Cognitive Dysfunction in Schizophrenia Convergence of ␥-Aminobutyric Acid and Glutamate Alterations

NEUROLOGICAL REVIEW Cognitive Dysfunction in Schizophrenia Convergence of ␥-Aminobutyric Acid and Glutamate Alterations

David A. Lewis, MD; Bita Moghaddam, PhD

mpairments in certain cognitive functions mediated by the dorsolateral prefrontal cortex, such as working memory, are core features of schizophrenia. Convergent findings suggest that these disturbances are associated with alterations in markers of inhibitory ␥-aminobutyric acid and excitatory glutamate neurotransmission in the dorsolateral prefrontal cor- Itex. Specifically, reduced ␥-aminobutyric acid synthesis is present in the subpopulation of ␥-aminobutyric acid neurons that express the calcium-binding protein parvalbumin. Despite presynaptic and postsynaptic compensatory responses, the resulting impaired inhibitory regulation of pyramidal neurons contributes to a reduction in the synchronized neuronal activity that is required for working memory function. Several lines of evidence suggest that these changes may be either secondary to or exacerbated by impaired signaling via the N-methyl-D-aspartate class of glutamate receptors. These findings suggest specific targets for therapeutic interventions to improve cognitive function in individuals with schizophrenia. Arch Neurol. 2006;63:1372-1376

Schizophrenia typically has its clinical on- sist throughout the illness, (2) exist in set during late adolescence or early adult- milder forms in the unaffected relatives of hood and is frequently associated with a individuals with schizophrenia, and (3) are lifetime of disability.1 A number of puta- the best predictor of long-term functional tive susceptibility genes for schizophre- outcome.5 Consequently, the develop- nia have been identified recently,2 al- ment of therapeutic interventions for these though the risk of illness associated with cognitive deficits is a major focus of re- any particular genetic variant seems small. search in schizophrenia. In addition, a range of adverse environ- Certain cognitive deficits in schizo- mental events, occurring from concep- phrenia reflect alterations in processes, tion through adolescence, seems to in- such as working memory, that are medi- crease the likelihood of developing ated by the circuitry of the dorsolateral pre- schizophrenia later in life.3 Thus, schizo- frontal cortex (DLPFC). Many individu- phrenia is thought to arise from altered als with schizophrenia perform poorly on neurodevelopmental trajectories because working memory tasks and exhibit al- of the interaction of genetic susceptibil- tered activation of the DLPFC when at- ity and environmental risk factors. tempting to perform such tasks.6 In con- Although psychosis is usually the most trast, these abnormalities are not present striking clinical aspect of schizophrenia, dis- in individuals with other psychotic disor- turbances in certain cognitive processes, ders.7 The altered activation of the DLPFC such as attention, certain types of memory, during working memory tasks predicts the and executive functions, are considered to severity of cognitive disorganization in be the core features of the illness.4 Such cog- subjects with schizophrenia,8 and re- nitive disturbances (1) can be present for duced working memory capacity may be years before the onset of psychosis and per- a key factor in the performance of other cognitive tasks in those with schizophre- Author Affiliations: Departments of Psychiatry and Neuroscience, University of nia.9 Because working memory deficits Pittsburgh, Pittsburgh, Pa. seem to be a central feature of schizophre-

