Abstracts

Rationale:
The role of neuroinflammation driven by microgliosis in the process of epileptogenesis is generally appreciated but not well understood. Part of the challenge lies with the difficulty to distinguish the spectrum of activation states within the microglial population. At the extremes these are termed pro-inflammatory (M1) and anti-inflammatory (M2). Analysis of an early transcriptome study of dentate granule cells in three status epilepticus models revealed dominating roles for the transcriptional repressor, EZH2, in driving persistent gene expression changes that oppose disease progression after status epilepticus (Khan et al. PLoS ONE 2019). Our goal is to extend the analysis of dentate granule cells to all cell types of the hippocampus.

Methods:
Hippocampal tissue was harvested four days after pilocarpine-induced status epilepticus (SE) in eight week old C57Bl/6NCrl male mice and processed for single nucleus RNAseq. Data were obtained from five pilocarpine-treated and four saline-treated mice. Altogether, full transcriptomes were obtained from 84,796 individual cells. Clusters of cells based on transcriptome similarity were visualized with Uniform Manifold Approximation and Projection (UMAP). A novel informatics tool was created that allowed visualization in UMAP space of cells that express high levels of any combination of transcripts. The MAGIC algorithm (Roopra, PLOS Comp. Biol 2020) was used to identify the major transcription factors responsible for regulation of gene expression within each of the identified cell types.

Results:
Well separated discrete cell clusters in UMAP projection space were obtained for more than 23 cell types, including dentate granule cells, CA1 and CA3 pyramidal cells, various interneurons, astrocytes, endothelial cells, myeloid cells and oligodendrocytes. Within each of these clusters only minimal overlap was seen between pilocarpine and saline-treated mice, which suggests that SE changes the phenotype of most major cell types in hippocampus. Whereas EZH2 is the primary epigenetic repressor of dentate granule cells after SE, REST and MIER1 serve this function for activated microglia. Upregulated genes in activated microglia are driven strongly by NFIC. The single Leiden cluster representing myeloid cells was disambiguated into eleven subclusters, most of which could be identified from their transcriptome profile as freshly infiltrated monocytes (two clusters), resting microglia, perivascular macrophages, and activated microglia (five clusters). The activated microglia were loosely classified as M1 (two subclusters) and M2 (three subclusters, two of which had characteristics of Disease-Associated Microglia). A minor population of replicating microglia was identified by high levels of TOP2A and MK167. Studies are underway to compare transcriptional drivers in the resting and activated microglia subclusters.

Conclusions:
The results demonstrate the appearance of multiple subtypes of activated microglia after SE and an approach to understanding the transcriptional drivers of all cell types in hippocampus after SE.

Funding:
Supported by CURE (AR, NHV), Lily’s Fund (AR), NIH grants NS108756 (AR, RD), NS112308 (RD,NHV), and NS112350 (NHV).