Glutathione precursor N-acetyl-cysteine modulates EEG synchronization in schizophrenia patients: a double-blind, randomized, placebo-controlled trial

Abstract

Glutathione (GSH) dysregulation at the gene, protein, and functional levels has been observed in schizophrenia patients. Together with disease-like anomalies in GSH deficit experimental models, it suggests that such redox dysregulation can play a critical role in altering neural connectivity and synchronization, and thus possibly causing schizophrenia symptoms. To determine whether increased GSH levels would modulate EEG synchronization, N-acetyl-cysteine (NAC), a glutathione precursor, was administered to patients in a randomized, double-blind, crossover protocol for 60 days, followed by placebo for another 60 days (or vice versa). We analyzed whole-head topography of the multivariate phase synchronization (MPS) for 128-channel resting-state EEGs that were recorded at the onset, at the point of crossover, and at the end of the protocol. In this proof of concept study, the treatment with NAC significantly increased MPS compared to placebo over the left parieto-temporal, the right temporal, and the bilateral prefrontal regions. These changes were robust both at the group and at the individual level. Although MPS increase was observed in the absence of clinical improvement at a group level, it correlated with individual change estimated by Liddle's disorganization scale. Therefore, significant changes in EEG synchronization induced by NAC administration may precede clinically detectable improvement, highlighting its possible utility as a biomarker of treatment efficacy.

Trial registration: ClinicalTrials.gov NCT01506765.

Figure 1. Enrollment, randomization, withdrawals, and completion of the 2 treatment phases (n = 11).

Figure 2. NAC vs. Placebo changes in population multivariate EEG synchronization.

The whole-head maps for the MPS show the surface topography of the NAC vs. Placebo effect. The maps are presented for CAR EEG data (top) and Laplacian data (bottom) for the theta (θ), alpha (α), beta (β), and gamma (γ) frequency bands. They are superimposed on the diagrams of the Geodesic 128-channel Sensor Net. The sensors corresponding to the International 10–20 System are shown with black circles. They are labeled in the upper left diagram. The large circles (irrespective of color) designate significant effect. The red sensors correspond to NAC>Placebo. All the effects are shown at FDR<0.05 and the sizes of effects are thresholded at 1 for CAR EEG and 0.7 for Laplacian. The colored surface (obtained by a trilinear interpolation from the three nearest electrodes) represents the effect size (see Materials and Methods for details).

Figure 3. NAC vs. Placebo changes in individual multivariate EEG synchronization.

The whole-head maps for the MPS show the surface topography of the NAC vs. Placebo effect for individual patients. The reported significant changes are restricted to the sensors and frequency bands that demonstrate a significant effect at the population level (Fig. 1), including four frequency bands for CAR EEG (θ, α, β, γ) and two frequency bands for Laplacian (β, γ). Patients are labeled as P2, P3, P4, P5, P6, P7, P9 and P13. Other designations are as in Figs. 2 and S1.

Figure 4. Surface topography of correlations between MPS and Liddle's factor of disorganization for NAC vs. Placebo contrast.

Significant Pearson correlations at FDR<0.05 are shown in the alpha (α) and beta (β) frequency bands for CAR EEG and Laplacian data. Blue circles designate the sensors with inverse correlations. The size of the correlations is reported as the coefficient of determination defined as the square of the Pearson correlation value (R²).