An overview of NMDA receptor physiology and clinical significance
The N-methyl-D-aspartate receptor (NMDA-R) is one sub-type of glutamate receptor that are increasingly being recognized for their critical role in the neurophysiology of important cognitive and psychological functions and the pathophysiology of diverse disease processes. As a neurotransmitter, glutamate mediates the majority of excitatory neurotransmission in the human brain, and an understanding NMDA-R physiology and clinical significance can help treat the diseases these receptors help to create.
NMDA-Rs have a role to play in diverse and complex processes of mind and brain including:
Memory Formation and Long term Potentiation
Excitotoxicity and Stroke Outcome
Ego and Frontal Lobe Function
Pain and Opiate Tolerance
Alcoholism and Drug Abuse
Schizophrenia and Depression
Excitatory Amino Acid receptors
NMDA-Rs are members of the larger class of Excitatory Amino Acid receptors (EEA-Rs) that register the neurotransmitter glutamate. Glutamate is a simple amino acid available to the body through ingestion or through biochemical synthesis from glutamine. Two broad classes of EEA-Rs exist:
1) The metabotropic receptors (m-Glu-Rs) – transmit signals to the inside of a neuron via 2nd messenger molecules. These do not allow ions to flux directly.
2) The ionotropic receptors – allow ions (charged elements) to flux across the cell membrane, exciting the neuron and/or activating protein kinases through the influx of calcium (Ca++). Ionotropic EEA-Rs mediate 90% of excitatory neurotransmission in the human brain. Three subtypes of ionotropic receptors exist, each named for the semi-specific agonists or activating molecules originally used to study them (though more highly specific agonists have now become available for research).
a) AMPA receptors (AMPA-Rs) – allow sodium (Na+), potassium (K+) and some Ca++ to flux across the membrane after binding the neurotransmitter glutamate. AMPA-Rs mediate an aspect of fast excitatory neurotransmission in most of the synapses in the CNS.
b) Kainate receptors – ubiquitous in the central nervous system (CNS), these receptors allow Na+, K+ and significantly more Ca++ to flux than AMPA-Rs.
c) NMDA Receptors – are the third type of ionotropic EAA-Rs. NMDA-Rs perform a slower but more sustained component of excitatory neurotransmission and have particular properties that make them substantially different than the others.
NMDA-Rs are highly permeable to Calcium. They are ubiquitous in the CNS and the peripheral nervous system (PNS), but different receptor subtypes localize to different areas. The highest densities are found in the CA1 field of the hippocampus, and in the frontal lobe.
Subunits: The NMDA-R is made of two subunits, the NR1 subunit and the NR2 subunit. The NR1 subunit is absolutely required while the NR2 subunit, of which there are four subtypes (named NR2A-D), is optional and confers upon the receptor specific characteristics. The NR2A subunit is common in the cerebral cortex, hippocampus, and cerebellum. NR2B is found mainly in the telencephalon. NR2C is only found in the cerebellum (Akbarian 1996).
Ligands: NMDA-R activation requires the simultaneous presence of two distinct natural agonists. The neurotransmitter Glycine is generally not considered to be the limiting neurotransmitter. Glutamate is usually the limiting factor in receptor activation. Natural glutamate site agonists include glutamate, aspartate, N-methy-D-aspartate and quinolinic acid, a metabolite of tryptophan (Stone et al., 2007).
N-methyl-D-aspartate Glutamate Glycine
Blockade: Many experimental and therapeutic drugs can block the receptor by acting as antagonists (blockers) at the phencyclidine (PCP) site. Ketamine, nitrous oxide (N20), dextromethorphan, and MK-801 all act in non-competitive fashion by blocking the open ion channel from the extracellular side. Magnesium (Mg++) is a naturally occurring antagonist that blocks the closed ion channel intracellularly, and increases the threshold stimulation needed to activate the receptor. Kynurenic acid is another natural NMDA-R antagonist. It is produced through catabolism of tryptophan and may block the glycine site on the receptor (Stone et al., 2007).
Glycine site agonists and antagonists also exist. Serine and alanine are glycine site agonists, and D-cycloserine is a partial agonist/antagonist. These agonists are being evaluated in schizophrenia research for their potential to increase glutaminergic neurotransmission without the danger of excitotoxicity that glutamate site agonists pose.
