Oxytocin Receptor Gene (OXTR) Impacts Salience Network Connectivity in Children with and without ASD

­­Background: Converging evidence suggests that common genetic variations in a number of Autism Spectrum Disorder (ASD) risk genes impact brain circuitry in individuals with ASD as well as neurotypical controls (e.g., Scott-Van Zeeland 2010). A single nucleotide polymorphism (SNP; rs53576) located in the oxytocin receptor gene (OXTR) has been linked to increased genetic risk for ASD (Wertmer 2009, Wu 2005), as well as with reductions in both grey matter volume and neural activity in components of the “salience network” in neurotypical adults (Tost 2010). The salience network plays an important role in the identification of the most homeostatically relevant stimuli, which includes socio-emotional processing (Seeley 2007). In particular, the right anterior insula is a critical hub of this network and altered activity and connectivity in this region has been implicated in autism (Uddin & Menon 2009).

Objectives: Despite research showing associations between OXTR, ASD genetic risk, and abnormalities in the salience network, no studies have examined how the presence of one or more OXTR risk alleles may impact functional connectivity in ASD. Here we sought to investigate how a common variant in OXTR (rs53576) may impact resting-state functional connectivity (rs-fcMRI) in typically developing children (TD) and children with ASD using the right anterior insula as a seed region. Given that additional SNPs in OXTR have been linked to ASD, we further examined how the presence of multiple OXTR risk alleles in other loci relate to social functioning in children and adolescents with ASD.

Methods: DNA was extracted from saliva samples and genotyped for rs53576 as well as 5 other oxytocin SNPs previously associated with ASD. rs-fcMRI scans were acquired in 70 children and adolescents (36 ASD, 34TD). Data were analyzed using a seed-based approach. After standard preprocessing, average timeseries were extracted from the right anterior insula for each subject and then correlated with activity across all other voxels in the brain. Individual correlation maps were combined at the group level and compared between oxytocin genotype groups stratified by diagnostic status.

Results: The rs53576 “A” risk allele was associated with decreased connectivity between the right anterior inslua and the right inferior frontal gyrus, bilateral amygdalae, and medial prefrontal cortex; reduced connectivity was most pronounced in individuals with ASD. Furthermore, social and communication phenotypes in the ASD group correlated with number of risk alleles for 6 OXTR SNPs (rs53576, rs237897, rs1042778, rs2254298, rs2268943, rs2268494), such that increased number of risk alleles were associated with higher scores on the ADOS social and communication subscales, and overall ASD symptom severity.

Conclusions: Our results show that a common variant in OXTR is associated with decreased functional connectivity between regions underlying social and emotional processing in ASD, and that multiple OXTR SNP risk alleles are associated with greater severity of ASD symptoms. These findings highlight a mechanism for how genetic risk might predispose to ASD and suggest that the use of an imaging-genetics approach combining multiple risk loci may help to further clarify the relationship between brain function and behavioral phenotypes in ASD.