Investigating the neurochemistry of depression and the novel mood stabilising agent, ebselen, using magnetic resonance spectroscopy

• Main Author: Masaki, Charles
• Other Authors: Harmer, Catherine ; Godlewska, Beata ; Cowen, Philip
• Published: University of Oxford 2016
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.729901

RATIONALE: Magnetic resonance spectroscopy (MRS) offers a tool for investigating in vivo neurochemical changes in depression, as well as for guiding targeted drug development. Using MRS, changes to the glutamatergic system have been demonstrated in depressed patients, although results from different studies have often been inconsistent. Elucidating the role of glutamate in depression can be helped by carrying out assessment with ultra-high field MRS and in addition to this, identifying patient characteristics that may contribute to variability in neurochemical findings. MRS can also be used to demonstrate in vivo effects of pharmacological manipulation, and by so doing, confirm mechanisms of drug action. Recently, a potential novel mood-stabilising and antidepressant agent, ebselen, has been identified based on preclinical evidence of lithium-mimetic effects. This drug offers the possibility of an alternative new treatment for mood disorders, without the side effects of lithium. OBJECTIVES: This thesis aimed to investigate glutamatergic neurochemistry in depression, as well as identify its clinical correlates. Further to this, the effects of ebselen treatment on neurochemistry and performance on neuropsychological tasks relevant to psychiatric treatment were investigated. METHODS: The first study involved 40 patients with major depressive disorder and 32 healthy controls. Subjects were scanned with 7-tesla Siemens scanner equipped with a 32 channel receive array head coil. High-resolution MP-RAGE structural images were acquired and used to guide the placement of MRS voxels. Proton-MRS was carried out sequentially on three voxels placed on the anterior cingulate cortex (ACC), occipital cortex and putamen, using a semi-LASER sequence (TE=36ms, TR=7s, NT=64). Metabolites were quantified with LCModel, using an unsuppressed water signal as a reference. The primary metabolites of interest were glutamate, glutamine and Glx. The second study featured exploratory analysis of associations between the concentrations of ACC glutamate, ACC Glx, and putamen glutamine with patient clinical parameters. In the third and fourth studies, 20 healthy volunteers were tested on two occasions following treatment with either ebselen (3600 mg in three doses over 2 days) or identically matching placebo, in a within-subject, double-blind, crossover design. Participants underwent 7-tesla MRS scanning 2 hours after the final dose of ebselen, with voxels placed over the ACC and occipital cortex. Scanning was performed with a STEAM sequence (TE = 11ms, TR=5s, NT = 64). Three hours after the final dose (immediately following the scan), participants completed the Cambridge Gambling Task (CGT) and the Facial Emotion Recognition Task (FERT). STATISTICS: Separate MANOVA for each voxel region were used to assess metabolite concentration differences between depressed patients and controls, with significant main effects followed up by comparisons with independent t-tests. Associations between neurochemistry and clinical variables were tested using Pearson's correlation tests for continuous variables and independent samples t-tests for categorical outcomes. Separate repeated measures MANOVA for each region were used to test for the effects of ebselen treatment on neurochemistry, with significant main effects followed up by comparisons with paired t-tests. Correlations were also determined between neurochemistry and neuropsychological task performance using Spearman's-rho test. RESULTS: There were main effects of gender on ACC metabolite levels, with depressed female patients having lower levels of glutamate (p = 0.005) and Glx (p = 0.010) relative to female controls, while depressed male patients had elevated levels of Glx (p = 0.046) relative to male controls. In the putamen, depressed patients had elevated levels of glutamine (p = 0.014) without gender effects. There were no group effects on occipital cortex metabolite levels. No significant associations were found between several patient variables and neurochemistry. Ebselen treatment lowered concentrations of inositol (p = 0.028), glutamate (p = 0.010), glutamine (p = 0.024), Glx (p = 0.001) and glutathione (p = 0.033) in the ACC. There were no effects of ebselen treatment on occipital cortex metabolite levels. On the CGT, ebselen treatment lowered the delay aversion score (p = 0.010), and increased reward seeking (p = 0.046). On the FERT, ebselen treatment increased the accuracy of recognition of positive facial expressions (p = 0.035). CONCLUSIONS: Glutamatergic changes are present in depression, although these may show gender specificity in the ACC. Ebselen treatment decreases brain inositol levels, suggestive of a lithium-mimetic effect. Ebselen treatment also decreases impulsivity, increases reward seeking and produces a positive bias in emotional processing, suggesting potential uses as an antidepressant and anti-impulsivity agent.