The Cingulate Structures in Autism Spectrum Disorder: Exploring Its Implication on Social Awareness and the Role of CNTNAP2

Backgrounds Although evidence suggests that the activity of the anterior cingulate cortex involves social cognition, there are inconsistent results regarding the aberrant cingulate gray matter (GM) and scanty evidence about altered cortical thickness and white matter (WM) of cingulate in individuals with autism spectrum disorder (ASD). Evidence supports the association between the autism risk gene CNTNAP2 variants and altered brain connectivity. This study investigated the cingulate substructure and its association with social awareness decits and the CNTNAP2 variants in individuals with ASD. Methods We assessed 122 individuals with ASD and 118 typically-developing controls (TDC) with MRI and clinical evaluation. The GM, WM volumes and cortical thickness of the cingulate gyrus were compared between ASD and TDC based on ne parcellation. Five SNPs of the CNTNAP2 linking to ASD and brain structural abnormality were genotyped. Results ASD individuals showed thinner cortical thickness in bilateral cingulate subregions than TDC but no signicant group differences in GM and WM volumes. The WM volume of the right anterior cingulate gyrus was correlated with social awareness decits in ASD. Two CNTNAP2 variants demonstrated the main effects on the WM volumes of the right middle cingulate gyrus and the cortical thickness of the right posterior ventral cingulate gyrus. Besides, the CNTNAP2 variants interacted with ASD status and age on the cortical thickness of the left anterior middle cingulate cortex. Limitations The CNTNAP2 variants selected in this study were based on the current literature that revealed genetic associations with ASD and brain structures. Future studies may consider ne mapping or sequencing of this gene. Besides, we measured social awareness decits by caregiver-report questionnaire. Although the observation from caregivers may reect social function in daily life, a social cognition task may better provide a standardized measurement. Conclusions Our ndings suggest that aberrant cingulate structure in ASD may be associated with autistic genetic