(REPRINTED) ARCH NEUROL / VOL 63, OCT 2006 WWW.ARCHNEUROL.COM 1372

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 nia, determining the nature of the underlying distur- of subjects with schizophrenia express decreased levels bances in DLPFC circuitry is essential for the identifica- of PV mRNA and undetectable levels of GAD67 and GAT1 tion of new drug targets. mRNAs, with the latter resulting in reduced GAT1 pro- Although other neurotransmitter systems are cer- tein in the axon cartridges of these neurons. In addi- tainly involved, findings from a number of investigations tion, the density of pyramidal neuron axon initial seg- ␥ ␣ suggest that disturbances in inhibitory -aminobutyric acid ments immunoreactive for the GABA type A (GABAA) 2 (GABA)–mediated10 or excitatory glutamate-medi- subunit is markedly increased in schizophrenia,19 appar- 11,12 ␣ ated neurotransmission may contribute to the cogni- ently reflecting higher levels of 2 subunits in the axon tive impairments of schizophrenia. In this article, we briefly initial segment. These changes are not found in subjects review this evidence, consider the potential relationships with other psychiatric disorders or in monkeys exposed that may exist between the GABA and glutamate abnor- long term to antipsychotic medications, suggesting that malities, and discuss the implications of these findings for they are specific to the disease process of schizophre- pharmacological interventions to improve cognitive per- nia.15-17,19 Thus, in the DLPFC of subjects with schizo- formance in individuals with schizophrenia. phrenia, GABAA receptors are up-regulated at pyramidal neuron axon initial segments in response to deficient ALTERED GABA NEUROTRANSMISSION GABA release from chandelier neuron axon terminals.10 AND COGNITIVE DISTURBANCES Understanding the contribution of these abnormali- IN SCHIZOPHRENIA ties to the cognitive deficits in schizophrenia depends on the demonstration of a pathophysiological process by Working memory depends on the coordinated and sus- which reduced chandelier cell inputs to pyramidal neu- tained firing of subsets of DLPFC pyramidal neurons be- rons could give rise to working memory impairments. tween the temporary presentation of the stimulus cue and Networks of PVϩ fast-spiking GABA neurons, formed the later initiation of the behavioral response, and fast- by chemical and electrical synapses, give rise to oscilla- spiking GABA neurons in the DLPFC seem essential for tory activity in the ␥-band range, the synchronized fir- such synchronization of pyramidal neuron activity.13 These ing of a population of neurons at 30 to 80 Hz.20 Inter- findings suggest that impairments in GABA-mediated in- estingly, ␥-band oscillations in the DLPFC increase in hibition in the DLPFC could contribute to the impair- proportion to working memory load,21 and in subjects ments in working memory present in schizophrenia. Con- with schizophrenia, DLPFC ␥-band oscillations are re- sistent with this interpretation, reduced expression of the duced during the delay period of a working memory task.22 messenger RNA (mRNA) for the 67-kDa isoform of glu- Thus, a deficit in the synchronization of pyramidal cells, ϩ tamic acid decarboxylase (GAD67), an enzyme that syn- resulting from impaired inhibition by PV GABA neu- thesizes GABA, is one of the most consistent findings in rons, might contribute to reduced ␥-band oscillations and, postmortem studies10 of individuals with schizophrenia. consequently, to working memory dysfunction in sub- 10 The mRNA expression of GAD67 is undetectable in a subjects with schizophrenia. population (about 25%-30%) of DLPFC GABA neurons,14,15 whereas most GABA neurons have normal lev- GLUTAMATE NEUROTRANSMISSION 15 els of GAD67 mRNA. Furthermore, in the same AND COGNITIVE DISTURBANCES individuals, the mRNA expression for the GABA mem- IN SCHIZOPHRENIA brane transporter (GAT1), a protein responsible for the reuptake of released GABA into nerve terminals, is simi- Glutamate mediates fast excitatory postsynaptic poten- larly decreased in a subpopulation of GABA neurons.16 tials by acting on the ionotropic receptors: ␣-amino-3- Thus, the synthesis and reuptake of GABA are reduced in hydroxy-5-methyl-4-isoxazole propionic acid (AMPA), a subset of DLPFC inhibitory neurons in schizophrenia. N-methyl-D-aspartate (NMDA), and kainate. Glutamate The affected GABA neurons include those that con- also exerts modulatory effects by acting on different sub- tain the calcium-binding protein parvalbumin (PV), which types of G protein–coupled metabotropic glutamate is present in approximately 25% of GABA neurons in the (mGlu) receptors. For example, the group 5 mGlu re- primate DLPFC, whereas the approximately 50% of GABA ceptors (mGlu5) potentiate the duration of NMDA re- neurons that express the calcium-binding protein cal- ceptor–dependent excitatory postsynaptic potentials, retinin (CR) are unaffected.17 The PV-positive (PVϩ) neu- whereas mGlu2 and mGlu3 receptors modulate the re- rons are also distinguishable from other cortical GABA lease of glutamate.12 The glutamate synapse is also in- neurons by their fast-spiking, nonadapting, firing pat- fluenced by efficient excitatory amino acid transport pro- tern, and subsets of these neurons can be identified by teins on glial cells and by a host of molecules that influence their morphological features.10 For example, the axons glutamate receptor trafficking and the intracellular sig- of the chandelier subclass of PVϩ GABA neurons give naling machinery associated with postsynaptic density. rise to linear arrays of terminals (termed cartridges) that The following lines of evidence implicate these glu- synapse exclusively on the axon initial segments of py- tamate receptors and glutamate receptor–associated ramidal neurons. In the DLPFC of subjects with schizo- molecules in the pathophysiological features of schizophrenia, the density of chandelier neuron axon car- phrenia (reviews have been performed by several re- tridges immunoreactive for GAT1 is significantly reduced, searchers11,23,24). First, postmortem studies25 show sig- whereas immunoreactivity for GAT1 in other popula- nificant, albeit modest, changes in glutamate receptor tions of axon terminals is unchanged.18 In concert, these binding, transcription, and subunit protein expression findings suggest that chandelier neurons in the DLPFC in the PFC, thalamus, and hippocampus of subjects with

