<<

Molecular Psychiatry (1999) 4, 418–428  1999 Stockton Press All rights reserved 1359–4184/99 $15.00

MECHANISMS OF ACTION Mechanisms of typical and atypical drug action in relation to and NMDA receptor hypofunction hypotheses of GE Duncan1, S Zorn2 and JA Lieberman1

1Department of Psychiatry and UNC Neuroscience Center, School of Medicine, University of North Carolina, Chapel Hill, NC 27599; 2NS Discovery, Pfizer Inc, Central Research Division, Pfizer Inc, Groton, CT 06340, USA

Available evidence indicates that is the most effective antipsychotic currently used for the pharmacotherapy of schizophrenia. Unfortunately, clozapine can cause serious side effects that limit the use of the drug. The therapeutic mechanism of action of clozapine is poorly understood, and accordingly, it has been difficult to design new with the advan- tageous therapeutic properties of clozapine. Based on hypotheses that and ser- otonergic receptor-blocking properties of clozapine account for its clinical efficacy, several novel antipsychotic drugs have been introduced recently. There is currently insufficient data to reach definitive conclusions regarding the efficacy of the newer ‘atypical’ in comparison to clozapine. However, most published studies, and general clinical impressions, suggest that none of the newer drugs are as effective as clozapine in treating patients resist- ant to drug . The present paper briefly reviews the clinical experi- ence with the newer ‘atypical’ antipsychotic drugs and then discusses clinical and preclinical data potentially relevant to mechanisms of action of clozapine in relation to the NMDA receptor hypofunction hypothesis of schizophrenia. Keywords: clozapine; ; ; ; ; antipsychotic; schizophrenia; dopamine; NMDA

The introduction of clozapine for the pharmacotherapy Table 1) but it is unclear whether actions at specific of schizophrenia represented a significant advance in , dopamine, or receptors, alone or the treatment of this devastating mental illness. Cloza- in combination, account for its clinical efficacy. pine is superior to typical neuroleptic drugs (eg A number of specific hypotheses concerning mech- ) in treating positive and negative symp- anisms of action of clozapine have directed drug dis- toms, and is effective in many patients who are refrac- covery efforts and led to clinical trials of drugs with tory to typical antipsychotics.1–3 In addition, clozapine widely different receptor-binding characteristics. The does not induce the extrapyramidal side effects (EPS) clinical experience with drugs whose development commonly caused by the typical agents. Because of was inspired by clozapine will be briefly reviewed and these properties, clozapine was termed atypical and then actions of clozapine and other antipsychotic drugs represents the of this class. Although will be addressed in the context of the NMDA hypo- clozapine is the most efficacious antipsychotic cur- function hypothesis of schizophrenia. rently available, serious side effects induced by the drug, including , impose substantial Clinical experience with clozapine-inspired limitations on its use. Concerted research and develop- putative antipsychotics in relation to therapeutic ment efforts have been made to produce an antipsy- mechanisms of action chotic drug with the therapeutic advantages of cloza- pine, without the properties contributing to its serious One of the earliest hypotheses of clozapine’s mech- side effects. However, the specific pharmacological anism of action related to the D1 dopamine antagon- characteristics of clozapine that confer its therapeutic istic properties of the drug.4 Clozapine binds to D1 properties are poorly understood. Clozapine binds to a receptors in vitro with modest affinity (see Table 1) and diverse number of neurotransmitter receptors (see at therapeutic doses occupies approximately 40–50% of D1 receptors in humans.5 Based on the hypothesis that the D1 antagonistic properties of clozapine dis- tinguished it from the typical antipsychotic drugs, Correspondence: GE Duncan, PhD, Neuroscience Center, CB # 7250, University of North Carolina School of Medicine, Chapel selective D1 antagonists were developed as potential Hill, NC 27599–7250, USA. E-mail: gduncanȰcss.unc.edu antipsychotic agents. Unfortunately, such drugs were Received 3 March 1999; revised and accepted 10 May 1999 not effective and there were indications that selective Antipsychotic drug action in schizophrenia GE Duncan et al 419 Table 1 Affinity of antipsychotic drugs for human neurotransmitter receptors

Receptor Clozapine Risperidone Olanzapine Ziprasidone Quetiapine Haloperidol M-100907

(Ki, nM)

D1 290 580 52 130 1300 120 1800 D2 130 2.2 20 3.1 180 1.4 1800 D3 240 9.6 50 7.2 940 2.5 D4 47 8.5 50 32 2200 3.3 5-HT1A 140 210 2100 2.5 230 3600 Ͼ4700 5-HT1Da 1700 170 530 2.0 Ͼ5100 Ͼ5000 Ͼ5100 5-HT2A 8.9 0.29 3.3 0.39 220 120 0.3 5-HT2C 17 10 10 0.72 1400 4700 68 5-HT6 11 2000 10 76 4100 6000 7400 5-HT7 66 3.0 250 9.3 1800 1100 2000 ␣1 4.0 1.4 54 13 15 4.7 68 ␣2 33 5.1 170 310 1000 1200 2200 H1 1.8 19 2.8 47 8.7 440 83 m1 1.8 2800 4.7 5100 100 1600

Values are geometric means of at least three determinations. aBovine.

D1 antagonists may have even exacerbated symptoms it is difficult to infer equal efficacy of the drugs from of schizophrenia and induced extrapyramidal side the study of Bondolfi et al.14 Further investigation is effects.6,7 Although selective D1 antagonists were not necessary to adequately compare the relative efficacy effective, it is possible that D1 antagonistic properties of risperidone and clozapine in treatment-resistant of clozapine, in combination with other actions of the patients. drug, contribute to, but are not sufficient for, its thera- The reduced EPS side effects associated with low peutic effects. dose risperidone treatment (4–6 mg day−1), even at high Clozapine has relatively high affinity for 5HT2A levels of D2 receptor occupancy, may be due to the receptors and is a potent antagonist of 5HT2A receptor- 5HT2A antagonistic properties of the drug.10,15 How- mediated responses in vivo.8,9 The hypothesis has been ever, at higher doses, risperidone produces EPS, indi- advanced that the combination of relatively high affin- cating that 5HT2A receptor antagonism alone cannot ity for 5HT2A receptors, and lower affinity for D2 completely eliminate EPS associated with high D2 receptors, account for the atypical therapeutic actions receptor blockade. The potential role of 5HT2A recep- of clozapine and other atypical antipsychotics.10 The tor antagonism in therapeutic responses to atypical balanced D2/5HT2 hypothesis was the driving force antipsychotic drugs may become more apparent when behind development of risperidone, which is a potent data from clinical trials are available for the selective D2 and 5HT2 antagonist, but with greater relative 5HT2A antagonists M100907. affinity for the 5HT2A receptor (Table 1). Consistent Risperidone, like clozapine, has relatively high with the receptor-binding characteristics of risperi- affinity for alpha-1 and alpha-2 adrenergic receptors. done, high occupancy of D2 and 5HT2 receptors is The potential therapeutic significance of the adrenergic observed clinically at therapeutic doses.11 receptor-blocking properties of clozapine and risperi- Although risperidone is an effective antipsychotic done are uncertain. Addition of the alpha-2 antagonist agent, and has a favorable side effect profile in com- to the regime of patients treated with the typi- parison to haloperidol, the