Introduction
Alterations of the cingulate structure are frequently reported in individuals with autism spectrum disorder (ASD) [1]. A meta-analysis of voxel-based morphometry (VBM) studies suggested that individuals with ASD showed a signi cant increase of grey matter (GM) volume in the subgenual anterior cingulate cortex (ACC) [1]. However, such associations were not supported by another two meta-analyses [2,3]. A recent meta-analysis focusing on adults with ASD, instead, reported reduced GM volume in the ACC [4]. These inconsistent ndings may be attributable either to different focuses within the cingulate substructures (i.e., subgenual ACC, dorsal/ventral ACC or the whole cingulate cortex) or different age populations across studies. Most existing studies only involved ACC but ignored other parts of the cingulate structure.
Given that the cingulate cortex is anatomically and functionally heterogeneous and consists of distinct cytoarchitectonic zones that differ in cellular structures indicative of functional subdivision [5,6], whether the subregions of the cingulate structure are altered in ASD in different patterns is worth investigation.
Regarding the function of the cingulate cortex, the ACC has been implicated in a broad range of behaviors and cognitive processes, including social cognition [7][8][9]. Using functional MRI to assess brain activation when doing the oddball social task, dorsal ACC activation was found to predict the severity of social impairments in a subset of ASD individuals [10]. By contrast, whether the structure of the cingulate cortex can predict the social cognitive impairment in individuals with ASD is largely unknown. A report has displayed that increased thickness in the rostral ACC was associated with more severe social impairment on the Autism Diagnostic Interview-Revised (ADI-R), suggests that cortical thickness of ACC may be correlated with the severity of social de cits in ASD [11]. Although there was evidence implying the role of ACC on regulating social behaviors [7], it is unclear whether the other subregions of cingulate also involves similar functions. New evidence supports that the middle cingulate cortex (MCC), previously noted as the dorsal ACC, may speci cally involve in social cognition [12], whereas the posterior cingulate cortex (PCC) may involve in attention and arousal state [13]. The reduced functional coupling between PCC and ventromedial prefrontal cortex was highlighted in a recent meta-analysis that provided an early insight into the multiple dimensions of functioning, including higher-order cognitive and complex social functions [14]. Given that MCC and PCC have been implicated as ASD-associated impairment, such as social cognition [12] and attention [13], whether these structural alterations are associated with social awareness de cits is of particular interest.
Recent studies have shown that brain structures such as total brain volume [15], cortical thickness [16][17][18], and measures of white matter (WM) integrity derived from diffusion tensor imaging [19] are all under moderate genetic control. Evidence suggests that several genes may play a role in regulating brain development. The contactin-associated protein-like 2 (CNTNAP2) gene encodes CASPR2, which is a member of the neurexin superfamily of transmembrane proteins that are responsible for voltage-gated K + channel clustering in juxtaparanodes [20,21] at the nodes of Ranvier [22]. As a cell adhesion molecule responsible for neuroblast migration and laminar organization [23][24][25], CASPR2 is suggested to play a role in the brain development. Performing cortical neuron cultures from mouse embryos, Caspr2 was lately demonstrated to play a dose-dependent role in axon growth in vitro [26]. A study of an Amish family reported that a deletion mutation in the CNTNAP2 gene manifested with many hallmarks of autism including seizures, language di culties, and social de cits [22]. Other studies suggested that homozygous mutations in exon 22 of the CNTNAP2 resulted in an ASD diagnosis in 67% of cases [22], while heterozygous mutations were associated with altered brain structure and functional connectivity in otherwise neurotypical subjects [27][28][29]. On the other hand, mouse models lacking Cntnap2 presented with autistic behaviors, including repetitive behaviors and impairments in social interactions and communication [30,31]. Loss of the CNTNAP2 produces abnormal neuronal migration that was recently highlighted in the neuropathology of ASD [32], decreased numbers of interneurons, and reduced cortical synchrony among the neuropathological abnormalities [31]. These ndings shed light on the functional consequences of CNTNAP2 disruption at the level of cortical structure.
The current study aimed to investigate the cingulate structures in individuals with ASD and typically developing controls (TDC). With a more delicate method of parcellation ( Fig. 1), the subregions of cingulate GM (i.e., ACC, anterior MCC, posterior MCC, ventral and dorsal PCC), WM volumes (rostral ACC, caudal ACC, PCC, and isthmus), and cortical thickness (i.e., ACC, anterior MCC, posterior MCC, ventral and dorsal PCC) were analyzed. We examined whether age, sex and autistic symptoms were signi cant correlates of cingulate structures. Besides, we analyzed the genetic associations between the cingulate structures and the CNTNAP2 variants (i.e., rs779475, rs759178, rs2710102, rs2538991, and rs2710126) that were repetitively shown to be associated with ASD and brain structure. We hypothesized that individuals with ASD might have abnormal cingulate structures compared to TDC, and these structural alterations were associated with the severity of social awareness de cits. We also hypothesized that ASD susceptibility gene, CNTNAP2, may play a role in modifying cingulate structures of ASD.

Participants
We collected brain MRI scans, clinical and genetic data from 118 individuals with ASD (mean 13.1 years, SD 4.6; male 113, 95.8%), and 122 TDC (mean 21.0, SD 9.7; male 75, 61.5%). Individuals with ASD were recruited from National Taiwan University Hospital, Taipei, Taiwan. They were clinically diagnosed as ASD by senior board-certi ed child psychiatrists based on the Diagnostic and Statistical Manual of Mental Disorders-5th edition (DSM-5) diagnostic criteria of autism spectrum disorder; the diagnosis was further con rmed by a structural interview using the Chinese version of the ADI-R [45,46]. Participants with major neuropsychiatric disorders or with a full-scale IQ lower than 70 were excluded. The TDC participants were recruited from schools in the same districts of the ASD participants. All the participants went on clinical evaluation and interviews; all their parents were also interviewed by using the Chinese version of the Kiddie epidemiologic version of the Schedule for Affective Disorders (K-SADS-E) interview [47,48] to exclude any current or lifetime ASD and other major psychiatric disorders including attention-de cit/hyperactivity disorder, schizophrenia, mood disorders, anxiety disorders, or neurodevelopmental disorders. The details of the psychometric properties of the K-SADS-E and interview training have been described elsewhere [47][48][49][50].

