Effects of antipsychotic treatments and D-serine supplementation on the electrophysiological...

Effects of Antipsychotic Treatmentsand D-Serine Supplementation on the

Electrophysiological Activation ofMidbrain Dopamine Neurons Induced

by the Noncompetitive NMDAAntagonist MK 801

STEPHEN BENNETT AND BENJAMIN GRONIER*

School of Pharmacy, De Montfort University, The Gateway, Leicester, United Kingdom

KEY WORDS VTA; A10; extracellular recording; schizophrenia; noncompetitiveNMDAantagonists; glycineB coagonists; neuroleptics; adjunct therapy

ABSTRACT The acute administration of the noncompetitive glutamate N-methyl-D-aspartate (NMDA) receptor antagonist dizocilpine (MK 801) is known to increasecentral dopaminergic activity in rats and to elicit schizophreniform behavior inhuman. The current study was undertaken to compare the effects of different acute orchronic neuroleptic treatments, on the response of ventral tegmental area dopamine(DA) neurons to MK 801, using the in vivo electrophysiological paradigm in anesthe-tized preparations. Sprague Dawley male rats were treated, acutely or chronically dur-ing 3 weeks, with saline, olanzapine (10 mg/kg), haloperidol (1 mg/kg) or the combina-tion of haloperidol with D-serine (1 mg/kg/300 mg/kg), a gliotransmitter coagonist ofthe NMDA receptor that has been shown to improve the efficacy of typical neurolep-tics. In control animals, the acute administration of MK 801 (0.5 mg/kg, i.v.) increasedsignificantly both the firing and burst activity of DA neurons by 20 and 26%, respec-tively, the latter effect being partially reversed by the selective 5-HT2A antagonistM 100,907 (0.4 mg/kg, i.v.). The acute preadministration of haloperidol (1 mg/kg, i.p.)and olanzapine (10 mg/kg, i.p.) failed to prevent or reverse the activatory effect of MK801 on firing activity. On the other hand, MK 801-induced burst activity, was partiallyprevented by olanzapine, but not by haloperidol pretreatment. All antipsychotic treat-ments, when administered chronically, prevent the activatory effect of MK 801 on bothfiring and burst activity, and occasionally convert the response to MK 801 on burstactivity to an inhibitory response, the latter occurring more predominantly in ratstreated with the combination haloperidol/D-serine. These results suggest that achronic antipsychotic regime alters the function of the NMDA receptors that tonicallycontrol the firing activity of midbrain dopaminergic neurons. Synapse 61:679–688,2007. VVC 2007 Wiley-Liss, Inc.

INTRODUCTION

Dizocilpine (MK 801), and other noncompetitiveantagonists of the glutamate N-methyl-D-aspartate(NMDA)-receptors, e.g., phencyclidine, can elicit aschizophreniform behavior in human, which is symp-tomatically indistinguishable from organic schizo-phrenia, and include the negative symptoms (Javittand Zukin, 1991), providing pharmacological supportfor the glutamate hypothesis of schizophrenia, inwhich it is proposed a hypofunctional state of the glu-tamate NMDA receptor in brain of schizophrenic

patients (Lindsley et al., 2006; Millan, 2002). Thishas prompted some investigators to examine the clini-cal effects of drugs that should facilitate glutamate/NMDA neurotransmission, such as the glycine co-

Contract grant sponsor: Royal Pharmaceutical Society of Great Britain.

*Correspondence to: Benjamin Gronier, School of Pharmacy, De MontfortUniversity, Leicester LE1 9BH, United Kingdom. E-mail: [email protected]

Received 19 December 2006; Accepted 20 February 2007

DOI 10.1002/syn.20413

Published online in Wiley InterScience (www.interscience.wiley.com).

VVC 2007 WILEY-LISS, INC.

