The Glutamate Hypothesis of Schizophrenia

Jonathan M. Meyer, MD

Department of Psychiatry, University of California, San Diego​

Dopamine dysfunction is clearly implicated in the etiology of schizophrenia. Excess presynaptic dopamine in the dorsal striatum and abnormal postsynaptic dopamine D2 receptor sensitivity have been shown to be related to positive psychotic symptoms.1 Simply put, enhanced dopamine leads to psychosis, which is alleviated by the D2 antagonism of antipsychotics. However, dysfunction at D2 receptors in the form of excess dopaminergic neurotransmission does not adequately account for all aspects of schizophrenia, especially negative and cognitive symptoms of the disorder. The glutamate hypothesis of schizophrenia fills in those blanks and posits that there may be a relationship of deficient activity at certain glutamate synapses, which may in turn account for negative and cognitive symptoms.2 Glutamate neurotransmission is complex at the circuit level, and an understanding of actions at the receptor level can help clinicians’ understanding of those glutamate circuits associated with schizophrenia.

Evidence for the Glutamate Hypothesis

The main evidence for the glutamate hypothesis emerged from findings related to NMDA glutamate receptor antagonists. In both animal and human studies, NMDA antagonists such as ketamine and PCP elicited effects that resemble the spectrum of schizophrenia symptoms, including auditory hallucinations, depersonalization, delusions, and at times, cognitive dysfunction. For example, in one study,3,19 healthy participants given ketamine exhibited acute effects similar to both positive and negative symptoms, among other behavioral and cognitive effects. In another study4 of patients with schizophrenia, ketamine induced a short, dose-related increase in positive symptoms (P ≤ .01) similar to the patients’ symptoms during uncontrolled periods of their illness.

A study5 of long-term administration of PCP in monkeys also confirms the hypothesis that glutamate is involved in psychosis. Monkeys given PCP twice a day for 14 days exhibited schizophrenia-like symptoms and poor performance on frontal lobe–sensitive tasks. These effects remained long after the PCP was stopped, and they were alleviated by antipsychotic treatment. Genetic manipulation studies6 also support the long-term effects of NMDA dysfunction. A mouse model using transgenic mice deficient in NMDA expression showed a pattern of hyperactivity, stereotypy, and social isolation that improved with the use of antipsychotics. This mouse study showed that possible defects in development may give rise to problems with connectivity that, in turn, may serve as a substrate for the later development of schizophrenia.

Neurotransmission of Glutamate

Glutamate has been studied since the 1970s as a primary excitatory neurotransmitter used by all pyramidal cortical cells.7 Overall, neurotransmission of glutamate is similar to that of other neurotransmitters. Presynaptic vesicles secrete glutamate into the synaptic cleft, and a presynaptic pump transports excess glutamate back into the presynaptic neuron via an excitatory amino acid transporter that exists throughout the brain.8

Glutamate receptors are divided into 2 families: ionotropic and metabotropic. Ionotropic receptors exist mainly on the postsynaptic neuron, while metabotropic receptors exist on both the presynaptic and postsynaptic neurons.8 The structure of these 2 types of receptors is quite different.

AV 1. Ionotropic Glutamate Receptor Subtypes (00:28)

Based on Tsapakis and Travis9

The ionotropic receptors, including NMDA, AMPA, and kainite, comprise a series of ion channels built from multi-subunit structures (AV 1).9 These structures exist around a central pore that controls the flow of calcium, sodium, and potassium ions in and out of the cell; once activated, they cause fast ion flow in a matter of milliseconds.

Metabotropic receptors include Group 1 receptors (mGlu1 and 5), Group 2 (mGlu2 and 3), and Group 3 (mGlu4 and 6 to 8); these are all connected to G-protein coupled receptors (GPCR). Metabotropic neurotransmission is slow, taking several seconds, and requires the GPCR to interact with a G-protein to either increase or decrease the level of a second messenger (such as cyclic AMP). Presynaptically, they act as autoreceptors, and postsynaptically, they are part of the second-messenger system propagating the signal.

Each of the glutamate receptors has its own properties, gene families, and preferred agonists. The 3 ionotropic receptors are colocalized on the postsynaptic neuron and are part of a complex interaction for NMDA-mediated neurotransmission. When AMPA and kainate are in a resting state, there is a low level of what is called constitutive activity, implying that a certain amount of sodium and potassium can move through their channels. However, when AMPA and kainate are bound by their primary agonist, glutamate, their channels open and the postsynaptic membrane is depolarized. The primary function of AMPA and kainate is to cause depolarization by allowing the influx of sodium and the efflux of potassium.


