How Ketamine Works in the Brain: A Complete Neuroscience Guide

How ketamine works in the brain starts with a single receptor: the NMDA glutamate receptor. Ketamine blocks it, which sets off a surge of glutamate that activates AMPA receptors, releases brain-derived neurotrophic factor (BDNF), and switches on the mTOR pathway that builds new synapses within hours. This chain of events explains why ketamine can ease depression symptoms in hours rather than the weeks typically needed by SSRIs and SNRIs, which act on serotonin, norepinephrine, and dopamine instead of glutamate.

For decades, the monoamine hypothesis, the idea that depression stems mainly from low serotonin, norepinephrine, or dopamine, shaped the first-line treatments that remain standard today. Ketamine's rapid antidepressant effect through the glutamate system forced researchers to reconsider how depression develops at the molecular level and how it might be treated. This guide walks through that mechanism step by step, from the first receptor interaction to the molecular cascades that reshape neural circuits.

The Glutamate System: The Brain's Primary Excitatory Network

Glutamate is the most abundant excitatory neurotransmitter in the central nervous system, and more than 80% of the brain's neurons are estimated to use it as their primary neurotransmitter. It underlies synaptic transmission, synaptic plasticity (the strengthening or weakening of neural connections that supports learning and memory), neural development, executive function, and mood regulation.

Glutamate acts on two receptor families. Ionotropic receptors open ion channels directly: NMDA receptors, named for the synthetic agonist N-methyl-D-aspartate, are ketamine's primary target; AMPA receptors handle fast synaptic transmission and are heavily involved in ketamine's downstream effects; kainate receptors modulate synaptic transmission and plasticity. Metabotropic glutamate receptors work through slower, intracellular signaling systems and fall into three pharmacological classes. Some of these receptors are being studied separately as depression treatment targets.

NMDA Receptor Blockade: The Initial Event

The NMDA receptor is a four-part protein complex, typically two GluN1 subunits paired with two GluN2 subunits (GluN2A, GluN2B, GluN2C, or GluN2D). Together they form a channel that lets sodium, potassium, and calcium ions cross the cell membrane. NMDA receptors are unusual because opening them requires two conditions at once: glutamate and a co-agonist, glycine or D-serine, must bind, and the postsynaptic membrane must already be depolarized enough to push out a magnesium ion that blocks the channel at rest. That dual requirement makes the NMDA receptor a coincidence detector, one that only fires when presynaptic signaling and postsynaptic state line up.

Ketamine blocks this channel in a specific way. It is a non-competitive, use-dependent, open-channel blocker: it binds a site inside the channel rather than competing with glutamate for the primary binding site, it can only enter once the channel is already open, and once inside it physically blocks calcium from passing through even when glutamate and glycine are bound. At the sub-anesthetic doses used in psychiatric treatment, typically 0.5 mg/kg given intravenously, ketamine does not block every NMDA receptor in the brain. It produces a partial blockade concentrated in specific neuronal populations, which helps explain why low doses produce antidepressant effects rather than anesthesia.

The GluN2B Subunit Preference

Research suggests ketamine may bind preferentially to NMDA receptors containing the GluN2B subunit. These receptors are heavily expressed on GABAergic interneurons in the prefrontal cortex, which helps explain why sub-anesthetic ketamine's effects concentrate in mood-related circuits rather than suppressing the brain broadly.

The Disinhibition Hypothesis: How Blocking Causes Activation

One of the more counterintuitive parts of ketamine's pharmacology is that blocking an excitatory receptor increases excitatory signaling overall. The explanation lies in prefrontal cortex circuitry, which contains two relevant cell types: GABAergic interneurons, inhibitory cells with high baseline firing rates and NMDA receptors that stay constantly active, and glutamatergic pyramidal neurons, the cortex's main excitatory output cells.

Because interneurons' NMDA receptors are constantly active, they are disproportionately vulnerable to ketamine's use-dependent blockade. When ketamine blocks these receptors, the interneurons fire less and release less inhibitory signal. With less inhibition holding them back, pyramidal neurons fire more freely and release a burst of glutamate, the glutamate surge that starts the rest of the cascade.

Several lines of evidence support this model. Rodent microdialysis studies show a transient rise in extracellular glutamate in the prefrontal cortex after sub-anesthetic ketamine. Blocking AMPA receptors prevents ketamine's antidepressant effects in animal models, confirming the glutamate surge matters. Optogenetic studies show that directly activating prefrontal pyramidal neurons reproduces some of ketamine's antidepressant effects, and human imaging studies show increased prefrontal cortex activity shortly after dosing.

