Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article


Master of Science


Medical Biophysics


Theberge, Jean

2nd Supervisor

Palaniyappan, Lena

Joint Supervisor


Schizophrenia is a neuropsychiatric illness without known etiology or cure. Current efforts for symptom treatment still seem to leave a large portion of affected individuals without proper symptom management, with those experiencing symptom relief still having to wrestle with potential side-effects from medication trials. There has been growing evidence suggesting that glutamate and glutathione abnormalities hold major roles in development and manifestation of schizophrenia symptoms.

Magnetic resonance spectroscopy (MRS) provides a non-invasive means to observe in-vivo brain chemistry, including glutamate and glutathione. By adding a functional component to an MRS paradigm (fMRS), such as the color-word Stroop task, it is possible to detect potential changes in brain metabolite levels in response to a cognitive stimulus.

The objectives of this thesis were to determine whether a conventional short echo-time or longer echo-time semi-LASER sequence would be more suitable for measurements of an fMRS paradigm at 7.0-Tesla, and to apply this finding to the study of a group of never-medicated, first-episode schizophrenia (FES) individuals to track any potential abnormal glutamate and glutathione dynamics.

Contrary to previous beliefs that the shortest achievable echo-time would produce the most measurement signal, results from this study found that a long echo time (TE=100ms) produced very similar quality of measurements, with further benefits of long echo time being the removal of any macromolecular signal contribution.

Comparison of healthy controls (n = 25) to a FES population (n = 21) revealed no significant difference in resting or dynamic glutamate levels. However, resting and dynamic glutathione level were significantly different, suggesting potential glutathione regulation abnormalities, and, in extension, although not observed in this study, potential glutamate and glutamine abnormalities.

Future fMRS studies should investigate glutamate and glutathione dynamics from longitudinal data of FES follow-ups as well as between specific sub-groups within FES.

Summary for Lay Audience

Schizophrenia is an illness that affects the human brain and mind. Patients often hear voices and hold delusional beliefs. They often lack motivation and have trouble with attention and memory. Currently, there is no known cure or clear knowledge of its cause(s). Symptom treatment typically involves administering medication that regulates a brain chemical called dopamine. However, a large portion of patients do not respond to these medications, and after a few unsuccessful trials, this subgroup, declared non-responders, will be given Clozapine, which indirectly acts on another brain chemical called glutamate. Because of its side effects, Clozapine is left as a last resort. For non-responders, finding their way to an effective medication can be long and costly, usually leading to prolonged patient suffering and disability.

Measurements of brain chemicals glutamate and glutathione early in the illness have been shown, at resting levels, to potentially distinguish responders from non-responders. However, these measurements had not taken into account that brain glutamate levels change depending on the amount of brain activity, and so, we used functional magnetic resonance spectroscopy (fMRS) to measure changes of glutamate and glutathione levels in the brain during the performance of a task, the same specific brain activity for all participants. We believe these measurements will improve our ability to identify non-responders even before treatment begins.

This work shows that never medicated, first-episode patients with schizophrenia did not change glutamate levels significantly during the task, but did change glutathione levels differently compared to a cohort of healthy control individuals, which gives us hints about the chemicals involved in task difficulties and how to possibly treat them.

We hope that future work descending from these results will help psychiatrists develop ways to identifying non-responders upon first admission, and choose the right medication without delay.