Introduction: Stem cell transplantation has emerged as a promising new experimental treatment for stroke; understanding its mechanism of action will facilitate the translation of stem cell therapy to the clinic. Previous work from our lab and others suggests that transplanted stem cells function by enhancing endogenous brain repair processes including structural brain plasticity. The ultimate change in brain plasticity is manifested at the synaptic level and thus we hypothesize that stem cells will enhance synaptic structural remodeling in the post-ischemic brain. To test this we use array tomography, a new high-resolution proteomic imaging method, to determine a) the number and subtype of glutamate and GABA synapses after stroke, and b) how these parameters are affected by transplantation of human neural progenitor cells (hNPCs). Method: Vehicle or hNPCs derived from fetal cortex were transplanted into the ischemic cortex of Nude rats at 7 days after distal middle cerebral artery occlusion. Neurological recovery was assessed weekly using a battery of behavioral tests. Small tissue was removed from the peri-infarct cortex at 4 weeks post-transplantation. The tissue was processed and ribbons, or arrays, of serial ultrathin sections (70 nm) were obtained using an ultramicrotome. Ribbons were stained with antibodies for the synaptic markers Synapsin1, VGlut1, VGlut2, PSD-95, GAD, VGAT, GABAAR-α1, and images taken in cortical layer 2/3 and layer 5. Computational analysis of the resultant staining pattern was used to identify and quantify subtypes of glutamatergic and GABAergic synapses. Results: Transplantation of hNPCs significantly improved behavioral recovery after stroke compared to vehicle-treated rats (4 weeks; p<0.01, n=9). There was an increase in the density and proportion of glutamatergic synapses expressing VGluT2 in layer5 at 4 weeks post-transplantation (Density: 0.090 vs 0.057 synapses/μm3. Proportion: 27.0 vs 22.6 %, n=4). No detectable differences in the density of GABAergic synapses were observed. Conclusions: These results suggest that stem cells alter synaptic remodeling after stroke and this is coincident with stem cell-induced functional recovery.