P12.07.B Getting to the core of microglia versus bone marrow-derived macrophages in glioblastoma

Abstract

Abstract Background Microglia and bone marrow-derived macrophages (BMDMs) are two ontogenetically distinct myeloid populations present within glioblastoma that can comprise 30-50% of the tumour mass. Historically, these cell types have been conflated and studied as a single population of ‘tumour-associated macrophages’. Recent advances in single-cell omics have allowed population delineation, suggesting microglia and BMDMs may play different roles within the tumour and subsequently differentially affect tumour progression. Despite building evidence for the unique functions of these cells within glioblastoma, the inherent heterogeneity of the tumour landscape has complicated such studies. Indeed, macrophages exist as phenotypically and functionally diverse populations that are polarised in a context-dependent manner. Hence, to understand the differences between microglia and BMDMs within glioblastoma, both ontogeny and spatial location must be considered. Material and Methods To elucidate the functional roles of microglia and BMDMs across the tumour landscape, a publicly available RNAseq dataset was utilised to classify myeloid cells into four populations based on spatial location and ontogeny. These were tumour core BMDMs and microglia, or tumour periphery BMDMs and microglia. Differential gene analysis was then performed to identify significant differentially expressed genes (DEGs) between classified myeloid populations. Tumour core DEGs were then compared against the Ivy Glioblastoma Atlas to define their expression across anatomical tumour regions. Finally, myeloid DEGs were validated at the protein level on human glioblastoma tissue through immunohistochemistry. Results Microglia and BMDMs showed different spatial distributions across the tumour landscape and displayed distinct functional expression profiles. Microglia held a more chemotactic and pro-inflammatory profile, whereas BMDMs held a more pro-tumoural profile. However, a comparison of microglia between the tumour core and periphery revealed that tumour microglia upregulate many pro-tumoural genes, including multiple genes that have previously been defined as ‘BMDM-enriched’. Moreover, we found myeloid DEGs identified within the tumour core cluster to distinct spatial tumour regions such as the vascular or hypoxic niche. Immunohistochemical staining reflected these spatial expression profiles, identifying a distinct population of phagocytic macrophages within the hypoxic niche. Conclusion Although microglia and BMDMs represent two ontogenetically distinct myeloid populations within glioblastoma, both cell types can adopt similar functional expression profiles within the tumour core, particularly within tumour niches. This indicates that myeloid cell function is strongly influenced by the tumour microenvironment, rather than ontogeny alone.