|Home | About | Journals | Submit | Contact Us | Français|
Recent reports of a rapid antidepressant effect of the glutamate N-methyl-D-aspartate (NMDA) receptor antagonist ketamine, even in treatment-resistant populations, has spurred translational therapeutic and neuroscience research aimed at elucidating ketamine’s mechanism of action. This article provides a concise overview of research findings that pertain to the effects of low-dose ketamine at the cellular, neurocircuitry and behavioral levels and describes an integrated model of the action of ketamine in depression.
Major depressive disorder (MDD) is a common medical illness associated with enormous morbidity and public health costs (1). As defined by the Diagnostic and Statistical Manual of Mental Disorders – Fourth Edition (DSM-IV) (2), MDD represents a heterogeneous clinical syndrome characterized by the core symptoms of pervasive, sustained low mood and/or loss of interest in the environment accompanied by a constellation of other symptoms involving alternations in sleep, appetite, energy level, psychomotor function and cognition. The devastating public health impact of MDD results in part because the illness tends to strike in young adulthood and run a chronic or recurrent course (1). Suicidal ideation and behavior is a particularly concerning component of MDD and suicide has become the third leading cause of death in individuals 15 to 24 years of age.
The advent of modern psychopharmacology for depression fundamentally transformed the practice and science of psychiatry and has relieved the suffering of untold numbers of patients. Modern clinical trials research, however, indicates that current treatments for MDD are in many cases only partially effective and in some cases not at all effective. The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, conducted in a large adult outpatient treatment-seeking sample with MDD (N = 3,671), found that only 36.8% of patients achieved remission following an optimized trial of the serotonin selective reuptake inhibitor (SSRI) citalopram for up to 12 weeks (3). Additional treatment steps in STAR*D yielded only modest incremental gains in remission rates and 33% of patients remained ill following up to four consecutive treatment steps involving antidepressant combination and augmentation strategies. In a second large-scale study, the Combining Medications to Enhance Depression Outcomes (CO-MED) trial compared two antidepressant combinations with SSRI monotherapy at 12 weeks and 7 months and found that remission rates for any strategy were modest (37.7% – 38.9%), similar to STAR*D, and did not differ from each other (4).
The sobering results of these large, well-designed clinical trials suggest that almost two-thirds of patients with MDD will remain ill despite an optimized trial of an antidepressant. This group of patients can be described as suffering from treatment-resistant depression (TRD), and as a group are more likely to suffer a higher symptom burden and a more chronic illness course compared to their non-TRD counterparts. For patients who eventually respond during a course of treatment, the long delay in the onset of therapeutic action (up to 12 weeks) inherent in current treatments furthers add to the illness burden and morbidity of MDD.
Despite the large public health burden of MDD, the pace of therapeutic discovery in this area has lagged significantly behind other areas of medicine. A major obstacle to the development of improved, more efficacious treatments is our limited understanding of the molecular mechanisms underlying the depressed state or treatment response (5). While the monoamine system has been the focus of research and treatment development in depression for more than 50 years, it is clear that novel treatment discovery in depression will require the advancement of new pathophysiological models of the disorder. Toward this end, the glutamate system and molecular mechanisms related to synaptic and neuronal plasticity are emerging as key components of a new generation of disease models of depression and antidepressant therapeutics (5,6).
In a novel proof-of-concept clinical trial targeting the glutamate system, Berman et al reported the first finding of a rapid antidepressant response to the glutamate N-methyl-D-aspartate (NMDA) receptor (NMDAR) antagonist ketamine in nine depressed patients in a randomized, saline-controlled cross-over design (7). Ketamine-related mood improvement was robust and peaked within 72 hours following a single low-dose infusion (0.5 mg/kg over 40 min). Importantly, the observed antidepressant effect was segregated in time from transient neurocognitive (including psychotomimetic and dissociative) effects, which peaked during the ketamine infusion and returned to baseline within two hours. These initial findings were then replicated and extended in a larger study of 18 inpatients with TRD utilizing a similar saline-controlled cross-over design (8). In that study, the drug-placebo effect size was very large the day following infusion and 71% of patients met response criteria at this time-point. In addition, a significant minority of patients (35%) maintained their response for at least 1 week. Following these original observations, multiple case reports, case series and several small-scale clinical trials of ketamine in depression have emerged, adding support to the hypothesis that ketamine possesses rapid-acting antidepressant properties (see Table 1).