Procedure
The Research Ethics Committee approved the study before its implementation (Approval number: 201201006RIB; ClinicalTrials.gov number, NCT01582256). After the purposes and procedures of the study were fully explained and con dentiality was assured, written informed consent was obtained from the participants and their parents. All the participants were then assessed with the brain MRI, and the Wechsler Intelligence Scale for Children (version III) or the Wechsler Adult Intelligence Scale (version IV) for IQ pro le according to their ages. The parents (mainly mothers) completed the following clinical measures about the participants.

Measures
The ADI-R [46] is a standardized, comprehensive, semi-structured, investigator-based interview for the caregivers of children with a mental age of 18 months into adulthood. It covers most developmental and behavioral aspects of ASD, including qualitative abnormalities in reciprocal social interaction, communication, and restricted, repetitive and stereotyped patterns of behaviors. The Chinese version of the ADI-R was approved by the Western Psychological Services in 2007 and has been widely used in several studies to validate the clinical diagnosis of ASD (e.g., [51][52][53]). The details of psychometric studies and interviewers training of using the ADI-R in Chinese have been described elsewhere [54,55].
The Social Responsiveness Scale (SRS) [56] is a widely-used quantitative measure of autistic traits in the general population. It includes 65 items to measure the severity of ASD symptoms in natural social settings over the past six months for children and adolescents aged 4-18 years. Items were rated by parents or caregivers on a 4-point Likert scale from "0" (not true) to "3" (almost always true). The SRS has been demonstrated to have good internal consistency, construct validity, inter-rater reliability, test-retest reliability, and discriminative validity in prior research [56]. Its Chinese version demonstrates a four-factor structure (i.e., social communication, stereotyped behaviors/interest, social awareness, and social emotion), but is better conceptualized as a one-factor model [54]. The Chinese SRS has been widely used in ASD research in Taiwan (e.g., [48,[57][58][59]. High internal consistency was found for the four subscales (Cronbach's alpha, .94-.95) and the total scale (Cronbach's alpha, .95). To test the association between cingulate structures and social cognition, we speci cally targeted the social awareness de cits subscale in this study.

Genotyping
Single nucleotide polymorphism (SNP) selection and genotyping.

MRI Data Acquisition
Brain images were acquired on a 3T MRI system (Trio, Siemens, Erlangen, Germany). Head movement was restricted with expandable foam cushions and was assessed immediately after image acquisition.
High-resolution T1-weighted MR images were acquired covering the whole head with a three-dimensional (3D) magnetization-prepared rapid gradient echo (MPRAGE) sequence, resulting in an isotropic spatial resolution of 1 mm 3 . Whole brain segmentation and cortical thickness calculation.
FreeSurfer V5.2.0 (https://surfer.nmr.mgh.harvard.edu/) was used on a 64-bit Linux operating system to reconstruct the cortical surface from the MPRAGE images [60]. Whole brain segmentation [61,62] was performed. The cortical parcellation units of the cortex were automatically identi ed and labeled according to the Desikan atlas [63] within the FreeSurfer automatic cortical parcellation routine. The cortical thickness was automatically calculated by computing the shortest distance between the WM boundary and the pial surface at each vertex [64]. The reliability of the cortical thickness calculated by FreeSurfer has been validated [65]. The automatic reconstruction and calculation were reprocessed after manually correcting the detected erroneous part. The automatic parcellation of cortical regions derived 74 brain regions in each hemisphere, and the thickness of each region was then calculated. This study focused on the cingulate structure, in which GM was divided into ve subregions, i.e., ACC, anterior and posterior parts of MCC, and dorsal and ventral parts of PCC, while WM was divided into four subregions, i.e., the rostral ACC, caudal ACC, , PCC, and isthmus ( Figure 1).

