Structural and functional connectivity in the default mode network in 22q11.2 deletion syndrome

Abstract

Background: The neural endophenotype associated with 22q11.2 deletion syndrome (22q11DS) includes deviant cortical development and alterations in brain connectivity. Resting-state functional magnetic resonance imaging (fMRI) findings also reported disconnectivity within the default mode network (DMN). In this study, we explored the relationship between functional and structural DMN connectivity and their changes with age in patients with 22q11DS in comparison to control participants. Given previous evidence of an association between DMN disconnectivity and the manifestation of psychotic symptoms, we further investigated this relationship in our group of patients with 22q11DS.

Methods: T1-weighted, diffusion, and resting-state fMRI scans were acquired from 41 patients with 22q11DS and 43 control participants aged 6 to 28 years. A data-driven approach based on independent component analysis (ICA) was used to identify the DMN and to define regions of interest for the structural and functional connectivity analysis. Prodromal psychotic symptoms were assessed in adolescents and adults using the positive symptom scores of the Structured Interview of Prodromal Syndromes (SIPS). Connectivity measures were compared between groups and correlated with age. Repeating the between-group analysis in three different age bins further assessed the presence of age-related alterations in DMN connectivity. Structural and functional connectivity measures were then correlated with the SIPS scores.

Results: A simultaneous reduction of functional and structural connectivity between core medial nodes of the DMN was observed. Furthermore, structural connectivity measures significantly increased with age in the control group but not in patients with 22q11DS, suggesting the presence of an age-related alteration of the DMN structural connections. No correlations were found between the DMN disconnectivity and expression of prodromal symptoms in 22q11DS.

Conclusions: These findings indicate the presence of functional and structural DMN disconnectivity in 22q11DS and that patients with 22q11DS fail to develop normal structural connections between medial DMN nodes. This suggests the presence of altered neurodevelopmental trajectories in 22q11DS.

Keywords: DTI; Maturation; Positive symptoms; Resting-state fMRI; Schizophrenia; Tractography.

Fig. 1

Nine resting-state networks (RSNs), which resulted from the independent component analysis (ICA)

Fig. 2

Structural connectivity between the anterior and the posterior node of the DMN (default mode network). DMN clusters were registered in DTI space (a) and used as regions of interest (ROIs) for the probabilistic tractography analysis (b). The connectivity map is represented in red/yellow colors and overlaid onto the fractional anisotropy (FA) map. The lighter the color, the greater the probability to have a connection. Structural connectivity measures (c) were significantly reduced in the 22q11DS population compared with controls (p < 0.05, indicated by *, p < 0.01, indicated by **)

Fig. 3

Correlation between structural connectivity and age. In the anterior-posterior (a, b) and anterior-left (c, d) default mode network (DMN) structural connectivity increases with age in control participants but not in patients with 22q11DS

Fig. 4

Structural connectivity in different age bins (6–12, 13–17, 18–28 years old). The mean number of tracts (a, c) and the mean connectivity values (b, d) in the anterior-posterior and anterior-left default mode network (DMN) connections were significantly reduced in the adult population (p < 0.05 indicated by *, p < 0.01, indicated by **)

Fig. 5

Functional connectivity analysis. The component corresponding to the default mode network (DMN) was isolated, unwrapped in subject space, and thresholded at z = 2 (a). Spheres of 8-mm radius were used as regions of interest (ROIs) for the functional connectivity analysis (b). Partial correlation revealed weaker functional connectivity between the anterior and the posterior DMN nodes (c) and between the left-right IPL (d) in the 22q11DS group compared to the control group (p < 0.01, indicated by **)