Figure 1: The NMDA Receptor
Image used with permission: http://www.profrontal.com/nmda.html
Voltage dependent blockade: NMDA-Rs are naturally subject to blockade by Mg++ at low voltages. Once the neuronal call membrane voltage rises a conformational change in the receptor’s physical shape ejects the Mg++ from the channel. It is likely that glutamate and glycine acting at NMDA-Rs is not enough to open these channels without other stimulation, specifically due to this effect of Mg++. The channel opens only if the neuron is already partially stimulated or depolarized either by glutamate acting at AMPA or Kainate receptors or by other neurotransmitter/receptor combinations.
In other words, some information in the form of neurotransmission must already have been received by the neuron for NMDA-Rs to have the potential to open. This prior neurotransmission may be received from the same presynaptic neuron via other types of receptors (e.g., AMPA-Rs) or may stem from converging input from different presynaptic neurons (Armstrong et al., 1993).
High Calcium Flux: The NMDA-R channel is highly permeability to Ca++, which leads to long-term potentiation (LTP) through increased protein kinase activity and resulting changes in the neuron. Ca++ appears to be essential for affecting change in neurons over time by signalling changes in receptor densities and profiles, secondary messenger molecules and cytosolic enzymes. NMDA-Rs are essential in this process.
Prolonged Activity: Once activated to a significant extent NMDA-Rs remain active and facilitate post-synaptic action potentials or neurotransmission long after a stimulating impulse has passed. The reasons for this sustained activity are not completely understood, but protein kinases that are activated by Ca++ do remain active for some time. There also may be sustained glutamate release by the presynaptic neuron as the result of an unknown process that may include retrograde messengers. Other post synaptic changes that make such sustained activity possible have been proposed as well (Bliss & Collingridge 1993).
This effect is similar to striking a piano key with the sustain pedal down. The initial strike of the string would symbolize initial AMPA-R facilitated action potentials, while the sustained tone that follows after the strike would symbolize the sustained NMDA-R medicated action potentials occurring for some time after the initial stimulus has passed.
Long Term Potentiation (LTP): The unique properties of the NMDA-R are generally believed to meet the requirements of a cellular mechanism called “coincidence detection,” which Hebb proposed as a requirement for memory formation (Bliss & Collingridge 1993). Because the NMDA-R opens only when multiple sources of information converge upon the post-synaptic neuron, an association between these many inputs can be formed and a fundamental memory unit may occur.
Increased protein kinase activity, caused by significant and sustained Ca++ influx, is believed to cause changes that ultimately strengthens synaptic connections and causes learning or LTP. This is believed by many to be the cellular basis of memory and neural plasticity. Synapses with NMDA-Rs are involved in relaying the importance of recent stimulus downstream.
There is short time delay in this process. A simple message is initially relayed as a nerve impulse by non-NMDA receptor activity, but if the information is granted importance by being related to other stimuli, a more powerful and sustained impulse mediated by NMDA-Rs follows 10 to 50 milliseconds later (Armstrong-James et al., 1993). In addition to marking the importance of a specific relational data set for downstream communication, LTP occurs if the circumstances merit and the future strength of signals relayed at this junction grows. In the future an identical stimulus causes a larger response.
LTP means that the connection between bits of information, or the association between data gets stored. The connection becomes stronger and faster the next time. Without LTP neurons wouldn’t learn or appropriately change response to the information they process.
Thus, NMDA-R activation represents a fundamental unit of association, because the receptor requires a convergence of information in order to become activated and creates change over time. NMDA-Rs help signify that “this is more important than that.” They also provide a process for registering the passage of time, because the sustained neurotransmission caused by NMDA-R activation continues to inform downstream processes even after a stimulus has passed.
Frontal Lobe Fuction and Ego: The frontal lobe has a high density of NMDA-Rs and can exhibit up to 94% more NMDA-R mRNA than the parietal-temporal lobe (Akbarian 1996). The high density and sustained activity of NMDA-Rs in the frontal lobe helps produce the phenomena of “working memory.”