general clinical impression cal neuroleptic fluphenazine improved treatment has been that risperidone is not as effective as cloza- responses in patients refractory to treatment with flu- pine in schizophrenic patients resistant to treatment phenazine alone.16 However, there has been no con- with typical antipsychotics. However, there is insuf- firmation of the effects of alpha-2 antagonists as ficient clinical data available to unequivocally estab- adjuncts to typical neuroleptic treatment and it has lish the superior efficacy of clozapine relative to risper- been suggested that alpha-2 agonists may actually be idone (for review see 3). Flynn et al12 and Breier et al13 useful for treating cognitive deficits of the disease.17 found that clozapine was more effective than risperi- Based on findings of elevated D3 dopamine receptors done in treatment-resistant patients. However, Bon- in drug-free schizophrenics, and a reduction in these dolfi et al found no difference between risperidone and receptors in patients treated with antipsychotics, Gure- clozapine in treatment-resistant patients.14 In this latter vich et al18 have suggested that normalization of D3 study, certain methodological issues may have led to receptors may be contribute to the efficacy of antipsy- an overestimation of the efficacy of both clozapine and chotic drugs. While risperidone and haloperidol are risperidone in truly resistant patients.3 Consequently, relatively potent D3 antagonists (Table 1), the affinity Antipsychotic drug action in schizophrenia GE Duncan et al 420 of clozapine for the D3 receptor is even lower than that doses of risperidone and olanzapine produce high for the D2 receptor. Thus antagonism of D3 receptors occupancy of D2 receptors suggest that D2 receptor by haloperidol and the newer ‘atypical’ drugs could antagonism could be a predominant mechanism of contribute to their therapeutic actions, but it is unlikely action of these new ‘atypical’ drugs.11,33,36 Although that the antipsychotic actions of clozapine can be clozapine does not exhibit high levels of D2 receptor explained on the basis of D3 receptor blockade. occupancy as assessed by standard approaches, See- D4 antagonism has also been sug- man and Tallerico37 have suggested that the drug can gested to contribute to the therapeutic action of cloza- indeed occupy high levels of the D2 receptors when pine.19,20 Clozapine has a relatively high affinity for the competing with endogenous dopamine, and not the D4 receptor compared to the D2 receptor (Table 1), radioligands used to assess in vivo receptor occupancy. which is enriched in the hippocampus and prefrontal Seeman and Tallerico37 hypothesize that the low EPS cortex. Furthermore, a number of clinically efficacious liability for clozapine is related to its relatively low antipsychotic drugs have high affinity for this receptor affinity for the D2 receptor and the antipsychotic effi- site (Table 1). An initial clinical study with the selec- cacy is due to effective antagonism of endogenous tive D4 antagonist L-745870 did not demonstrate effi- dopamine. Although this theoretical position could cacy in the treatment of schizophrenia.21,22 However, explain the lack of EPS in response to clozapine treat- since only one small dose was tested, it is difficult to ment, it does not explain the greater efficacy of cloza- draw firm conclusions regarding the potential utility of pine in relation to other currently used antipsychotic 23 D4 antagonists as antipsychotic agents. Several other drugs. If a predominant action of clozapine is not D2 selective D4 antagonists have shown promise in pre- antagonism, this would provide an explanation for the 24–26 clinical pharmacological screens and are currently superior efficacy of clozapine in the patients resistant being tested in patients with schizophrenia. Results of to typical neuroleptics (as well as to olanzapine and the clinical studies could allow a more complete risperidone), for which D2 receptor antagonism may be assessment of the efficacy of D4 antagonists in the treat- a predominant therapeutic mechanism of action. ment of schizophrenia. The high level of 5HT2A receptor occupancy pro- Olanzapine is a drug closely related in chemical duced by ‘atypical’ antipsychotic drugs such as risperi- structure to clozapine (Figure 1), and the two drugs done and olanzapine suggests that antagonism of this have many common receptor-binding characteristics receptor may be involved in their clinical (Table 1). Olanzapine was selected for development in actions.5,11,33,36 Preclinical studies show that 5HT2A large part because of its relatively potent antagonistic receptor antagonists attenuate cataleptic effects of D2 effects at both D2 and 5HT2A receptors.27,28 Like cloza- receptor antagonists.38–41 These data suggest that the pine, olanzapine is more potent at 5HT2 than D2 recep- 5HT2A receptor antagonism may contribute to the low tors. In addition, the pharmacologic profile of olanza- EPS liability of risperidone and olanzapine, and also pine in regard to other dopaminergic, , the recently introduced drugs quetiapine and ziprasi- , and subtypes is simi- done. lar to clozapine, but there are also some notable differ- Clozapine, risperidone and olanzapine occupy ences (see Table 1). For example, clozapine has sub- Ͼ80% of 5HT2A receptors in the therapeutic dose stantially higher affinity for 5HT1A and 5HT7 5,11,33,36,42 receptors in comparison to olanzapine. Extensive con- range in humans. Although 5HT2A receptor trolled studies have proven olanzapine to be an effec- antagonism is likely to be involved in the low EPS liab- tive antipsychotic that has very low EPS liability.29–32 ility of risperidone and olanzapine, the role of this mol- EPS side effects are minimal even at doses that produce ecular action in therapeutic responses to clozapine is D2 receptor occupancy of 70–80%.33,34 Although uncertain. If a high level of 5HT2A receptor antagon- olanzapine is highly effective in many patients, clinical ism was responsible for the efficacy of clozapine, then experience with currently suggested doses indicates risperidone and olanzapine would be expected to be as that it is not as effective as clozapine in patients refrac- effective as clozapine in treatment-resistant patients, 3,12,13,35 tory to effects of typical antipsychotics.35 There is but to date this has not been demonstrated. insufficient clinical data for ziprasidone, quetiapine, The available basic science and clinical data suggest and M-100907 to compare the efficacy of these newer that antagonistic actions at D2/D3 and 5HT2A recep- atypical antipsychotics with clozapine. tors are insufficient to explain the superior therapeutic efficacy of clozapine in relation to other antipsychotic agents. Whether the combination of clozapine’s actions Potential relationships between D2 and 5HT2 at other serotonergic receptors (ie 5HT1A, 5HT2C, receptor occupancy and therapeutic mechanisms 5HT6, 5HT7) or adrenergic receptors is responsible for of drugs the drug’s efficacy may become apparent when more Although further studies are required to firmly estab- extensive clinical data are available for ziprasidone lish the relative efficacy of the new generation of atypi- and quetiapine, which exhibit widely varying affinities cal antipsychotic drugs in comparison to clozapine, for these receptors. No definite conclusions can be general clinical impressions suggest that clozapine has reached regarding the relative efficacy of the new atypi- greater efficacy than new ‘atypical’ antipsychotic cal antipsychotics, until adequately controlled clinical drugs. The imaging studies showing that therapeutic studies are conducted that compare the new atypical Antipsychotic drug action in schizophrenia GE Duncan et al 421