Statistical Analysis
We used SAS v. 9.4 (SAS Institute Inc, Cary NC, USA) to perform the statistical analyses. Age and IQ pro les were compared by analysis of variance, while SRS subscores were compared between the ASD and TDC groups by the generalized linear model controlling for sex and age. In an age-and sexcompatible subsample (88 ASD & 51 TDC), we compared the volume of the GM and WM, and cortical thickness of cingulate structures for each subregion between ASD and TDC, controlling for sex, age, full-scale IQ, and handedness. The relationships between cingulate structures and social awareness de cits were examined by Pearson's correlation analyses in the ASD group, partial out the effects of age, sex, and full-scale IQ. The genetic effects of the CNTNAP2 variants on the cingulate structures were rstly examined by the main effect of each SNP on cingulate structures (i.e., GM and WM volumes, and cortical thickness of each subregion). Then, we tested the interactions between the CNTNAP2 variants and age or diagnosis on each cingulate structure, controlling for age, sex, and full-scale IQ, as well as the main effects of each SNP and diagnosis. False discovery rate (FDR) was applied to correct for multiple comparisons. FDR q-value < 0.05 was set as statistical signi cance. As for sensitivity power analysis, with a total sample size of 240, the statistical tests had a power 0.8 to detect a difference with the effect size of 0.27.

Results
The IQ pro les of the whole sample were within the normal range, while the ASD group had signi cantly lower full-scale IQ (100.8 ± 19.9) compared to the TDC (116 ± 10.9) (   (Table 2). By contrast, ASD individuals had thinner cortical thickness than TDC in the following subregions: the right ACC, right anterior MCC, bilateral post MCC, and bilateral ventral and dorsal PCC ( Table 2). The most signi cant difference was found in the right posterior MCC thickness (Cohen's d = -0.578). Correlations between cingulate structures and autistic symptoms in ASD Correlation analyses between cingulate structures and social awareness de cits revealed a signi cant negative correlation between the WM volume of the right caudal ACC and social awareness de cits on SRS (r = -0.313, p = 0.001) (Fig. 2). There were no signi cant correlations between social awareness and GM volumes or cortical thickness.

