAAT2 and GFAP by WB at passage 1. One representative gel is shown. (Mean ± SEM; n = 3 independent differentiations.) Source data are available for this figure: SourceDataF1.CC BY · PubMed Central
Figure S1. Characterization of KO iPSCs and iPSC-NSCs. (A) Loss of RNASEH2B expression in RNASEH2BKO iPSCs was confirmed by WB and comparison to RNASEH2BKO hTERT-RPE1 cells. (n = 3 independent experiments; panel also shown in Fig. 1 B.) (B) SOX2 and OCT4 pluripotency markers were detected by gene expression in full KO clones and WT compared with the healthy human fibroblasts (h Fibro) previously used to reprogram these iPSCs and normalized on HPRT1 housekeeping gene. (Mean ± SEM; n = 3 independent experiments.) (C) Lolliplot of TP53 variants in WT, T−/−, and R−/− KO clones. Only single nucleotide polymorphisms (SNPs) and insertion–deletion mutations (INDELs) with a frequency higher than the 20% (see y-axis scale) are reported, and variants located in exonic vs. intronic regions are colored in a different way (light blue and bright blue, respectively). TES, transcription end site. (D) Karyotype of the WT and KO clones. (E) Although cellular RNase H2 activity was substantially reduced in RNASEH2BKO iPSCs, iPSC-NSCs, and astrocytes, as well as in AGS2 (RNASEH2BA177T/A177T) primary fibroblasts, iPSCs and iPSC-NSC KO cells retained low-level nuclease activity against substrate (DNA duplex with a single embedded ribonucleotide), above that seen for independently generated RNase H2–null hTERT-RPE1 cells. Individual data points for n = 2–6 independent differentiations. (Mean ± SEM; n = 2–6 independent experiments.) (F) Representative results for alkaline gel electrophoresis of RNase H2–digested genomic DNA from WT, RNASEH2BKO (R−/−), and RNASEH2BA177T/A177T (AGS2) cells (primary fibroblasts, iPSCs, iPSC-NSCs, and astrocytes). hTERT-RPE1 RNase H2–null (RPE1 KO) DNA was included as a positive control for increased fragmentation as a consequence of embedded ribonucleotides. (n = 2–3 technical replicates.) (G) Densitometry plots for gels in A show no apparent shift in DNA fragment size distribution for mutant RNASEH2B fibroblasts, iPSCs, iPSC-NSCs, or astrocytes, in contrast with independent RNase H2–null RPE1 cells. (H) OCT4 and PAX6 were detected by gene expression in differentiated WT and KO clones in comparison with WT iPSCs and normalized on HPRT1 housekeeping gene. (Mean ± SEM; n = 3 independent experiments.) Source data are available for this figure: SourceDataFS1.
Figure S1. Characterization of KO iPSCs and iPSC-NSCs. (A) Loss of RNASEH2B expression in RNASEH2BKO iPSCs was confirmed by WB and comparison to RNASEH2BKO hTERT-RPE1 cells. (n = 3 independent experiments; panel also shown in Fig. 1 B.) (B) SOX2 and OCT4 pluripotency markers were detected by gene expression in full KO clones and WT compared with the healthy human fibroblasts (h Fibro) previously used to reprogram these iPSCs and normalized on HPRT1 housekeeping gene. (Mean ± SEM; n = 3 independent experiments.) (C) Lolliplot of TP53 variants in WT, T−/−, and R−/− KO clones. Only single nucleotide polymorphisms (SNPs) and insertion–deletion mutations (INDELs) with a frequency higher than the 20% (see y-axis scale) are reported, and variants located in exonic vs. intronic regions are colored in a different way (light blue and bright blue, respectively). TES, transcription end site. (D) Karyotype of the WT and KO clones. (E) Although cellular RNase H2 activity was substantially reduced in RNASEH2BKO iPSCs, iPSC-NSCs, and astrocytes, as well as in AGS2 (RNASEH2BA177T/A177T) primary fibroblasts, iPSCs and iPSC-NSC KO cells retained low-level nuclease activity against substrate (DNA duplex with a single embedded ribonucleotide), above that seen for independently generated RNase H2–null hTERT-RPE1 cells. Individual data points for n = 2–6 independent differentiations. (Mean ± SEM; n = 2–6 independent experiments.) (F) Representative results for alkaline gel electrophoresis of RNase H2–digested genomic DNA from WT, RNASEH2BKO (R−/−), and RNASEH2BA177T/A177T (AGS2) cells (primary fibroblasts, iPSCs, iPSC-NSCs, and astrocytes). hTERT-RPE1 RNase H2–null (RPE1 KO) DNA was included as a positive control for increased fragmentation as a consequence of embedded ribonucleotides. (n = 2–3 technical replicates.) (G) Densitometry plots for gels in A show no apparent shift in DNA fragment size distribution for mutant RNASEH2B fibroblasts, iPSCs, iPSC-NSCs, or astrocytes, in contrast with independent RNase H2–null RPE1 cells. (H) OCT4 and PAX6 were detected by gene expression in differentiated WT and KO clones in comparison with WT iPSCs and normalized on HPRT1 housekeeping gene. (Mean ± SEM; n = 3 independent experiments.) Source data are available for this figure: SourceDataFS1.CC BY · PubMed Central