Figure 1 Structures of clozapine, haloperidol and recently introduced ‘atypical’ antipsychotic drugs. drugs in head to head comparison with each other and The drugs were selected primarily for their antagonist with clozapine. actions at dopamine and serotonin receptors and their ability to antagonize dopamine- and serotonin- mediated effects in pharmacological screens. Since that Potential new directions for antipsychotic drug strategy of drug development has apparently not discovery yielded an antipsychotic that is as efficacious as cloza- The interaction of clozapine with a diverse array of pine, it may be prudent to consider other hypotheses neurotransmitter receptors complicates identification of schizophrenia to guide alternative strategies for the of the molecular characteristics of the drug that confer discovery of novel antipsychotic agents. In this regard, its atypical therapeutic properties. It is possible that the NMDA receptor hypofunction hypothesis of none of the newer ‘atypical’ drugs have the appropriate schizophrenia may provide a novel theoretical frame- combination and relative proportion of generally recog- work for investigating mechanisms of action of antipsy- nized pharmacological actions of clozapine at specific chotic drugs. neurotransmitter receptors (Table 1). Alternatively, it is possible that a currently unrecognized neurochemical The NMDA receptor hypofunction hypothesis of action of clozapine contributes to its therapeutic effi- schizophrenia cacy and none of the newer ‘atypical’ drugs have been modeled on the critical molecular property to create a In early clinical investigations, the PCP and drug with an efficacy profile comparable to clozapine. were observed to induced a state resembling As noted above, the rationale for development of the schizophrenia in 40–50% of individuals given the antipsychotic drugs recently introduced, and currently drugs.43–48 In the 1970s, the abuse of PCP was fre- under development, has been based predominantly on quently associated with psychotic-like responses (for dopamine and serotonin hypotheses of schizophrenia. review see49). In retrospective analyses of admission Antipsychotic drug action in schizophrenia GE Duncan et al 422 records from PCP-induced adverse reactions, there was responses after systemic administration, suggests that little distinction between symptoms presented by these the excitatory effects of ketamine result from disrup- patients and acute psychotic reactions in schizo- tion of inhibitory neural circuits (ie from phrenic patients. The discovery that ketamine and PCP disinhibition). are non-competitive antagonists of the NMDA receptor Consistent with excitatory effects demonstrated in prompted the hypothesis that schizophrenia involves electrophysiological investigations, NMDA antagonists reduced NMDA receptor function.49–52 Competitive have been observed to induce neuroanatomically selec- antagonists of the NMDA receptor also induced psych- tive activation of brain glucose utilization. In response otomimetic effects,53 further supporting the hypothesis to doses of ketamine, increased 2-deoxyglu- that endogenous may result from reduced cose (2-DG) uptake was observed in the hippocampus, NMDA function. and decreased uptake observed in the medial genicu- Recent studies with healthy human volunteers dem- late, inferior colliculus and isocortical regions.69,70 onstrated that ketamine induces a spectrum of Since subanesthetic doses of ketamine induce schizo- responses that resemble positive, negative, and cogni- phrenic-like reactions, we examined subanesthetic tive symptoms of schizophrenia.54–56 Furthermore, in doses of the NMDA antagonist on 2-DG uptake in rats. stabilized schizophrenic patients, ketamine can pre- Robust increases in 2-DG uptake were observed in spe- cipitate psychotic reactions that in some cases re-create cific hippocampal subregions, limbic cortical regions specific psychotic symptoms experienced during active (medial prefrontal, cingulate, retrosplenial cortices), phases of their illness.57–59 These recent studies con- nucleus accumbens, anterior ventral thalamic nucleus, firm the early clinical investigations and provide con- and basolateral nucleus of the amygdala.71,72 In vincing support for the hypothesis that symptoms of addition, laminar patterns of 2-DG uptake in isocortical schizophrenia may be associated with NMDA recep- regions were dramatically altered. In contrast to the tor hypofunction. marked activation of limbic brain regions after subane- The ‘psychotomimetic’ effects induced by NMDA sthetic doses of ketamine, 2-DG uptake was substan- antagonists and psychostimulants (eg ) tially reduced in the inferior colliculus, medial genicu- are distinctly different. For example, unlike ketamine, late, and auditory cortex. The effects of subanesthetic single injections of amphetamine rarely induced psy- doses of ketamine on 2-DG were almost identical to chotic reactions in healthy subjects. After repeated those induced by the more specific antagonist MK- treatments, amphetamine can induced positive schizo- 801,73,74 indicating that the effects observed for keta- phrenic-like reactions in healthy individuals, such as mine were due to the NMDA antagonistic properties of and hostility, but the withdrawal responses the drug. These findings of increased functional and cognitive deficits induced by ketamine are not activity in limbic regions, and decreased activity in observed.60,61 Similarly, in schizophrenic patients, sensory pathways, suggest a dramatic remodeling of precipitate positive symptoms but do not functional circuitry in response to a subanesthetic dose induce negative or cognitive symptoms.62 The ability of ketamine. of ketamine and other NMDA antagonists to induce a Functional imaging studies in humans after subanes- behavioral state that resembles several dimensions of thetic doses of ketamine have also shown neuroana- schizophrenia suggests that paradigms involving tomically specific functional activation. Lahti et al57 assessment of behavioral and neurochemical conse- found that ketamine increased flow in the quences of NMDA antagonists administration may con- anterior and Breier et al75 demon- stitute valid experimental models of some aspects of strated increased 18F-flurodeoxyglucose (FDG) in the schizophrenia. after infusion of subanesthetic doses of ketamine. Vollenweider et al76,77 observed global increases in FDG uptake in response to subanesthetic Neurobiological consequences of NMDA doses of ketamine, but found greater increases in pre- antagonist administration frontal cortex in comparison to other regions. Thus, Ketamine and PCP block NMDA receptors by binding studies in rats and humans indicate that subanesthetic to a site within the calcium channel of the receptor. doses of ketamine can alter patterns of brain activity Although, this molecular action of ketamine and PCP and suggest a functional neuroanatomical basis for psy- is well documented, mechanisms responsible for trans- chotomimetic effects of reduced NMDA receptor func- lation of this cellular and molecular action into the tion. psychotomimetic effects of the drugs are poorly under- stood. The electrophysiological consequences of sys- Potential mechanisms for NMDA antagonist- temic