Patterns of Cortical Folding Associated with Autistic Symptoms in Carriers and Noncarriers of the 22q11.2 Microdeletion

Abstract

22q11.2 deletion syndrome (22q11.2DS) is a genetic condition accompanied by a range of psychiatric manifestations, including autism spectrum disorder (ASD). It remains unknown, however, whether these symptoms are mediated by the same or distinct neural mechanisms as in idiopathic ASD. Here, we examined differences in lGI associated with ASD in 50 individuals with 22q11.2DS (n = 25 with ASD, n = 25 without ASD) and 81 individuals without 22q11.2DS (n = 40 with ASD, n = 41 typically developing controls). We initially utilized a factorial design to identify the set of brain regions where lGI is associated with the main effect of 22q11.2DS, ASD, and with the 22q11.2DS-by-ASD interaction term. Subsequently, we employed canonical correlation analysis (CCA) to compare the multivariate association between variability in lGI and the complex clinical phenotype of ASD between 22q11.2DS carriers and noncarriers. Across approaches, we established that even though there is a high degree of clinical similarity across groups, the associated patterns of lGI significantly differed between carriers and noncarriers of the 22q11.2 microdeletion. Our results suggest that ASD symptomatology recruits different neuroanatomical underpinnings across disorders and that 22q11.2DS individuals with ASD represent a neuroanatomically distinct subgroup that differs from 22q11.2DS individuals without ASD and from individuals with idiopathic ASD.

Keywords: 22q11.2 deletion syndrome; autism spectrum disorder; brain anatomy; cortical folding; local gyrification index.

Figure 1

Results of the categorical analysis. Significant differences in local gyrification for (A) the main effect of 22q11.2DS, (B) the main effect of ASD, and (C) the 22q11.2DS-by-ASD interaction effect. Displayed are the unthresholded t-maps (left panel) and the random field theory (RFT)-based cluster corrected (P < 0.05, 2-tailed) t-maps (right panel). Here, increased parameter estimates in 22q11.2DS or ASD relative to their respective counterparts are marked in red to yellow, and decreased parameters are marked in blue to cyan.

Figure 2

CCA results for idiopathic ASD versus TD controls. (A) CCs (subplot) and CVs sorted in descending order, and based on the percentage of clinical variance explained; (B) clinical canonical loadings plot depicting correlations between each of the five clinical CVs and each of the five SRS subdomain scores in social awareness (SAW), social cognition (SCG), social communication (SCM), social motivation (SM), and restricted and repetitive behaviors (RRB). CVs are sorted in descending order based on the percentage of explained clinical variance as indicated in shades of green; (C) scatter plot depicting individual observations based on their scores on the first CV, which explained the largest percentage of clinical variance. Data points are colored by group (ASD vs. TD controls) and sized by the individual’s total SRS score; (D) canonical loadings of lGI on the first CV.

Figure 3

CCA results for 22q11.2DS individuals. (A) CCs (subplot) and CVs sorted in descending order, and based on the percentage of clinical variance explained; (B) clinical canonical loadings plot depicting correlations between each of the five clinical CVs and each of the five SRS subdomain scores in social awareness (SAW), social cognition (SCG), social communication (SCM), social motivation (SM), and restricted and repetitive behaviors (RRB). CVs are sorted in descending order based on the percentage of explained clinical variance as indicated in shades of green; (C) scatter plot depicting individual observations based on their scores on the third CV, which explained the largest percentage of clinical variance. Data points are colored by group (22q11.ASD vs. 22q11.nonASD) and sized by the individual’s total SRS score; (D) canonical loadings of lGI on the third CV.

Figure 4

Comparison of brain loadings between 22q11.2DS and non-22q11.2DS. Neuroanatomical loadings for lGI on the 1st and 3rd CV, which explained the largest percentage of clinical variance in each group, i.e., CV1 for the non-22q11.2DS group (first row) and CV3 for the 22q11.2DS group (second row). The third row shows the difference in loadings between the groups for each region. Differences in neuroanatomical loadings are displayed for (A) Left hemisphere and (B) right hemisphere. Numbers indicate a set of brain regions: 1 banks superiortemporal, 2 caudalanteriorcingulate, 3 caudalmiddlefrontal, 4 cuneus, 5 entorhinal, 6 fusiform, 7 inferiorparietal, 8 isthmuscingulate, 9 lateraloccipital, 10 lingual, 11 medialorbitofrontal, 12 middletemporal, 13 parahippocampal, 14 parsopercularis, 15 pericalcarine, 16 posteriorcingulate, 17 precuneus, 18 rostralanteriorcingulate, 19 rostralmiddlefrontal, 20 superiorparietal, 21 frontalpole, 22 insula, 23 caudalmiddlefrontal, 24 cuneus, 25 entorhinal, 26 fusiform, 27 inferiorparietal, 28 isthmuscingulate, 29 lateraloccipital, 30 lateralorbitofrontal, 31 lingual, 32 medialorbitofrontal, 33 parahippocampal, 34 pericalcarine, 35 precuneus, 36 rostralanteriorcingulate, 37 rostralmiddlefrontal, 38 superiorfrontal, 39 superiorparietal, 40 frontalpole. Canonical loadings displayed on the cortical surface (“fsaverage” brain) highlighting the set of brain regions where the multivariate association between neuroanatomical and clinical CVs differed (Δ) between groups.