Bone marrow-derived myeloid cells transiently colonize the brain during postnatal development and interact with glutamatergic synapses

Abstract

Although the roles of embryonic yolk sac-derived, resident microglia in neurodevelopment were extensively studied, the possible involvement of bone marrow-derived cells remains elusive. In this work, we used a fate-mapping strategy to selectively label bone marrow-derived cells and their progeny in the brain (FLT3⁺IBA1⁺). FLT3⁺IBA1⁺ cells were confirmed to be transiently present in the healthy brain during early postnatal development. FLT3⁺IBA1⁺ cells have a distinct morphology index at postnatal day(P)0, P7, and P14 compared with neighboring microglia. FLT3⁺IBA1⁺ cells also express the microglial markers P2RY12 and TMEM119 and interact with VGLUT1 synapses at P14. Scanning electron microscopy indeed showed that FLT3⁺ cells contact and engulf pre-synaptic elements. Our findings suggest FLT3⁺IBA1⁺ cells might assist microglia in their physiological functions in the developing brain including synaptic pruning which is performed using their purinergic sensors. Our findings stimulate further investigation on the involvement of peripheral macrophages during homeostatic and pathological development.

Keywords: Biological sciences; Cellular neuroscience; Developmental neuroscience; Natural sciences; Neuroscience.

Graphical abstract

Figure 1

Infiltrating BMDC density in the developing brain and their fractal-based morphology analysis (A–D) Representative images of infiltrating BMDC at P0, P7, P14, and P90 with representative image (inset) of the serial sections disposition used in the experiment, FLT3⁺IBA1⁺ cells are highlighted with white arrows. (E–G) Representative illustrations of the localization of some of the clusters of FLT3⁺IBA1⁺ cells found in the brain in P0, P7, and P14 animals. P90 animals did not present any FLT3⁺IBA1⁺ cell clusters. (H) Non-parametric Kruskal-Wallis statistical analysis of the density of BMDC in the brain across multiple ages (n = 6–8 animals). (I) Illustrations explaining the fractal dimension and lacunarity analysis. (J–L) Paired Student’s t test statistical analysis of the fractal dimension morphology analysis on BMDCs in the brain across ages compared to microglia (n = 12–15 cells per age, N = 7 animals). (M–O) Paired Student’s t test statistical analysis of the fractal lacunarity morphology analysis on FLT3⁺IBA1⁺ cells in the brain across ages compared to FLT3^-IBA1⁺ microglia (n = 12–15 cells per age, N = 7 animals). p < 0.05 was considered as statistically significant (∗ = p < 0.05, ∗∗ = p < 0.01, ∗∗∗ = p < 0.001, ∗∗∗∗ = p < 0.0001). Scale bar, 20 μm.

Figure 2

BMDCs are morphologically distinct from microglia, yet they express the microglial markers P2RY12 and TMEM119 and prune VGLUT1 synapses (A–D) Representative images using confocal microscopy of the FLT3⁺IBA1⁺ cells across ages. (E) Representative image of the P2RY12 staining in confocal microscopy at P14, FLT3⁺IBA1⁺ cells are highlighted with white arrows. (F) Representative image of the TMEM119 staining in confocal microscopy at P14, FLT3⁺IBA1⁺ are highlighted with white arrows. (G) Non-parametric Student’s t test with Welch’s correction statistical analysis of the cells with TMEM119 colocalization over the cells non-colocalized with TMEM119 in the brain at P14 (n = 12–15 cells, N = 3 animals). (H) Non-parametric Student’s t test with Welch’s correction statistical analysis of the cells with P2RY12 colocalization over the cells non-colocalized with P2RY12 in the brain at P14 (n = 12–15 cells, N = 4 animals). p < 0.05 was considered as statistically significant (∗∗∗∗ = p < 0.0001). Scale bar, 20 μm. (I and K) Representative image of the VGLUT1 staining on FLT3^-IBA1⁺ microglia in confocal microscopy at P14 (n = 1–3 cells, N = 5 animals) Scale bar, 5 μm. (J and L) Representative image of the VGLUT1 staining on FLT3⁺IBA1⁺ cells in confocal microscopy at P14 (n = 1–3 cells, N = 5 animals). Scale bar, 5 μm. Zoom-in was made to highlight the pruning event.

Figure 3

Light and scanning electron microscopy imaging of the FLT3⁺ cells in the developing brain parenchyma presenting thin processes contacting and engulfing synaptic elements (A) Brightfield images of FLT3⁺ cells in the brain parenchyma at P14 with zoomed in images of a ramified FLT3⁺ cell with macrophage features in (B–D). (E) Representative image of a FLT3⁺ cell in the piriform cortex of a P14 male mice with zoomed in images in (F–K). (F–I) Engulfed pre-synaptic axon terminals within FLT3⁺ cell body and processes. Altered mitochondria are also observed within the cytoplasm of the FLT3⁺ cell. (J and K) FLT3⁺ processes are shown directly juxtaposing a neuronal cell body. Red outline = cytoplasmic membrane, blue outline = nuclear membrane, orange pseudo-coloring = pre-synaptic axon terminal, purple pseudo-coloring = post-synaptic dendritic spine, red pseudo-coloring = phagosome, green pseudo-coloring = non-altered mitochondria, blue pseudo-coloring = altered mitochondria. (B) Scale bar, 200 μm; (C) Scale bar, 50 μm; (D) Scale bar, 25 μm; (E) Scale bar, 10 μm; (F-K) Scale bar, 5 μm.

Figure 4

Infiltrating FLT3⁺ cells exhibit a distinct molecular signature through the expression of peripheral proteins including CD45, Ly6c and CD11b (A) Representative flow cytometry analysis of the infiltrating FLT3⁺ cells in the control P14 and P90 mousebrain (n = 6 animals). (B) Representative flow cytometry analysis of the infiltrating FLT3⁺ cells in the FLT3; cre P14, and P90 mouse brain (n = 6 animals). Numbers represent percentage from parent gate (arrow).