Reason: Access restricted by the author. A copy can be requested for private research and study by contacting your institution's library service. This copy cannot be republished
The influence of biological sex on neurobiological mechanisms underlying autism spectrum disorder: an investigation of neural activity and connectivity
thesisposted on 22.02.2017, 02:59 authored by Kirkovski, Melissa
Autism spectrum disorder (ASD) is characterised by impairment in social communication and interaction, and the presence of restricted and repetitive patterns of behaviours and interests. Though a behaviourally defined condition, a growing amount of literature provides substantial support for neurobiological abnormality underlying a vast array of impairments and characteristics associated with ASD. Of particular relevance to this thesis is the notion of abnormal or impaired neural connectivity (NC) in ASD. There is typically a differentiation between structural connectivity (SC; structural integrity of connections within and between brain regions) and functional connectivity (FC; synchronous activity between brain regions), and abnormalities in both of these have been noted among individuals with ASD compared to unaffected, neurotypical (NT) controls. Often overlooked in the literature at a neurobiological level, but also more broadly, is the role of biological sex in ASD. The limited literature investigating females with ASD is scarce and inconsistent, however some common themes do emerge. At a behavioural level, females with ASD appear to experience greater social difficulty, yet have the ability to mask or camouflage their symptoms. Structural brain imaging further supports this notion of greater abnormality among females with ASD, while accounting for expected sexual dimorphisms. To date, the functional underpinnings of these observations has not been investigated. This research examined whether biological sex influences the neurobiological abnormalities observed in ASD. Broadly, this study aimed to 1) investigate structural mechanisms (i.e., white matter [WM] pathways) involved in NC in ASD, 2) to provide insight into the functional underpinnings of key processes known to be impaired in ASD, and 3) to further understand electrophysiological processes in cortical regions known to be functionally impaired in ASD. Participants attended three experimental testing sessions. Session 1 involved clinical, behavioural and neurocognitive assessment. In session 2, participants underwent a magnetic resonance imaging (MRI) scan. Finally, session 3 involved a combined transcranial magnetic stimulation and electroencephalography (TMS-EEG) paradigm. Broadly, it was hypothesized that individuals with ASD would have impaired SC and FC compared to unaffected NT controls. It was further hypothesized biological sex would mediate this finding in ASD beyond expected sexual dimorphisms, and that neurobiological abnormality would be more severe in females with ASD. Firstly, we present an investigation of SC of key WM tracts throughout the brain, as there is strong evidence to suggest differences in WM architecture and structure in individuals with ASD compared to NT controls. We used diffusion tensor imaging (DTI) to investigate WM aberration in adults with high-functioning ASD and age, sex and IQ matched controls. Tract-based spatial statistics (TBSS) was used to explore differences in WM in major tract bundles. The effects of biological sex were also investigated. TBSS revealed no differences in fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD), or axial diffusivity (AD) between groups. There were no effects of biological sex. Next, FC was explored in a resting state paradigm, as atypical FC of the default mode network (DMN) has previously been identified in ASD. We investigated resting state FC, within the key DMN structures and also the wider resting state network, in a sample of high functioning males and females with ASD, and age, sex and IQ matched NT controls. FC within the DMN and broader resting state network did not differ between groups. Individuals with ASD did however have increased FC between a) the cingulate and primary somatosensory cortices, and b) the posterior superior temporal gyrus and primary somatosensory cortex. This effect appeared to be largely driven by affected males, who show increased FC relative to neurotypical males. Females with ASD did not differ in any regard to neurotypical females. Subsequently, the functional role of the mentalising network (primarily comprising the medial prefrontal cortex [mPFC] and the right temporo-parietal junction [rTPj]) was investigated, as this network is suggested to be highly involved in social understanding, a core impairment in ASD. Functional abnormalities have also been previously observed in this network among individuals with ASD. While observing a series of silent animations depicting social interactions between two geometric shapes, hemodynamic (blood oxygen level dependent; BOLD) response and FC were investigated. There were no differences in neural activity or connectivity in the mPFC or rTPj between groups during social processing. Sex-stratified analysis showed decreased activity in the right posterior superior temporal sulcus, and decreased functional connectivity between this site and the bilateral ventral posterior cingulate cortex in males with ASD during social processing. This pattern was not observed in the female sub-sample. Finally, electrophysiological measures of brain function were assessed using a novel approach combining transcranial magnetic stimulation and electroencephalography (TMS-EEG). TMS was applied to the right primary motor cortex, right dorsolateral prefrontal cortex and right TPj in separate blocks while EEG was concurrently recorded. Power spectra oscillations (activity) or phase-synchrony (connectivity) in the alpha, beta or gamma frequency bands did not differ statistically between groups, nor when data were stratified by sex. There are two interlinked considerations that may contribute to the overall findings of this thesis. Firstly, the sample presented in this thesis was comprised of high-functioning (i.e. IQ > 70) adults. For two reasons, chronological age may mediate the presented findings as a) it is well established that patterns of (structural) brain abnormality seen in younger individuals with ASD appear to normalise with age, and b) individuals (particularly well adapted high-functioning) with ASD may adapt their behaviours or learn to understand social cues, a process which may have an effect on brain function. A second consideration is the role of varying levels of impairment in ASD, and the extent to which this may influence the aforementioned consideration as well as the findings presented in this thesis. Future research should involve a large scale, multi-site investigation comparing males and females with ASD of various ages and levels of functioning. The findings of this thesis are, in many aspects, contradictory to our proposed hypothesis. Overall, we did not identify any neurobiological differences between high-functioning adults with ASD compared to age, sex and IQ matched neurotypical (NT) controls. Further, while it was expected that females with ASD may be more impaired neurobiological compared to affected males, the findings of our fMRI investigations suggest that in fact affected males may experience greater neurobiological impairment as high-functioning adult females with ASD displayed a functional neurobiological profile akin to their NT counterparts, while affected males did not. Moreover, from an electrophysiological perspective, our TMS-EEG investigation provides trend level suggestion that both males and females differ neurobiological from matched controls, however in seemingly opposite directions from one another. Considered together, the overall finding presented in this thesis provides support that males and females with ASD do differ neurobiologically, however the direction and function of this abnormality remains unclear, raising an imperative consideration for future research.