SYNAPSE 61:679–688 (2007)

agonists glycine and D-serine. These drugs have beenshown to successfully increase the efficacy of typicalneuroleptics toward positive and negative symptoms,and to improve also cognitive deficit of schizophrenia(Javitt, 1999; Lane et al., 2005; Tsai et al., 1998).

Noncompetitive NMDA antagonists, including MK801 have been shown to increase the firing and burstactivity of midbrain ventral tegmental area (VTA) do-pamine (DA) neurons when recorded extracellularlyin vivo (French et al., 1993). In addition, these drugspromote the release of DA in DA terminal areas ofthe mesolimbic dopaminergic system (Imperato et al.,1990; Mathe et al., 1998, 1999; Schmidt and Fadayel,1996; Yan et al., 1997), in a manner that dependsdirectly on the impulse-flow activity of the VTA DAneurons (Mathe et al., 1999) and that is associatedwith an increase in locomotor activity (Mathe et al.,1998).

At present it is not fully understood by whichmechanism noncompetitive NMDA antagonists inducetheir stimulatory effects on DA neuronal activity. Fewelectrophysiological studies have demonstrated thatlocally, there is possibly a loss of inhibitory controlfrom the VTA GABAergic interneurons that controlthe activity of midbrain DA cells (Zhang et al., 1993).However, according to some neurochemical studiesthe stimulatory effects of MK 801 on accumbens DArelease seems to also involve a complex circuitry thatincludes ultimately the activation of a glutamate exci-tatory drive to the VTA (Mathe et al., 1998), possiblythrough an activation of serotonin 5-HT2A receptors(Schmidt and Fadayel, 1996).

Interestingly, most of the behavioral manifestationsinduced by NMDA antagonists can be abolished byatypical antipsychotics or by the administration ofa selective 5-HT2A antagonists, such as M 100,907(Bakshy and Geyer, 1995; Carlsson, 1995; Corbettet al., 1995; O’Neill and Shaw, 1999; Varty et al.,1999).

It is not clear at present if typical or atypical anti-psychotics can amend the electrophysiological effectsof NMDA antagonists. In the present study, we haveexamined some of the electrophysiological correlatesof the behavioral and neurochemical effects of MK801, using the technique of electrophysiology in vivo.

First, we have examined if the selective 5-HT2A an-tagonist M 100,907, as well as drugs that facilitateglutamate NMDA neurotransmission, such as the gly-cineB coagonists D-serine, can reverse the acute elec-trophysiological effects of MK 801 on midbrain DAneurons. Second, our investigations have focused onwhether the acute or chronic administration of dif-ferent antipsychotic treatments can alter the responseof VTA DA neurons to the acute administration ofMK 801. Treatments include the atypical neurolepticolanzapine and the typical neuroleptic haloperidol,the latter administered alone or in conjunction with

the endogenous agonist of the glycineB site of theNMDA receptor D-serine. Finally, we have examined,in parallel, if the coadministration of D-serine withhaloperidol alters the activity of VTA DA neurons ina distinctive way to the chronic administration ofhaloperidol alone.

METHODSubjects

Experiments were performed on male Sprague-Dawley rats (Charles River, UK) weighing 250–350 g.The animals were housed in groups of 3–5 and freeaccess to food and water was provided. Environmen-tal conditions were checked daily and maintainedunder constant temperature and humidity in a roomwith a regulated 12-h light/dark cycle (lights on at06:00). Experiments were approved and performed inaccordance with the guidelines of the ethical commit-tee of De Montfort University and Home Office regu-lation. Every effort was made to minimize the num-ber of animal used and their suffering.

