Who are you, ketamine? Good, evil, or dose‐ and context‐dependent?

Background

In 1962, Calvin Lee Stevens synthesized ketamine as a safer anesthetic alternative to phencyclidine, its structural precursor. Ketamine is a non‐selective NMDA receptor (NMDAR) antagonist that remains widely used as an anaesthetic (Acevedo & Siegel, 2022). However, the clinical use of ketamine has expanded, and the drug has been repurposed for treatment‐resistant major depressive disorder. Additionally, due to its cognitive, emotional and sensory effects, ketamine is also widely used as a recreational drug. Moreover, ketamine has long been used to induce a pharmacological model of schizophrenia based on the NMDAR hypofunction hypothesis (Acevedo & Siegel, 2022). Thus, among ketamine's diverse uses, considerable interest surrounds its effects, mechanisms and impact on neurodevelopment. Addressing this issue in a recent article published in The Journal of Physiology, Guiffa‐Gómez et al. (2026) examined the effects of adolescent ketamine exposure (to induce NMDAR hypofunction) on electrophysiological parameters of neurotransmission and plasticity in the adult prelimbic medial prefrontal cortex (mPFC).

To assess the long‐term effects of adolescent ketamine exposure on neurotransmission and plasticity, Guiffa‐Gómez et al. (2026) administered ketamine (30 mg/kg, KET group) or vehicle (CTRL group) to male mice for 7 days (postnatal days (P)45–51). Layer 2/3 pyramidal neurons (PyNs) and parvalbumin‐positive interneurons (PV‐INs) of the mPFC were evaluated in adulthood (P60–90). Analysis of this microcircuit is particularly relevant in the context of NMDAR hypofunction, as adolescence is a critical period for mPFC inhibitory maturation and both cell types receive NMDAR‐mediated excitatory inputs. PyNs are the principal excitatory output neurons of the mPFC, whereas PV‐INs provide axosomatic inhibition that regulates cortical excitability and excitation/inhibition (E/I) balance. Together, this microcircuit shapes cortical output and supports mPFC‐dependent cognitive, executive and social functions. Spike timing‐dependent plasticity (STDP), a physiologically relevant form of activity‐dependent synaptic plasticity, depends on coordinated excitatory and inhibitory circuit function and therefore serves as a sensitive measure of network integrity and plasticity.

Electrophysiological recordings from adult PyNs

Analysis of STDP in PyNs showed that the KET group exhibited a reversal of STDP polarity. Paired‐pulse ratio (PPR) remained stable between baseline and post‐STDP induction, with no changes in coefficient of variation (CV), indicating that STDP alterations are likely postsynaptic in origin. In the KET group, (2R)‐amino‐5‐phosphonovaleric acid (APV, an NMDAR antagonist) abolished the reversal of STDP polarity, whereas STDP polarity in the CTRL group remained unchanged. In the presence of APV, no group differences in PPR were observed, suggesting that NMDAR signalling is required for the STDP reversal induced by adolescent ketamine exposure.

Application of APV to isolate AMPA/kainate‐ and GABA_A‐mediated currents showed that the E/I ratio calculated from spontaneous events was increased in the KET group. Spontaneous inhibitory postsynaptic current (sIPSC) frequency was reduced and the inter‐event interval (IEI) was altered in the KET group, with no changes in amplitude or amplitude distribution, consistent with the altered E/I balance. Spontaneous excitatory postsynaptic current frequency, amplitude, IEI and amplitude distribution were unchanged, indicating preserved excitatory transmission. To isolate inhibitory currents, APV and 6‐cyano‐7‐nitro‐quinoxaline‐2,3‐dione (CNQX; an AMPA receptor antagonist) were applied. Pharmacological inhibition reduced sIPSC frequency and IEI was altered in the KET group, without changes in amplitude or amplitude distribution. Following tetrodotoxin (TTX; a voltage‐gated sodium channel blocker) application, mIPSC frequency was reduced and IEI altered in the KET group, without changes in amplitude or amplitude distribution. These results indicate long‐lasting alterations in inhibitory transmission induced by ketamine.

Picrotoxin (PTX; a GABA_A receptor antagonist) was used to reduce GABA_A‐mediated transmission by approximately 40–50%. PTX did not reproduce the STDP alterations induced by ketamine exposure, suggesting that this effect is not driven by a global reduction in inhibition. Finally, DAMGO (a μ‐opioid receptor (MOR) agonist that suppresses GABA release from specific PV‐INs) reduced evoked IPSC amplitude in the CTRL group but not in the KET group, demonstrating reduced sensitivity of PV‐IN‐mediated inhibition to MOR modulation.

Electrophysiological recordings from adult PV‐INs

PV‐INs from the CTRL and KET groups showed no differences in passive membrane properties or in action potential (AP) firing induced by suprathreshold current pulses. However, the KET group exhibited reduced mean and maximal AP firing frequencies. Temporal analysis revealed sustained firing in CTRL PV‐INs, whereas KET PV‐INs fired predominantly during the initial phase of stimulation. First AP properties showed increased amplitude in KET neurons, along with a reduced last/first AP amplitude ratio and increased last/first AP width ratio, suggesting progressive attenuation of AP amplitude during sustained firing. Although mean AP upstroke and downstroke velocities were unchanged, last/first AP ratios for both parameters were reduced in the KET group. Together, these findings indicate impaired ability of PV‐INs to sustain high‐frequency firing.