Figure S2. KO iPSC-NSCs and iPSC-derived neurons do not show strong activation of inflammation or DNA damage responses. (A) Heatmap visualizing the significantly enriched GSEA terms in iPSC-derived NSCs against the Hallmark gene set (Molecular Signatures Database). GSEA was performed on logFC preranked gene lists obtained from T−/− and R−/− gene expression compared with WT. (NES, normalized enrichment score; adjusted P < 0.05.) (B and C) Violin plots showing the distribution of IFN score (B) and inflammation score (C) in T−/− and R−/− KO cells measured from the RNA-seq expression data in NSCs of 32 ISGs and 30 IRGs, respectively, compared with WT. (D) IL8 and ISG15 mRNA expression levels in iPSC-derived iNSCs stimulated with poly(I:C) for 6 h. Data are expressed as fold difference vs. untreated condition and normalized to HPRT1 housekeeping gene. (Mean ± SEM; n = 3 technical replicates; ctrl, control; UT, untreated.) (E) IF of WT and KO iPSC–derived NSCs stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments.) (F) Percentages of cells expressing Tuj1 neuronal marker and GFAP glial marker were measured and quantified upon neuronal differentiation of WT and KO iPSC-NSCs. Three slides from two independent differentiations were analyzed. One representative image per genotype is shown. Scale bar, 50 µm. (Mean ± SEM; n = 2 independent experiments.) (G) IF of WT and KO iPSC–derived neurons stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments.) (H and I) Violin plots showing the distribution of IFN score (H) and inflammation score (I) in T−/− and R−/− KO cells measured from the RNA-seq expression data in neurons of 32 IFN-RGs and 30 IRGs, respectively, compared with WT.
Figure S2. KO iPSC-NSCs and iPSC-derived neurons do not show strong activation of inflammation or DNA damage responses. (A) Heatmap visualizing the significantly enriched GSEA terms in iPSC-derived NSCs against the Hallmark gene set (Molecular Signatures Database). GSEA was performed on logFC preranked gene lists obtained from T−/− and R−/− gene expression compared with WT. (NES, normalized enrichment score; adjusted P < 0.05.) (B and C) Violin plots showing the distribution of IFN score (B) and inflammation score (C) in T−/− and R−/− KO cells measured from the RNA-seq expression data in NSCs of 32 ISGs and 30 IRGs, respectively, compared with WT. (D) IL8 and ISG15 mRNA expression levels in iPSC-derived iNSCs stimulated with poly(I:C) for 6 h. Data are expressed as fold difference vs. untreated condition and normalized to HPRT1 housekeeping gene. (Mean ± SEM; n = 3 technical replicates; ctrl, control; UT, untreated.) (E) IF of WT and KO iPSC–derived NSCs stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments.) (F) Percentages of cells expressing Tuj1 neuronal marker and GFAP glial marker were measured and quantified upon neuronal differentiation of WT and KO iPSC-NSCs. Three slides from two independent differentiations were analyzed. One representative image per genotype is shown. Scale bar, 50 µm. (Mean ± SEM; n = 2 independent experiments.) (G) IF of WT and KO iPSC–derived neurons stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments.) (H and I) Violin plots showing the distribution of IFN score (H) and inflammation score (I) in T−/− and R−/− KO cells measured from the RNA-seq expression data in neurons of 32 IFN-RGs and 30 IRGs, respectively, compared with WT.CC BY · PubMed Central
Figure 2. KO iPSC–derived astrocytes spontaneously activate type I IFN and proinflammatory responses. (A and B) Heatmap visualizing the significant enriched GSEA terms in iPSC-derived neurons (A) and iPSC-derived astrocytes (B) against the Hallmark gene set (Molecular Signatures Database). GSEA was performed on logFC preranked gene lists obtained from T−/− and R−/− gene expression compared with WT. (NES, normalized enrichment score; *, adjusted P < 0.05; **, adjusted P < 0.01 for upregulated pathways.) (C and D) Enrichment plots showing the GSEA results of the hallmark IFNα response gene set (C) and hallmark IFNγ response gene set (D) resulting from the comparison of genes preranked based on logFC values in T−/− KO cells vs. WT (upper plot) and R−/− KO cells vs. WT (lower plot) in astrocytes. The top part of each plot shows the enrichment score representing the running-sum statistic calculated by walking down the ranked list of genes. The middle part shows the position of a member of the gene set in the ranked list of genes. The bottom part depicts the ranking metric that measures a gene’s correlation with the biological function. (E and F) Heatmaps depicting the expression levels of genes belonging to IFNα (E) and IFNγ (F) gene sets upregulated in T−/− cells and R−/− cells. Gene expression, in rows, was row-scaled (z-scores) for visualization. The color scale differentiates values as high (red), medium (yellow), and low (blue) expression. (G and H) Violin plots showing the distribution of IFN score (G) and the inflammation score (H) in T−/− and R−/− KO cells in astrocytes measured from the RNA-seq expression data of 32 IFN-RGs and 30 IRGs, respectively, compared with WT. (Wilcoxon rank-sum test; *, P < 0.05; **, P < 0.01; ****, P < 0.0001.) (I and J) Violin plots showing the IFN score (I) and inflammation score (J) measured in T−/− and R−/− KO cells from qRT-PCR quantification of the median FC of six ISGs and six IRGs, respectively. (n = 5 independent experiments; Wilcoxon rank-sum test; ***, P < 0.001; ****, P < 0.0001.) (K) Heatmap showing the expression level of genes belonging to the NLRP3 inflammasome. Gene expression, in rows, was row-scaled (z-scores) for visualization. (L and M) Violin plots showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with the STING inhibitor H151 (L) or NLRP3 inhibitor MCC950 (M). (n = 3 independent experiments; Wilcoxon rank-sum test; *, P < 0.05; **, P < 0.01.)