Surgery

Animals were initially deeply anesthetized using400 mg/kg chloral hydrate i.p., thereafter an i.v. can-nula was implanted in one lateral vein of the tail.Subjects were then placed in a Kopf stereotaxic framein preparation for electrode implantation surgery,body temperature was maintained at 378C. An inci-sion was made across the top of the head and theedges of the skin drawn back to reveal the cranium.Bregma was marked and a hole was drilled throughthe bone at the coordinates of the VTA according tothe atlas of Paxinos and Watson (1998). Electrodeswere manufactured in house from borosilicate capilla-ries (1.5 mm, Harvard Apparatus, UK) pulled on aPP-830 electrode puller (Narishige, Japan) and filledby hand with an electrolyte solution (in mM: NaCl147, KCL 4, adjusted to pH 6). Typical resistance wasin the range 3–6 MO. Outputs from the electrodewere sent to a Neurolog AC preamplifier and ampli-fier (Digitimer, UK). Signals were filtered and sent toan audio amplifier, a Tektronix 2201 digital storageoscilloscope, and a computer running Spike 2, fordata capture and analysis.

DA neurons were identified by their location withinthe VTA (or A10 region; 4.8–6.0 mm posterior tobregma, 0.3–1 mm lateral to the midline, and 6.5–9 mm below the cortical surface), and well establishedelectrophysiological criteria (Grace and Bunney, 1984a)including: (1) spontaneous firing rate between 5 and90 spikes/10 s occurring sometimes in bursts; (2) tri-phasic or biphasic waveforms, with an initial positivedeflection (usually notched) followed by a prominentnegative phase; (3) long duration action potentials (2–

680 S. BENNETT AND B. GRONIER

Synapse DOI 10.1002/syn

4 ms); (4) low pitch sound when monitored by anaudioamplifier.

For cell/track measurement, the number of sponta-neously active DA neurons was determined in 5–8stereotaxic electrode descents. Electrode tracks wereseparated from each other by 200 mm and made in a

slow uniform speed. The mean number of descentsper animal did not differ in control and treated rats.

Treatment and drugs

Olanzapine, haloperidol, and D-serine were admin-istered intraperitonally for 1 or 21 days. The last day

TABLE I. Table showing for each treatment group the proportion and percentage (between brackets) of dopamine neurons that respond byan increase, a decrease, or no change in firing and burst activity to the administration of MK 801 at the dose 0.5 mg/kg, i.v.

Treatment group

Response

Activated Insensitive/low response Inhibited

Firing Burst Firing Burst Firing Burst

Saline/control 11/13 (85%)* 8/13 (62%)* 2/13 (15%) 4/13 (30%) 0/13 (0%) 1/13 (8%)Haloperidol acute 8/8 (100%)* 7/8 (88%)* 0/8 (0%) 1/8 (12%) 0/8 (0%) 0/8 (0%)Olanzapine acute 6/9 (67%)* 4/9 (45%) 3/9 (33%) 4/9 (45%) 0 (0%) 1/9 (10%)Haloperidol 21 days 1/7 (14%) 1/7 (14%) 5/7 (72%) 3/7 (43%) 1/7 (14%) 3/7 (43%)Hal/D-ser 21 days 1/8 (12%) 1/8 (12%) 5/8 (68%) 2/8 (25%) 2/8 (25%) 5/8 (63%)Olanzapine 21 days 3/10 (30%) 2/10 (20%) 6/10 (60%) 4/10 (40%) 1/10 (10%) 4/10 (40%)D-serine 21 days 4/5 (80%)* 4/5 (80%)* 1/5 (20%) 1/5 (20%) 0 (0%) 0 (0%)Saline/control 11/13 (85%)* 8/13 (62%)* 2/13 (15%) 4/13 (30%) 0/13 (0%) 1/13 (8%)

Neurons were considered activated or inhibited if their baseline values changed by more than 15% following MK 801 (0.5 mg/kg, i.v.).*Significantly different than pre-drug value, as indicated in Figures 3 and 4.