AMPA Receptor Activation and the AMPA-to-NMDA Ratio

The glutamate released by disinhibited pyramidal neurons acts mainly on AMPA receptors on neighboring neurons rather than more NMDA receptors, since AMPA receptors respond to glutamate faster. Chronic stress and depression are associated with a shift toward reduced AMPA signaling relative to NMDA signaling, sometimes described as the AMPA-to-NMDA ratio. Ketamine appears to reset that balance by reducing NMDA activity through direct blockade while increasing AMPA activity through the glutamate surge, a rebalancing that may underlie its speed.

BDNF and the Neurotrophic Cascade

AMPA receptor activation triggers a rapid release of brain-derived neurotrophic factor (BDNF), a protein in the neurotrophin family that supports neuronal survival, growth, and synaptic plasticity. People with major depression consistently show lower BDNF levels in blood and brain tissue, and chronic stress, a major depression risk factor, suppresses BDNF expression in the prefrontal cortex and hippocampus. Conventional antidepressants also raise BDNF levels, but typically over weeks. Ketamine raises BDNF within hours, a speed difference that lines up with the gap between ketamine's and conventional antidepressants' clinical timelines and points to BDNF release as a key driver of ketamine's fast onset. For more on this protein's broader role in mood and cognition, see our guide to BDNF.

The TrkB Receptor

Released BDNF binds to its receptor, tropomyosin receptor kinase B (TrkB), which activates several intracellular pathways: PI3K/Akt for cell survival and growth, Ras/MAPK/ERK for gene expression tied to synaptic plasticity, and PLCgamma for intracellular calcium signaling. Together these pathways promote neuronal growth and new synaptic connections.

Genetic experiments confirm BDNF's necessity in this pathway. Mice carrying the Val66Met variant, which blocks activity-dependent BDNF release, do not show an antidepressant response to ketamine, and pharmacologically blocking TrkB receptors produces the same failure to respond. Both findings establish BDNF-TrkB signaling as a required part of the mechanism, not just a side effect of it.

The mTOR Pathway and Synaptogenesis

Downstream of BDNF-TrkB signaling, ketamine activates the mechanistic target of rapamycin (mTOR) pathway, a kinase that acts as a master regulator of protein synthesis in the cell. Once active, mTOR drives production of proteins the cell needs to build and maintain synapses. Ketamine activates mTOR quickly, with detectable increases in active (phosphorylated) mTOR in the prefrontal cortex within 30-60 minutes, peaking around two hours.

This activation drives synthesis of several key synaptic proteins: PSD-95, a scaffolding protein that organizes the postsynaptic machinery at excitatory synapses; GluA1, an AMPA receptor subunit that increases the density of functional AMPA receptors at synapses; synapsin I, involved in neurotransmitter vesicle release; and Arc/Arg3.1, an activity-regulated protein involved in synaptic plasticity.

That protein synthesis leads to synaptogenesis, the formation of new synapses. Imaging studies show ketamine increases the number and function of dendritic spines, the small protrusions where synapses form, in the prefrontal cortex within 24 hours. This matters because chronic stress and depression are linked to reduced dendritic spine density and synaptic loss in that same region; our synaptogenesis guide covers this process in more detail. Ketamine appears to directly reverse that loss rather than simply masking symptoms.

The Rapamycin Experiment

A pivotal 2010 study published in Science showed that pretreating animals with rapamycin, a drug that specifically blocks mTOR, completely prevented ketamine's antidepressant effects, direct evidence that mTOR-driven synaptogenesis is necessary rather than merely correlated with recovery. A later human study found the opposite: co-administering rapamycin with ketamine prolonged the antidepressant effect instead of blocking it. That discrepancy suggests ketamine's mechanism in humans is more complex than the rodent model alone explains, and it remains an active area of research.

Effects on Neural Networks

The Default Mode Network

The default mode network (DMN) is a set of connected brain regions, including the medial prefrontal cortex, posterior cingulate cortex, angular gyrus, and medial temporal regions, that is most active during rest, self-referential thought, and mind-wandering. In depression, the DMN often shows hyperconnectivity linked to rumination, trouble disengaging when attention should shift to a task, and excessive focus on negative self-evaluation. Functional MRI studies show ketamine temporarily reduces DMN hyperconnectivity, particularly between the medial prefrontal cortex and posterior cingulate cortex, and patients with larger DMN changes tend to report greater symptom improvement. Our default mode network guide covers this in more depth.