Figure 2. KO iPSC–derived astrocytes spontaneously activate type I IFN and proinflammatory responses. (A and B) Heatmap visualizing the significant enriched GSEA terms in iPSC-derived neurons (A) and iPSC-derived astrocytes (B) against the Hallmark gene set (Molecular Signatures Database). GSEA was performed on logFC preranked gene lists obtained from T−/− and R−/− gene expression compared with WT. (NES, normalized enrichment score; *, adjusted P < 0.05; **, adjusted P < 0.01 for upregulated pathways.) (C and D) Enrichment plots showing the GSEA results of the hallmark IFNα response gene set (C) and hallmark IFNγ response gene set (D) resulting from the comparison of genes preranked based on logFC values in T−/− KO cells vs. WT (upper plot) and R−/− KO cells vs. WT (lower plot) in astrocytes. The top part of each plot shows the enrichment score representing the running-sum statistic calculated by walking down the ranked list of genes. The middle part shows the position of a member of the gene set in the ranked list of genes. The bottom part depicts the ranking metric that measures a gene’s correlation with the biological function. (E and F) Heatmaps depicting the expression levels of genes belonging to IFNα (E) and IFNγ (F) gene sets upregulated in T−/− cells and R−/− cells. Gene expression, in rows, was row-scaled (z-scores) for visualization. The color scale differentiates values as high (red), medium (yellow), and low (blue) expression. (G and H) Violin plots showing the distribution of IFN score (G) and the inflammation score (H) in T−/− and R−/− KO cells in astrocytes measured from the RNA-seq expression data of 32 IFN-RGs and 30 IRGs, respectively, compared with WT. (Wilcoxon rank-sum test; *, P < 0.05; **, P < 0.01; ****, P < 0.0001.) (I and J) Violin plots showing the IFN score (I) and inflammation score (J) measured in T−/− and R−/− KO cells from qRT-PCR quantification of the median FC of six ISGs and six IRGs, respectively. (n = 5 independent experiments; Wilcoxon rank-sum test; ***, P < 0.001; ****, P < 0.0001.) (K) Heatmap showing the expression level of genes belonging to the NLRP3 inflammasome. Gene expression, in rows, was row-scaled (z-scores) for visualization. (L and M) Violin plots showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with the STING inhibitor H151 (L) or NLRP3 inhibitor MCC950 (M). (n = 3 independent experiments; Wilcoxon rank-sum test; *, P < 0.05; **, P < 0.01.)CC BY · PubMed Central
Figure 3. Single-cell transcriptomics reveals AGS phenotypes in patient-derived neuronal/glial mixed populations. (A) scRNA-seq experimental scheme (created with BioRender.com) performed with iPSC-derived cells from patients harboring mutations in TREX1 (AGS1) or RNASEH2B (AGS2). (B) UMAP plot of single-cell data with Seurat. Cells are clustered in two dimensions using the UMAP dimensionality reduction technique and annotated by cell type. (C) Hierarchically clustered average gene expression heatmap for genes overexpressed across the different cell types grouped according to the Seurat classification (A, astrocytes; IA, immature astrocytes; IN, immature neurons; N, neurons; O, oligodendrocytes; OPC, oligodendrocyte precursor cells). Yellow, high expression; purple, low expression. Scaled-in normalized gene expression data are shown. (D–F) Heatmap visualizing the enriched GSEA terms in astrocytes (D), neurons (E), and oligodendrocytes (F) against the Hallmark gene set (Molecular Signatures Database). GSEA was performed on logFC preranked gene lists obtained from AGS1 and AGS2 gene expression compared with WT within each cell type. (NES, normalized enrichement score; *, adjusted P < 0.05; **, adjusted P < 0.01; ***, adjusted P < 0.001.) (G–J) Violin plots showing the distribution of IFN scores and inflammation scores in astrocytes (G and H) and neurons (I and J) calculated from the expression of 32 IFN-RGs and 30 IRGs, respectively, in AGS1 and AGS2 patients compared with WT. (Wilcoxon rank-sum test; *, P < 0.05; ****, P < 0.0001.)
