Authors :
Presenting Author: Marek Rajman, PhD – UCB
Anouk Verboven, PhD – UCB
Josephine Lenfant, MSc – UCB
Margot Fournier, PhD – GliaPharm SA
Sylvain Lengacher, PhD – GliaPharm SA
Charles Finsterwald, PhD – GliaPharm SA
Liesbeth Francois, PhD – Muna Therapeutics
Christian Wolff, PhD – UCB
Rationale:
Astrocytes play an important role in different CNS disorders including epilepsy. Recently we have identified gene co-expression modules enriched in glial cells including astrocytes which show differential expression in brain tissue from drug resistant epilepsy patients (mTLE+HS) (François et al., Nat Commun. 2024; 15(1); 2180). These modules were enriched in pathways involved in immune function, extracellular matrix and energy metabolism. We hypothesized that reversal of disease astrocyte mRNA signatures could attenuate astrogliosis, restore glio-transmission and metabolic state of the astrocytes and lead to seizure suppression in patients. Using causal reasoning analytical framework for target discovery (CRAFT) (Srivastava et al., Nat Commun. 2018; 9(1); 3561) p65 was identified as one of the predicted regulators of patient glial mRNA signatures. p65, member of canonical NFKB pathway, drives reactive astrocyte phenotype in CNS. To determine the role of p65 we have developed an in vitro primary mouse reactive astrocyte system which partially recapitulates patient mRNA signatures and showed that p65 knockdown (KD) reverses predicted regulatory pathways and restores astrocyte functions.Methods:
Primary hippocampal or cortical mouse astrocytes were subjected to 7 different inflammatory conditions (TNFa; IL1b; C1q; IFNg; C1q/IL1b/TNFa; IFNg/IL1b/TNFa; TGFb/IL1b/TNFa) over 3 time points (1-3-5 days). RNAseq was performed to assess the transcriptional signatures. High content imaging system / analysis (Revvity) and immunoassays (MSD) were used to characterize reactive astrocyte phenotype. KD of p65 was performed by LV-shRNAmirs or by siRNA smart pools.Results:
All inflammatory conditions except C1q alone induced global changes in gene expression. Furthermore, combination of different cytokines had the biggest impact on astrocyte gene expression. The three cytokine cocktails show the highest expression of reactive astrocyte markers. Increased expression of two human glial mTLE+HS modules (mTLE10.o and mTLE.20.o) was recapitulated in the mouse reactive astrocytes, with the most pronounced effects in the three cytokine cocktails. Finaly, IFNg/IL1b/TNFa cocktail was selected to test the role of p65 in mouse reactive astrocytes. KD of p65 by 3 different LV-shRNAmirs interfered with several CRAFT predictions for activation of mTLE.20.o module. Furthermore, partial reversal of immune pathways and A1-reactive astrocyte marker expression was observed. Changes in mRNA signatures reversed some of the astrocyte assessed reactive phenotypes – decreased Il6 secretion, restored astrocyte morphology (Gfap) and connectivity (Cx43) and attenuated aerobic glycolysis (Pkm2).
Conclusions:
We have developed an in vitro mouse model of reactive astrocytes which partially replicates mTLE+HS patient glial mRNA signatures and can be used to explore new therapeutic target hypothesis. Funding: No