Grey matter volume changes and corresponding cellular metrics identified in a longitudinal in vivo imaging approach

Magnetic resonance imaging (MRI) of the brain combined with voxel-based morphometry (VBM) has revealed structural changes of grey and white matter in a range of neurological and psychiatric disorders. However, the cellular basis of volume changes observed with VBM has remained unclear. We devised an approach to systematically correlate changes in grey matter volume (GMV) with cellular composition. Mice were alternately examined with structural MRI and two-photon in vivo microscopy at three time points, taking advantage of age-dependent changes in brain structure. We chose to image fluorescently labelled cell nuclei, because these can be readily imaged in large tissue volumes and allow inferences on several structural parameters: (1) the physical volume as determined from a subset of nuclei used to generate a geometrically defined space, (2) the number of cells, (3) the nearest neighbour distance measured between all nuclei as an indicator of cell clustering, and (4) the volume of the cell nuclei. Using this approach, we found that physical volume did not significantly correlate with GMV change, whereas mean nuclear volume was inversely correlated. When focusing on layers within the imaging volume, positive correlations of GMV were found with cell number near the cortical surface and nearest neighbour distance in deeper layers. Thus, the novel approach introduced here provided new insights into the factors underlying grey matter volume changes.