Figure 3. Single-cell transcriptomics reveals AGS phenotypes in patient-derived neuronal/glial mixed populations. (A) scRNA-seq experimental scheme (created with BioRender.com) performed with iPSC-derived cells from patients harboring mutations in TREX1 (AGS1) or RNASEH2B (AGS2). (B) UMAP plot of single-cell data with Seurat. Cells are clustered in two dimensions using the UMAP dimensionality reduction technique and annotated by cell type. (C) Hierarchically clustered average gene expression heatmap for genes overexpressed across the different cell types grouped according to the Seurat classification (A, astrocytes; IA, immature astrocytes; IN, immature neurons; N, neurons; O, oligodendrocytes; OPC, oligodendrocyte precursor cells). Yellow, high expression; purple, low expression. Scaled-in normalized gene expression data are shown. (D–F) Heatmap visualizing the enriched GSEA terms in astrocytes (D), neurons (E), and oligodendrocytes (F) against the Hallmark gene set (Molecular Signatures Database). GSEA was performed on logFC preranked gene lists obtained from AGS1 and AGS2 gene expression compared with WT within each cell type. (NES, normalized enrichement score; *, adjusted P < 0.05; **, adjusted P < 0.01; ***, adjusted P < 0.001.) (G–J) Violin plots showing the distribution of IFN scores and inflammation scores in astrocytes (G and H) and neurons (I and J) calculated from the expression of 32 IFN-RGs and 30 IRGs, respectively, in AGS1 and AGS2 patients compared with WT. (Wilcoxon rank-sum test; *, P < 0.05; ****, P < 0.0001.)CC BY · PubMed Central
Figure S3. Spontaneous activation of multiple inflammatory pathways identified by scRNA-seq in AGS patient–derived astrocytes. (A) Number of cells quantified from single-cell data by CellRanger in the different cell type clusters across the studied samples. (B) UMAP plot showing the clustering of cells across samples with Seurat. Cells are clustered in two dimensions using the UMAP dimensionality reduction technique and annotated by sample type. (C) Heatmaps depicting the expression of genes from the GSEA terms identified from the different comparisons between AGS1 and AGS2 iPSC-derived astrocytes compared with WT. Gene expression, in rows, was row-scaled (z-scores) for visualization. The color scale differentiates values as high (red), medium (yellow), and low (blue) expression.
Figure S3. Spontaneous activation of multiple inflammatory pathways identified by scRNA-seq in AGS patient–derived astrocytes. (A) Number of cells quantified from single-cell data by CellRanger in the different cell type clusters across the studied samples. (B) UMAP plot showing the clustering of cells across samples with Seurat. Cells are clustered in two dimensions using the UMAP dimensionality reduction technique and annotated by sample type. (C) Heatmaps depicting the expression of genes from the GSEA terms identified from the different comparisons between AGS1 and AGS2 iPSC-derived astrocytes compared with WT. Gene expression, in rows, was row-scaled (z-scores) for visualization. The color scale differentiates values as high (red), medium (yellow), and low (blue) expression.CC BY · PubMed Central
Figure S4. AGS patient–derived and KO iPSCs efficiently differentiate into proinflammatory astrocytes and spontaneously activate DNA damage responses. iPSC-derived proinflammatory astrocytes from healthy control (WT) or patients harboring mutations in TREX1 (AGS1) or RNASEH2B (AGS2) were analyzed. (A) Percentages of S100b- and EAAT2-expressing and double-positive cells were measured by IF and quantified. One representative image per genotype is shown. Five fields from three slides of two independent differentiation experiments were acquired. Scale bar, 50 µm. (Mean ± SEM; n = 2 independent experiments.) (B) GFAP astrocyte marker was detected by gene expression in AGS patient–derived clones and WT at passage 1 after the iPSC-derived astrocyte enrichment step and expressed as fold difference vs. day 0, normalized on HPRT1 housekeeping gene. (Mean ± SEM; n = 3 independent experiments.) (C) Expression and ImageJ quantification of the astrocyte markers EAAT2 and GFAP by WB. One representative gel is shown. (Mean ± SEM; n = 3 independent experiments.) (D) IF of WT and KO iPSC–derived proinflammatory astrocytes stained for phosphorylated p53, quantified by measuring area intensity normalized on number of nuclei. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments; one-tailed Mann–Whitney U test; ****, P < 0.0001.) (E) IF of AGS patient–derived iPSC-derived proinflammatory astrocytes stained for 53BP1, quantified by cell profiler. Each dot corresponds to one field from three slides of three independent differentiations. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; ****, P < 0.0001.) Scale bar, 50 µm. (F) Violin plot showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells transduced with a lentiviral vector expressing GSE56. (n = 2 independent experiments; Wilcoxon rank-sum test; *, P < 0.05.) (G) p21 mRNA levels were detected by gene expression in astrocytes upon transduction with lentiviral vector expressing GFP (LV GFP) or GSE (LV GSE56) in the presence or not of Camptothecin (CPT), a topoisomerase inhibitor. Data are expressed as fold vs. untransduced cells (UT) and normalized to HPRT1 housekeeping gene. (Mean ± SD; n = 2 technical replicates.) Source data are available for this figure: SourceDataFS4.