Research has begun to investigate relapse-prevention and therapeutic maintenance strategies for patients who manifest an antidepressant response to ketamine (9,10). In a first study of 26 patients with TRD, the neuroprotective agent riluzole was tested as a relapse-prevention approach following ketamine (9). In this study, seventeen patients (65%) met response criteria and thirteen patients (50%) met remission criteria 24 hours following a single low-dose ketamine infusion. Patients who continued to maintain their response 3 days following the infusion (54%) proceeded to participate in a four-week, double blind, placebo-controlled, continuation trial of riluzole where the main outcome measure was time-to-relapse. This study did not support a specific role for riluzole for relapse prevention following ketamine, in part because both groups maintained a relatively prolonged response (> 3 weeks). In another study, repeated infusions of low-dose ketamine were administered three days a week following an electroconvulsive therapy (ECT)-like schedule (10). In this study nine patients with TRD underwent six infusions of ketamine over two weeks. The procedures were well tolerated and resulted in a sustained response during the two-week treatment period and for an average for nearly three weeks following the end of treatment. In a particularly striking case, a woman with severe and chronic TRD experienced a sustained remission from her symptoms for three months following the end of the acute treatment period in the absence of concurrent pharmacotherapy (11).
Suicidal ideation and behavior is a chief concern related to MDD and there is an urgent and unmet need for rapidly acting medical interventions for this worrisome condition. In contrast to conventional antidepressant agents that may actually worsen suicidal ideation in the short-term, preliminary evidence suggests that ketamine may possess acute anti-suicidal ideation (SI) properties (12–14). An initial study reported rapid reduction in both explicit and implicit measures of SI in response to ketamine with 81% of patients with TRD free of SI 24 hours following infusion (12). A second study reported decreases in SI as early as 40 minutes after initiation of ketamine (13) and a third study reported rapid reductions in SI following ketamine in patients presenting to the emergency department with SI and depression (14).
Ketamine is a dissociative anesthetic that causes transient altered mental function at the sub-anesthetic doses that have been studied in patients with depression, including dissociative and psychotomimetic effects. These effects are usually mild to moderate, although they can be more pronounced in a minority of cases. Ketamine is also a drug of abuse and both clinical and preclinical studies have raised concerns regarding the potential for neurotoxicity of ketamine and other NMDAR antagonists, particularly when given at high doses and over extended periods of time. Given these considerations, future clinical research with ketamine must proceed cautiously and weigh the potential risks and benefits for patients with TRD.
The observed antidepressant effects of ketamine in clinical populations provides compelling rationale for investigating NMDAR antagonists and other modulators of the glutamate system as novel targets for therapeutic development in MDD. In addition to NMDAR antagonism, ketamine potentiates glutamate transmission at alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors (AMPAR) and has additional inhibitory effects at muscarinic acetylcholine receptors. NMDARs, AMPARs and kainate receptors are ionotropic glutamate receptors localize throughout the central nervous system (CNS) and play a major role in signaling at excitatory synapses. Trullas and Skolnick first articulated the hypothesis that NMDAR antagonists may represent a new class of antidepressants, based in part on the observation that inescapable stress led to disruptions in hippocampal neuronal long-term potentiation (LTP) – an NMDAR-dependent process – in parallel with the observed syndrome of behavioral depression (15). This line of research demonstrated the ability of NMDAR antagonists, including dizolcipine (also known as MK-801), to exert antidepressant effects in animal models (15) and showed that traditional monoaminergic antidepressants modified NMDARs following chronic (but not acute) treatment, suggesting that NMDAR modulation may represent a downstream effect of monoamine drugs (16). Taken together, these data suggested that targeting the NMDAR complex directly may represent a strategy leading to improved, faster-acting antidepressant agents (6).
A series of preclinical studies of ketamine in animal models of depression are beginning to yield important insights into the cellular and molecular mechanisms of ketamine’s antidepressant action and the downstream consequences of NMDAR antagonism (Table 2). In particular, several recent well-designed studies have provided compelling evidence that ketamine induces rapid enhancement of synaptic structure and function in cortical regions concomitantly with antidepressant behavioral effects in rodents (17–19). Li et al reported that ketamine rapidly activated the ubiquitous mammalian target of rapamycin (mTOR) pathway, leading to increased synaptic signaling proteins as well as an increased number and function of new spine synapses in the prefrontal cortex (PFC) of rats (17). Specifically, ketamine was observed to transiently increase levels of phosphorylated eukaryotic initiation factor 4E binding protein 1 (4E-BP1), p70S6 kinase (p70S6K) and mTOR, all members of the mTOR-signaling pathway that function as key regulators of protein translation. Ketamine was also observed to increase the activation of other ubiquitous mediators of translation and synaptic plasticity, namely extracellular signal–regulated kinase (ERK) and protein kinase B (PKB/Akt). In fact, blockade of either the ERK or Atk pathway abolished the impact of ketamine on the mTOR pathway, emphasizing the interaction between plasticity-related signally pathways. Ketamine also resulted in elevate levels of several synaptic proteins (PSD95, GluR1, synapsin I), increased spine density and enhanced excitatory postsynaptic current (EPSC) frequency responses – all suggestive of a rapid enhancement of the structure and function of cortical synapses by ketamine. Blockade of mTOR signaling prevented both ketamine-induced synaptogenesis and antidepressant behavioral responses.
