Neuronal network screening and predictive profiling of pharmacological efficacy in a rat model of autism spectrum disorder

Autism spectrum disorders (ASDs) and intellectual disabilities (IDs) are complex, heterogeneous disorders with poorly understood underlying cellular and circuit pathophysiology. Monogenic mutations are, however, known to account for a large proportion of cases where individuals present with ASD and co-occurring moderate to severe ID. Clinically, anxiety disorders and phobias are highly prevalent in individuals with ASD and are often used to inform diagnosis1.
Mutations in the gene encoding for transmembrane protein neuroligin-3 are highly correlative with ASD and ID. Recently, we reported that rats lacking the Nlgn3 gene (Nlgn3-/y) exhibit robust, fear/anxiety related phenotypes2. This behavioral abnormality has been linked to a region of the brain critically involved in the orchestration of appropriate survival behaviors, the periaqueductal gray (PAG). We have shown that neurons in the dorsal part of the PAG are intrinsically hyper-excitable in Nlgn3-/y rats compared to wild type controls and that this cellular abnormality may underlie the imbalance in flight-freeze behaviors observed2. To our knowledge, this is the first time that PAG abnormality has been linked to ASD and as such highlights the critical need for improved understanding of brain circuitry involved in neurodevelopmental disorders to allow development of improved, precision therapeutic tools.
Therefore, Aim 1 of this project is to further understand the neural circuit mechanisms driving aberrant fear responses in Nlgn3-/y rats. By simultaneously recording deep-brain LFP from PAG and a range of anatomically linked structures, alongside electroencephalography (EEG), the student will examine how intrinsic PAG hyper-excitability manifests in large-scale fear network interaction and how this correlates to behavioral responses. Functional links/correlations between deep brain LFP and surface EEG signals will also be quantified to deliver translatable biomarkers.
Pharmacological treatment studies in people with ASD have met been variable in success3. This variation is treatment outcome is a major issue clinically. Therefore, Aim 2 of this project is to examine the effect of Arbaclofen (a potent GABA B agonist) treatment on observed, robust phenotypes in Nlgn3-/y rats. To do this, the drug will be applied at various time points during development, either systemically or to local brain regions (e.g. PAG) while measuring its impact upon neural network activity and behavioral responses.
The variable efficacy of Arbacolfen treatment in ASD3 highlights the unmet need for targeted approaches to identifying individuals with higher likelihood of treatment responsivity. Recent work from Dr Kristoffer Månsson (Karolinska Institutet) has revealed that moment-to-moment brain signal variability can reliably predict psychiatric treatment outcomes in patients with social anxiety4, a common ASD co-morbidity. Therefore, Aim 3 is to test how effectively neuronal network biomarkers can predict pharmacological efficacy. To do this, the student will first investigate whether LFP/EEG signal variability can be used to predict Arbaclofen treatment outcomes in the Nlgn3-/y rat model. Importantly, the student will have opportunity to expand the application of these analyses to a range of behavioral and pharmacological treatment data available from other monogenic rat models of ASD.
The results of this project will deepen our understanding of brain networks underlying the behavioural deficits observed in Nlgn3-/y rat model and give insight into targeted drug treatment for subpopulations of people with ASD and ID. The findings will drive development of analytical methods to optimize the transformation of biological data into predictive models.