Structure-function coupling in white matter uncovers the abnormal brain connectivity in Schizophrenia

Abstract

Schizophrenia is characterized by dysconnectivity syndrome. Evidence of widespread impairment of structural and functional integration has been demonstrated in schizophrenia. Although white matter (WM) microstructural abnormalities have been commonly reported in schizophrenia, the dysfunction of WM as well as the relationship between structure and function in WM remains uncertain. In this study, we proposed a novel structure-function coupling measurement to reflect neuronal information transfer, which combined spatial-temporal correlations of functional signals with diffusion tensor orientations in the WM circuit from functional and diffusion magnetic resonance images (MRI). By analyzing MRI data from 75 individuals with schizophrenia (SZ) and 89 healthy volunteers (HV), the associations between structure and function in WM regions in schizophrenia were examined. Randomized validation of the measurement was performed in the HV group to confirm the capacity of the neural signal transferring along the WM tracts, referring to quantifying the association between structure and function. Compared to HV, SZ showed a widespread decrease in the structure-function coupling within WM regions, involving the corticospinal tract and the superior longitudinal fasciculus. Additionally, the structure-function coupling in the WM tracts was found to be significantly correlated with psychotic symptoms and illness duration in schizophrenia, suggesting that abnormal signal transfer of neuronal fiber pathways could be a potential mechanism of the neuropathology of schizophrenia. This work supports the dysconnectivity hypothesis of schizophrenia from the aspect of circuit function, and highlights the critical role of WM networks in the pathophysiology of schizophrenia.

Fig. 1. The flowchart of calculating FCT-DT consistency.

a fMRI images were coregistered to DWI space using the transformation matrix acquired during the registration from fMRI images to DWI images, and then functional correlation tensor was computed. b The individual diffusion tensor was calculated in native DWI space derived from DWI images. c For each voxel, the FCT-DT consistency was estimated by measuring the Euclidean distance between the functional correlation tensor and diffusion tensor, which consequently generated the FCT-DT consistency map of the whole brain.

Fig. 2. The group differences in FCT-DT consistency in the WM tracts and its correlation to clinical symptoms.

a The abnormal WM tracts in SZ group. The color of WM tracts corresponds to the T value. SS_L left sagittal stratum, SLF superior longitudinal fasciculus, PTR posterior thalamic radiation, CST_R right corticospinal tract. b The correlation between FCT-DT consistency and clinical symptoms in SZ. Partial correlation was used with controlling sex, age, education and mean FD of fMRI. The significance level was set at P < 0.05 (uncorrected). RIC_L left retrolenticular part of internal capsule, PCR_L left posterior corona radiata, CGC_L left cingulum (cingulate gyrus), #: uncorrected.