administration of ketamine are complex, with induced functional activation some regions showing inhibition and other regions exhibiting excitation that can progress to -like One potential mechanism by which NMDA receptor activity.47,63,64 By contrast, iontophoretic application of antagonism could result in functional activation would ketamine to neurons uniformly antagonizes NMDA- be to reduce excitation of inhibitory neurons. In sup- evoked excitatory responses.65–68 The well documented port of a disinhibitory action involving GABAergic ability of ketamine to inhibit excitatory actions of transmission, NMDA receptors on hippocampal inhibi- NMDA at the cellular level, and induce excitatory tory interneurons were demonstrated to be more sensi- Antipsychotic drug action in schizophrenia GE Duncan et al 423 tive to an NMDA antagonist, in comparison to NMDA Differential effects of typical and atypical receptors on CA1 pyramidal neurons.78 In addition, antipsychotics on NMDA antagonist-induced , which potentiate GABA-mediated alterations in brain metabolic activity neurotransmission, are effective in reducing adverse We have demonstrated dramatic differences in the emergence reactions following ketamine anesthesia.79 effects of clozapine in comparison to other antipsy- However, ketamine-induced psychosis was not attenu- chotic drugs on ketamine-induced alterations in 2-DG ated by subhypnotic doses of , and cognitive uptake. Clozapine blocked the effects of the NMDA deficits induced by ketamine were worsened by the antagonist on 2-DG uptake in all brain regions, whereas .80 Thus, components of ketamine- haloperidol did not block, and in fact potentiated the induced behavioral effects may be due in part, but not metabolic activation in most brain regions.71 Pretreat- wholly, to disinhibitory effects via reduced activity of ment of rats with risperidone did not block the effects GABA-containing neurons. of ketamine on 2-DG uptake, but unlike haloperidol, In addition to disinhibition of GABAergic neurons, did not potentiate the effects of the NMDA antagon- reduced autoregulation of glutamate release by antag- ist.100 In preliminary studies, olanzapine (5 mg kg−1) onism of presynaptic NMDA receptors is another effectively blocked ketamine-induced increases in 2- potential mechanism for ketamine-induced functional DG uptake in the medial prefrontal cortex, but not in activation. In support of this latter possibility, Liu and the hippocampus (Duncan et al, unpublished Moghaddam81 found that local infusion of NMDA observations). However, at a dose of 10 mg kg−1 olanza- antagonists into the and hippocampus pine blocked the effects of ketamine in all brain increased extracellular glutamate, suggesting a tonic regions. At a dose of 5 mg kg−1 of olanzapine, there inhibitory role of NMDA receptors on glutamate would be a maximally effective blockade of 5HT2 and release. Also, systemic administration of ketamine and D2 receptors.101 Therefore an action of olanzapine PCP increased glutamate release in the prefrontal cor- other than D2 or 5HT2 receptor blockade may be tex.82,83 This excessive release of glutamate could then involved in the drug’s effect on ketamine-induced 2DG activate non-NMDA glutamate receptors (ie AMPA and uptake. The results we have obtained with different kainate receptors). A role for increased glutamate classes of antipsychotic drugs suggest that the para- release in the actions of NMDA antagonists is suggested digm of ketamine-induced alterations in brain 2-DG by the observations that AMPA/kainate receptor antag- uptake may constitute a promising model to explore onists partially block behavioral effects induced by the mechanisms of action of atypical antipsychotic drugs, drugs.82,84–86 Furthermore, a metabotropic Group II/III and to discriminate drugs with clozapine-like agonist, that inhibits glutamate release, antagonized properties. behavioral activation induced by PCP.83 Differential effects of typical and atypical antipsychotics on electrophysiological responses to Effects of antipsychotic drugs on responses to NMDA and NMDA antagonists NMDA antagonists in preclinical paradigms Electrophysiological studies examining the effects of Effects of typical and atypical drugs on clozapine indicate that the atypical antipsychotic may neurotoxicological and behavioral responses to potentiate NMDA-mediated responses. Clozapine was NMDA antagonists shown to potentiate electrophysiological activation of There is an increasing body of evidence that atypical neurons in the medial prefrontal cortex induced by antipsychotic drugs can antagonize effects of ketamine stimulation of the forceps minor of the corpus callo- and other NMDA antagonists. Although ketamine and sum.102 By contrast, haloperidol inhibited responses in the more selective NMDA antagonist MK-801 have the same experimental paradigm. These data are con- neuroprotective effects in ischemic conditions, these sistent with the opposite effects of colzapine on keta- drugs can induce neuronal damage in select brain mine-induced 2-DG uptake. However, when effects of regions.87–89 Both typical and atypical antipsychotic haloperidol and clozapine were examined on drugs can antagonize MK-801-induced neurotoxic responses induced by direct application of NMDA, responses.90,91 In addition, MK-801-induced locomotor both haloperidol and clozapine potentiated NMDA- activation can be blocked by typical and atypical anti- evoked responses, although clozapine was more potent psychotics.92–94 However, certain behavioral responses in this action.102 The differential effects observed in the induced by NMDA antagonists are differentially affec- NMDA-evoked responses, in comparison to excitation ted by atyptical drugs. For example, clozapine and induced by stimulation of the corpus callosum, was olanzapine blocked PCP-induced deficits in social explained by the observation that haloperidol, but not interactions, whereas haloperidol and risperidone were clozapine, inhibited AMPA-activated responses. ineffective.92 Also, clozapine, quetiapine and olanza- In accord with the action of clozapine to potentiate pine blocked NMDA antagonists-induced deficits in NMDA-evoked electrophysiological responses are the .95–97 While both typical and atypi- findings that the atypical antipsychotic was able to cal antipsychotics can reverse disrupting effects of reverse the inhibitory effect of PCP in the prefrontal dopamine agonists on prepulse inhibition,98 the typical cortex.103 However, haloperidol and (a drugs do not block effects of NMDA antagonists in potent D2 antagonist) did not prevent the PCP-induced this paradigm.99 inhibition of NMDA-mediated electrophysiological Antipsychotic drug action in schizophrenia GE Duncan et al 424 activation.103 These electrophysiological findings are treatment blunted ketamine-induced increases in consistent with the ability of clozapine (but not thought disturbance in schizophrenic patients.105 haloperidol) to block ketamine-induced increases in 2- These findings in humans parallel the observed effects DG uptake. of haloperidol and clozapine on responses to NMDA antagonists in preclinical models, and provide support Potential mechanisms for clozapine-induced for the contention that such preclinical paradigms may alterations in responses to NMDA antagonists have utility for understanding the therapeutic basis of Further study is required to determine whether the atypical antipsychotic drugs action. action of clozapine to alter brain metabolic and electrophysiological