Figure S4. AGS patient–derived and KO iPSCs efficiently differentiate into proinflammatory astrocytes and spontaneously activate DNA damage responses. iPSC-derived proinflammatory astrocytes from healthy control (WT) or patients harboring mutations in TREX1 (AGS1) or RNASEH2B (AGS2) were analyzed. (A) Percentages of S100b- and EAAT2-expressing and double-positive cells were measured by IF and quantified. One representative image per genotype is shown. Five fields from three slides of two independent differentiation experiments were acquired. Scale bar, 50 µm. (Mean ± SEM; n = 2 independent experiments.) (B) GFAP astrocyte marker was detected by gene expression in AGS patient–derived clones and WT at passage 1 after the iPSC-derived astrocyte enrichment step and expressed as fold difference vs. day 0, normalized on HPRT1 housekeeping gene. (Mean ± SEM; n = 3 independent experiments.) (C) Expression and ImageJ quantification of the astrocyte markers EAAT2 and GFAP by WB. One representative gel is shown. (Mean ± SEM; n = 3 independent experiments.) (D) IF of WT and KO iPSC–derived proinflammatory astrocytes stained for phosphorylated p53, quantified by measuring area intensity normalized on number of nuclei. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments; one-tailed Mann–Whitney U test; ****, P < 0.0001.) (E) IF of AGS patient–derived iPSC-derived proinflammatory astrocytes stained for 53BP1, quantified by cell profiler. Each dot corresponds to one field from three slides of three independent differentiations. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; ****, P < 0.0001.) Scale bar, 50 µm. (F) Violin plot showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells transduced with a lentiviral vector expressing GSE56. (n = 2 independent experiments; Wilcoxon rank-sum test; *, P < 0.05.) (G) p21 mRNA levels were detected by gene expression in astrocytes upon transduction with lentiviral vector expressing GFP (LV GFP) or GSE (LV GSE56) in the presence or not of Camptothecin (CPT), a topoisomerase inhibitor. Data are expressed as fold vs. untransduced cells (UT) and normalized to HPRT1 housekeeping gene. (Mean ± SD; n = 2 technical replicates.) Source data are available for this figure: SourceDataFS4.CC BY · PubMed Central
Figure 4. AGS KO and patient-derived astrocytes show signs of increased DNA damage that contributes to inflammation. iPSC-derived proinflammatory astrocytes from healthy control (WT), TREX1 KO, RNASEH2B KO, or patients harboring mutations in TREX1 (AGS1) or RNASEH2B (AGS2) were analyzed. (A) IF of WT and KO proinflammatory astrocytes stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; **, P < 0.01.) (B) IF of WT or patient-derived proinflammatory astrocytes stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of three independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; **, P < 0.01; ****, P < 0.0001.) (C) Micronuclei counts in WT, KO, and AGS patient iPSC–derived astrocytes. Each dot corresponds to one field from three slides of three (AGS patients) and five (WT and KO) independent iPSC-derived astrocyte differentiation experiments. (Mean ± SEM; n = 3–5 independent experiments; Kruskal–Wallis test; **, P < 0.01; ****, P < 0.0001.) (D) γH2AX-positive micronuclei counts in WT, KO, and AGS patient–derived astrocytes. Each dot corresponds to one field from three slides of three (AGS patients) and five (WT and KO) independent astrocyte differentiation experiments. (Mean ± SEM; n = 3–5 independent experiments; Kruskal–Wallis test; *, P < 0.05; **, P < 0.01.) (E) Representative images of micronuclei and γH2AX-positive micronuclei (white arrowheads) for each sample. Scale bar, 50 µm. (F) IF of WT and KO proinflammatory astrocytes treated or not (UT) with ATM inhibitor (ATMi) and stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of three independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; **, P < 0.01; ****, P < 0.0001.) (G) Violin plot showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with ATM inhibitor (ATMi). (n = 3 independent experiments; Wilcoxon rank-sum test; *, P < 0.05.)