Fig. 1. A: Representative example of firing activity of a VTA DAneurons treated at the point indicated in the figure with the NMDAantagonist MK 801 (two doses of 0.5 mg/kg, i.v.) and the 5-HT2A an-tagonist M 100,907 (0.4 mg/kg, i.v.). Each histogram represents thenumber of action potential generated in 10 s. B: Representativeburst train of the same neuron corresponding to 1 min recordingobtained in baseline condition, after the i.v. administration of the

second dose of MK 801 (1 mg/kg) and after the administration ofM100,907 (0.4 mg/kg). C: Interspike interval histograms of thesame neuron assessed before (50–150 s), and after the subsequentadministration of MK 801 (340–420 s) and M 100,907 (460–550 s).The figure shows that the firing and burst activity is enhanced byMK 801. Only the stimulatory effect on burst activity is partiallyreversed by the 5-HT2A antagonist.

681ANTIPSYCHOTICS AND SENSITIVITY TO MK 801


of treatment, the drug was administered 1–2 h beforethe electrophysiological experiment. At least five ani-mals were studied in each group of treatment.M 100,907 was a gift from Eli Lilly Research Labora-tories, olanzapine and haloperidol were purchasedfrom Sequoia International (Oxford, UK), MK 801and D-serine were obtained from Sigma. M 100,907 wasdissolved in 25% DMSO, other drugs were directlydissolved in saline 0.8%.

Calculations and statistical analysis

All values are expressed as the mean 6 standarderror of the mean (SEM). Mean basal firing activityand basal % of spikes occurring in burst were eval-uated after neuron had attained a stable firing rate,after at least 5 min of recording. Burst analysis wasmade by use of a slightly modified version of theSpike II program. The onset of a burst was defined byan interspike interval shorter than 80 ms and the ter-mination of a burst by the next interval longer than160 ms (Grace and Bunney, 1984b). For i.v. drugadministration: predrug values of firing rate and % inburst firing were obtained by averaging them over aperiod of 2 min immediately prior the i.v. administra-tion, post drug values were obtained by averagingthem during a period of 2 min immediately after thefirst minute following the administration. In general,subsequent i.v. administrations of drugs were sepa-rated by 3 min intervals. Changes in firing rate or inthe proportion of burst firing in DA neurons wereassessed, using repeated measures ANOVA (followedby post hoc test if appropriate) or Student’s t-test, asappropriate. Probabilities smaller than 0.05 were con-sidered as significant.

RESULTS

A total of 210 VTA dopaminergic neurons werestudied. VTA dopaminergic neurons were recognizedaccording to the criteria described in the earliermethod. Only the neurons that met these criteriawere included in this study.

Effects of an acute administration ofMK 801 in drug-naı̈ve rats

The control group was made of naı̈ve (n ¼ 5) and21-day saline-treated (n ¼ 8) rats. Since the effectsproduced by MK 801 were similar in both groups,their data were combined together. It can be seenfrom Figures 1–3 that the acute administration of adose of 500 lg/kg of MK 801 (via an i.v. cannula) incontrol rats produced a significant increase in the fir-ing rate and bursting activity of DA neurons in theVTA. The second dose did not produce any furtherincrease in firing and burst activation. The mean fir-ing rate increased by 20% (range þ10 to þ90%) and

mean burst activity by 26% (variation range þ26 toþ90%). A small proportion of neurons were not re-sponsive to MK 801 (see Table I) and in one case, MK801 produced a large inhibitory effect on burst activ-ity (26%), and a slight reduction of firing activity(10%), which were not reversed by the selective 5-HT2A antagonist M 100,907. Neither the lack ofresponse, nor the magnitude of the response to theNMDA antagonist, could be predicted from the firingcharacteristics of the neurons in basal conditions. Forinstance, a neuron with low firing and low burst ac-tivity could be as responsive to the drug as a neuronexhibiting a high firing and high bursting activity.Acute MK 801 also produces a nonsignificant trend(P < 0.1) towards an increase in the number of spikesper burst and a slight decrease in the variation coeffi-cient (not shown).

