Mechanisms of ketamine action as an antidepressant

Major depressive disorder affects a large proportion of the population and many patients remain treatment-resistant to standard monoaminergic therapies, which also typically require weeks to months to achieve full effect. Clinical trials have shown that a single sub‑anesthetic infusion of ketamine produces rapid (hours) and, in many patients, sustained (days) antidepressant effects, but ketamine's dissociative effects, abuse potential and intravenous route restrict its routine clinical use. Zanos and colleagues set out to synthesise preclinical and clinical evidence concerning the molecular and circuit mechanisms that underlie ketamine's rapid antidepressant actions. The review focuses on hypotheses that centre on N‑methyl‑D‑aspartate receptor (NMDAR) inhibition (including extra‑synaptic GluN2B‑selective blockade, inhibition of interneuron NMDARs and blockade of spontaneous NMDAR‑mediated transmission), suppression of lateral habenula burst firing, downstream plasticity pathways (BDNF, eEF2, mTOR, GSK‑3) and NMDAR‑independent routes involving ketamine enantiomers and hydroxynorketamine (HNK) metabolites. The authors emphasise that these mechanisms are not mutually exclusive and may converge to produce rapid changes in synaptic plasticity required for antidepressant responses.

This paper is a narrative review synthesising clinical and preclinical findings on ketamine's antidepressant mechanisms. The extracted text presents experimental results and clinical trial outcomes across multiple lines of evidence—behavioural pharmacology, electrophysiology, molecular signalling and metabolite pharmacology—rather than reporting a formal systematic review protocol. The reviewers organise the literature by mechanistic hypothesis: direct NMDAR inhibition (including extra‑synaptic and spontaneous NMDAR transmission), preferential inhibition of NMDARs on GABAergic interneurons (the disinhibition hypothesis), inhibition of NMDAR‑dependent burst firing in the lateral habenula, downstream effectors (AMPAR activation, BDNF, eEF2, mTOR, GSK‑3) and NMDAR‑independent effects mediated by ketamine enantiomers and HNK metabolites. Clinical trial data and animal model results are integrated to evaluate support for each hypothesis. The extracted text does not report a reproducible search strategy, inclusion/exclusion criteria, databases searched, or a formal risk‑of‑bias assessment. Consequently, the review should be read as an expert synthesis of available lines of evidence rather than as a systematic review with pre‑specified methods.

Clinical evidence reviewed indicates that a single intravenous sub‑anesthetic ketamine infusion (commonly 0.5 mg kg^-1 over 40 minutes) produces antidepressant effects within hours and can persist for days. Early controlled trials reported rapid benefit within 2–4 hours and that 35% of patients maintained response for at least 7 days after a single infusion. Use of an active psychoactive comparator (midazolam) reduced functional unblinding and still showed greater response with ketamine (64% versus 28%). Ketamine has also produced rapid antidepressant effects in bipolar depression and reduced suicidal ideation and anhedonia in major depression. Evidence addressing NMDAR‑dependent mechanisms is mixed. Several NMDAR antagonists and manipulations yield antidepressant‑relevant effects in rodents, supporting an NMDAR‑related hypothesis. Specific proposals include preferential blockade of NMDARs on fast‑spiking GABAergic interneurons leading to pyramidal neuron disinhibition and increased cortical glutamate release; blockade of spontaneous (resting) NMDAR‑mediated miniature EPSCs (NMDAR‑mEPSCs) which tonically suppress protein synthesis; and inhibition of extra‑synaptic GluN2B‑containing NMDARs that are tonically activated by ambient glutamate. Preclinical support includes increased extracellular glutamate and glutamate cycling in prefrontal cortex after ketamine, electrophysiological data of initial interneuron suppression followed by pyramidal activation, and rapid increases in markers of excitatory transmission. However, alternative or challenging findings exist: drugs that block the NMDAR but do not affect NMDAR‑mEPSCs (for example memantine) have repeatedly failed in clinical depression trials. Certain GluN2B‑preferring antagonists produced delayed or inconsistent antidepressant responses in small human studies (for example CP‑101,606 showed effects at 5–8 days but not at 2 days; MK‑0657 showed modest mood effects on one scale but not others and a larger trial was negative). Manipulations intended to mimic interneuron disinhibition do not uniformly reproduce ketamine's effects in rodents, and some data are inconsistent with a sole interneuron‑disinhibition mechanism. Data supporting NMDAR‑independent mechanisms are substantial in preclinical models. Rodent studies indicate that the (R)‑enantiomer can produce more potent and longer‑lasting antidepressant behavioural effects than (S)‑ketamine despite (S) having higher NMDAR potency, arguing that NMDAR inhibition alone may not fully account for efficacy. Ketamine is metabolised to hydroxynorketamine (HNK) stereoisomers, primarily (2S,6S;2R,6R)‑HNK, and several lines of evidence implicate HNKs in efficacy: hindering metabolism to HNK (by chemical deuteration at C6) abolishes ketamine's antidepressant action in mice; (2R,6R)‑HNK (derived from R‑ketamine) exerts dose‑dependent antidepressant effects in multiple rodent paradigms; and female rodents, which show greater antidepressant responses to ketamine, have higher brain HNK levels. Pharmacologically relevant concentrations of (2R,6R)‑HNK do not potently inhibit the NMDAR: displacement of [3H]‑MK‑801 indicates an affinity near 4,100 μM for (2R,6R)‑HNK versus 7–20 μM for (2S,6S)‑HNK, and functional assays show no NMDAR inhibition at ~10 μM, with modest inhibition only at higher concentrations (~50 μM). Behaviourally, (2R,6R)‑HNK lacks ketamine‑like anaesthetic or dissociative side effects in rodents at doses that are antidepressant. Across lines of evidence, activation of AMPA receptors (AMPARs) emerges as a consistent downstream requirement. AMPAR antagonism with NBQX blocks ketamine's antidepressant‑like actions across multiple rodent tests and prevents ketamine‑induced mTOR signalling changes. Ketamine and (2R,6R)‑HNK increase AMPAR‑mediated synaptic transmission and upregulate AMPAR subunits GluA1/GluA2 in hippocampus and medial prefrontal cortex within hours to a day. GluA2 deletion prevents ketamine‑induced synaptic potentiation and behavioural responses in mice. Electrophysiologically, ketamine and (2R,6R)‑HNK increase high‑frequency gamma power, an effect abolished by AMPAR blockade. Several intracellular plasticity pathways are implicated. Ketamine and (2R,6R)‑HNK rapidly increase brain‑derived neurotrophic factor (BDNF) levels and TrkB phosphorylation; genetic or local depletion of BDNF abolishes ketamine's antidepressant effects in mice. The eukaryotic elongation factor 2 kinase (eEF2K) pathway is proposed as a mechanism linking NMDAR inhibition at rest to increased BDNF translation: ketamine reduces eEF2 phosphorylation (de‑repressing translation) and eEF2K knockout prevents ketamine‑induced BDNF increases and behavioural responses. The mechanistic target of rapamycin complex 1 (mTORC1) signalling cascade is activated rapidly (within ~30 minutes) after ketamine in rodents and rapamycin pre‑treatment blocks ketamine‑induced synaptic and behavioural effects. Glycogen synthase kinase‑3 (GSK‑3) phosphorylation (inactivation) appears necessary for ketamine‑induced mTOR activation and behavioural responses; mice with non‑phosphorylatable GSK‑3 fail to show antidepressant effects. Finally, inhibition of excessive NMDAR‑dependent burst firing in lateral habenula (LHb) neurons is presented as a circuit mechanism: ketamine reduces LHb burst firing in animal models with congenital helplessness and this acute reduction is associated with rapid antidepressant‑like effects. The extract notes that long‑term behavioural consequences of LHb modulation require further testing.