Figure 4. AGS KO and patient-derived astrocytes show signs of increased DNA damage that contributes to inflammation. iPSC-derived proinflammatory astrocytes from healthy control (WT), TREX1 KO, RNASEH2B KO, or patients harboring mutations in TREX1 (AGS1) or RNASEH2B (AGS2) were analyzed. (A) IF of WT and KO proinflammatory astrocytes stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of four independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 4 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; **, P < 0.01.) (B) IF of WT or patient-derived proinflammatory astrocytes stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of three independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; **, P < 0.01; ****, P < 0.0001.) (C) Micronuclei counts in WT, KO, and AGS patient iPSC–derived astrocytes. Each dot corresponds to one field from three slides of three (AGS patients) and five (WT and KO) independent iPSC-derived astrocyte differentiation experiments. (Mean ± SEM; n = 3–5 independent experiments; Kruskal–Wallis test; **, P < 0.01; ****, P < 0.0001.) (D) γH2AX-positive micronuclei counts in WT, KO, and AGS patient–derived astrocytes. Each dot corresponds to one field from three slides of three (AGS patients) and five (WT and KO) independent astrocyte differentiation experiments. (Mean ± SEM; n = 3–5 independent experiments; Kruskal–Wallis test; *, P < 0.05; **, P < 0.01.) (E) Representative images of micronuclei and γH2AX-positive micronuclei (white arrowheads) for each sample. Scale bar, 50 µm. (F) IF of WT and KO proinflammatory astrocytes treated or not (UT) with ATM inhibitor (ATMi) and stained for γH2AX quantified by cell profiler. Each dot corresponds to one field from three slides of three independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; **, P < 0.01; ****, P < 0.0001.) (G) Violin plot showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with ATM inhibitor (ATMi). (n = 3 independent experiments; Wilcoxon rank-sum test; *, P < 0.05.)CC BY · PubMed Central
Figure 5. AGS KO astrocytes accumulate R-loops. (A) Slot blot analysis of genomic DNA extracted from WT, R−/−, and T−/− cells treated with the recombinant RNase H (RNH) enzyme for 2.5 h at 37°C and probed with S9.6 (top) and ssDNA (bottom) antibodies. Slot blot probed with the antibody recognizing ssDNA was used as a loading control. (Mean ± SEM; n = 3 independent experiments; two-tailed unpaired t test; *, P < 0.05; **, P < 0.01.) (B) Scheme of DRIP-qPCR experiment. (C) DRIP-qPCR ± RNH treatment for the indicated genes in WT, R −/−, and T−/− cells. Dashed line represents background signal. Data are expressed as percentage of input. (Mean ± SEM; n > 3 independent experiments; two-tailed unpaired t test; *, P < 0.05; **, P < 0.01; ***, P < 0.001.) Source data are available for this figure: SourceDataF5.
Figure 5. AGS KO astrocytes accumulate R-loops. (A) Slot blot analysis of genomic DNA extracted from WT, R−/−, and T−/− cells treated with the recombinant RNase H (RNH) enzyme for 2.5 h at 37°C and probed with S9.6 (top) and ssDNA (bottom) antibodies. Slot blot probed with the antibody recognizing ssDNA was used as a loading control. (Mean ± SEM; n = 3 independent experiments; two-tailed unpaired t test; *, P < 0.05; **, P < 0.01.) (B) Scheme of DRIP-qPCR experiment. (C) DRIP-qPCR ± RNH treatment for the indicated genes in WT, R −/−, and T−/− cells. Dashed line represents background signal. Data are expressed as percentage of input. (Mean ± SEM; n > 3 independent experiments; two-tailed unpaired t test; *, P < 0.05; **, P < 0.01; ***, P < 0.001.) Source data are available for this figure: SourceDataF5.CC BY · PubMed Central
Figure 6. KO iPSC–derived astrocytes secrete neuroinflammatory mediators. (A) Heatmap representation showing gene expression analysis of interleukin genes in T−/− and R−/− KO cells compared with WT. (B) ELISA results for inflammatory cytokines (IL8, IL6, IL1β, TNFα, and CXCL10) in the supernatant of WT and KO astrocytes at steady state. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05.) (C and D) Volcano plots highlighting the differentially secreted proteins in R−/− (C) and T−/− (D) compared with WT samples. Log2(FC) and adjusted P values are reported for each protein. Upregulated common secreted proteins between the two comparisons are highlighted in blue. Proteins common with published neurotoxic factors are highlighted in red. (E) Barplot highlighting the significant enriched functional pathways resulting from the WikiPathways database of secreted proteins common with published neurotoxic factor. (F) Bubble plot showing the differentially secreted neurotoxic factors involved in the identified enriched pathways.
