A comparison of microglial morphological complexity in adult mouse brain samples using 2-dimensional and 3-dimensional image analysis tools

Characterizing cell morphology has been an essential aspect of neuroscience for over a century to provide essential insights into cellular function and dysfunction. Microglia, the resident innate immune cells of the central nervous system, undergo drastic changes in morphology in response to various stimuli, with many classifications proposed in recent years. Increased availability of advanced analysis software to study microglial morphology represents a step forward in the field. However, whether the use of advanced analysis tools provides equivalent or varied outcomes remains undetermined. This study re-analyzed raw data, previously processed using a standard 2D microglial morphology analysis method, using 3D analysis methods. The published article observed significant changes in microglial morphology in the mouse ventral hippocampus after administration of a ketogenic diet and exposure to repeated social defeat stress in young adult male mice. Overall, we observed different statistical outcomes in the 3D dataset compared to the previously published 2D results, with both maintained and new findings. This may indicate that the 3D analysis method is better able to capture minute changes in morphology. However, overall conclusions on microglial morphology changes remain consistent between methods. Lastly, we highlight the difference between a nested statistical design, which considers within animal variability, and a non-nested design. Overall, we highlight and discuss differences between 2D and 3D microglial morphology analysis and explore the contribution of individual cell and animal variability to statistical outcomes. Main PointsO_LI3D analysis method generates similar and novel results compared to a 2D method. C_LIO_LIverall microglial morphological changes to stimuli are comparable between methods. C_LIO_LInested statistical design produces distinct significant differences. C_LI