(REPRINTED) ARCH NEUROL / VOL 63, OCT 2006 WWW.ARCHNEUROL.COM 1373

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 schizophrenia. These include decreases in the NR1 sub- neurons (Figure). First, PVϩ cells receive a larger units of the NMDA receptor in the hippocampus and complement of excitatory inputs. In the rodent hippo- DLPFC, decreased AMPA receptor expression in the hip- campus, the total number of excitatory synapses onto PVϩ pocampus, high expression of excitatory amino acid trans- neurons is nearly an order of magnitude greater than the port in the thalamus, decreased kainate receptor bind- number onto CRϩ neurons.30 Consistent with these ob- ing in the DLPFC, and robust changes in the NMDA and servations, the density of excitatory synapses on PVϩ den- AMPA receptor–affiliated intracellular proteins, such as drites in monkey DLPFC is significantly greater than on postsynaptic density protein 95 and synapse-associated CRϩ dendrites, although the magnitude of the differ- protein 102, in the DLPFC and thalamus. ence is less striking than in rodent hippocampus.31 Second, levels of the amino acids N-acetylaspartate and Second, subpopulations of GABA neurons differ in the N-acetylaspartylglutamate, and the activity of the en- complement of glutamate receptor subunits that they ex- zyme N-acetyl-␣-linked acidic peptidase, which cleaves press. In the human temporal cortex, approximately 90% N-acetylaspartate to N-acetylaspartylglutamate and glu- of PVϩ cells, but only 20% of CRϩ cells, are immunore- tamate, are altered in the cerebrospinal fluid and brain active for the glutamate receptor 1 subunit, and this ratio specimens of subjects with schizophrenia. N-acetyl- is nearly completely reversed for the glutamate receptor aspartylglutamate is an endogenous ligand for the mGlu3 2 or 4 subunit.32 Across regions of monkey neocortex, im- subtype of glutamate receptor. Furthermore, reduced N- munoreactivity for the NR1 subunit was detected in most acetylaspartate levels are thought to reflect decreased glu- (50%-90%) PVϩ neurons, but in less than 10% of CRϩ tamate availability.11 neurons,33 although the difference between these cell types Third, many of the genes recently associated with an in- was less marked in the human temporal cortex.32 creased risk for schizophrenia can influence the function Third, PVϩ GABA neurons are particularly sensitive of modulatory sites on the NMDA receptor or intracellular- to the effects of NMDA receptor antagonists, suggesting receptor interacting proteins that link glutamate recep- a high level of NMDA receptor tonic activation in these tors to signal transduction pathways.23,24 These genes in- neurons. For example, following long-term phencycli- clude neuregulin 1, which can influence the expression of dine exposure, the expression level of PV mRNA per neu- NMDA receptors through activation of ErbB4 receptors; ron was decreased by 25% in rat PFC, but the density of and the type 3 metabotropic glutamate receptor gene, which PV mRNA–positive neurons was unchanged.34 These find- encodes the mGlu3 subtype of mGlu receptors. ings are strikingly similar to the pattern of PV mRNA ex- Fourth, exposure to NMDA receptor antagonists, such pression changes observed in the DLPFC of subjects with as phencyclidine or ketamine, produces “schizophrenia- schizophrenia.17 Taken together, these findings suggest like” symptoms in healthy individuals and profoundly that the alterations in GABA neurotransmission selec- exacerbates preexisting symptoms in patients with schizo- tive for PVϩ neurons might be a downstream conse- phrenia.26 In particular, direct comparison of healthy vol- quence of impaired NMDA receptor–mediated glutama- unteers receiving subanesthetic doses of ketamine and tergic inputs to these neurons. of individuals with schizophrenia reveal similar disrup- tions in working memory and thought disorder.27 TREATMENT IMPLICATIONS Together, these findings strongly suggest that deficient activation of NMDA receptors may be a critical com- The value of understanding these alterations in GABA ponent of the cognitive deficits of schizophrenia. This and glutamate neurotransmission in schizophrenia rests conclusion is consistent with a large body of animal stud- in the extent to which they inform the identification of ies suggesting that NMDA receptors within the PFC are novel targets for pharmacological intervention. Al- critical for sustaining working memory and other cog- though dopamine D2 receptor antagonists effectively sup- nitive functions that are impaired in schizophrenia.28 press the psychotic features of schizophrenia, typical antipsychotic drugs might, in fact, impair cognitive func- WHAT IS THE RELATIONSHIP tions.35 Furthermore, while atypical antipsychotic drugs, BETWEEN ALTERED GABA which target several subtypes of dopamine and seroto- AND GLUTAMATE NEUROTRANSMISSION nin receptors, may improve cognitive function, this effect IN SCHIZOPHRENIA? is relatively small and has not been reproducible across laboratories.36 Thus, research aimed at the improve- Because the activity of cortical GABA neurons is, in part, ment of cognitive symptoms must move beyond mono- regulated by glutamate inputs, the alterations in GABA amine-based treatment strategies. neurotransmission might reflect an abnormal glutama- The following glutamate receptor–related strategies have tergic drive onto these neurons. For example, the defi- been proposed to mitigate deficient glutamate neurotrans- 12,37 cit in GAD67 mRNA expression could represent an activity- mission in schizophrenia. First, a class of drugs called dependent change in response to reduced glutamatergic ampakines reduces the rapid rate of desensitization of activity in projections to the DLPFC from the thala- AMPA receptors and, hence, prolongs AMPA- and NMDA- mus14 or hippocampus.29 However, given the ubiqui- mediated excitatory postsynaptic potentials. Preliminary tous nature of glutamate neurotransmission, a critical trials38,39 of ampakines in patients with schizophrenia have question is why changes in glutamate function might dif- produced mixed results, but larger clinical trials are un- ferentially affect the PVϩ, and not the CR-positive (CRϩ), der way. Second, positive modulation of the NMDA re- subclass of GABA neurons. Interestingly, the glutama- ceptor by stimulating the glycine site on the NMDA chan- tergic drive seems notably stronger to PVϩ than CRϩ nel has been tried,37 including direct activation of this site