Figure 6. KO iPSC–derived astrocytes secrete neuroinflammatory mediators. (A) Heatmap representation showing gene expression analysis of interleukin genes in T−/− and R−/− KO cells compared with WT. (B) ELISA results for inflammatory cytokines (IL8, IL6, IL1β, TNFα, and CXCL10) in the supernatant of WT and KO astrocytes at steady state. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05.) (C and D) Volcano plots highlighting the differentially secreted proteins in R−/− (C) and T−/− (D) compared with WT samples. Log2(FC) and adjusted P values are reported for each protein. Upregulated common secreted proteins between the two comparisons are highlighted in blue. Proteins common with published neurotoxic factors are highlighted in red. (E) Barplot highlighting the significant enriched functional pathways resulting from the WikiPathways database of secreted proteins common with published neurotoxic factor. (F) Bubble plot showing the differentially secreted neurotoxic factors involved in the identified enriched pathways.CC BY · PubMed Central
Figure 7. KO and AGS patient iPSC–derived astrocytes mediate neurotoxicity. (A–C) WT neurons expression of cleaved Caspase-3 (cC3) and phosphorylated γH2AX after 48-h exposure to KO iPSC–derived astrocyte conditioned medium, measured by IF staining. Representative pictures and quantification. Each dot corresponds to one field from three slides of three independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 3 independent experiments; Kruskal–Wallis test; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.) CM, conditioned medium. (D) Reduced γH2AX upon exposure of WT neurons to conditioned medium collected from WT and TREX1- or RNASEH2B-deficient astrocytes and then supplemented with Complement C5 inhibitor eculizumab or conditioned medium collected from WT and TREX1- or RNASEH2B-deficient astrocytes overexpressing dominant-negative p53-inhibiting peptide GSE56. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; ****, P < 0.0001.) (E) Reduced γH2AX and cleaved Caspase-3 immunopositive area upon exposure of WT neurons to conditioned medium collected from astrocytes treated with anakinra (ANA, IL-1β inhibitor), infliximab (IFX, anti-TNFα antibody), or Z-Vad (pan-caspase inhibitor) for 24 h. γH2AX and cleaved Caspase-3 levels expressed as over WT condition. (Mean ± SEM; n = 3 independent experiments; Kruskal–Wallis test; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.) (F) Reduced γH2AX immunopositive area upon exposure of WT neurons to conditioned medium collected from astrocytes supplemented with anti-TNFα (infliximab, IFX), anti–IL-1β (canakinumab, CNK), or anti-IL8 for 48 h. γH2AX expressed as fold over WT condition. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.)
Figure 7. KO and AGS patient iPSC–derived astrocytes mediate neurotoxicity. (A–C) WT neurons expression of cleaved Caspase-3 (cC3) and phosphorylated γH2AX after 48-h exposure to KO iPSC–derived astrocyte conditioned medium, measured by IF staining. Representative pictures and quantification. Each dot corresponds to one field from three slides of three independent differentiations. Scale bar, 50 µm. (Mean ± SEM; n = 3 independent experiments; Kruskal–Wallis test; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.) CM, conditioned medium. (D) Reduced γH2AX upon exposure of WT neurons to conditioned medium collected from WT and TREX1- or RNASEH2B-deficient astrocytes and then supplemented with Complement C5 inhibitor eculizumab or conditioned medium collected from WT and TREX1- or RNASEH2B-deficient astrocytes overexpressing dominant-negative p53-inhibiting peptide GSE56. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; ****, P < 0.0001.) (E) Reduced γH2AX and cleaved Caspase-3 immunopositive area upon exposure of WT neurons to conditioned medium collected from astrocytes treated with anakinra (ANA, IL-1β inhibitor), infliximab (IFX, anti-TNFα antibody), or Z-Vad (pan-caspase inhibitor) for 24 h. γH2AX and cleaved Caspase-3 levels expressed as over WT condition. (Mean ± SEM; n = 3 independent experiments; Kruskal–Wallis test; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.) (F) Reduced γH2AX immunopositive area upon exposure of WT neurons to conditioned medium collected from astrocytes supplemented with anti-TNFα (infliximab, IFX), anti–IL-1β (canakinumab, CNK), or anti-IL8 for 48 h. γH2AX expressed as fold over WT condition. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.)CC BY · PubMed Central
Figure S5. Impact of the different treatments on KO astrocyte inflammatory profiles. (A) Line 1 (5′ UTR) mRNA expression levels were measured at different time points during differentiation protocol in WT, KO, and AGS patient–derived cells, expressed as fold difference vs. WT and normalized to the GAPDH housekeeping gene. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; **, P < 0.01.) (B) Representative FACS plot and fold reduction quantification of GFP + iPSC-derived astrocytes 5 d after lentiviral vector transduction at MOI 10 with and without the RTi cocktail. (Mean ± SD; n = 2 independent experiments; one-tailed Mann–Whitney U test; ****, P < 0.0001.) (C) Violin plots showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with RTi (n = 3 independent experiments: Wilcoxon rank-sum test; *, P < 0.05). (D) WT neuron expression of cleaved Caspase-3 (cC3) upon 48-h exposure to KO iPSC–derived astrocyte conditioned medium treated with the RTi cocktail measured by IF staining. Quantification of two slides of two independent differentiations. (Mean ± SD; n = 2 independent experiments.) (E) Violin plots showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with infliximab (IFX), anakinra (ANA), or Zvad. (n = 5–6 independent experiments.) (F and G) Representative images of WT neurons treated or not with conditioned medium collected from AGS patient–derived and KO astrocytes ± ANA, IFX, or Zvad. One image per genotype per condition shown. Scale bar, 50 µm.
