Neuroplasticity is the capacity of the nervous system to modify its organization as a result of a dynamic internal or external environment. These modifications occur in different levels and timescales. Structural morphological changes such as formation of new synapses, gliogenesis, and changes in oligodendrocytes and myelin formation are considered long scale plasticity modifications. In this study we aim to create a framework for identification and localization of plasticity dynamics in the mouse brain as a result of a task with two degrees of difficulty, using non-invasive imaging techniques (magnetic resonance imaging-MRI). We focused on diffusion tensor imaging (DTI) which enables us to visualize and characterize white matter fasciculi and gray matter regions. These examinations were accompanied by behavioral tests. Since our focus is on the dynamic aspect of plasticity, we seek a paradigm which includes a longer acquisition and consolidation period of memory and learning. In order to achieve that goal, we assessed motor learning ability using a running wheel with irregularly spaced rungs ("complex wheel").

Behavioral paradigm:

8 mice were placed in cages equipped with a running wheel with evenly spaced rungs ("regular wheel", one per cage). The mice were first allowed free access to the regular wheel for three weeks. Following that some of the wheel rungs were removed to create a more complex motor task (complex wheel). The mice were allowed free access to the complex wheel for an additional week. MRI was performed before they were placed in the wheel cages, following the first learning period (3-weeks) and following the complex motor learning period (1 month).

Imaging parameters:

Mice underwent MRI scans on a Bruker 7T system (Biospec 30/70). MRI was performed before the mice were placed in the wheel cages, following the first learning period (3-weeks) and following the complex motor learning period (1 month). The imaging protocol included a diffusion tensor imaging (DTI) protocol consisting of 35 diffusion weighted images (DWIs) sampled at b=1000 s/mm2 (32 directions) and b=0 (3 replicas). The DTI dataset was analyzed in ExploreDTI to produce fractional anisotropy (FA) and mean diffusivity (MD) maps. These maps were registered and normalized to the mouse brain template, before a voxel-by-voxel repeated measure ANOVA was performed (1 group, 3 measuring time points). The statistical threshold was set to p<0.005 and, for voxels that passed this threshold, post-hoc analysis was performed to reveal which of the time points contributed to the effect.

DTI statistical analysis revealed a significant effect of regular wheel training, which induced micro-structural changes in several brain regions. The MD analysis revealed reductions in MD in the entorhinal cortex, medullary lamina, raphe nucleus, trigeminal root, the corpus callosum, the caudate/putamen, S1/S2 cortex, the geniculate bodies, cerebral peduncles (part of the cortico-spinal tract) and M1/M2 cortex (Fig. 1). The FA analysis revealed changes in M1/M2, nucleus accumbens, corpus callosum, hippocampus, thalamus and piriform cortex (Fig. 2). Post hoc analysis revealed that the regular wheel learning affected mainly sub-cortical and brain stem regions including the medial lemniscus, raphe nucleus, trigeminal root, Caudate/putamen and cerebral peduncles. The complex wheel affected cortical regions including entorhinal cortex, S1/S2, M1/M2 and corpus callosum. Diffusion MRI revealed significant micro-structural changes in multiple brain areas expected to be affected by this task. The most important of those are related to the motor domain (e.g. M1/M2), sensory perception regions (S1/S2) and white matter tracts (cortico-spinal tract and corpus callosum). Our preliminary results suggest that long acquisition of learning and consolidation period induced structural plasticity.