Discussion
Our work is one of the rst studies to report delicate parcellation on the cingulate structures in individuals with ASD, including GM and WM volumes and cortical thickness and is also the rst to explore the relationships between cingulate structures and the CNTNAP2 gene. The major ndings are three folds. First, individuals with ASD showed thinner cortical thickness of bilateral cingulate subregions, greater WM volume of the isthmus, and no difference of the cingulate GM volume. Second, decreasing WM volumes of the right caudal anterior cingulate gyrus was associated with increasing social awareness de cits. Third, the CNTNAP2 variants were associated with the WM volumes of the right caudal anterior cingulate gyrus and cortical thickness of the right ventral PCC, and interacted with ASD status on the cortical thickness of the left anterior MCC.
Our nding that ASD youths did show a greater WM volume of isthmus than TDC that has not been reported in previous studies. Given the cingulate body connected to the parahippocampal gyrus via the isthmus, whether the greater WM volume of isthmus here re ects impaired limbic connectivity like previously suggested [67] is worth further investigation. Meanwhile, although there was no signi cant difference in cingulate GM volumes that was consistent with some previous studies [2,3], cortical thickness of most subregions was signi cantly thinner in ASD compared to TDC. This nding was consistent with a recent study with ASD adults that showed thinner caudal ACC thickness in both the discovery (n = 301) and the replication sample (n = 61) [68], yet contradictory to previous studies showing increased cortical thickness of the left cingulate in 19 adults with high-functioning autism [69], bilateral PCC in 15 young adult males with autism [70], and the left MCC in 22 children with ASD (mean age 9.2 ± 2.1 years) [71]. Our results were based on a comparison between 88 ASD and 51 TDC with careful adjustment on sex, age, full-scale IQ, and handedness. Notably, age was a signi cant covariate for Two SNPs, rs759178 and rs2538991, were signi cantly associated with the WM volume of the right caudal anterior cingulate [rs759178: GG (n = 78) 3187.5 ± 531.7, GT (n = 127) 2951.3 ± 422.7, TT (n = 31) 3197.3 ± 559.4 (F = 6.76, p = 0.0014); rs2538991: GG (n = 79) 3188.7 ± 528.4, GT (n = 127) 2950.5 ± 422.1, TT (n = 31) 3197.3 ± 559.4 (F = 6.75, p = 0.0014); post-hoc analysis, GG, TT > GT, Bonferroni correction p < 0.05] (Fig. 3a, 3b). However, the heterozygotes (GT), but not the homozygotes (GG and TT), of both SNPs showed the smallest WM volume. Given that the two SNPs had quite similar genotype frequency, they were suspected in linkage disequilibrium. Meanwhile, the SNP rs779475 was associated with the cortical thickness of the right ventral PCC, showing that participants who carried TT had thinner cortex at the right ventral PCC compared to those carrying AA or AT [AA + AT (n = 35), 3.49 ± 0.33; TT (n = 201), 3.34 ± 0.34; F = 7.13, p = 0.0081] (Fig. 3c).
We further conducted multiple regression analyses separately including each with the main effect of each SNP, two-way interactions (SNP × age, SNP × diagnosis) and three-way interaction (SNP × age × diagnosis) were tested for each SNP. For the model containing both two-way interactions and three-way interaction, the same SNPs rs759178 (Supplementary Figure S1a), rs2538991 (Supplementary Figure  S1b), and rs2710102 (Supplementary Figure S1c) interacted with age and diagnosis group on the cortical thickness of the same region, i.e., the left anterior MCC (Supplementary Table S2).
cortical thickness in most cingulate subregions in our study (except PCC), in support of a previous argument that ASD is characterized by complex alterations in lifetime trajectories of several brain regions that underpin social-cognitive function [72]. Whether the cortical thinning in the cingulate gyrus was abnormal in the ASD population, speci cally during adolescence, warrants more research. In contrast to previous ndings of reduced cingulate GM volume in adults with ASD [4,66], in our sample of youths aged 8-20, we did not nd group differences in the cingulate GM volume. Whether different age groups of the study samples explained the inconsistent ndings warrants further investigation.
Our nding that the WM volumes of the caudal anterior cingulate gyrus (MCC) was negatively correlated with social awareness de cits worths more research. Although the function of cingulate gyrus has implicated in social cognition, only a few studies investigated the relationship between cingulate structure and autistic symptoms. One study reported that increased thickness in the rostral ACC was associated with poorer social scores [11]. Another study reported that social awareness de cits were correlated with increased functional connectivity between the dorsal ACC (namely MCC in this study) and right superior temporal gyrus, and decreased functional connectivity between the dorsal ACC and right putamen and thalamus [73]. Combining with our ndings that ASD youths had a thinner cingulate cortex and that smaller right MCC WM volume was correlated with more severe social awareness de cits, such quantitative trait of social cognition may be associated with cingulate structural variations linked to ASD. These ndings, if veri ed, implied a potential role of the cingulate gyrus in modulating the core symptom of ASD.
As the rst study to focus on the genetic association between the CNTNAP2 variants and cingulate structures, we found that the CNTNAP2 variants were associated with the right MCC WM volume (rs759178 and rs2538991) and right ventral PCC thickness (rs779475). The common variant of the CNTNAP2 at intron 2, rs7794745 (risk allele T), has been shown to be associated with reduced GM volume in the occipital gyrus, fusiform gyrus, and frontal cortices [37,39]. Our nding suggests that this SNP (TT genotype) may also associate with thinner ventral PCC independent of diagnostic status.
Besides, several studies support that the CNTNAP2 plays a role in regulating brain connectivity. Children carrying the autism risk allele (C) of rs2710102 showed reduced long-range functional connectivity (i.e., fronto-occipital) in the prefrontal cortex but increased short-range functional connectivity [28]. Another study used whole-brain tractography and graph metrics to compare structural connectivity between genotypes of this SNP; the result also supported the association between this SNP and the reduced longrange brain connectivity [27]. A neighbor SNP, rs2710126, was also reported to associate with structural connectivity of the uncinate fasciculus [40]. Several SNPs located in intron 13 (rs2710102, rs759178, and rs2538991) were found nominally associated with axial diffusivity of the dorsal cingulum bundle [40].
Our result provided new evidence to support an association between the two SNPs in intron 13 (rs759178 and rs2538991) and MCC WM volume, implying that the CNTNAP2 variants may possibly in linkage disequilibrium to the genetic markers that regulate cingulate structure. The role of intron 13 of the CNTNAP2 on the neural development of cingulate WM warrants further research. Intriguingly, the heterozygote genotypes of these two SNPs had the smallest MCC WM volume. Evidence has shown that heterozygous mutations of CNTNAP2 were associated with altered brain structure and functional connectivity in neurotypical subjects [27][28][29]. Most disruptions of the CNTNAP2 are heterozygous, suggesting that loss of a single allele could be su cient to cause a disorder [74]. The mechanism of heterozygote of common variants warrants further investigation.
Given the distinct developmental trajectories in ASD brain particularly the GM volume of the MCC [72], we speci cally tested whether the CNTNAP2 variants interacted with age on the cortical thickness of two components of MCC, the anterior and posterior MCC, separately. We found a group × age × SNP interaction on the left anterior MCC only for the three SNPs in intron 13 of the CNTNAP2 (i.e., rs2538991, rs2710102, and rs759178), which might suggest the effect of intron 13 on cortical thinning differs between ASD and TDC. The nding that ASD participants who carried TT genotype showed a faster cortical thinning with age than TDC counterparts with the same genotype (Fig. 3) needs to be replicated in a larger sample with an adequate number of minor alleles. Also, as a cross-sectional study, age effect can hardly be addressed by the developmental approach. A longitudinal study may further help to elucidate the effect of the allele on cortical thinning.

