Dynamic A-to-I RNA editing during acute neuroinflammation in sepsis-associated encephalopathy

Abstract

Introduction: The activation of cerebral endothelial cells (CECs) has recently been reported to be the earliest acute neuroinflammation event in the CNS during sepsis-associated encephalopathy (SAE). Importantly, adenosine-to-inosine (A-to-I) RNA editing mediated by ADARs has been associated with SAE, yet its role in acute neuroinflammation in SAE remains unclear.

Methods: Our current study systematically analyzed A-to-I RNA editing in cerebral vessels, cerebral endothelial cells (CECs), and microglia sampled during acute neuroinflammation after treatment in a lipopolysaccharide (LPS)-induced SAE mouse model.

Results: Our results showed dynamic A-to-I RNA editing activity changes in cerebral vessels during acute neuroinflammation. Differential A-to-I RNA editing (DRE) associated with acute neuroinflammation were identified in these tissue or cells, especially missense editing events such as S367G in antizyme inhibitor 1 (Azin1) and editing events in lincRNAs such as maternally expressed gene 3 (Meg3), AW112010, and macrophage M2 polarization regulator (Mm2pr). Importantly, geranylgeranyl diphosphate synthase 1 (Ggps1) and another three genes were differentially edited across cerebral vessels, CECs, and microglia. Notably, Spearman correlation analysis also revealed dramatic time-dependent DRE during acute neuroinflammation, especially in GTP cyclohydrolase1 (Gch1) and non-coding RNA activated by DNA damage (Norad), both with the editing level positively correlated with both post-LPS treatment time and edited gene expression in cerebral vessels and CECs.

Discussion: The findings in our current study demonstrate substantial A-to-I RNA editing changes during acute neuroinflammation in SAE, underlining its potential role in the disease.

Keywords: RNA editing; acute neuroinflammation; cerebral endothelial cells; cerebral vessels; microglia; sepsis-associated encephalopathy.

Figure 1

Overall changes in A-to-I RNA editing activities in cerebral vessels during acute neuroinflammation. (A,B) Expression of editing enzyme Adar and Adarb1 in cerebral vessels. (C) The average A-to-I RNA editing level and (D) number of editing events in cerebral vessels are shown. (E) Spearman correlation between the relative expression level and the avarge A-to-I RNA editing level in cerebral vessels. (F) Venn plot comparing A-to-I editing events detected in cerebral vessels among groups. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001.

Figure 2

Differential A-to-I RNA editing during acute neuroinflammation. (A–C) Manhattan plots showing GLM p-values of individual A-to-I RNA editing events in cerebral vessels, CECs, and microglia. The solid horizontal line indicates p = 1.0 × 10⁻⁷, and the dashed horizontal line indicates p = 1.0 × 10⁻⁴. (D) Functional category distribution of DRE events in cerebral vessels, CECs, and microglia. (E,F) Differential RNA editing events associated with Adar and Adarb1 expressions in cerebral vessels. (G–I) Wordcloud plots showing the frequency of RBPs with binding sites overlapping with DRE sites. (J) Venn plot showing six RBPs shared by the top 10 frequent RBPs in the cerebral vessels, CECs, and microglia shown in panels (G–I). GLM, generalized linear model; DRE, differential RNA editing; RBPs, RNA binding proteins; CECs, cerebral endothelial cells.

Figure 3

Genes differentially edited across cerebral vessels, CECs, and microglia during acute neuroinflammation. (A) Venn plot showing four common differentially edited genes in the tree tissues or cell types. (B) GO analysis showing biological processes significantly enriched by the four common differentially edited genes. CECs, cerebral endothelial cells; GO, gene ontology.

Figure 4

Time-dependent A-to-I RNA editing in cerebral vessels, CECs, and microglia during acute neuroinflammation. (A) Venn plot comparing the A-to-I editing events that highly correlated with time after LPS treatment (Spearman p < 0.05) in cerebral vessels, CECs, and microglia. (B,C) Spearman correlation of the top 25 differentially edited events strongly correlated with time in CECs and cerebral vessels. (D) RNA editing of Gch1:chr14:47155050 in cerebral vessels. (E) Spearman correlation between the editing level of Gch1:chr14:47155050 and Gch1 expression in cerebral vessels. (F) RNA editing of Gch1:chr14:47155050 in CECs. (G) Spearman correlation between the editing level of Gch1:chr14:47155050 and Gch1 expression in CECs. (H) RNA editing level of Norad:chr2:156390419 in cerebral vessels. (I) Correlation between levels of Norad:chr2:156390419 editing and Norad expression in cerebral vessels. (J) RNA editing level of Norad:chr2:156390419 in CECs. (K) The correlation between editing efficiency at Norad:chr2:156390419 editing and Norad expression in CECs. CECs, cerebral endothelial cells; Gch1, GTP cyclohydrolase1; Norad, non-coding RNA actived by DNA damage.