Working memory is the brain’s immediate memory cache. Generally, working memory lasts for around one hour and holds on to information that is currently relevant. Heimer says that working memory is responsible for the “moment to moment awareness and retrieval of stored information for the purpose of making informed decisions” (Heimer 1995, pp 448-449). For example, working memory acting in the service of the executive may allow the shift in quality and meaning from, “It is raining,” to, “It was raining when I last looked outside so I had better bring an umbrella.”
Perhaps counter-intuitively, PET scans of brain activity under NMDA-R blockade show increased activity in the prefrontal cortex (Breirer et al., 1997) and executive processes are clearly overwhelmed functionally. This region of the brain becomes so active that the result is a significantly decreased capacity for decisions, planning and intellect. In research using medications to block NMDA-Rs, working memory and executive cognitive function in humans appears to be particularly impaired (Goldman-Rakic 1995).
It appears that the increase in prefrontal activity under NMDA receptor blockade is actually mediated by a loss of local inhibitory tone. Gamma Aminobutyric Acid, or GABA, is an inhibitory neurotransmitter, but the GABAergic neurons in the frontal lobe that release this neurotransmitter are actually stimulated to produce their inhibitory action potentials by glutamate at NMDA-Rs. In the frontal lobe, GABAergic inhibitory processes powerfully regulate neurotransmission and whittle down the myriad of possibilities in executive processing and decisions. Loss of this essential reducing valve can overwhelm frontal processes.
The NMDA-R’s slow gating kinetics produces prolonged signal amplification that likely gives the frontal lobe much of its executive “tone” from moment to moment. These receptors remain active long after a stimulating impulse has passed, especially in the frontal lobe. This powerful inhibitory force may be a large part of what makes a person who he or she “is” simply by reducing the myriad possibilities to a small enough stream to make decisions and actions possible.
In many ways, as far as ego and executive consciousness is concerned, it might be accurate to say, “One IS what one is NOT NOT.”
It’s a bit of a puzzle, but in the end what “is” as regards the smaller, ego-driven center of consciousness, may only reflect what remains after most of the vast possible responses, thoughts, emotions and impulses are reduced or suppressed. Rather than a synthesis of our whole being, what “is” on the everyday, ego-driven level may reflect only what hasn’t been cancelled by inhibition.
Religious Experience and Transcendence: The subjective internal experience of the NMDA-R antagonists nitrous oxide and ketamine can be powerfully spiritual. Subjective reports of a transcendent euphoria resulting from these medications can be found easily on the internet and in the writings of such philosophers and scientists as William James and Stanislov Grof. Those reporting upon the spiritual side of the experience describe a powerful but gentle psychological freedom, predominately positive when appropriate preparations are made.
Evidence also points to the existence of endogenous blockers of the NMDA-R that may participate in spontaneous experiences of transcendence or near death experiences, especially during periods of danger and risk due to oxygen deprivation such as a heart attack or stroke (Jansen 1990b). One molecule with possible NMDA-R antagonist properties is dynorphin (Brauneis et al., 1996), one of the body’s natural opiates that may be released in response to damage and stress (Cechetto 1994). Dynorphin also may decrease glutaminergic influence from (primary afferent) sensory nerves by decreasing Ca++ and increasing K+ influx, which would make it more difficult for excitatory neurotransmission to occur (Lai et al., 2001).
Standard medical models have often chosen to label these states of mind as “psychotic,” but subjects often feel that they have discovered a deep and lasting truth, or that they have spent some timeless period in a transpersonal or archetypal realm. Many subjective accounts of NMDA-R blockade are essentially indistinguishable from historical mystical accounts of spontaneous ecstasy.
Indeed, in his chapter on mysticism, the lauded American philosopher William James makes no distinction between his own mystical revelations brought about by inhaling nitrous oxide and those occurring spontaneously (James 1902, p 387). It took one hundred years since James was doing his experiments with nitrous oxide, commonly known as “laughing gas,” to discover that it causes anesthesia by blocking NMDA-R activity (Mennerick et al., 1998). For a long time it was thought to exert it’s action through a GABAergic mechanisms, but drugs acting directly on the GABAergic system do not change the quality or content of consciousness this profoundly.
