Glutamate and gamma-aminobutyric acid systems in the pathophysiology of major depression and antidepressant response to ketamine

Earlier preclinical and clinical work has implicated abnormalities in excitatory and inhibitory neurotransmission—principally involving glutamate (Glu) and gamma-aminobutyric acid (GABA)—in the pathophysiology of major depressive disorder (MDD) and bipolar disorder (BD). Animal models using pharmacologic and stress-induction paradigms have shown cortical Glu alterations that are reversible with conventional antidepressant interventions, motivating a glutamatergic hypothesis that extends beyond monoaminergic explanations. Human neurochemical studies using proton magnetic resonance spectroscopy (1H-MRS), positron emission tomography (PET), and functional imaging (EEG, MEG, fMRI) have reported region-specific changes in Glu, Gln, Glx (combined Glu and Gln), and GABA, and concurrent evidence has linked these neurochemical alterations with aberrant large-scale network functional connectivity in depressed patients. Lener and colleagues present a narrative review that synthesises neurochemical and functional imaging literature to explore links between Glu/GABA disturbances, network-level dysfunction, and the rapid antidepressant effects of subanesthetic-dose ketamine. The review highlights ketamine as a molecular probe for glutamatergic modulation in humans because of its reproducible rapid antidepressant effects and relatively lower side-effect burden compared with other glutamatergic agents. The authors argue for combined multimodal imaging approaches to better characterise biologically homogeneous subgroups and to identify biomarkers predictive of treatment response to glutamatergic interventions such as ketamine.

The extracted text does not provide a formal Methods section describing literature search strategy, inclusion/exclusion criteria, dates, databases searched, or systematic review procedures. From the content, the paper appears to be a narrative review that integrates findings from animal studies and clinical investigations using 1H-MRS, 13C-MRS, PET, fMRI, MEG, and EEG, and also discusses interventional ketamine studies (both clinical trials and imaging/pharmacodynamic experiments). The authors organise the review by modality and topic: they summarise rodent and human neurochemical studies of Glu and GABA (including the Glu/Gln metabolic cycle and mitochondrial energetics), PET studies of Glu receptors (for example mGluR5), 1H-MRS findings across brain regions, functional connectivity abnormalities from resting-state fMRI and MEG/EEG, and imaging/pharmacologic studies examining ketamine's effects on neurochemistry and circuitry. Where available, the review reports sample sizes and study designs of highlighted studies (for example, early double-blind placebo-controlled ketamine trials and small MRS or PET investigations), but no systematic quality assessment or meta-analytic pooling methods are described in the extracted text.

