Understanding the GABAergic Mechanisms of Early Psychosis: A Cross-Species Approach

Student thesis: Doctoral ThesisDoctor of Philosophy


Psychosis is mental health syndrome consisting of symptoms such as delusions and hallucinations. This syndrome is a core feature of psychotic disorders, such as schizophrenia, that are accompanied by debilitating social and occupational dysfunction. Although elevated dopamine plays a role in psychotic symptoms, dopamine-targeting treatments do not alleviate other detrimental symptom dimensions, specifically negative and cognitive symptoms. Prior research suggests that abnormalities in the GABAergic system, which leads to disinhibition, play a central role in psychosis risk and schizophrenia pathogenesis. This PhD thesis aimed to gain a better understanding of GABAergic system abnormalities in the context of psychosis risk using translational neuroimaging approaches in a preclinical and clinical setting.

In Chapter 1.1, I introduced the concepts of psychosis, the psychosis spectrum, psychotic disorders such as schizophrenia, and the need for improved treatment. Following this, in Chapter 1.2, I introduced magnetic resonance imaging techniques as a non-invasive method to study the brain in vivo. Neural activity may be measured via blood-oxygen-level-dependent functional magnetic resonance imaging. Additionally, neurochemistry can be quantified using proton magnetic resonance spectroscopy. As neurochemistry and neural activity are related at a microscopic level, magnetic resonance imaging measures may be related too. In Chapter 1.2.6, Paper 1, I showed, through systematic review and meta-analysis, that neural activity and the inhibitory and excitatory neurotransmitters – GABA and glutamate – are associated at a macroscopic scale in healthy individuals. Specifically, GABA and local neural activity are anticorrelated when measured using magnetic resonance techniques.

Next, I explored the relationship between GABA and brain function in the context of psychosis. In Chapter 2.1, I briefly outlined the current consensus of dopamine system dysregulation in psychosis, and the need to look beyond this neurotransmitter system. Multiple avenues of research suggest that the dopamine system is dysregulated by upstream neurochemical system in the hippocampus, specifically by an abnormal balance between excitation and inhibition. In Chapter 2.2, I detailed hippocampal neuroimaging abnormalities that suggest that GABAergic inhibition is abnormal in individuals at clinical high risk for psychosis and patients with frank psychosis in Chapter 2.2. However, due to the correlational nature of human research, preclinical research using animal models of psychosis (Chapter 2.3) can help elucidate the involvement of cellular GABAergic system abnormalities in hippocampal neuroimaging signals. For example, we used the Erbb4 knockout mouse model (Chapter 2.4) to demonstrate how disrupted inhibitory parvalbumin-expressing interneurons can contribute towards neuroimaging signatures related to psychotic disorders in Chapter 2.6, Paper 2. In this study, we found that disruption of parvalbumin-expressing interneurons via genetic manipulation yields neuroimaging abnormalities similar to those observed in individuals at clinical high-risk of psychosis and schizophrenia patients.

Having established a link between dysfunctional GABAergic neurotransmission (parvalbumin- expressing interneuron dysfunction) and neuroimaging phenotypes, my next aim was to understand whether inhibitory deficits could be modulated in an animal model of schizophrenia and in individuals at clinical high-risk of psychosis. Firstly, I presented the methylazoxymethanol acetate (MAM) developmental model of schizophrenia in Chapter 3.1, and the predictions it makes about a disinhibited corticolimbic circuit putatively involved in all three symptom domains of schizophrenia in Chapter 3.2. Research in MAM rats suggests that GABAA receptors (Chapter 3.3) may be a potential target for psychosis phenotype prevention via the action of GABAA receptor positive allosteric modulators such as diazepam (Chapter 3.4). However, the cellular mechanisms of diazepam efficacy remain unclear. We addressed this question in Chapter 3.6, Paper 3 in the MAM model by investigating (1) GABAA and NMDA receptor alterations compared to controls, and (2) the effects of repeated peripubertal diazepam administration on these receptors. In this study, I found that GABAA and NMDA receptor abnormalities are present in adult MAM rats and associated with schizophrenia- like behavioural phenotypes. However, in our study we did not replicate diazepam efficacy in preventing schizophrenia-relevant behaviours. To forward-translate the potential modulation of GABAergic transmission via diazepam to humans, in Chapter 3.7 I tested whether an acute diazepam challenge can modulate glutamate and glutamine metabolite levels in the hippocampus and anterior cingulate cortex in individuals at clinical high risk for psychosis. In our preliminary sample, we did not identify glutamatergic metabolite reductions in either region.

Finally, in my general discussion (Chapter 4), I placed findings from my four empirical studies into the context of GABAergic system abnormalities, psychosis development, multimodal neuroimaging studies and preventative strategies for psychosis. I considered limitations to my studies and suggest future directions to address these limitations as well as remaining knowledge gaps in the field.
Date of Award1 Oct 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorGemma Modinos (Supervisor), James Stone (Supervisor) & Cathy Davies (Supervisor)

Cite this