A second study by the same group extended these findings to a chronic mild stress (CMS) paradigm and demonstrated that ketamine was able to reverse CMS-induced deficits in behavior as well as in synaptic proteins, spine density and EPSC frequency/amplitude in PFC (19). The behavioral and molecular effects of ketamine were again blocked by inhibition of the mTOR signaling pathway. In another recent study, Autry et al reported that the antidepressant effect of ketamine was dependent on the rapid synthesis of brain-derived neurotrophic factor (BDNF) in a mouse model (18). Ketamine resulted in elevated levels of BDNF in the hippocampus (but not the nucleus accumbens) and knockouts of either BDNF and or the neurotrophic tyrosine kinase receptor (TrkB) were insensitive to the antidepressant effects of ketamine. Extending these results further, the authors found that ketamine blockade of NMDARs deactivated eukaryotic elongation factor 2 (eEF2) kinase, which resulted in reduced eEF2 phosphorylation and subsequent enhancement of BDNF translation. Several classes of antidepressants, including ECT, have also been demonstrated to increase BDNF in the hippocampus, although generally over a longer time period. A clinical study found that peripheral measures of BDNF did not increase following ketamine (20), however the significance of this finding is unclear given the potential uncoupling of peripheral and central functioning of BDNF. Finally, a study of repeated daily doses of ketamine over seven days in a CMS model found that ketamine reversed CMS-induced weight loss and normalized neuroendocrine measures but did not alter hippocampal BDNF protein levels (21).
Taken together, these studies provide compelling evidence for the involvement of synaptic plasticity and neurotrophic signaling in the mechanism of action of ketamine. The observation of mTOR pathway or BDNF pathway activation by NMDAR antagonism provides a critical link between NMDAR modulation and neurotrophic and synaptic plasticity-related theories of depression and antidepressant action (22). All three studies described above reported that the molecular and behavioral effects of ketamine were blocked by administration of the AMPAR antagonist NBQX, extending a previous report (23). Preclinical work has linked enhancement of AMPA signaling with synaptic plasticity and these findings support to a model whereby ketamine may act to both decrease NMDAR signaling and enhance AMPAR signaling with a net effect of enhancement of synaptic plasticity and neurotrophic signaling (6). Of potential therapeutic relevance, a novel class of compounds that allosterically enhances the activity of AMPARs (referred to as AMPA potentiators) has demonstrated antidepressant properties in animal models (6) and may represent a promising avenue for novel treatment development based on the glutamate system.
Current neurocircuitry models of depression posit relatively deficient dorsal and lateral PFC and anterior cingulate cortex (ACC) regulation over ventral cortical and subcortical regions that govern emotion generation and stress responses (24,25). Dorsolateral and dorsomedial regions of PFC (DLPFC and DMPFC, respectively) have been consistently found to be underactive during a major depressive episode, often in parallel with overactivity of ventral cortical structures (e.g. subgenual ACC [SGACC], orbital frontal cortex [OFC]) or subcortical limbic structures (e.g. amygdala) (24). This perturbation of neural activity results in the cognitive, emotional and behavioral manifestations of depression, while normalization of aberrant activity patterns is posited to underlie amelioration of clinical symptoms.
The question of how ketamine affects the brain at the circuit level to bring about a rapid antidepressant response remains an area of active research. A series of studies found that ketamine and other NMDAR antagonists enhance glutamateric signaling in the cortex of rodents, potentially through inhibition of GABAergic interneurons and subsequent disinhibition of cortical pyramidal neurons (26,27). Enhancement of activity at pyramidal glutamatergic synapses by ketamine would be consistent with the observations of enhanced cortical synaptic plasticity and function described above. Neuroimaging studies in humans likewise suggest that subanesthetic doses of ketamine result in elevated cortical activity, including in regions of PFC and ACC (28–31). A functional MRI (fMRI) study found that ketamine resulted in decreased activity in ventromedial PFC (VMPFC), OFC and SGACC accompanied by increased activity in posterior cingulate and other cortical regions (32). Studies utilizing proton magnetic resonance spectroscopy (1H-MRS) to examine the impact of ketamine on the glutamate system in vivo have yielded mixed results (33,34), although the lack of robust findings may be attributed to limitations inherent in current 1H-MRS methodology. It should be noted that the neuroimaging studies described above have all been conducted in healthy populations and therefore extrapolating the findings to depression remains speculative.