(REPRINTED) ARCH NEUROL / VOL 63, OCT 2006 WWW.ARCHNEUROL.COM 1374

Glu Afferents From the Hippocampus, Thalamus, or Cerebral Cortex Glu

Pyramidal Neuron

GABA PV+

Glu Glu mGlu2 or mGlu3 NAAG Glu Glu Glu

Glu ++ Glu Glu Mg Glu Gly Glu NR2 NR1 NRG1 GluR1-GluR4 GluR5-GluR7 Glu mGlu5 Subunit Subunit ErbB4 KA1-KA2 NAAG Gq mGlu2 or PSD mGlu3 + NMDA AMPA Kainate

GAD67 GABA

Figure. Glutamate neurotransmission. The glutamatergic drive seems notably stronger to parvalbumin-positive (PVϩ) than calretinin-positive neurons. AMPA indicates ␣-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid; ErbB4, epidermal growth factor receptor tyrosine kinase 4; GABA, ␥-aminobutyric acid; GAD, glutamic acid decarboxylase; Glu, glutamate; GluR1 through GluR7, Glu receptor 1 subunit through Glu receptor 7 subunit; Gly, glycine; Gq, family G protein; KA1 and KA2, kainate receptor subunits 1 and 2, respectively; mGlu2, mGlu3, and mGlu5, groups 2, 3, and 5 metabotropic glutamate, respectively; Mgϩϩ, magnesium; NAAG, N-acetylaspartylglutamate; NMDA, N-methyl-D-aspartate; NRG1, neuregulin 1 gene; PSD, postsynaptic density protein; and PV−, PV negative. The straight white arrow indicates that the mGlu2 and mGlu3 receptors modulate the release of Glu. The curved arrow indicates that activation of the group 5 mGlu receptor potentiates the duration of signaling through NMDA receptors.

