Structural and Functional Connectivity of Human Thalamo-Cortical Communication

Introduction: To enable a precise thalamo-cortical communication the thalamus is organized in a number of nuclei, which are closely connected with each other and project to defined subcortical and cortical areas in a sophisticated feedback and feedforward wiring manner (Douglas and Martin, 2007). So far, the human thalamo-cortical connectivity has mainly examined using cortical areas as seed regions, while reverse approaches are limited due to the lack of validly given thalamic substructures. Therefore, we studied the structural and functional properties of thalamo-cortical communication using 15 thalamic clusters derived from an instantaneous correlation parcellation method (van Oort et al., 2014). Methods: Volunteers: Structural, resting state functional, and diffusion MR data of 40 volunteers (20 male 20 female; mean age 31 years) were obtained from the Human Connectome Project (Van Essen et al., 2012). MR Data specification: rsfMRI (resting state fMRI): Four sessions; multiband multi-slice-EPI, 1200 scans/session, duration: 14:33 min, TR: 720 ms, TE: 33.1 ms, voxel size: 2 mm isotropic, 72 slices, multiband factor: 8. Diffusion Spectrum Imaging (DSI): The DSI data were acquired using a spin-echo EPI sequence, TR: 5520 ms, TE: 89.5 ms, flip angle; 78 degree; voxel size: 1.25 mm isotropic, 111 slices, multiband factor: 3, echo spacing: 0.78 ms, b-values: 1000, 2000, and 3000. Structural-Connectivity: Motion and distortion corrected data were reconstructed using the multi-shell model (FSL) (Jenkinson et al., 2012). The probabilistic tractography was performed with FSL-probtrakx using each cluster as a seed region to the ipsilateral hemispheric. The tractograms were normalized and group fixed effects were computed across all subjects. Finally, all tracks in each hemisphere were combined and the strongest determined using the winter-takes-it-all (WTA) approach. WTA maps were visualized using BrainVoyager (s. Fig. 1-2). Functional-Connectivity: For the functional connectivity each cluster served as a seed mask to extract the time courses from 4 rsfMRI sessions in each subject and then a partial regression map was calculated for each cluster. Finally, group maps of functional connected cortical areas were computed using a fixed effect analysis. The strongest cortical correlation of each cluster was determined again using the WTA approach. Results: The fiber maps showed an organized pattern with a compact parcellation of the whole cortex into eight larger compartments (s. Fig 1a-b). In addition, a rough hemispheric symmetry was found. The functional connectivity of the WTA maps reveal in contrast a disseminated pattern, which was not confined certain lobes. Therefore no clear assignment of clusters to cortical areas could be established. However, by applying a minimal threshold of 0.002- 0.005 this pattern was reduced and more confined to specific cortical regions (s. Fig. 2). Conclusions: The structural connections showed compact and specific projections to various cortical and subcortical areas, which roughly concur with reported thalamo-cortical projections (Mai and Forutan, 2012; Nieuwenhuys et al., 2008). However, the functional connections were much more wide-spread and not confined certain lobes. This striking differences between structural and functional connectivity emphasizes that hard wired connections determined by diffusion imaging are of limited use to mirror the functional variability of neuronal connections to and within the cortex. Finally, in routing thalamo-cortical connections one has keep in mind that this connectivity is not restricted to the cortex, but that numerous communications are devoted to the basal ganglia, brainstem and cerebellum