Figure S5. Impact of the different treatments on KO astrocyte inflammatory profiles. (A) Line 1 (5′ UTR) mRNA expression levels were measured at different time points during differentiation protocol in WT, KO, and AGS patient–derived cells, expressed as fold difference vs. WT and normalized to the GAPDH housekeeping gene. (Mean ± SEM; n = 3 independent experiments; one-tailed Mann–Whitney U test; *, P < 0.05; **, P < 0.01.) (B) Representative FACS plot and fold reduction quantification of GFP + iPSC-derived astrocytes 5 d after lentiviral vector transduction at MOI 10 with and without the RTi cocktail. (Mean ± SD; n = 2 independent experiments; one-tailed Mann–Whitney U test; ****, P < 0.0001.) (C) Violin plots showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with RTi (n = 3 independent experiments: Wilcoxon rank-sum test; *, P < 0.05). (D) WT neuron expression of cleaved Caspase-3 (cC3) upon 48-h exposure to KO iPSC–derived astrocyte conditioned medium treated with the RTi cocktail measured by IF staining. Quantification of two slides of two independent differentiations. (Mean ± SD; n = 2 independent experiments.) (E) Violin plots showing the distribution of IFN scores calculated from the median FC of six ISGs in T−/− and R−/− KO cells treated with infliximab (IFX), anakinra (ANA), or Zvad. (n = 5–6 independent experiments.) (F and G) Representative images of WT neurons treated or not with conditioned medium collected from AGS patient–derived and KO astrocytes ± ANA, IFX, or Zvad. One image per genotype per condition shown. Scale bar, 50 µm.CC BY · PubMed Central

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 131 references

Research works printed but not linked: 2

  1. Stetson, 2008, Trex1 prevents cell-intrinsic initiation of autoimmunity, Cell, 134, 587, 10.1016/j.cell.2008.06.032 · DOI · Paper page
  2. Tumienė, 2017, Inflammatory myopathy in a patient with Aicardi-Goutières syndrome, Eur. J. Med. Genet., 60, 154, 10.1016/j.ejmg.2016.12.004 · DOI

Cited by

See all 84

Metadata

Authors
Giordano, A. M. S., Luciani, M., Gatto, F., Abou Alezz, M., Beghè, C., Della Volpe, L., Migliara, A., Valsoni, S., Genua, M., Dzieciatkowska, M., Frati, G., Tahraoui-Bories, J., Giliani, S. C., Orcesi, S., Fazzi, E., Ostuni, R., D’Alessandro, A., Di Micco, R., Merelli, I., Lombardo, A., Reijns, M. A. M., Gromak, N., Gritti, A., & Kajaste-Rudnitski, A.
Affiliations
  • Vita-Salute San Raffaele University (IT)
  • The San Raffaele Telethon Institute for Gene Therapy (IT)
  • San Raffaele Telethon Institute for Gene Therapy, Istituto di Ricovero e Cura a Carattere Scientifico San Raffaele Scientific Institute, Milan, Italy 1
  • University of Oxford (GB)
  • Department of Biochemistry and Molecular Genetics, University of Colorado Anschutz Medical Campus, Aurora, CO 4
  • Azienda Socio Sanitaria Territoriale degli Spedali Civili di Brescia (IT)
  • University of Pavia (IT)
  • University of Colorado Anschutz (US)
  • Institute of Genetics and Cancer (GB)
Venue
Journal of Experimental Medicine
Published
2022-03-09
Volume
219
Issue
4
ISSN
  • 0022-1007 (ISSN-L)
  • 1540-9538
OpenAlex
W4220867373
Cited by
76
References
131
Citations by year
2026: 9 · 2025: 11 · 2024: 29 · 2023: 20 · 2022: 7
FWCI
6.39
Type
journal article
Language
en
Open access
Locations
  • The Journal of Experimental Medicine (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Archivio Istituzionale della Ricerca - Università Vita-Salute San Raffaele (submittedVersion, cc-by)
  • Institutional Research Information System (Università degli Studi di Brescia) (submittedVersion, other-oa)
  • Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, cc-by)
Funders
  • Fondazione Telethon
  • Royal Society (UF150656, RGF\EA\180023)
  • Medical Research Council (MC_UU_00007/5)
Topics
  • Life Sciences › Immunology and Microbiology › Immunology › interferon and immune responses
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Inflammasome and immune disorders
Keywords
Aicardi-Goutières syndrome, astrocytes, DNA damage, neuroinflammation, type I interferon, TREX1, RNASEH2B, R-loops, micronuclei, STING signaling, NLRP3 inflammasome, pro-inflammatory activation, patient-derived cells, neurotoxicity, single-cell transcriptomics, genome-wide transcriptomics
MeSH terms
Astrocytes, DNA Damage, Humans, Inflammation, Nervous System Malformations, Autoimmune Diseases of the Nervous System

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: DNA damage contributes to neurotoxic inflammation in Aicardi-Goutières syndrome astrocytes. https://app.cochat.ai/papers/10.1084/jem.20211121
The paper
Giordano, A. M. S., Luciani, M., Gatto, F., Abou Alezz, M., Beghè, C., Della Volpe, L., Migliara, A., Valsoni, S., Genua, M., Dzieciatkowska, M., Frati, G., Tahraoui-Bories, J., Giliani, S. C., Orcesi, S., Fazzi, E., Ostuni, R., D’Alessandro, A., Di Micco, R., Merelli, I., . . . Kajaste-Rudnitski, A. (2022). DNA damage contributes to neurotoxic inflammation in Aicardi-Goutières syndrome astrocytes. Journal of Experimental Medicine. https://doi.org/10.1084/jem.20211121

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.