In healthy brains, the DMN and the task-positive network, regions active during focused, goal-directed behavior, work in reciprocal balance: when one is active, the other quiets down. In depression this anti-correlation weakens, so the DMN stays active even during tasks that need external focus. Ketamine appears to help restore that balance, which may explain why patients often report better concentration and motivation after treatment.

Prefrontal-Limbic Connectivity

Ketamine also affects connectivity between the prefrontal cortex and limbic structures like the amygdala and hippocampus. In depression, the prefrontal cortex often exerts too little top-down control over the amygdala's emotional processing. Ketamine appears to strengthen that connection, potentially supporting better emotional regulation.

Beyond NMDA: Additional Mechanisms

Ketamine's effects are not limited to the glutamate cascade above. Depression is increasingly understood to involve neuroinflammation, elevated levels of pro-inflammatory cytokines such as IL-1beta, IL-6, and TNF-alpha in blood and brain tissue. Preclinical studies show ketamine reduces these cytokine levels and modulates microglial activation, suggesting an anti-inflammatory effect that may work alongside the glutamate mechanism. Our guide on ketamine and neuroinflammation covers this research in more depth.

A 2018 study found that pretreating patients with naltrexone, which blocks opioid receptors, reduced ketamine's antidepressant effect in some participants, raising the possibility of partial opioid system involvement. Later studies produced mixed results, and researchers generally agree that opioid receptor activation is not ketamine's primary antidepressant mechanism.

Ketamine also interacts with sigma-1 receptors, located on the endoplasmic reticulum and involved in cellular stress responses and neuroprotection, and with hyperpolarization-activated cyclic nucleotide-gated (HCN) channels that regulate neuronal excitability and firing patterns. Both interactions are documented, but their clinical significance is not yet established.

Enantiomer-Specific Mechanisms: Esketamine vs. Arketamine

Ketamine has two mirror-image forms, or enantiomers: esketamine (S-ketamine) and arketamine (R-ketamine). They appear to reach their antidepressant effects through partly different routes. Esketamine is the form the FDA approved as Spravato in 2019 for treatment-resistant depression, available only through a restricted program at certified health care settings. Arketamine remains investigational. See our guides on esketamine and racemic ketamine vs. esketamine for how these differences affect treatment choice.

The Ketamine Metabolite Hypothesis

Ketamine's metabolites, the molecules the body produces as it breaks ketamine down, add another layer to the mechanism. According to a 2016 study by researchers at the National Institute of Mental Health published in Nature, one metabolite, (2R,6R)-hydroxynorketamine (HNK), produces antidepressant effects in rodent models without blocking NMDA receptors or causing dissociation. HNK appears to act mainly through AMPA receptor-dependent pathways.

If this translates to humans, it would mean some of ketamine's antidepressant benefit comes from a metabolite rather than the parent drug, which has spurred development of HNK-based compounds aimed at antidepressant effects without dissociation or misuse potential. The clinical relevance of HNK in humans is still debated, and phase 1 trials of synthetic HNK are underway to test whether it produces meaningful antidepressant effects in patients.

Why Some Patients Do Not Respond

Understanding this mechanism helps explain why ketamine does not work equally well for everyone. Roughly 30-40% of patients do not show a strong response, and, as summarized in a 2016 review in the journal Depression and Anxiety, researchers point to several possible biological explanations: differences in NMDA receptor subunit composition that change how ketamine binds, variation in GABAergic interneuron function that alters the disinhibition response, genetic differences in BDNF expression or TrkB signaling such as the Val66Met variant, differences in how individuals metabolize ketamine into active compounds like HNK, and varying levels of neuroinflammation or existing synaptic damage. Our guide on why some patients don't respond to ketamine for treatment-resistant depression looks at this from a clinical angle.

Bringing It Together

This is also why many clinicians recommend combining ketamine with ketamine-assisted psychotherapy rather than using it in isolation, and why its mechanism looks so different from SSRIs and SNRIs. See our comparison of ketamine vs. SSRIs for how the two approaches differ in practice, and our complete guide to ketamine for how this mechanism translates into treatment.