A Meta-Analysis of Microglial Heterogeneity and Gene Coexpression Networks in Aging and Neurodegeneration

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Microglia are a key brain cell population pivotal for homeostasis, and in brain aging and neurodegeneration they exhibit both adaptive phagocytic and maladaptive proinflammatory responses. To understand this dynamics, I independently analyzed five human datasets, totaling 83,190 microglial nuclei and covering Alzheimer’s disease, Parkinson’s disease, amyotrophic lateral sclerosis, frontotemporal lobar degeneration, and aging, as well as one Macaca fascicularis hippocampal aging dataset with 39,589 nuclei. Homeostatic microglia subpopulations in humans exhibited marked heterogeneity in GRID2 and CCDC26 expression. The expansion of GPNMB/IQGAP2-expressing clusters of phagocytic microglia was found to be the most conservative and reproducible response of microglia across pan-neurodegenerative conditions and aging, and the gene PTPRG from this signature was reproducibly evelated in pseudobulk; notably, the entire GPNMB⁺ signature is highly restricted to perivascular macrophages in macaques. Microglia subpopulations expressing ribosomal genes at high levels, along with FTL/H1, C1QA/B/C, TPT1, PLEKHA7, etc., although sometimes considered pathogenic, tended to be youth/health-enriched, and their coexpression modules were strongly increased in young and healthy samples. I also observed ST18⁺ and NRG3⁺ subpopulations, which reproduced across all human datasets and in monkeys, although their nature is not yet completely understood: some marker genes arise from myelin phagocytosis (e.g., ST18), while others may reflect intrinsic microglial expression (e.g., PCDH9). Gene coexpression network analysis revealed SORL1, MAML3, ELMO1, MEF2A/C, ADAM28, ABCC4, and several other genes as key putative regulators of evolutionarily conserved homeostatic modules, while identifying TANC2, NHSL1, FMN1, ETV6, and particularly DPYD as potential central players in activation-related networks. Transcriptional factor analysis revealed FOXP1 as a positive regulator of activation-related genes (e.g., DPYD and LRRK2), highlighted the role of TCF4/12 in activation (via TMEM163 and LRRK2), showed MITF and POU3F1 as regulators of numerous genes within the GPNMB-IQGAP2 signature, and identified FOXP2, FOXN3, FLI1, and KLF12 as key homeostatic regulators, alongside the well-studied MEF2A/C. Although many of these results are highly preliminary, this may shed light on microglial physiology in health and disease.

Article activity feed