Abstract:
Microglia are the resident immune cells of the brain and play an important role in regulating brain function in health and disease. They are organized in a brain-wide network, with each cell having its own territory. It is generally recognized that microglia are long-lived cells, and individual cortical microglia can survive for a lifetime in a laboratory mouse. How the microglial network is maintained and how aging and disease affect the homeostasis of the network is not yet fully understood.
To investigate the homeostasis of the microglial network, I pursued two approaches. First, I describe my efforts to develop a new mouse model to study the response of the microglial network after ablation of individual microglia in vivo. The mouse model co-expresses the diphtheria toxin receptor (iDTR) and tdTomato dependent on Cre-recombination in a small percentage of microglial cells, rendering tdTomato-positive microglia susceptible to diphtheria toxin-induced cell death. Unfortunately, Cre-recombination of tdTomato and iDTR rarely, if ever, occurred in the same cell. Most likely, differences in length between the two loxP sites flanking the STOP cassette hampered success of this in vivo approach.
In the second part of the thesis, I describe the subsequent development of a novel hippocampal slice culture model as a simplified in vitro model to study microglia network homeostasis. In this model, the endogenous murine microglia were replaced by human induced pluripotent stem cells (iPSC)-derived microglia (iMics), facilitating the discovery of human microglial network changes. iMics in these chimeric hippocampal slice cultures differentiated and matured into microglia with a highly ramified morphology, transcriptional profile and network organization reminiscent of human microglia. In response to lipopolysaccharide stimulation or focal laser injury, iMics secrete pro-inflammatory cytokines or shield the injury site with their processes, respectively. Surprisingly, human colony-stimulating factor 1 (CSF1) was not required for iMic differentiation and survival in these chimeric hippocampal slice cultures, which contrasts with existing xenotransplantation models that express human CSF1. The observation that loss-of-function CSF1 receptor mutations diminish the integration of iMics into mouse brain slices suggests that cross-species ligand-receptor interactions of mouse CSF1 or interleukin 34 are sufficient for the differentiation and survival of iMics in the mouse brain slices.
To investigate how proteopathic lesions affect the homeostasis of this microglial network, chimeric slice cultures were combined with a recently developed -synucleinopathy hippocampal slice culture seeding model. Similar to what has been observed in mouse models of -synucleinopathy, also iMics in chimeric slice cultures develop -synuclein inclusions that accumulate over time and show a transcriptional response associated with neurodegeneration such as upregulation of the inflammatory response and increased phagocytosis.
While the investigation of in vivo microglial network homeostasis will require further adjustments of the mouse model for targeted ablation of individual microglia, the here developed chimeric slice cultures provide an easily accessible and scalable platform for in vitro study of human microglia under both homeostatic as well as diseased conditions.