Schizophrenia and Psychosis: Symptoms in schizophrenia, especially the negative symptoms of the disorder, may stem from reduced NMDA-R function that develops and accelerates in the late teenage years. This period is marked by a large increases in NMDA-R function (Hirsch et al., 1997; Olney & Farber 1995; Tsai et al., 1995) and this may help explain why schizophrenia rarely develops before adolescence.
Adolescence is marked by a large and rapid increase in either the number or strength of NMDA-R mediated neuronal connections. For this reason blockade of NMDA-Rs with ketamine can cause profound perceptual changes in adult patients leading to anxiety and distress, especially in hospital circumstances. These disturbances are either significantly reduced in preadolescent patients or they are less likely to cause conscious distress (Farber et al., 1995). Because ego related disturbances with ketamine become more profound in patients over 16 years of age it is rarely used as a solo agent in adults, despite its reputation as an extremely effective and safe short-term anesthetic.
The theory that low NMDA-R function contributes to schizophrenia symptoms is named the “Glutaminergic/NMDA-R Hypofunction” (NRH) theory of schizophrenia. It is not in conflict with the “Dopaminergic Theory of Schizophrenia,” because dopaminergic neurons are known to be inhibited by GABAergic neurons excited by glutamate at NMDA-Rs. Therefore a decrease in NMDA-R activity causes a downstream increase in dopamine activity that also may contribute to symptoms.
It is possible that schizophrenia with primarily positive symptoms is due more to dopamine signaling, while schizophrenia with primarily negative symptoms may be more due to NRH (Olney & Farber 1995). NRH may occur in numerous ways. Decreased NR2 subunit density (Akbarian 1996), over production of endogenous blockers such as the kynurenic acid metabolite of serotonin, direct receptor dysfunction, and decreased production of glutamate are just a few of the proposed mechanisms.
Clinically, this theory is behind attempts to treat schizophrenics with glycine site agonists designed to help increase NMDA-R activation without the risk of excitotoxicity that glutamate presents. Glycine, serine and sarcosine are all potential treatments in this category.
Tolerance and Behavioral Sensitization: NMDA receptors are fundamental to the development of “tolerance” as a byproduct of LTP. Tolerance is the process by which a given reaction in response to specific stimulus shifts downward in amplitude over time. For example, a loud or repetitive noise changes in its ability to startle us as NMDA-Rs activate changes (Oye et al., 1992). Alternatively, drugs such as alcohol (Morato et al., 1996) and morphine (Wiesenfeld-Hallin 1998) change in their ability to cause euphoria over time through NMDA-R activity.
Similarly, NMDA-Rs have been widely implicated in the phenomena of behavioral sensitization to psychomotor stimulants such as amphetamine. As successive doses of stimulants are given to animals, certain stereotyped behaviors such as tics become augmented over time. This augmentation is blocked by the administration of NMDA-R antagonists (Wolf 1998).
Essentially tolerance is “learning” or LTP, and this learning is mediated in large part by NMDA-Rs. When the “learning” is undesirable, for example in the development of opiate tolerance, NMDA-R antagonism can inhibit this process. This technique is employed to decrease the dosages of morphine required to achieve sufficient pain control by decreasing the development of tolerance. With NMDA-R blockade neurons do not “learn” to ignore the morphine as they would normally.
Stoke and Excitotoxicity: Many neurons are lost during stroke through apoptosis due to specific activation of cellular suicide programs. The process of apoptosis is partly due to uncontrolled influx of Ca++ into neurons through NMDA-Rs. Excessive Ca++ activates protein kinase activity in an uncontrolled fashion and increases free radical formation. With oxygen supplies diminished, the antioxidant species needed to quell these runaway processes become severely depleted, and activation of cell death programs is often inevitable.
It is also possible that glutamate reuptake pumps that serve in concert with NMDA-Rs may work in reverse in low oxygen conditions, leading to increased release and post-synaptic binding of glutamate and excessive influx of Ca++.
NMDA-R antagonists can decrease neuronal call death during hypoxic conditions such as stroke. Animal data with NMDA-R antagonists are quite promising, but getting these medications delivered quickly is difficult due to functional limits in clinical care. Human trials have been disappointing, partly because antagonists are usually delivered after a majority of damage has already occurred.
