Microglia are the immune cells of the central nervous system. They are suspected to play important roles in adult synaptogenesis and in the development of the neuronal network. Microglial cells originate from progenitors in the yolk sac. Although it was suggested that they invade the cortex at early developmental stages in the embryo, their invasion pattern remains largely unknown. To address this issue we analyzed the pattern of cortical invasion by microglial cells in mouse embryos at the onset of neuronal cell migration using in vivo immunohistochemistry and ex vivo time-lapse analysis of microglial cells. Microglial cells begin to invade the cortex at 11.5 days of embryonic age (E11.5). They first accumulate at the pial surface and within the lateral ventricles, after which they spread throughout the cortical wall, avoiding the cortical plate region in later embryonic ages. The invasion of the cortical parenchyma occurs in different phases. First, there is a gradual increase of microglial cells between E10.5 and E14.5. From E14.5 to E15.5 there is a rapid phase with a massive increase in microglia, followed by a slow phase again from E15.5 until E17.5. At early stages, many peripheral microglia are actively proliferating before entering the parenchyma. Remarkably, activated microglia accumulate in the choroid plexus primordium, where they are in the proximity of dying cells. Time-lapse analysis shows that embryonic microglia are highly dynamic cells.

In this study, the effect of glycine receptor (GlyR) α3 alternative RNA splicing on the distribution of receptors in the membrane of human embryonic kidney 293 cells is investigated using optical super-resolution microscopy. Direct stochastic optical reconstruction microscopy is used to image both α3K and α3L splice variants individually and together using single- and dual-color imaging. Pair correlation analysis is used to extract quantitative measures from the resulting images. Autocorrelation analysis of the individually expressed variants reveals clustering of both variants, yet with differing properties. The cluster size is increased for α3L compared to α3K (mean radius 92 ± 4 and 56 ± 3 nm, respectively), yet an even bigger difference is found in the cluster density (9,870 ± 1,433 and 1,747 ± 200 μm(-2), respectively). Furthermore, cross-correlation analysis revealed that upon co-expression, clusters colocalize on the same spatial scales as for individually expressed receptors (mean co-cluster radius 94 ± 6 nm). These results demonstrate that RNA splicing determines GlyR α3 membrane distribution, which has consequences for neuronal GlyR physiology and function.