Brain white matter microstructure associations with blood markers of the GSH redox cycle in schizophrenia

• Glutathione dysregulation may impact the brain white matter through oxidative or reductive stress in schizophrenia. • Advanced diffusion MRI (e.g., DKI, biophysical models) improves the sensitivity to demyelination and neuroinflammation as compared to DTI. • In healthy controls, blood markers of elevated reductive stress correlated with reduced brain microstructure complexity. • Blood markers of oxidative or reductive stress did not account for white matter abnormalities in our schizophrenia cohort. In groups of patients suffering from schizophrenia (SZ), redox dysregulation was reported in both peripheral fluids and brain. It has been hypothesized that such dysregulation, including alterations of the glutathione (GSH) cycle could participate in the brain white matter (WM) abnormalities in SZ due to the oligodendrocytes’ susceptibility to oxidative stress. In this study we aim to assess the differences between 82 schizophrenia patients (PT) and 86 healthy controls (HC) in GSH-redox peripheral blood markers: GSH peroxidase (GPx), reductase (GR) enzymatic activities and their ratio (GPx/GR-ratio), evaluating the hypotheses that alterations in the homeostasis of the systemic GSH cycle may be associated with pathological mechanisms in the brain WM in PT. To do so, we employ the advanced diffusion MRI methods: Diffusion Kurtosis Imaging (DKI) and White Matter Tract Integrity-Watson (WMTI-W), which provide excellent sensitivity to demyelination and neuroinflammation. We show that GPx levels are higher (p = 0.00041) in female control participants and decrease with aging (p = 0.026). We find differences between PT and HC in the association of GR and mean kurtosis (MK, p < 0.0001). Namely, lower MK was associated with higher blood GR activity in HC, but not in PT, suggesting that high GR activity (a hallmark of reductive stress) in HC was linked to changes in myelin integrity. However, GSH-redox peripheral blood markers did not explain the WM anomalies detected in PT, or the design of the present study could not detect subtle phenomenon, if present.