Alcoholism: Alcohol is a molecule with a dual combination of neurotransmitter effects. First, it is GABAergic, helping to relax and inhibit neurotransmission. Second, it blocks NMDA-Rs, which directly contributes to some of alcohol’s famous disinhibiting effects. Over time, blocking these two receptors actually increases their total numbers as synaptic membranes increase receptor densities to amplify signalling that has been artificially suppressed by chronic alcohol use. When a chronic heavy drinker stops using alcohol the result is increased physical tension, anxiety and agitation, sleeplessness, depression and sometimes seizures.
The loss of GABAergic tone is usually managed in detox protocols with medications that increase the strength of GABA-mediated neurotransmission (e.g., diazepam, lorazepam). But there is usually little to be done about the increase in NMDA-R activity, despite the likelihood that this issue heavily contributes to the early abstinence symptom stages characterized by emotional angst, irritability, depression and intense self-critique and doubt. Acamprosate, an FDA-approved medication for alcohol dependence, may provide some relief from this process through NMDA-R effects. Magnesium supplementation may also help, because it naturally blocks the NMDA-R and can become severely depleted in chronic alcoholism.
Depression: It is becoming clear that depression in mood suffered so commonly by humans is at least partially mediated by glutamate and NMDA-R function. A significant portion of ego awareness and frontal lobe function is due to NMDA-R activation of inhibitory GABAergic neural nets in the frontal lobe, and it may be that that Depression also result from this process as well.
Clinical research and medication trials in this area are occurring rapidly and results are very promising. For example, seven patients with major depression were given ketamine (0.5 mg/kg) vs saline in a DBPC crossover study. The relief of depressive symptoms was rapid and significant in those receiving ketamine, with the mean Hamilton Depression Rating Scale reduced by 14 points with active treatment (Berman et al., 2000). In 2006 these study results were essentially repeated with 17 patients in another DBPC crossover study. 71% of these patients met response criteria and 29% of patients met criteria for remission within hours of the infusion. One week later many improvements in the HAM-D rating scale were holding, though some were fading (Zarate et al., 2006).
A different treatment approach was assessed by Kudoh et al., (2002) who looked at the potential benefits of ketamine in 70 depressed orthopedic surgery patients. Controls had their anesthetic induction with fentanyl and propofol only, while active treatment included low doses of ketamine as well. The patients receiving low dose ketamine reported significantly less postoperative depressive symptoms and pain than controls.
Interestingly, there was a clue as to the antidepressant effects of ketamine buried in the medical literature in 1979. A research group used ketamine to increase physiologic arousal in a study using controlled anxiety induction as a treatment for Generalized Anxiety. Ketamine is known to increase sympathetic drive, and the authors hoped that this would increase the intensity of the exposure and subsequent treatment response. But, instead they found that while ketamine did reduce the negative emotion or affect experienced by patients during stressful situations, most of this effect was due to a decrease in depressive symptoms rather than increased subjective emotional arousal (Sappington et al., 1979) during the exposure. They may have inadvertently stumbled upon an effect of ketamine we are only now exploring, almost 30 years later.
Over the last few years there have been a number of papers hypothesizing upon the potential antidepressant mechanism of ketamine. Maeng and Zarate (2007) argue that blocking NMDA-Rs increases the relative contribution of AMPA-Rs to fast excitatory neurotransmission, creating an antidepressant effect. Others argue that downstream dopamine release is the key. Regardless of the mechanism, it is known that altered glutamate neurotransmission is a likely part of the etiology of both depression and anxiety, and it appears that ketamine can intervene through NMDA-R blockade. More research is needed to understand why there may be fading effect size to this treatment over time (Liebrenz et al., 2007).
Dementia and epilepsy: A lifetime of NMDA-R activation and Ca++ influx may be at the heart of neural aging and loss of memory. The neurons in the hippocampus are generally required for memory formation and consolidation, and this area of the brain relies heavily upon NMDA-Rs, especially in the CA1 field. Protecting these neurons from excessive excitation in patients with significant memory loss is the goal of pharmaceutical treatment with memantine, a mild NMDA-R antagonist. Blocking excessive activation of NMDA-R may save neurons over time and may even help to distinguish signal from noise more efficiently.
