Abnormal Resting-State Connectivity in a Substantia Nigra-Related Striato-Thalamo-Cortical Network in a Large Sample of First-Episode Drug-Naïve Patients With Schizophrenia

Abstract

Objective: The dopamine hypothesis is one of the most influential theories of the neurobiological background of schizophrenia (SCZ). However, direct evidence for abnormal dopamine-related subcortical-cortical circuitry disconnectivity is still lacking. The aim of this study was therefore to test dopamine-related substantia nigra (SN)-based striato-thalamo-cortical resting-state functional connectivity (FC) in SCZ.

Method: Based on our a priori hypothesis, we analyzed a large sample resting-state functional magnetic resonance imaging (fMRI) dataset from first-episode drug-naïve SCZ patients (n = 112) and healthy controls (n = 82) using the SN as the seed region for an investigation of striato-thalamo-cortical FC. This was done in the standard band of slow frequency oscillations and then in its subfrequency bands (Slow4 and Slow5). Results: The analysis showed in SCZ: (1) reciprocal functional hypo-connectivity between SN and striatum, with differential patterns for Slow5 and Slow4; (2) functional hypo-connectivity between striatum and thalamus, as well as functional hyper-connectivity between thalamus and sensorimotor cortical areas, specifically in Slow4; (3) correlation of thalamo-sensorimotor functional hyper-connectivity with psychopathological symptoms. Conclusions: We demonstrate abnormal dopamine-related SN-based striato-thalamo-cortical FC in slow frequency oscillations in first-episode drug-naive SCZ. This suggests that altered dopaminergic function in the SN leads to abnormal neuronal synchronization (as indexed by FC) within subcortical-cortical circuitry, complementing the dopamine hypothesis in SCZ on the regional level of resting-state activity.

Fig. 1.

FC alteration in SCZ compared to HC in the different frequency bands (seed: SN). The t maps were thresholded at corrected P < .05. The color bar shows voxel-wise t values. Reduced FC is shown in blue, while increased FC is shown in red. In SCZ, the SN showed hypo-connectivity with the striatum in SFB (bilaterally), as well as in Slow5 (right SN) and Slow4 (left SN). Abbreviations: FC, functional connectivity; SN, Substantia Nigra; SCZ, schizophrenia; HC, healthy controls; SFB, Standard Frequency Band (0.01–0.08 Hz); Slow4 (0.027–0.073 Hz); Slow5 (0.01–0.027 Hz).

Fig. 2.

FC alteration in SCZ compared to HC in the different frequency bands (seed: St). The t maps were thresholded at corrected P < .05. The color bar shows voxel-wise t values. Reduced FC is shown in blue, while increased FC is shown in red. In SCZ, the striatum showed hypo-connectivity with the SN and thalamus in SFB, and mainly in Slow4 (SN and thalamus) rather than Slow5 (SN only). Abbreviations: FC, functional connectivity; St, Striatum; SCZ, schizophrenia; HC, healthy controls; SFB, Standard Frequency Band (0.01–0.08 Hz); Slow4 (0.027–0.073 Hz); Slow5 (0.01–0.027 Hz).

Fig. 3.

FC alteration in SCZ compared to HC in the different frequency bands (seed: Thal). The t maps were thresholded at corrected P < .05. The color bar shows voxel-wise t values. Reduced FC is shown in blue, while increased FC is shown in red. In SCZ, the thalamus showed hypo-connectivity with the striatum and SN as well as hyper-connectivity with the sensorimotor cortical areas in SFB, and Slow4, while no significant results were found in Slow5. Abbreviations: FC, functional connectivity; Thal, Thalamus; SCZ, schizophrenia; HC, healthy controls; SFB, Standard Frequency Band (0.01–0.08 Hz); Slow4 (0.027–0.073 Hz); Slow5 (0.01–0.027 Hz).

Fig. 4.

Functional disconnectivity in the SN-related striato-thalamo-cortical network and its clinical correlation in SCZ. Left side of the figure. The violin plots show the significant differences in mean FC in HC and SCZ separately, as well as the significant differences between the 2 groups, at the various relay stations of the SN-related striato-thalamo-cortical network in SFB. The SCZ group showed reduced positive FC between the SN and striatum as well as between the striatum and thalamus, when compared to highly positive FC in HC; by contrast, the SCZ group showed abnormal positive coupling between the thalamus and sensorimotor cortical areas, when compared to no FC in HC. ###P < .001 (1-sample t tests against zero); ***P < .001 (2-sample t tests). Right side of the figure. The scatter plots show in SCZ the partial correlation analysis (with age, gender and movement as covariates) between FC alterations at the various relay stations of the SN-related network in SFB and PANSS total score. A significant correlation between thalamo-sensorimotor FC alterations and psychopathological severity score was detected in SCZ. **P < .01 (correlation analysis). Abbreviations: FC, functional connectivity; SN, Substantia Nigra; St, striatum; Thal, Thalamus; SM Areas, SensoriMotor Areas; SCZ, schizophrenia; HC, healthy controls; SFB, Standard Frequency Band (0.01–0.08 Hz); PANSS, positive and negative syndrome scale.

Fig. 5.

Schema. In this model of SCZ, the dopamine-related SN functional disconnectivity leads to desynchronization in the activity within striatum and between striatum and thalamus with subsequent abnormal coupling between thalamus and sensorimotor cortical areas, in the slow frequency oscillations. This, in turn, leads to abnormal input and output processing by baseline intrinsic neuronal activity, finally leading to psychopathological symptoms. Abbreviations: FC, functional connectivity; SN, Substantia Nigra; St, Striatum; Thal, Thalamus; SM Areas, SensoriMotor Areas; SCZ, schizophrenia; HC, healthy controls.