effects induced by NMDA antag- Therapeutic potential of agonists of the glycine onists results from direct modulatory actions on site on the NMDA receptor in schizophrenia NMDA receptors, or involves interactions with other patients neurochemical systems and circuits. Although there is Glycine is a positive allosteric modulator and co-agon- no evidence for direct effects of clozapine on MK-801 ist at the NMDA receptor (for review see106). The allo- binding, Banerjee et al104 found that clozapine steric glycine regulatory site on the NMDA receptor enhanced glycine-stimulated MK-801 binding to stria- represents a potential target for drugs to augment tal membranes, indicating that the drug could potenti- NMDA-mediated neurotransmission. There have been ate NMDA-induced responses. Haloperidol produced a a number of clinical studies to test effects of different similar effect so it is difficult to explain the differences glycine site agonists in schizophrenic patients. The in clinical and preclinical effects of these prototype earliest studies in this regard used glycine in doses of typical and atypical antipsychotic drugs by that action. − 5–15 g day 1 and obtained inconsistent results.107,108 In It will be of interest to compare effects of different anti- more recent work with glycine, higher doses were psychotics on glycine-stimulated MK-801 binding in − administered (30–60 g day 1) and more robust and con- regions other than the striatum, such as medial pre- sistent effects were found, primarily in the improve- frontal cortex and hippocampus. ment of negative symptoms.109–111 d-Cycloserine, a par- In regard to the observed effects of clozapine on tial agonist at the glycine regulatory site on the NMDA electrophysiological responses to NMDA and PCP, receptor, has also been tested in schizophrenic the selective 5HT2A M100907 patients. In a very narrow dose range, d-cycloserine produces similar effects as clozapine.103 Those data was demonstrated to improve negative symptoms suggest that the 5HT2A antagonistic properties of when administered alone112 and when added to con- clozapine could account for the observed electrophysi- ventional neuroleptic treatment regimes.113,114 How- ological effects of the drug. Results of ongoing clinical ever, when d-cycloserine was administered in conjunc- trials with M100907 will determine the therapeutic tion with clozapine, the negative symptoms of the potential of selective 5HT2A antagonism for schizo- patients were worsened.115,116 A similar worsening of phrenia. negative symptoms in schizophrenic patients who The ability of clozapine to block the brain metabolic were treated with clozapine was observed after admin- effects of ketamine in vivo may not relate directly to istration of glycine.117 Defining mechanisms that 5HT2A antagonism, since risperidone did not mimic account for the worsening of negative symptoms after the effects of clozapine in that paradigm.100 However, administration of glycine and d-cycloserine to cloza- it is possible that the 5HT2A blocking properties of clo- pine-treated patients could contribute to the under- zapine contribute to its action, perhaps in concert with standing of clozapine’s therapeutic mechanism of other pharmacological properties of the drug. Assess- action. ment of the effects of M100907, alone and in combi- The poor penetration of the blood–brain barrier by nation with other drugs selective for specific dopamine glycine, and the partial agonistic properties of d-cyclo- and serotonin receptors, could resolve the importance serine, make these agents less than optimal for provid- of various receptor-blocking properties of clozapine on ing pharmacological agonism of the glycine regulatory the robust antagonism of ketamine-induced metabolic site on the NMDA receptor. d-Serine is a full agonist activation. of the glycine regulatory site118 and has recently been demonstrated to improve both positive and negative Effects of clozapine and haloperidol on responses symptoms when given in conjunction with antipsy- to ketamine in schizophrenia patients chotic in treatment-resistant patients.119 These data, together with the results of the clinical Although limited information is available from human investigations with glycine and d-cycloserine, offer studies, reported effects of antipsychotic drugs on promise for the therapeutic potential of enhancing responses to ketamine in schizophrenic patients are NMDA receptor function as a strategy for the pharma- consistent with results from experimental animals, cotherapy of schizophrenia. indicating that clozapine selectively modifies responses to NMDA antagonists. Patients on haloperi- dol exhibited greater increases in ketamine-induced Conclusions psychosis ratings relative to baseline, compared to a A number of antipsychotic drugs designated as ‘atypi- haloperidol-free condition.58 By contrast clozapine cal’ have been introduced, or are in advanced stages of Antipsychotic drug action in schizophrenia GE Duncan et al 425 clinical development. The rationale for the develop- 8 Fink H, Morgenstern R, Oelssner W. Clozapine—a serotonin ment of these drugs was to mimic the favorable actions antagonist? Pharmacol Biochem Behav 1984; 20: 513–517. 9 Nash JF, Meltzer HY, Gudelsky GA. Antagonism of serotonin of clozapine, while minimizing its adverse side effects. receptor mediated neuroendocrine and temperature responses by The discovery strategy which led to this new and atypical neuroleptics in the rat. Eur J Pharmacol 1988; 151: emerging class of atypical drugs has been based largely 463–469. on antagonistic activity at D2 and 5HT2A receptors. 10 Meltzer HY, Matsubara S, Lee JC. Classification of typical and Although insufficient data are available to reach atypical antipsychotic drugs on the basis of dopamine D1, D2 and Serotonin2 pKi values. J Pharmacol Exp Ther 1989; 251: 238–246. definitive conclusions, the clinical experience with ris- 11 Farde L, Nyberg S, Oxenstierna G, Nakashima Y, Halldin C, Erics- peridone and olanzapine to date suggests that neither son B. Positron emission tomography studies on D2 and 5-HT2 drug is as efficacious as clozapine in treatment-resist- receptor binding in risperidone-treated schizophrenic patients. J ant schizophrenic patients. Risperidone and olanza- Clin Psychopharmacol 1995; 15: 19S–23S. pine are highly effective antipsychotics with low EPS 12 Flynn SW, MacEwan GW, Altman S, Kopala LC, Fredrikson DH, Smith GN et al. An open comparison of clozapine and risperidone liability at appropriate doses, but available data suggest in treatment-resistant schizophrenia. Pharmacopsychiatry 1998; that these new ‘atypical’ drugs are not as effective as 31: 25–29. clozapine.13,35,120 Since olanzapine and risperidone are 13 Breier AF, Malhotra AK, Su TP, Pinals DA, Elman I, Adler CM et potent antagonists at D2 and 5HT2 receptors, it is al. Clozapine and risperidone in chronic schizophrenia: effects on unlikely that those same properties of clozapine symptoms, parkinsonian side effects, and neuroendocrine response. Am J Psychiatry 1999; 156: 294–298. account for its superior therapeutic efficacy. There is 14 Bondolfi G, Dufour H, Patris M, May JP, Billeter U, Eap CB et al. currently insufficient clinical data to compare efficacy Risperidone versus clozapine in treatment-resistant chronic of clozapine with the newer ‘atypicals’ such as ziprasi- schizophrenia: a randomized double-blind study. The Risperi- done, M-100907, and quetiapine. done Study Group. Am J Psychiatry 1998; 155: 499–504. A growing