by endogenous or exogenous agonists. These clinical trials receptors are only in the preclinical developmental stages, have also produced mixed results, but the relative lack of this class of modulators holds promise for improving work- specificity and solubility of the drugs used limits the in- ing memory dysfunction associated with NMDA recep- terpretation of these studies. Animal work with more spe- tor deficiency. cific glycine site agonists or glycine transporter blockers If, as previously suggested, reduced GABA neuro- suggests that these drugs may be effective in ameliorating transmission in PVϩ neurons is secondary to altered the adverse effects of NMDA deficiency.36 Third, mGlu2 NMDA receptor function, then abnormal GABA neuro- or mGlu3 agonists reduce the working memory deficits transmission could represent a “final common path- elicited by NMDA antagonists, but the effects of these drugs way” to prefrontal dysfunction in schizophrenia. Thus, on cognitive functions in schizophrenia have not yet been drugs targeted to mitigate the disturbances in inhibition evaluated.11 Fourth, animal studies40 suggest that tonic might be particularly effective in improving cognitive per- stimulation of mGlu5 receptors is critical for working formance in schizophrenia. For example, positive allo- memory functioning. This class of receptors also plays a steric modulators selective for GABAA receptors contain- ␣ ␣ critical role in burst activity of PFC neurons and en- ing 2 subunits (eg, a GABAA 2-selective benzodiazepine), hances the function of NMDA receptors.41 Although li- by enhancing the response of pyramidal neurons to the gands that allosterically enhance the function of mGlu5 release of GABA from PVϩ chandelier cell axons, would