The molecular and neuroimaging studies summarized herein are consistent with the hypothesis that ketamine exerts its rapid antidepressant effects via rapid enhancement of PFC and ACC function (see Figure 1), although this hypothesis remains to be tested directly. Intriguingly, direct stimulation of medial PFC (MPFC) using optogenetic techniques in mice resulted in an immediate antidepressant effect in a social defeat stress paradigm (35). The same report described reduced expression of immediate early genes, reliable markers of neuronal activity, in both mouse MPFC following social defeat stress and in PFC tissue obtained from depressed individuals postmortem, further supporting the hypothesis of reduced functioning of PFC regions in depression. A series of studies utilizing magnetoencephalographic (MEG) recordings in patients with TRD prior to infusion of a single low-dose of ketamine provide initial support for a role for the ACC in antidepressant response (36,37). In these studies, pretreatment ACC activity was associated with subsequent antidepressant response to ketamine. A significant gap in the current literature is that no study has investigated changes in neural markers following ketamine in a depressed population. Direct testing of the role of the ACC, PFC or other regions in the mechanism of action of ketamine will require optimized pre- and post-intervention imaging designs in clinical populations utilizing a placebo control condition.
Despite progress in drug discovery for MDD, there remains an unmet public health need to identify novel, rapidly acting agents for patients with treatment-resistant forms of depression. The observations of a rapid antidepressant effect of low-dose ketamine has generated considerable interest in the scientific and medical community alike, representing a proof-of-principle in targeting the NMDAR, and glutamate system more broadly, as a novel treatment approach in depression. Low-dose ketamine appears to enhance the strength of cortical synapses through NMDAR and AMPAR-dependent and neurotrophic mechanisms and may rapidly reverse prefrontal cortical deficits in depression at the neurocircuit level, leading to an amelioration of clinical symptoms. Models linking ketamine’s effects at the cellular, circuit and behavioral levels in depression, however, remain to be tested directly. For example, no study to date has investigated the impact of ketamine on functional neural markers in patients with MDD.
The interpretation of the available clinical outcome data concerning the efficacy of ketamine as an antidepressant is limited by small sample sizes and the lack of an adequate control condition in some cases. To date, no large-scale study of ketamine in MDD has been completed that utilizes an “active” control condition (e.g. a different anesthetic agent) that would mimic some of ketamine’s acute neuropsychiatric effects and thereby strengthen the integrity of the study blinding procedures, although these studies are underway (ClinicalTrials.gov NCT00768430). Concerns regarding potential adverse behavioral and neurotoxic effects of ketamine and its potential for abuse demand a cautious approach to treatment development. Never the less, ketamine may represent an important addition to the armamentarium of therapeutic interventions for severe and treatment refractory cases of MDD. Future research may investigate alternative dosing or delivery strategies in order to enhance the safe use of ketamine.
In parallel with clinical research designed to optimize the safety and efficacy of ketamine for TRD, translational neuroscience approaches using ketamine as a model of rapid antidepressant action hold considerable potential to advance treatment discovery for MDD. In particular, the identification of biomarkers of rapid antidepressant response using neuroimaging or other in vivo approaches is hoped to speed novel therapeutic discovery in depression by providing surrogate endpoints for testing novel candidate therapeutics.
Support and Disclosures
Dr. Murrough is a full-time employee of Mount Sinai School of Medicine and is currently supported by NIH/NIMH (1K23MH094707-01), NARSAD (Young Investigator Award), and the American Foundation For Suicide Prevention (YIG-0-10-223). The content of this article is the responsibility of the author and does not necessarily reflect the views or opinions of the NIH or other funding agencies. Within the past two years Dr. Murrough completed a research fellowship at Mount Sinai School of Medicine that was supported in part by an educational grant from AstraZeneca to Mount Sinai.
Dr. Charney, Dean of Mount Sinai School of Medicine, has been named as an inventor on a use-patent of ketamine for the treatment of depression. If ketamine were shown to be effective in the treatment of depression and received approval from the Food and Drug Administration for this indication, Dr. Charney and Mount Sinai School of Medicine could benefit financially.