Zanos and colleagues interpret the assembled evidence as indicating that multiple, potentially complementary mechanisms underlie ketamine's rapid antidepressant actions rather than a single exclusive pathway. While many hypotheses emphasise direct NMDAR inhibition—at synaptic versus extra‑synaptic sites, on interneurons versus pyramidal neurons, or by blocking spontaneous NMDAR‑mediated transmission—preclinical and clinical inconsistencies suggest that NMDAR inhibition is not solely sufficient to explain ketamine's unique efficacy and time course. The authors highlight growing support for NMDAR‑independent contributions, particularly metabolism to HNK metabolites and enantiomer‑specific effects. Preclinical findings that (R)‑ketamine and (2R,6R)‑HNK produce antidepressant‑relevant electrophysiological, molecular and behavioural changes without potent NMDAR inhibition are central to this argument. A consistent theme across mechanisms is the requirement for AMPAR activation and subsequent engagement of neuroplasticity pathways—BDNF translation/release, eEF2K/eEF2 regulation, mTORC1 signalling and GSK‑3 phosphorylation—that together promote sustained strengthening of excitatory synapses in mood‑relevant cortico‑mesolimbic circuits. The authors acknowledge important limitations and uncertainties reported in the literature: alternative NMDAR antagonists have failed to reproduce ketamine's rapid and durable clinical effects; some preclinical studies do not replicate mTOR activation or the necessity of specific components; translational gaps remain, for example a lack of direct clinical comparisons of (R)‑ versus (S)‑ketamine and incomplete characterisation of HNK pharmacodynamics in humans. They also note that many preclinical results are acute and that the durability of certain mechanisms (for instance LHb burst suppression) for producing long‑lasting behavioural change requires further investigation. In terms of implications, the review argues that understanding these convergent mechanisms can guide development of next‑generation rapid‑acting antidepressants that retain ketamine's therapeutic properties while minimising dissociative, anaesthetic and abuse‑related adverse effects. The authors emphasise that full NMDAR blockade has limiting side effects, so targeting downstream pathways or HNK‑like actions may offer safer long‑term therapeutic strategies. They also point out additional mechanisms not covered in depth—ketamine's monoaminergic and anti‑inflammatory effects—which may contribute to its antidepressant profile and merit further study.

The review concludes that ketamine's antidepressant actions are likely mediated by a combination of mechanisms rather than by a single mode of action. While inhibition of NMDARs has long been central to mechanistic models, accumulating evidence supports important roles for ketamine metabolism to HNKs, enantiomer‑specific effects, AMPAR activation and downstream plasticity signalling (BDNF, eEF2, mTORC1, GSK‑3). The authors state that recognising and disentangling these complementary pathways is essential for identifying novel therapeutic targets capable of producing rapid, durable antidepressant effects without the adverse consequences of broad NMDAR blockade.