At rest, the NMDA receptor does not have constitutive activity, as a magnesium cap blocks the central pore of the receptor. The magnesium also inhibits binding of glutamate and its co-agonist, glycine. However, once AMPA and kainate receptors are depolarized and open, the central magnesium cap is displaced, allowing complete binding of glutamate and glycine as well as the transmission of both calcium and sodium down the central pore. PCP and ketamine have binding sites in the central channel of the NDMA receptor and, thus, inhibit glutamate neurotransmission.10

AV 2. Neurotransmission of Glutamate via Ionotropic Receptors

The NMDA receptor is somewhat unique among monoamine neurotransmitter receptors in that it requires the co-agonist glycine for the receptor to fully open. Glycine may be produced from a separate glycinergic neuron or from a glial cell. Glycine and glutamate bind at distinct sites on the NMDA receptor. Reuptake of glutamate occurs on the presynaptic membrane of the glutamatergic neuron, whereas for glycine, reuptake can occur either at the glycinergic neuron or at the glial cell. NMDA receptors also utilize another co-agonist, at the glycinergic (but not glutamatergic) site in the form of d-serine (AV 2). d-Serine is one of the few d amino acids that exist in mammalian systems, and it is converted from l-serine by a racemase enzyme.11 At the glycinergic site, d-serine has about 80% of the intrinsic activity of glycine.12 Both d-serine and glycine have low penetration across the blood-brain barrier,13 presenting a challenge to conducting human research with these compounds due to the high dosages that would be required to obtain adequate central nervous system levels. However, the compound d-cycloserine, which has about 50% of the intrinsic activity of glycine at the glycinergic site,12 has much higher blood-brain penetration,14 and can be used at much lower doses.


The glycine sites, with these 2 co-agonists, provide potential targets for pharmacologic manipulation through inhibition of reuptake pumps for glycine or inhibition of the enzyme that metabolizes d-serine. The glutamate hypothesis provides not only possible new targets for pharmacotherapy, but it also serves as a reminder that, although dopamine may be intimately involved in schizophrenia, other receptor systems may be implicated in both the pathogenesis and possible treatment of this illness. An understanding of actions at the receptor level is helpful in developing new conceptualizations of psychosis and schizophrenia, but gaps still exist in how receptor level dysfunction might translate to more global issues with the circuits that they impact.

Clinical Points

  • Dopamine dysfunction does not account for all the different types of symptoms in schizophrenia
  • The glutamate hypothesis may provide keys to understanding and treating cognitive and negative symptoms in schizophrenia
  • Understanding the neurobiology of glutamate is helpful in developing a new understanding of schizophrenia and its possible etiology and potential treatments


AMPA = amino-3-hydroxy-5-methyl-4-isoxazole propionic acid
D2 = dopamine D2 receptor
GPCR = G-protein coupled receptor
NMDA = N-methyl-d-aspartate
PCP = phencyclidine


  1. 1. Abi-Dargham A, Rodenhiser J, Printz D, et al. Increased baseline occupancy of D2 receptors by dopamine in schizophrenia. Proc Natl Acad USA. 2000;97(14):8104–8109. PubMed
  2. 2. Kantrowitz JT, Javitt DC. N-methyl-D-aspartate (NMDA) receptor dysfunction or dysregulation: the final common pathway on the road to schizophrenia. Brain Res Bull. 2010;83(3–4):108–121. PubMed
  3. 3. Krystal JH, Karper LP, Seibyl JP, et al. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans: psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch Gen Psychiatry. 1994;51(3):199–214. PubMed
  4. 4. Lahti AC, Koffel B, LaPorte D, et al. Subanesthetic doses of ketamine stimulate psychosis in schizophrenia. Neuropsychopharmacology. 1995;13(1):9–19. PubMed
  5. 5. Jentsch JD, Redmond DE Jr, Elsworth JD, et al. Enduring cognitive deficits and cortical dopamine dysfunction in monkeys after long-term administration of phencyclidine. Science. 1997;277(5328):953–955. PubMed
  6. 6. Mohn AR, Gainetdinov RR, Caron MG, et al. Mice with reduced NMDA receptor expression display behaviors related to schizophrenia. Cell. 1999;98(4):427–436. PubMed
  7. 7. Curtis DR, Johnston GAR. Amino acid transmitters in the mammalian central nervous system. In: Adrian RH, Helmreich E, Holzer H, et al, eds. Reviews of Physiology: Biochemistry and Experimental Pharmacology, vol 69. New York: Springer-Verlag; 1974:97–188.
  8. 8. Stahl MS. Stahl’s Essential Psychopharmacology: Neuroscientific Basis and Practical Applications. 3rd ed. New York, NY: Cambridge University Press; 2008.
  9. 9. Tsapakis EM, Travis MJ. Glutamate and psychiatric disorders. Adv Psychiatr Treat. 2002;8(3):189–197. Full Text
  10. 10. Hashimoto K. Glycine transport inhibitors for the treatment of schizophrenia. Open Med Chem J. 2010;4:10–19. PubMed
  11. 11. Stahl SM. Novel therapeutics for schizophrenia: targeting glycine modulation of NMDA glutamate receptors. CNS Spectr. 2007;12(6):423–427. PubMed
  12. 12. Monahan JB, Handelmann GE, Hood WF, et al. D-cycloserine, a positive modulator of the N-methyl-D-aspartate receptor, enhances performance of learning tasks in rats. Pharmacol Biochem Behav. 1989;34(3):649-653. PubMed
  13. 13. Oldendorf WH. Brain uptake of radiolabeled amino acids, amines, and hexoses after arterial injection. Am J Physiol. 1971;222(6):1629–1639. PubMed
  14. 14. Holdiness MR. Clinical pharmacokinetics of the antituberculosis drugs. Clin Pharmacokinet. 1984;9(6):511–544. PubMed