body of data implicates involvement of 15 Gerlach J. New antipsychotics: classification, efficacy, and adverse effects. Schizophr Bull 1991; 17: 289–309. NMDA receptors in the etiology of schizophrenia and 16 Litman RE, Su TP, Potter WZ, Hong WW, Pickar D. Idazoxan and mechanisms of action of clozapine. Exploration of the response to typical neuroleptics in treatment-resistant schizo- effects of antipsychotic drugs in experimental animal phrenia. Comparison with the A typical neuroleptic, clozapine. and human models of reduced NMDA receptor func- Br J Psychiatry 1996; 168: 571–579. tion may be useful for defining mechanisms of atypical 17 Friedman JI, Temporini H, Davis KL. Pharmacologic strategies for augmenting cognitive performance in schizophrenia. Biol Psy- antipsychotic drug action, and distinguishing drugs chiatry 1999; 45: 1–16. with clozapine-like qualities. Furthermore, if reduced 18 Gurevich EV, Bordelon Y, Shapiro RM, Arnold SE, Gur RE, Joyce NMDA receptor function is a component of the patho- JN. Mesolimbic dopamine D3 receptors and use of antipsychotics physiology of schizophrenia, new therapeutic in patients with schizophrenia. A postmortem study. Arch Gen approaches could involve potentiating NMDA- Psychiatry 1997; 54: 225–232. mediated responses using allosteric modulators such 19 Van Tol HH, Bunzow JR, Guan HC, Sunahara RK, Seeman P, Niz- 119,121,122 nik HB et al. Cloning of the gene for a human dopamine D4 recep- as glycine site agonists or other approaches tor with high affinity for the antipsychotic clozapine. Nature 1991; 123 such as inhibition of glycine uptake. 350: 610–614. 20 Seeman P, Corbett R, Van Tol HH. Atypical neuroleptics have low affinity for dopamine D2 receptors or are selective for D4 recep- tors. 1997; 16: 93–110. References 21 Kramer MS, Last B, Getson A, Reines SA. The effects of a selective D4 dopamine receptor antagonist (L-745 870) in acutely psychotic 1 Breier A, Buchanan RW, Kirkpatrick B, Davis OR, Irish D, Sum- inpatients with schizophrenia. D4 Group. merfelt A et al. Effects of clozapine on positive and negative Arch Gen Psychiatry 1997; 54: 567–572. symptoms in outpatients with schizophrenia. Am J Psychiatry 22 Bristow LJ, Kramer MS, Kulagowski J, Patel S, Ragan CI, Seabrook 1994; 151: 20–26. GR. Schizophrenia and L-745 870, a novel dopamine D4 receptor 2 Kane J, Honigfeld G, Singer J, Meltzer H. Clozapine for the treat- antagonist. Trends Pharm Sci 1997; 18: 186–188. ment-resistant schizophrenic. A double-blind comparison with 23 Mansbach RS, Brooks EW, Sanner MA, Zorn SH. Selective dopa- . Arch Gen Psychiatry 1988; 45: 789–796. mine D4 receptor antagonists reverse -induced 3 Bradford DW, Chakos MH, Sheitman BB, Lieberman JA. Atypical antipsychotic drugs in treatment-refractory schizophrenia. Psy- blockade of prepulse inhibition. 1998; 135: chiatric Annals 1998; 28: 618. 194–200. 4 Altar CA, Boyar WC, Wasley A, Gerhardt SC, Liebman JM, Wood 24 Faraci WS, Zorn SH, Sanner MA, Fliri A. The discovery of potent PL. Dopamine neurochemical profile of atypical antipsychotics and selective dopamine D4 receptor antagonists. Curr Opin Chem resembles that of D-1 antagonists. Naunyn-Schmiedebergs Arch Biol 1998; 2: 535–540. Pharmacol 1988; 338: 162–168. 25 Feldpausch DL, Needham LM, Stone MP, Althaus JS, Yamamoto 5 Farde L, Nordstrom AL, Wiesel FA, Pauli S, Halldin C, Sedvall BK, Svensson KA et al. The role of dopamine D4 receptor in the G. Positron emission tomographic analysis of central D1 and D2 induction of behavioral sensitization to amphetamine and dopamine receptor occupancy in patients treated with classical accompanying biochemical and molecular adaptations. J Pharma- neuroleptics and clozapine. Relation to extrapyramidal side col Exp Ther 1998; 286: 497–508. effects. Arch Gen Psychiatry 1992; 49: 538–544. 26 Merchant KM, Gill GS, Harris DW, Huff RM, Eaton MJ, Look- 6 Karlsson P, Smith L, Farde L, Harnryd C, Sedvall G, Wiesel FA. ingland K, et al. Pharmacological characterization of U-101387, a Lack of apparent antipsychotic effect of the D1-dopamine receptor dopamine D4 receptor selective antagonist. J Pharmacol Exp Ther antagonist SCH39166 in acutely ill schizophrenic patients. Psych- 1996; 279: 1392–1403. opharmacology 1995; 121: 309–316. 27 Bymaster FP, Calligaro DO, Falcone JF, Marsh RD, Moore NA, Tye 7 Den Boer JA, van Megen HJ, Fleischhacker WW, Louwerens JW, NC et al. Radioreceptor binding profile of the atypical antipsy- Slaap BR, Westenberg HG. Differential effects of the D1–DA recep- chotic olanzapine. Neuropsychopharmacology 1996; 14: 87–96. tor antagonist SCH39166 on positive and negative symptoms of 28 Bymaster FP, Rasmussen K, Calligaro DO, Nelson DL, DeLapp schizophrenia. Psychopharmacology 1995; 121: 317–322. NW, Wong DT et al. In vitro and in vivo biochemistry of olanzap- Antipsychotic drug action in schizophrenia GE Duncan et al 426 ine: a novel, atypical antipsychotic drug. J Clin Psychiatry 1997; 50 Olney JW, Farber NB. dysfunction and schizo- 58 Suppl 10: 28–36. phrenia. Arch Gen Psychiatry 1995; 52: 998–1007. 29 Beasley CMJ, Tollefson GD, Tran PV. Efficacy of olanzapine: an 51 Deutsch SI, Mastropaolo J, Schwartx BL, Rosse R, Morihisa JM. A overview of pivotal clinical trials. J Clin Psychiatry 1997; 58 Suppl ‘glutamatergic hypothesis’ of schizophrenia. Clin Neuropharma- 10: 7–12. col 1989; 12: 1–13. 30 Tollefson GD, Beasley CMJ, Tamura RN, Tran PV, Potvin JH. 52 Coyle JT. The glutamatergic dysfunction hypothesis for schizo- Blind, controlled, long-term study of the comparative incidence phrenia. Harv Rev Psychiatry 1996; 3: 241–253. of treatment-emergent tardive with olanzapine or halo- 53 Yenari MA, Bell TE, Kotake AN, Powell M, Steinberg GK. Dose peridol. Am J Psychiatry 1997; 154: 1248–1254. escalation safety and tolerance study of the competitive NMDA 31 Tollefson GD, Beasley CMJ, Tran PV, Street JS, Krueger JA, antagonist selfotel (CGS 19755) in neurosurgery patients. Clin Tamura RN et al. Olanzapine versus haloperidol in the treatment Neuropharmacol 1998; 21: 28–34. of schizophrenia and schizoaffective and schizophreniform dis- 54 Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, Bremner orders: results of an international collaborative trial. Am J Psy- JD et al. Subanesthetic effects of the noncompetitive NMDA antag- chiatry 1997; 154: 457–465. onist, ketamine, in humans. Psychotomimetic, perceptual, cogni- 32 Tran PV, Dellva MA, Tollefson GD, Beasley CMJ, Potvin JH, tive, and neuroendocrine responses. Arch Gen Psychiatry 1994; Kiesler GM. and tolerability of olanzap- 51: 199–214. ine versus haloperidol in the acute treatment of schizophrenia. J 55 Malhotra AK, Pinals DA, Weingartner H, Sirocco K, Missar CD, Clin Psychiatry 1997; 58: 205–211. Pickar D et al. NMDA receptor function and human cognition— 33 Kapur S, Zipursky RB, Remington G, Jones C, DaSilva J, Wilson the effects of ketamine in healthy volunteers. Neuropsychopharm- AA et al. 5-HT2 and D2 receptor occupancy of olanzapine in acology 1996; 14: 301–307. schizophrenia: a PET investigation. Am J Psychiatry 1998; 155: 56 Adler CM, Goldberg TE, Malhotra AK, Pickar D, Breier A. Effects 921–928. of ketamine on , working memory, and semantic 34 Tauscher J, Kufferle B, Asenbaum S, Fischer P, Pezawas L, Barnas memory in healthy volunteers. Biol Psychiatry 1998; 43: 811–816. C et al. In vivo I-123 IBZM SPECT imaging of striatal dopamine- 57 Lahti AC, Holcomb HH, Medoff DR, Tamminga CA. Ketamine acti- 2 receptor occupancy in schizophrenic patients treated with vates psychosis and alters limbic blood flow in schizophrenia. olanzapine in comparison to clozapine and haloperidol. Psycho- Neuroreport 1995; 6: 869–872. 