Dosis sola facit venenum

‘The dose makes the poison’ – a phrase attributed to Paracelsus – remains relevant and applicable. Ketamine has emerged as a valuable pharmacological tool for studying NMDAR hypofunction. Conversely, it also produces rapid antidepressant effects. This apparent paradox raises an important question: why use a drug capable of inducing long‐lasting neural disturbances to treat depression? The answer lies, in part, in the treatment regimen. Chronic administration protocols (7 days or longer, typically ≥30 mg/kg) are commonly used to induce NMDAR hypofunction and schizophrenia‐like phenotypes (Guiffa‐Gómez et al. 2026; Pérez et al. 2019), whereas single‐dose administration (10–20 mg/kg) is generally associated with antidepressant‐like effects (Fitzgerald et al. 2019; Ng et al. 2018), despite substantial variability across studies (Acevedo & Siegel, 2022). Recently, interest in the antidepressant effects of single‐dose ketamine has increased, particularly given concerns regarding its abuse liability and addictive potential.

Does biological context shape ketamine effects?

Ketamine's effects appear to depend not only on dose but also on the pre‐existing biological context. In naïve animals (lacking a depressive‐like state), ketamine induces behavioural and neurobiological alterations relevant to schizophrenia research (Pérez et al. 2019), whereas in animals with a depressive‐like state, it may exert antidepressant effects.

As demonstrated by Guiffa‐Gómez et al. (2026), adolescent ketamine exposure altered STDP polarity in layer 2/3 PyNs of the mPFC. Underlying these changes, reduced GABAergic transmission and decreased PV‐IN output were observed, consistent with the dependence of PV‐IN activity on NMDAR activation. Consequently, PyNs receive reduced inhibition, shifting the E/I balance toward excitation. In a previous study, this same research group showed that the identical ketamine administration regimen induced behavioural alterations relevant to schizophrenia research, including hyperlocomotion, reduced sociability, cognitive impairment and increased anxiety‐like behaviour (Pérez et al. 2019).

Conversely, ketamine may exert beneficial effects in animals previously exposed to environmental adversity. Fitzgerald et al. (2019) showed that a single dose of ketamine (10 or 30 mg/kg) increased immobility in the tail suspension test in non‐stressed animals but reduced immobility in animals exposed to chronic unpredictable stress. These findings indicate that ketamine increased depression‐like behaviour in naïve animals while producing antidepressant‐like effects in chronically stressed animals. In stress‐exposed animals, ketamine exerts prominent effects on PFC circuits. Ng et al. (2018) demonstrated that 7 days of restraint stress increased dendritic spine elimination in layer 5 PyNs of the PFC, an effect prevented by daily ketamine administration throughout the stress protocol (10 mg/kg). Moreover, restraint stress increased axonal bouton elimination in layer 2/3 PV‐INs of the PFC, whereas ketamine administration increased axonal bouton formation in stress‐exposed animals. This effect appears to depend on ketamine‐induced enhancement of PV‐IN activity, reflected by increased evoked AP firing rates.

Adolescence as a sensitive period

Ketamine effects may also depend on the stage of neurodevelopment. This issue is particularly relevant given growing interest in its antidepressant effects during adolescence, a period of profound brain maturation. In rodents, a single low dose of ketamine administered during adolescence appears to produce beneficial effects with few adverse outcomes in animals displaying neurobiological and behavioural alterations induced by early‐life adversity (Acevedo & Siegel, 2022). In contrast, chronic or subchronic ketamine exposure during adolescence induces substantial alterations, as demonstrated by Guiffa‐Gómez et al. (2026) and Pérez et al. (2019). These observations also underscore the potential risks of recreational ketamine use, particularly during adolescence. Overall, the literature remains inconsistent regarding the beneficial and detrimental effects of adolescent ketamine exposure (Acevedo & Siegel, 2022), underscoring the need for further research and caution when translating preclinical findings to adolescent clinical populations.

Future directions

Future studies should consider including both male and female animals in their experimental design. Future studies should evaluate ketamine's antidepressant effects in animals with an established depression‐like state, rather than in naïve animals.

The contribution of MORs to ketamine's effects remains debated and, as demonstrated by Guiffa‐Gómez et al. (2026), ketamine produces long‐lasting alterations in MOR‐mediated modulation. Future studies should clarify the contribution of MORs to both ketamine's therapeutic effects and NMDAR hypofunction‐related phenotypes.

Even in animals previously exposed to stress, ketamine may potentiate stress‐induced disturbances, as observed following social isolation (Shangase et al. 2026). Future studies may distinguish the isolated effects of stress or ketamine exposure from their combined effects. Different treatment regimens may help clarify the duality between ketamine's beneficial and harmful effects in a dose‐ and context‐dependent manner.

Conclusion

The study by Guiffa‐Gómez et al. (2026) demonstrates the long‐lasting effects of subchronic ketamine exposure during adolescence and sheds light on the PFC mechanisms underlying the behavioural alterations observed in the NMDAR hypofunction model. Despite these detrimental effects, ketamine exerts important antidepressant actions, and further studies are needed to better define the biological contexts, developmental stages and treatment regimens that determine whether ketamine produces beneficial or detrimental outcomes.

Additional information

Competing interests

No competing interests declared.

Author contributions

Sole author.

Funding

E.G.O. hold scholarships from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).

Generative AI statement

No generative AI was used in the preparation of this manuscript.

Supporting information

Peer Review History