Elucidating the Molecular Heterogeneity of Microglia

Neuroinflammationassociated with brain diseases, such as chronic neurodegeneration or brain tumours, has specific propertiesthat distinguish it from inflammation in the periphery (1, 2). Microglia, the main local immune effectors in the nervous system, are a unique type of resident macrophagesthatdiffer from their counterparts in the periphery, both in terms of origin, and in terms of activation profile (3). These cells can express, at least in vitro, an array of common, but also unique “resting” markers, as well as pro-and anti-inflammatory mediators, and these observations led to the assumption that there exists 3 microglial states: resting (M0),pro-inflammatory (M1), and anti-inflammatory (M2), each with distinct functional properties (4, 5). The situation in vivois considerably more complex though, where a combination of propertiesare more typical (6). Research on microglia in vivohas experienced a resurgence in the last 5 years since the emergence of protocols to isolate these cells out of adult tissues(human, or animal models), and theuse of profiling (“omics”)approaches to understand their biology (7).One key aspect of microglial biology that has emerged from these lines of research is that ofmicroglial heterogeneity. Microglia show variation in density, morphology, and, importantly, molecular properties across different brain regions (8, 9). However, within individual brain regions, no information is available on microglial heterogeneity at the molecular level. Such information is essential though to gain insights in the functionality of microglial cell sub-populations and their differential responses to disease in susceptible brain regions (10). We believe we are uniquely positioned to tackle this complex issue by the combination of 4 key areas of expertise: mouse brain anatomy and pathology (Buttini), microglia isolation and biology (Michelucci), single-cell RNA-seq profiling (Poovathingal/Skupin), andbioinformatics (Glaab). In the present project, which shall cover one year, we propose to look at the molecular heterogeneity of different mouse brain regions at baseline(naive mice): midbrain (with substantia nigra) and striatum as regions most susceptible to Parkinson’s disease(PD), hippocampus and cortex as regions most susceptible to Alzheimer’s disease(AD).Malignant gliomascan occur in any brain regions, hence all the baseline information obtained in all these regions will be of value for this line of research. These first investigations will pave the way for awider effort in 2018 aimed at analyzing the molecular response of microglial sub-populations in the context of specific brain diseasesalong two main axis: brain tumours (gliomas), and neurodegenerative diseases (PD, AD), which are the focus of LIH and LCSB, respectively. Pilot results obtained by single-cell RNA-seq profiling of microglia isolated from different mouse glioma models show specific microglial transcriptional signatures associated with different tumour features. We believe that uncovering microglial heterogeneity at the single-cell level will reveal brain-region specific microglial sub-populations whose reactionto disease and injury may modulate disease outcome and response to therapeutic intervention.