1999; 141: 175–181. 58 Lahti AC, Koffel B, LaPorte D, Tamminga CA. Subanesthetic doses 35 Conley RR, Tamminga CA, Bartko JJ, Richardson C, Peszke M, Lin- of ketamine stimulate psychosis in schizophrenia. Neuropsycho- gle J et al. Olanzapine compared with chlorpromazine in treat- pharmacology 1995; 13: 9–19. ment-resistant schizophrenia. Am J Psychiatry 1998; 155: 914– 59 Malhotra AK, Pinals DA, Adler CM, Elman I, Clifton A, Pickar D 920. et al. Ketamine-induced exacerbation of psychotic symptoms and 36 Kapur S, Zipursky RB, Remington G. Clinical and theoretical cognitive impairment in neuroleptic-free schizophrenics. Neuro- implications of 5-HT2 and D2 receptor occupancy of clozapine, psychopharmacology 1997; 17: 141–150. risperidone, and olanzapine in schizophrenia. Am J Psychiatry 60 Angrist BM, Gershon S. The phenomenology of experimentally 1999; 156: 286–293. induced amphetamine psychosis—preliminary observations. Biol 37 Seeman P, Tallerico T. Antipsychotic drugs which elicit little or Psychiatry 1970; 2: 95–107. no bind more loosely than dopamine to brain D2 61 Griffith J, Cavanaugh J, Held N, Oates J. d-Amphetamine evalu- receptors, yet occupy high levels of these receptors. Mol Psy- ation of psychotomimetic properties in man. Arch Gen Psychiatry chiatry 1998; 3: 123–134. 1972; 26: 97–100. 38 Wadenberg ML. Serotonergic mechanisms in neuroleptic-induced 62 Lieberman JA, Kane JM, Alvir JAJ. Provocative tests with psychos- in the rat. Neurosci Biobehav Rev 1996; 20: 325–339. timulant drugs in schizophrenia. Psychopharmacology 1987; 91: 39 Costall B, Fortune DH, Naylor RJ, Mardsen CD, Pycock C. Sero- 415–433. tonergic involvement with neuroleptic catalepsy. Neuropharmac- 63 Ferrer-Allado T, Brechner VL, Dymond A, Cozen H, Crandall P. ology 1975; 14: 859–868. Ketamine induced electroconvulsive phenomena in human limbic 40 Waldmeier PC, Delini-Stula AA. Serotonin–dopamine interactions and thalamic regions. Anesthesiology 1973; 38: 333–344. in the nigrostriatal system. Eur J Pharmacol 1979; 55: 363–373. 64 Winters WD, Ferrer-Allado T, Guzman-Flores C, Alcaraz M. The 41 Kapur S. 5-HT2 antagonism and EPS benefits: is there a causal cataleptic state induced by ketamine—a review of the neurophar- connection? Psychopharmacology 1996; 124: 35–39. macology of anesthesia. Neuropharmacology 1972; 11: 303–315. 42 Nordstrom AL, Farde L, Nyberg S, Karlsson P, Halldin C, Sedvall 65 Anis NA, Berry SC, Burton NR, Lodge D. The anes- G. D1, D2, and 5-HT2 receptor occupancy in relation to clozapine thetics, ketamine and , selectively reduce excitation serum concentration: a PET study of schizophrenic patients. Am of central mammalian neurons by N-methyl-aspartate. Br J Pharm- J Psychiatry 1995; 152: 1444–1449. acol 1983; 79: 565–575. 43 Cohen BD, Rosenbaum G, Luby ED, Gottlieb JS. Comparison of 66 Thomson AM, West DC, Lodge D. An N-methyl-aspartate receptor- phencyclidine hydrochloride (sernyl) with other drugs: simul- mediated synapse in rat cerebral cortex: a site of action of keta- ation of schizophrenic performance with phencyclidine hydroch- mine? Nature 1985; 313: 479–481. loride (sernyl)lysergic acid diethylamide (LSD-25), and amobarbi- 67 White PF, Way WL, Trevor AJ. Ketamine—its pharmacology and tal (Amytal) sodium, II—symbolic and sequential thinking. Arch therapeutic uses. Anesthesiology 1982; 56: 119–136. Gen Psychiatry 1962; 6: 79–85. 68 Sinclair JG, Tien AF. Neuronal responses to ketamine adminis- 44 Luby ED, Cohen BD, Rosenbaum G, Gottilieb JS, Kelley R. Study tered microiontophoretically or intraperitoneally in the rat. Gen of a new schizophrenomimetic drug—sernyl. Arch Neurol Psych Pharmacol 1979; 10: 51–55. 1959; 81: 363–369. 69 Nelson SR, Howard RB, Cross RS, Samson F. Ketamine-induced 45 Bakker CB, Amini FB. Observations on the psychotomimetic changes in regional glucose utilization in the rat brain. Anesthesi- effects of sernyl. Compr Psych 1961; 2: 269–280. ology 1980; 52: 330–334. 46 Davies BM, Beech HR. The effect of 1- 70 Hammer RP, Herkenham M. Altered metabolic activity in the cer- (sernyl) on twelve normal volunteers. J Ment Sci 1960; 106: ebral cortex of rats exposed to ketamine. J Comp Neurol 1983; 220: 912–924. 396–404. 47 Corssen G, Domino EF. Dissociative anesthesia: further pharmaco- 71 Duncan GE, Leipzig JN, Mailman RB, Lieberman JA. Differential logic studies and first clinical experience with phencyclidine effects of clozapine and haloperidol on ketamine-induced brain derivative C1–581. Anesth Analg 1966; 45: 29–40. metabolic activation. Brain Res 1998; 812: 65–75. 48 Domino EF, Chodoff P, Corssen G. Pharmacological effects of CI– 72 Duncan GE, Moy SS, Knapp DJ, Mueller RA, Breese GR. Metabolic 581, a new dissociative anesthetic in man. Clin Pharmacol Ther mapping of the rat brain after subanesthetic doses of ketamine: 1965; 6: 279–291. potential relevance to schizophrenia. Brain Res 1998; 787: 181– 49 Javitt DC, Zukin SR. Recent advances in the phencyclidine model 190. of schizophrenia. Am J Psychiatry 1991; 148: 1301–1308. 73 Kurumaji A, Nehls DG, Park CK, McCulloch J. Effects of NMDA Antipsychotic drug action in schizophrenia GE Duncan et al 427 antagonists, MK-801 and CPP, upon local cerebral glucose use. methyl-d-aspartate antagonist-induced behaviors. Psychopharma- Brain Res 1989; 496: 268–284. cology 1995; 120: 67–74. 74 Duncan GE, Miyamoto S, Leipzig JN, Lieberman JA. Comparison 93 Tiedtke PI, Bischoff C, Schmidt WJ. MK-801-induced stereotypy of brain metabolic activity patterns induced by ketamine, MK801 and its antagonism by neuroleptic drugs. J Neural Trans 1990; 81: and amphetamine in rats: support for NMDA receptor involve- 173–182. ment in responses to subanesthetic doses of ketamine. Brain Res 94 Hoffman DC. Typical and atypical neuroleptics antagonize MK- (in press). 801-induced locomotion and stereotypy in rats. J Neural Trans 75 Breier A, Malhotra AK, Pinals DA, Weisenfeld NI, Pickar D. 1992; 89: 1–10. Association of ketamine-induced psychosis with focal activation 95 Bakshi VP, Geyer MA. Antagonism of phencyclidine-induced of the prefrontal cortex in healthy volunteers. Am J Psychiatry deficits in prepulse inhibition by the putative atypical antipsy- 1997; 154: 805–811. chotic olanzapine. Psychopharmacology 1995; 122: 198–201. 76 Vollenweider FX, Leenders KL, Oye I, Hell D, Angst J. Differential 96 Bakshi VP, Swerdlow NR, Geyer MA. Clozapine antagonizes psychopathology and patterns of cerebral glucose utilisation pro- phencyclidine-induced deficits in sensorimotor gating of the duced by (S)- and (R)-ketamine in healthy volunteers using posi- startle response. J Pharmacol Exp Ther 1994; 271: 787–794. tron emission tomography (PET). Eur Neuropsychopharmacol 97 Swerdlow NR, Bakshi VP, Geyer MA. Seroquel restores sensori- 1997; 7: 25–38. motor gating in phencyclidine-treated rats. J Pharmacol Exp Ther 77 Vollenweider FX, Leenders KL, Scharfetter C, Antonini A, Mag- 1996; 279: 1290–1299. uire P, Missimer J et al. Metabolic hyperfrontality and psycho- 98 Swerdlow NR, Keith VA, Braff DL, Geyer MA. Effects of spip- pathology in the ketamine model of psychosis using positron erone, raclopride, SCH 23390 and clozapine on apomorphine emission tomography (PET) and [18F]fluorodeoxyglucose (FDG). inhibition of sensorimotor gating of the startle response in the rat. Eur Neuropsychopharmacol 1997; 7: 9–24. J Pharmacol Exp Ther 1991; 256: 530–536. 