Limitations
This study had several limitations. First, the ASD participants of the whole sample were male predominant and older than controls. Therefore, cingulate substructures were further compared in a subsample whereby the sex and age were compatible between the ASD and TDC. Second, we measured social awareness de cits by caregiver-report questionnaire rather than a social cognition task. Although the observation from caregivers may better re ect the social function of participants with ASD, particularly in their youth stage, in daily life, a social cognition task that targets the speci c social de cit of ASD (e.g., [75]) may provide a standard objective measure and thus should be considered. Third, although the number of our sample is one of the large-scale imaging genetic studies in ASD, the sample size of the current study may not have adequate power to detect small effect differences in the scale of a genetic study. A larger independent sample is needed to verify our results. Third, the candidate SNPs selected in this study were based on previous imaging genetic studies of the CNTNAP2. Future studies may consider ne mapping or sequencing of this gene.

Conclusions
The ndings of this study may suggest that youths with ASD showed comparable GM and WM volume

Consent for publication
Not applicable Availability of data and material The datasets generated and/or analyzed during the current study are not publicly available due to the research ethics regulation in Taiwan but are available from the corresponding author on reasonable request.
YLC and SSG designed the study. SSG collected the clinical, image, and genetic data. YCC analysed the brain image data. YLC did the statistical analysis. YLC and SSG interpreted the results. YLC drafted the manuscript, and SSG did rigorous revision on the manuscript. All authors read and approved the nal manuscript. Correlation between cingulate structure and social awareness de cits in individuals with autism spectrum disorder. Right anterior cingulate white matter volume was negatively correlated with social awareness de cits (r = -0.313, p = 0.001).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. CNTNAP2ASDCTCingulateSupplementaryTables.docx