(REPRINTED) ARCH NEUROL / VOL 63, OCT 2006 WWW.ARCHNEUROL.COM 1375

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 be predicted to increase the synchronization of pyrami- ing by disinhibition: GABAA blockade of prefrontal cortical neurons engaged by ␥ working memory. J Neurosci. 2000;20:485-494. dal cell firing at frequencies and, thus, improve work- 14. Akbarian S, Kim JJ, Potkin SG, et al. Gene expression for glutamic acid decar- ing memory function in schizophrenia.10 In contrast, drugs boxylase is reduced without loss of neurons in prefrontal cortex of schizophrenics. ␣ Arch Gen Psychiatry. 1995;52:258-266. that directly activate 2-containing GABAA receptors in- 15. Volk DW, Austin MC, Pierri JN, Sampson AR, Lewis DA. Decreased GAD67 mRNA dependent of the presence of GABA or those that gen- expression in a subset of prefrontal cortical GABA neurons in subjects with erally increase the firing rate of chandelier cells might schizophrenia. Arch Gen Psychiatry. 2000;57:237-245. disrupt the critical timing of inhibition necessary to syn- 16. Volk D, Austin M, Pierri J, Sampson A, Lewis D. GABA transporter-1 mRNA in the prefrontal cortex in schizophrenia: decreased expression in a subset of neurons. chronize pyramidal neuron firing. In addition, available Am J Psychiatry. 2001;158:256-265. benzodiazepines have activity at GABAA receptors con- 17. Hashimoto T, Volk DW, Eggan SM, et al. Gene expression deficits in a subclass ␣ ␣ of GABA neurons in the prefrontal cortex of subjects with schizophrenia. J Neurosci. taining other subunits (eg, 1 or 5) whose activation can 2003;23:6315-6326. impair cognitive function. 18. Woo T-U, Whitehead RE, Melchitzky DS, Lewis DA. A subclass of prefrontal ␥-ami- Inconclusion,advancesinunderstandingthepathophysi- nobutyric acid axon terminals are selectively altered in schizophrenia. Proc Natl Acad Sci U S A. 1998;95:5341-5346. ology of cognitive dysfunction in schizophrenia are provid- 19. Volk DW, Pierri JN, Fritschy JM, Auh S, Sampson AR, Lewis DA. Reciprocal al- ing a rational basis for the development of novel pharma- terations in pre- and postsynaptic inhibitory markers at chandelier cell inputs to cologicalinterventionsthatmayimprovethelong-termfunc- pyramidal neurons in schizophrenia. Cereb Cortex. 2002;12:1063-1070. 20. Whittington MA, Traub RD. Interneuron diversity series: inhibitory interneurons tional outcome of individuals with schizophrenia. and network oscillations in vitro. Trends Neurosci. 2003;26:676-682. 21. Howard MW, Rizzuto DS, Caplan JB, et al. Gamma oscillations correlate with working memory load in humans. Cereb Cortex. 2003;13:1369-1374. Accepted for Publication: October 14, 2005. 22. Cho RY, Konecky RO, Carter CS. Impaired task-set maintenance and frontal cor- Correspondence: David A. Lewis, MD, Department of Psy- tical ␥-band synchrony in schizophrenia. Paper presented at: Cognitive Neuro- science Society Annual Meeting; April 20, 2004; San Francisco, Calif. chiatry, University of Pittsburgh, 3811 O’Hara St, W1651 23. Harrison PJ, Weinberger DR. Schizophrenia genes, gene expression, and neuro- Biomedical Science Tower, Pittsburgh, PA 15213 (lewisda pathology: on the matter of their convergence. Mol Psychiatry. 2005;10:40-68. @upmc.edu). 24. Moghaddam B. Bringing order to the glutamate chaos in schizophrenia. Neuron. 2003;40:881-884. Author Contributions: Study concept and design: Lewis 25. Konradi C, Heckers S. Molecular aspects of glutamate dysregulation: implica- and Moghaddam. Acquisition of data: Lewis and Mog- tions for schizophrenia and its treatment. Pharmacol Ther. 2003;97:153-179. haddam. Analysis and interpretation of data: Lewis and Mo- 26. Krystal JH, Karper LP, Seibyl JP, et al. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans: psychotomimetic, perceptual, cognitive, ghaddam. Drafting of the manuscript: Lewis and Mo- and neuroendocrine responses. Arch Gen Psychiatry. 1994;51:199-214. ghaddam. Critical revision of the manuscript for important 27. Adler CM, Malhotra AK, Elman I, et al. Comparison of ketamine-induced thought disorder in healthy volunteers and thought disorder in schizophrenia. Am J intellectual content: Lewis and Moghaddam. Obtained fund- Psychiatry. 1999;156:1646-1649. ing: Lewis and Moghaddam. Administrative, technical, and 28. Verma A, Moghaddam B. NMDA receptor antagonists impair prefrontal cortex material support: Lewis and Moghaddam. function as assessed via spatial delayed alternation performance in rats: modulation by dopamine. J Neurosci. 1996;16:373-379. Funding/Support: This study was supported by grants 29. Lipska BK, Lerman DN, Khaing ZZ, Weickert CS, Weinberger DR. Gene expres- MH045156 (Dr Lewis), MH051234 (Dr Lewis), MH043784 sion in dopamine and GABA systems in an animal model of schizophrenia: ef- (Dr Lewis), MH065026 (Dr Moghaddam), and MH048404 fects of antipsychotic drugs. Eur J Neurosci. 