99 Keith VA, Mansbach RS, Geyer MA. Failure of haloperidol to 78 Grunze HC, Rainnie DG, Hasselmo ME, Barkai E, Hearn EF, block the effects of phencyclidine and on prepulse McCarley RW et al. NMDA-dependent modulation of CA1 local inhibition of startle. Biol Psychiatry 1991; 30: 557–566. circuit inhibition. J Neurosci 1996; 16: 2034–2043. 100 Duncan GE, Leipzig JN, Miyamoto S, Mailman RB, Lieberman JA. 79 Coppel DL, Bovill JG, Dundee JW. The taming of ketamine. Anes- Regional brain metabolic activity is affected differentially by typi- thesia 1973; 28: 293–296. cal and atypical antipsychotic drugs in rats, Schizophr Res 1999; 80 Krystal JH, Karper LP, Bennett A, D’Souza DC, Abi-Dargham A, 36: 307. Morrissey K et al. Interactive effects of subanesthetic ketamine 101 Zhang W, Bymaster FP. The in vivo effects of olanzapine and other and subhypnotic lorazepam in humans. Psychopharmacology antipsychotic agents on receptor occupancy and antagonism of 1998; 135: 213–229. dopamine D-1, D-2, D-3, 5HT(2A) and muscarinic receptors. Psy- 81 Liu J, Moghaddam B. Regulation of glutamate efflux by excitatory chopharmacology 1999; 141: 267–278. amino acid receptors evidence for tonic inhibitory and phasic 102 Arvanov VL, Liang X, Schwartz J, Grossman S, Wang RY. Cloza- excitatory regulation. J Pharmacol Exp Ther 1995; 274: 1209– pine and haloperidol modulate N-methyl-d-aspartate- and non-N- 1215. methyl-d-aspartate receptor-mediated neurotransmission in rat 82 Moghaddam B, Adams B, Verman A, Daly D. Activation of glutam- prefrontal cortical neurons in vitro. J Pharmacol Exp Ther 1997; atergic neurotransmission by ketamine—a novel step in the path- 283: 226–234. way from NMDA receptor blockade to dopaminergic and cognitive 103 Wang RY, Liang X. M100907 and clozapine, but not haloperidol disruptions associated with the prefrontal cortex. J Neurosci 1997; or raclopride, prevent phencyclidine-induced blockade of NMDA 17: 2921–2927. responses in pyramidal neurons of the rat medial prefrontal 83 Moghaddam B, Adams BW. Reversal of phencyclidine effects by a cortical slice. Neuropsychopharmacology 1998; 19: 74–85. group II metabotrophic glutamate receptor agonist in rats. Science 104 Banerjee SP, Zuck LG, Yablonsky-Alter E, Lidsky TI. Glutamate 1998; 281: 1349–1352. agonist activity: implications for antipsychotic drug action and 84 Hauber W, Andersen R. The non-NMA glutamate receptor antag- schizophrenia. Neuroreport 1995; 6: 2500–2504. onist GYKI Y1–52466 counteracts locomotor stimulation and 105 Malhotra AK, Adler CM, Kennison SD, Elman I, Pickar D, Breier anticataleptic activity induced by the NMDA antagonist dizocilp- A. Clozapine blunts N-methyl-d-aspartate antagonist-induced psy- ine. Naunyn Sch Arch Pharmacol 1993; 348: 486–490. chosis: a study with ketamine. Biol Psychiatry 1997; 42: 664–668. 85 Bubser M, Tzschentke T, Hauber W. Behavioural and neurochem- 106 Leeson PD, Iversen LL. The glycine site on the NMDA receptor: ical interactions of the AMPA antagonist GYKI Y1–52466 and the structure-activity relationships and therapeutic potential. J Med non-competitive NMDA antagonist dizocilpine in rats. J Neural Chem 1994; 37: 4053–4067. Trans 1995; 101: 115–126. 107 Rosse RB, Theut SK, Banay-Schwartz M, Leighton M, Scarcella E, 86 Willins DL, Narayanan S, Wallace LJ, Uretsky NJ. The role of Cohen CG et al. Glycine to conventional neuro- dopamine and AMPA/kainate receptors in the nucleus accumbens leptic treatment in schizophrenia: an open-label, pilot study. Clin in the hypermotility response to MK801. Pharmacol Biochem Neuropharmacol 1989; 12: 416–424. Behav 1993; 46: 881–887. 108 Costa J, Khaled E, Sramek J, Bunney WJ, Potkin SG. An open trial 87 Fix AS, Horn JW, Wightman KA, Johnson CA, Long GG, Storts of glycine as an adjunct to neuroleptics in chronic treatment- RW et al. Neuronal vacuolization and necrosis induced by the refractory schizophrenics. J Clin Psychopharmacol 1990; 10: 71– noncompetitive N-methyl-d-aspartate (NMDA) antagonist 72. MK(+)801 (dizocilpine maleate): a light and electron microscopic 109 Heresco-Levy U, Javitt DC, Ermilov M, Mordel C, Horowitz A, evaluation of the rat retrosplenial cortex. Exp Neurol 1993; 123: Kelly D. Double-blind, -controlled, crossover trial of gly- 204–215. cine adjuvant therapy for treatment-resistant schizophrenia. Br J 88 Olney JW, Labruyere J, Wang G, Wozniak DF, Price MT, Sesma Psychiatry 1996; 169: 610–617. MA. NMDA antagonist neurotoxicity—mechanism and protection. 110 Javitt DC, Zylberman I, Zukin SR, Heresco-Levy U, Lindenmayer Science 1991; 154: 1515–1518. JP. Amelioration of negative symptoms in schizophrenia by gly- 89 Horvath ZC, Czopf J, and Buzsaki G. MK-801-induced neuronal cine. Am J Psychiatry 1994; 151: 1234–1236. damage in rats. Brain Res 1997; 753: 181–195. 111 Heresco-Levy U, Javitt DC, Ermilov M, Mordel C, Silipo G, Licht- 90 Farber NB, Price MT, Labruyere J, Nemnich J, St Peter H, Wozniak enstein M. Efficacy of high-dose glycine in the treatment of endur- DF et al. Antipsychotic drugs block phencyclidine receptor- ing negative symptoms of schizophrenia. Arch Gen Psychiatry mediated neurotoxicity. Biol Psychiatry 1993; 34: 119–121. 1999; 56: 29–36. 91 Farber NB, Foster J, Duhan NL, Olney JW. Olanzapine and fluper- 112 van Berckel BN, Hijman R, van der Linden JA, Westenberg HG, lapine mimic clozapine in preventing MK-801 neurotoxicity. van Ree JM, Kahn RS. Efficacy and tolerance of d-cycloserine in Schizophr Res 1996; 21: 33–37. drug-free schizophrenic patients. Biol Psychiatry 1996; 40: 92 Corbett R, Camacho F, Woods AT, Kerman LL, Fishkin RJ, Brooks 1298–1300. K et al. Antipsychotic agents antagonize non-competitive N- 113 Goff DC, Tsai G, Levitt J, Amico E, Manoach D, Schoenfeld DA et Antipsychotic drug action in schizophrenia GE Duncan et al 428 al. A placebo-controlled trial of d-cycloserine added to conven- 119 Tsai G, Pinchen Y, Chung L-C, Lange N, Coyle JT. d-Serine added tional neuroleptics in patients with schizophrenia. Arch Gen Psy- to antipsychotics for the treatment of schizophrenia. Biol Psy- chiatry 1999; 56: 21–27. chiatry 1998; 44: 1081–1089. 114 Goff DC, Tsai G, Manoach DS, Coyle JT. Dose-finding trial of d- 120 Henderson DC, Nasrallah RA, Goff DC. Switching from clozapine cycloserine added to neuroleptics for negative symptoms in to olanzapine in treatment-refractory schizophrenia: safety, clini- schizophrenia. Am J Psychiatry 1995; 152: 1213–1215. cal efficacy, and predictors of response. J Clin Psychiatry 1998; 115 Goff DC, Tsai G, Manoach DS, Flood J, Darby DG, Coyle JT. d- 59: 585–588. Cycloserine added to clozapine for patients with schizophrenia. 121 Farber NB, Newcomer JW, Olney JW. Glycine agonists—What can Am J Psychiatry 1996; 153: 1628–1630. they teach us about schizophrenia? Arch Gen Psychiatry 1999; 56: 116 Goff DC, Henderson DC, Evins AE, Amico E. A placebo-controlled 13–17. d crossover trial of -cycloserine added to clozapine in patients 122 Goff DC, Wine L. Glutamate in schizophrenia: clinical and with schizophrenia. Biol Psychiatry 1999; 45: 512–514. research implications. Schizophr Res 1997; 27: 157–168. 117 Potkin SG, Jin Y, Bunney BG, Costa J, Gulasekaram B. Effect of 123 Bergeron R, Meyer TM, Coyle JT, Greene RW. Modulation of N- clozapine and adjunctive high-dose glycine in treatment-resistant methyl-d-aspartate receptor function by glycine transport. Proc schizophrenia. Am J Psychiatry 1999; 156: 145–147. Natl Acad Sci USA 1998; 95: 15730–15734. 118 Hashimoto A, Oka T. Free d-aspartate and d-serine in the mam- malian brain and periphery. Prog Neurobiol 1997; 52: 325–353.