Preclinical and human neurochemical findings: Rodent paradigms that induce depressive-like behaviour are associated with cortical Glu alterations that can be normalised by monoaminergic antidepressants and electroconvulsive therapy. In human 1H-MRS studies, patients with MDD commonly show reduced Glu, Glx, and Gln in dorsolateral, dorsomedial, dorsoanterolateral prefrontal cortex (PFC) regions and the anterior cingulate cortex (ACC), with increased Glu/Glx reported in occipital cortex in some studies. A meta-analysis of 17 1H-MRS studies found reductions of Glx in the PFC that correlated with the number of failed antidepressant treatments, suggesting a relation with illness chronicity, but did not find isolated Glu reductions, implicating astrocytic metabolic dysfunction as a potential mechanism. One in vivo 13C-MRS plus 1H-MRS study reported a 26% reduction in mitochondrial energy production in glutamatergic neurons in patients with MDD versus healthy controls; however, no difference in measured Glu/Gln cycle rate was observed, and the reported negative association between Glu concentrations and number of depressive episodes was a post hoc, uncorrected finding. In BD, results are more heterogeneous: two meta-analyses noted increased Glx in the PFC irrespective of mood state and increased ACC Glu in depressed states, and elevated ACC Glu correlated with episode counts in one study. Medication confounding and clinical heterogeneity in BD samples complicate interpretation. GABA-related findings: 1H-MRS studies in MDD tend to report lower GABA in PFC, ACC, and occipital cortex, with some evidence of greater reductions in melancholic presentations. In BD the literature is inconsistent, with reports of decreased, increased, or unchanged GABA. Overall, the pattern supports that an excitatory/inhibitory imbalance—often manifesting as reduced excitatory neurotransmission or altered Glu/GABA ratios—may characterise a subset of depressed patients rather than a universal GABA deficit. PET receptor studies and other molecular findings: Two PET studies reported reduced mGluR5 availability in patients with MDD. The review notes the need for more specific ligands for metabotropic Glu receptors and ionotropic receptor subunits to clarify receptor-level pathology. Functional connectivity and electrophysiology: Resting-state fMRI studies identify network abnormalities in MDD, including hyperconnectivity within the default mode network (DMN) and increased coupling between subgenual ACC (sgACC) and medial PFC (mPFC), alongside reduced frontoparietal connectivity. MEG studies (including beta- and gamma-band analyses) show concordant connectivity disruptions; for example, Lener and colleagues' group reported decreased connectivity between the sgACC and a precentral motor/precuneus network and increased limbic connectivity in MDD. Combined MRS–fMRI data link regional Glu reductions with decreased functional connectivity and blunted BOLD responses to emotional stimuli, suggesting that local amino-acid neurochemistry may relate to specific network dysfunctions. Ketamine clinical and imaging findings: The initial double-blind placebo-controlled study cited administered 0.5 mg/kg intravenous ketamine over 40 minutes and observed an antidepressant response (>=50% HAMD reduction) in 4 of 8 patients (7 completers). Subsequent studies replicated rapid antidepressant effects in MDD and BD: symptom reduction begins within 2 hours, peaks around 24 hours, and can persist up to 1 week. Mechanistically, preclinical data indicate ketamine antagonises NMDA receptors on GABAergic interneurons (leading to disinhibition of glutamatergic neurons) and on postsynaptic neurons (promoting intracellular growth factor synthesis such as BDNF), with downstream activation of mTOR signalling and promotion of synaptogenesis. Neurochemical imaging during ketamine: Human studies are inconsistent. Some healthy volunteer studies observed an acute rise in Glu during ketamine infusion (a proposed "glutamate surge") with Glu returning to baseline after infusion; other double-blind studies found no change in mPFC/ACC Glx, Glu, or GABA 40 minutes after infusion. PET with an mGluR5 ligand in 10 healthy subjects detected reduced ligand binding across cortical and limbic regions after ketamine, though ligand characteristics limit interpretation. Clinical 1H-MRS studies in patients with MDD (small samples) reported mixed results: of three studies summarised, two found no specific neurochemical signature predictive of antidepressant response, while one reported that a higher pretreatment Glx/Glu ratio in dorsomedial/dorsal anterolateral PFC associated with clinical improvement. Some studies detected transient increases in Glx/water and GABA/water ratios during infusion, indicating target engagement, but these changes did not consistently predict clinical response. Ketamine and functional circuitry: Pretreatment activation in the rostral ACC has been associated with antidepressant response across modalities. Ketamine acutely alters ACC–mPFC connectivity in healthy subjects (increasing acutely, decreasing by 24 hours). PET studies link ketamine to altered glucose metabolism in dorsal ACC and PFC regions; sgACC hypermetabolism has been reported to predict ketamine response in BD. MEG studies in healthy volunteers (N=25 men) show ketamine-induced increases in anterior theta and gamma power and decreases in posterior low-frequency power, with frontoparietal connectivity changes and reduced NMDA/AMPA-mediated frontoparietal connectivity. The extent to which electrophysiological gamma-band changes directly index glutamatergic signalling in depressed patients remains unresolved, as most electrophysiological ketamine work to date has been in healthy samples.

The investigators interpret the accumulated evidence as supportive of a role for glutamatergic dysfunction in a subset of patients with MDD and BD, with GABAergic abnormalities being less consistently observed. They emphasise that altered glutamatergic metabolism and impaired mitochondrial energy production in glutamatergic neurons could contribute to an excitatory/inhibitory imbalance and to network-level connectivity perturbations evident in resting-state fMRI and electrophysiology. Ketamine is framed as a valuable translational probe: its rapid antidepressant effects and mechanistic profile (NMDA antagonism, disinhibition of glutamatergic neurons, BDNF and mTOR pathway engagement) provide a means to test how modulation of Glu/GABA systems affects functional circuitry. However, the authors note that clinical neurochemical and electrophysiological studies to date are limited by small sample sizes, heterogeneous patient populations, variable imaging platforms and field strengths, differences in voxel placement, and medication confounds—factors that contribute to inconsistent findings across studies. To address these limitations, the authors advocate for multimodal, combinatorial imaging approaches that pair MRS or PET measures of amino-acid neurochemistry with functional imaging (resting-state and pharmacodynamic fMRI, MEG/EEG) before, during, and after ketamine infusion. Specific recommendations include using higher magnetic field strengths (for example 7T) and focusing on targeted regions of interest such as the sgACC, dACC, mPFC, frontoparietal cortices, and limbic structures. They also highlight resting-state MEG—particularly beta- and gamma-band analyses—and pharmacodynamic fMRI as promising methods for identifying biomarkers predictive of ketamine response. Finally, the authors underscore the need to identify biologically homogeneous subgroups within MDD and BD to improve diagnostic specificity and to personalise glutamatergic treatments.