Loss of hippocampal neurons via excitotoxicity or other insults may contribute to temporal lobe epilepsy. After a loss of connections between neurons new growth often occurs and can lead to abnormal circuits that can cause seizures. Because these circuit loops are especially common in neural tissue rich with NMDA-Rs, it is felt that these receptors contribute to seizure circuitry through the damage they create. Many anti-seizure medications work by blocking the neurotransmitter Glutamate.
Electroconvulsive Therapy (ECT): The full and indiscriminate activation of NMDA-Rs during ECT probably contributes to the common post-treatment side effect of transient memory loss and dysfunction. During ECT, NMDA-Rs are maximally stimulated by the supplied electricity; afterward, their response to normal stimuli is capped, having been maximal activated already, and memory formation is impaired (Jeffrey et al., 1997). It has been show in vitro that blocking NMDA-Rs before ECT can decrease the random activation that occurs with the procedure, but whether this would decrease the effectiveness of the treatment is unknown.
NMDA-R Hypofunction (NRH) and “Olney’s Lesions”: Olney & Farber (1995) found that prolonged blockade of NMDA-R activity in rats can lead to excitotoxic lesions in the retrospenial cortex and the posterior cingulate. In PET scans, these areas are highly active metabolically when NMDA-R blockade occurs, probably due to loss of GABAergic inhibitory tone. Because GABAergic neurons are strongly excited through NMDA-R activation, blocking NMDA-Rs leads to a profound loss of normal inhibitory influences.
Decreasing the excitation of these areas with GABAergic drugs blocks both these excitatory lesions and prevents the changes in consciousness that occurs with NMDA-R blockade. It is ironic that NMDA-R antagonists, known to protect neurons from hyperexcitatory apoptosis, may actually cause lesions in focal regions of the rat brain when given for extended periods. It is unclear whether these lesions occur in humans.
Akbarian S, Sucher NJ, Bradley D, Tafazzoli A, Trinh D, Hetrick WP, Potkin SG, Sandman CA, Bunney Jr. WE, Jones EG (1996) Selective alterations in gene expression for NMDA receptor subunits in prefrontal cortex of schizophrenics. Journal of Neuroscience 16(1):19-30
Armstrong-James M, Welker E, Callahan CA (1993) The contribution of NMDA and non-NMDA receptors to fast and slow transmission of sensory information in the rat SI barrel cortex. Journal of Neuroscience 13(5):2149-2160
Berman RM, Cappiello A, Anand A, Oren DA, Heninger GR, Charney DS, Krystal JH (2000) Antidepressant effects of ketamine in depressed patients. Biol Psychiatry 47(4):351-4.
Bliss TVP, Collingridge GL (1993) A synaptic model of memory: long term potentiation in the hippocampus. Nature 361(7 Jan):31-39
Brauneis U, Oz M, Peoples RW, Weight FF, Zhang L (1996) Differential sensitivity of recombinant N-methyl-D-aspartate receptor subunits to inhibition by dynorphin. Journal of Pharmacology and Experimental Therapeutics 279(3):1063-1068.