2003;18:391-402. 30. Gulya´s AI, Megı´as M, Emri Z, Freund TF. Total number and ratio of excitatory (Dr Moghaddam) from the National Institutes of Health; and inhibitory synapses converging onto single interneurons of different types and by the National Alliance for Research on Schizophre- in the CA1 area of the rat hippocampus. J Neurosci. 1999;19:10 082-10 097. nia and Depression (Dr Moghaddam). 31. Melchitzky DS, Lewis DA. Pyramidal neuron local axon terminals in monkey prefrontal cortex: differential targeting of subclasses of GABA neurons. Cereb Cortex. 2003;13:452-460. REFERENCES 32. Gonza´lez-Albo MC, Elston GN, DeFelipe J. The human temporal cortex: charac- terization of neurons expressing nitric oxide synthase, neuropeptides and calcium- binding proteins, and their glutamate receptor subunit profiles. Cereb Cortex. 1. Lewis DA, Lieberman JA. Catching up on schizophrenia: natural history and 2001;11:1170-1181. neurobiology. Neuron. 2000;28:325-334. 33. Huntley GW, Vickers JC, Morrison JH. Quantitative localization of NMDAR1 re- 2. Owen MJ, Williams NM, O’Donovan MC. The molecular genetics of schizophre- ceptor subunit immunoreactivity in inferotemporal and prefrontal association cor- nia: new findings promise new insights. Mol Psychiatry. 2004;9:14-27. tices of monkey and human. Brain Res. 1997;749:245-262. 3. Lewis DA, Levitt P. Schizophrenia as a disorder of neurodevelopment. Annu Rev 34. Cochran SM, Kennedy M, McKerchar CE, Steward LJ, Pratt JA, Morris BJ. In- Neurosci. 2002;25:409-432. duction of metabolic hypofunction and neurochemical deficits after chronic in- 4. Elveva˚g B, Goldberg TE. Cognitive impairment in schizophrenia is the core of the termittent exposure to phencyclidine: differential modulation by antipsychotic disorder. Crit Rev Neurobiol. 2000;14:1-21. drugs. Neuropsychopharmacology. 2003;28:265-275. 5. Gold JM. Cognitive deficits as treatment targets in schizophrenia. Schizophr Res. 35. Epstein JI, Keefe RS, Roitman SL, Harvey PD, Mohs RC. Impact of neuroleptic 2004;72:21-28. medications on continuous performance test measures in schizophrenia. Biol 6. Callicott JH, Mattay VS, Verchinski BA, Marenco S, Egan MF, Weinberger DR. Psychiatry. 1996;39:902-905. Complexity of prefrontal cortical dysfunction in schizophrenia: more than up or 36. Keefe RS, Silva SG, Perkins DO, Lieberman JA. The effects of atypical antipsy- down. Am J Psychiatry. 2003;160:2209-2215. chotic drugs on neurocognitive impairment in schizophrenia: a review and 7. MacDonald AW III, Carter CS, Kerns JG, et al. Specificity of prefrontal dysfunc- meta-analysis. Schizophr Bull. 1999;25:201-222. tion and context processing deficits to schizophrenia in never-medicated pa- 37. Coyle JT, Tsai G. The NMDA receptor glycine modulatory site: a therapeutic tar- tients with first-episode psychosis. Am J Psychiatry. 2005;162:475-484. get for improving cognition and reducing negative symptoms in schizophrenia. 8. Perlstein WM, Carter CS, Noll DC, Cohen JD. Relation of prefrontal cortex dys- Psychopharmacology (Berl). 2004;174:32-38. function to working memory and symptoms in schizophrenia. Am J Psychiatry. 38. Goff DC, Leahy L, Berman I, et al. A placebo-controlled pilot study of the ampa- 2001;158:1105-1113. kine CX516 added to clozapine in schizophrenia. J Clin Psychopharmacol. 2001; 9. Silver H, Feldman P, Bilker W, Gur RC. Working memory deficit as a core neu- 21:484-487. ropsychological dysfunction in schizophrenia. Am J Psychiatry. 2003;160: 39. Marenco S, Egan MF, Goldberg TE, et al. Preliminary experience with an ampa- 1809-1816. kine (CX516) as a single agent for the treatment of schizophrenia: a case series. 10. Lewis DA, Hashimoto T, Volk DW. Cortical inhibitory neurons and schizophrenia. Schizophr Res. 2002;57:221-226. Nat Rev Neurosci. 2005;6:312-324. 40. Homayoun H, Stefani MR, Adams BW, Tamagan GD, Moghaddam B. Functional 11. Coyle JT. The GABA-glutamate connection in schizophrenia: which is the proxi- interaction between NMDA and mGlu5 receptors: effects on working memory, mate cause? Biochem Pharmacol. 2004;68:1507-1514. instrumental learning, motor behaviors, and dopamine release. Neuropsycho- 12. Moghaddam B. Targeting metabotropic glutamate receptors for treatment of the pharmacology. 2004;29:1259-1269. cognitive symptoms of schizophrenia. Psychopharmacology (Berl). 2004;174: 41. Homayoun H, Moghaddam B. Bursting of prefrontal cortex neurons in awake rats 39-44. is regulated by metabotropic glutamate 5 (mGlu5) receptors: rate-dependent in- 13. Rao SG, Williams GV, Goldman-Rakic PS. Destruction and creation of spatial tun- fluence and interaction with NMDA receptors. Cereb Cortex. 2006;16:93-105.

(REPRINTED) ARCH NEUROL / VOL 63, OCT 2006 WWW.ARCHNEUROL.COM 1376