Imaging-Genetics of Gender Differences in ASD: Sex-Specific Additive Effects of Oxytocin Receptor Gene Polymorphisms on Reward Circuitry

Background: Sex differences in the prevalence of Autism Spectrum Disorder (ASD) are well documented; yet, the mechanisms underlying this gender bias remains unknown. As the ASD phenotype is defined primarily by social impairments, exploring sex differences in brain networks and genes associated with social behavior is a logical first step. Allelic variations on the oxytocin receptor gene (OXTR) have been associated with increased rates of ASD (LoParo 2014), and animal models suggest OXTR signaling in the nucleus accumbens (NAcc) is crucial for typical social behavior (Keebaugh 2015). In a predominantly male sample, we recently showed that OXTR risk-allele-dosage was associated with decreased NAcc-reward circuit connectivity in ASD youth, whereas typically developing (TD) carriers of disease-associated variants showed increased NAcc-frontal connectivity, suggesting a compensatory mechanism in the face of increased genetic risk (Hernandez 2017). Importantly, however, evidence from animal models indicates gender-specific expression of the OXTR in the brain (Olazábal 2016) and gender-specific associations between OXTR methylation and brain volume have been documented in humans with psychiatric disorders (Rubin 2016).

Objectives: To examine the moderating effects of gender on the relationship between OXTR risk-allele-dosage and functional connectivity of the brain’s reward network in males and females with ASD (ASD-M, ASD-F) and matched TD controls (TD-M, TD-F).

Methods: DNA was genotyped for four ASD-associated OXTR SNPs (rs53576/rs237887/rs2254298/rs1042778). Participants were 32 ASD-F, 37 ASD-M, 33 TD-F, and 34 TD-M, ages 9-17 (males were the same subjects as in Hernandez 2017). Children completed a resting-state fMRI scan. ASD-F/TD-F data were processed according to the pipeline used for ASD-M/TD-M in Hernandez 2017. Data were motion scrubbed, activity was extracted from bilateral-NAcc and correlated with all other brain voxels to create resting-state maps. Single-subject resting-state maps were combined and compared at the group level, modeling the number of OXTR risk alleles as a covariate of interest. Results were thresholded at z>3.1 (p<.001), corrected for multiple comparisons at p<.05.

Results: Greater OXTR-risk-allele-dosage was associated with greater connectivity between NAcc and ventro-medial prefrontal cortex in TD-F, and greater connectivity between NAcc and subcortical brain regions in ASD-F. Comparing ASD-F to ASD-M, a significant interaction was detected such that as risk-allele-dosage increased, ASD-F showed an increase in connectivity between the NAcc and frontal and subcortical brain regions whereas ASD-M showed a decrease in connectivity with these same brain regions. Further, in ASD-F greater NAcc-frontal connectivity was associated with better social cognition measured by the Social Responsiveness Scale (p=.001); mirroring the brain-behavior relationship we previously reported in TD-M (Hernandez 2017).

Conclusions: In the face of increased genetic risk on the OXTR, unlike ASD-M, ASD-F showed increased NAcc-frontal connectivity, the same pattern observed in TD-M. Importantly, this increase in NAcc-frontal connectivity was associated with better social cognition in both ASD-F and TD-M groups. This neurobiological compensatory mechanism supports a female protective model whereby a greater number of genetic/environmental risk-factors are required for ASD-F to display altered brain connectivity and to exhibit high levels of ASD-associated symptomatology, which may ultimately help to explain observed sex differences in the prevalence of ASD.