Breier A, Malhtra AK, Pinals DA, Weisenfeld NI, Pickar D (1997) Association of ketamine-induced psychosis with focal activation of the prefrontal cortex in healthy volunteers. American Journal of Psychiatry 154(6):805-81
Cechetto DF (1994). Identification of a cortical site for stress-induced cardiovascular dysfunction. Integrative Physiological and Behavioral Science 29(4):362-373
Farber NB, Wozniak DF, Price MT, Labruyere J, Huss J, St. Peter H, Olney JW (1995) Age-specific neurotoxicity in the rat associated with NMDA receptor blockade: potential relevance to schizophrenia? Biological Psychiatry 38(12):788-796
Goldman-Rakic PS (1995) Cellular basis of working memory. Neuron 14:477-485
Heimer (1995) The Human Brain and Spinal Cord. Springer-Verlag New York, New York, NY
Hirsch SR, Das I, Garey LJ, de Belleroche J (1997) A pivotal role for glutamate in the pathogenesis of schizophrenia, and its cognitive dysfunction. Pharmacology, Biochemistry and Behavior 56(4):797-802
James W (1902) The Varieties of Religious Experience. Viking Penguin Inc., New York, NY (1995)
Jansen, KL (1990) Neuroscience and the near-death experience: roles for the NMSA-PCP receptor, the sigma receptor and the endopsychosins. Medical Hypotheses 1:25-29
Jeffrey KJ, Reid IC (1997) Modifiable neuronal connections: an overview for psychiatrists. American Journal of Psychiatry 152(2):156-164
Kudoh A, Takahira Y, Katagai H, Takazawa T. (2002) Small-dose ketamine improves the postoperative state of depressed patients. Anesth Analg. 95(1):114-8,
Lai J, Ossipov MH, Vanderah TW, Malan Jr TP, Porreca F (2001) Neuropathic Pain: the paradox of dynorphin. Mol. Interv. 1:160-167
Liebrenz M, Stohler R, Borgeat A. (2007) Repeated intravenous ketamine therapy in a patient with treatment-resistant major depression. World J Biol Psychiatry. Jul 10:1-4.
Maeng S, Zarate CA Jr. (2007) The role of glutamate in mood disorders: results from the ketamine in major depression study and the presumed cellular mechanism underlying its antidepressant effects. Curr Psychiatry Rep. 9(6):467-74.
Mennerick S, Jevtovic-Todorovic V, Todorovic SM, Shen W, Olney JW, Zroumski CF (1998) Effect of nitrous oxide on excitatory and inhibitory synaptic transmission in hippocampal cultures. Journal of Neuroscience 18(23):9716-9726
Morato GS, Khanna JM (1996) N-methyl-D-aspartate receptors, nitric oxide, and ethanol tolerance. Brazilian Journal of Medical and Biological Research 11:1415-1426
Olney JW, Farber NB (1995) NMDA antagonists as neurotherapeutic drugs, psychotogens, neurotoxins, and research tools for studying schizophrenia. Neuropsychopharmacology 13(4):335-345
Oye I, Paulsen O, Maurset A (1992) Effects of ketamine on sensory perception: evidence for a role of N-Methyl-D-Aspartate receptors. Journal of Pharmacology and Experimental Therapeutics 260(3):1209-1213
Salvadore G, Cornwell BR, Colon-Rosario V, Coppola R, Grillon C, Zarate CA Jr, Manji HK. (2008) Increased Anterior Cingulate Cortical Activity in Response to Fearful Faces: A Neurophysiological Biomarker that Predicts Rapid Antidepressant Response to Ketamine. Biol Psychiatry. Sep 24.
Sappington AA, Corssen G, Becker AT, Tavakoli M (1979) Ketamine-facilitated induced anxiety therapy and its effect upon clients’ reactions to stressful situations. J Clin Psychol. 35(2):425-9.
Stefanczyk-Sapieha L, Oneschuk D, Demas M. (2008) Intravenous ketamine “burst” for refractory depression in a patient with advanced cancer. J Palliat Med. 11(9):1268-71
Stone TW, Forrest CM, Mackay GM, Stoy N, Darlington LG (2007) Tryptophan, adenosine, neurodegeneration and neuroprotection. Metab Brain Dis. Dec;22(3-4):337-52
Tsai G, Passani LA, Slusher BS, Carter R, Baer L, Kleinman JE, Coyle JT (1995) Abnormal excitatory neurotransmitter metabolism in schizophrenic brains. Archives of General Psychiatry 52(10):829-836
Wiesenfeld-Hallin Z (1998) Combined opioid-NMDA antagonist therapies: What advantages do they offer for the control of pain syndromes? Drugs 55(1):1-4
Wolf ME (1998) The role of excitatory amino acids in behavioral sensitization to psychomotor stimulants. Progress in Neurobiology 54:679-720
Zarate CA Jr, Singh JB, Carlson PJ, Brutsche NE, Ameli R, Luckenbaugh DA, Charney DS, Manji HK. (2006) A randomized trial of an N-methyl-D-aspartate antagonist in treatment-resistant major depression. Arch Gen Psychiatry. 63(8):856-64.