Fig. 2. Mean firing activity (A) and mean burst activity (B) ofVTA DA neurons in control rats subsequently treated with twodoses of MK 801 (0.5 mg/kg, i.v.), followed by a dose of the 5-HT2Aantagonist M 100,907 (0.4 mg/kg, i.v.). The same neurons wererecorded in the complete sequence. The number shown in the top ofeach figure represents the number of neurons tested. The figureshows that both firing and burst activity are significantly enhancedby MK 801. The activatory effect of MK 801 on burst activity, butnot on firing activity, was partially reversed by the selective 5-HT2Aantagonist. ** P < 0.001, * P < 0.01 compared with respective pre-treatment value, þP < 0.05 compared with corresponding valuesobtained immediately prior M 100,907 administration, Newman–Keuls test after significant repeated measure ANOVA.



Acute reversal of MK 801-induced effects by5-HT2A antagonist and lack of effects of

glycineB agonists

The acute i.v. administration of D-serine (n ¼ 5) orsarcosine (a naturally occurring glycine uptake inhibi-tor, n ¼ 3), at doses up to 300 mg/kg (i.v.), did notchange the firing rate or burst firing of dopaminergicneurons in the VTA (data not shown), nor did itreverse the effect of MK 801 on dopaminergic neuronactivity (n ¼ 5, data not shown). However, the selec-tive 5-HT2A antagonist M 100,907 (0.4 mg/kg, i.v.),which had no effect on firing and burst activity on itsown (n ¼ 4, data not shown) was found to produce apartial but significant reversal of the increase inburst activity induced by MK 801, without alteringthe effect of MK 801 on firing activity (Figs. 1 and 2).

Acute neuroleptic administration

The acute administration of haloperidol (1 mg/kg,i.p.) administered 30–60 min before the recordingmeasurement, did not alter the sensitivity of VTA DAneurons to acute MK 801 (Fig. 3). The acute adminis-

tration of the atypical neuroleptic olanzapine (10 mg/kg, i.p.) administered 30–60 min before the electro-physiological recording does not prevent the effect ofacute MK 801 on firing activity. However, burst activ-ity was no longer significantly increased followingMK 801 administration, though MK 801 was stillincreasing burst activity by more than 15% in fourneurons out of the nine tested (Table I).

Effect of chronic antipsychotic administration

The sensitivity of VTA DA cells to the administra-tion of MK 801 was dramatically altered following thedifferent chronic antipsychotic treatments (Fig. 4).

In contrast to what was observed with the controlgroup, i.v. administration of MK 801 lacked any effectin the haloperidol-treated group. Only one cell out ofthe seven neurons tested in this group shows anenhancement of firing and burst activity by morethan 15% following MK 801 administration (Table I).Three cells out of the seven tested showed a signifi-cant reduction (more than 15%) in their bursting ac-tivity, with little change on their firing activity (TableI). Similarly, in the combination treatment group, themean firing and burst activity of DA neurons did notincrease in response to the two challenge doses of MK801 in all but one cell (Fig. 4). However, it was foundthat a significant proportion (5/8) of the neuronestested had their bursting activity reduced followingthe administration of the drug (Table I). The decreasewas more dramatic in three neurones (more than 50%of their baseline levels, see example in Fig. 5). If thedata from the haloperidol and haloperidol/D-serinegroups were combined together, the reduction of burstactivity induced by MK 801 was significant (P < 0.04,paired Student’s t-test, for the first dose comparedwith the predrug value, n ¼ 15). In the olanzapine-treated group, the mean firing activity and mean per-centage of burst activity remain also unchanged afterthe administration of MK 801 (Fig. 4). However, theindividual neuronal response to MK 801 was morevariable than in the other groups (Table I). Neverthe-less, burst activity decreased by more than 15% infour cells out of the 10 cells investigated.

Contrarily to what was observed in the othergroups of animals treated with antipsychotics, thechronic administration of D-serine only did not alterthe response of DA cells to the acute administrationof MK 801 on firing and burst activity relative to controlcondition (Fig. 4). Four neurons out of five had theirfiring and burst activity increased following MK 801.

Effect of chronic D-serine supplementation onfiring characteristics of DA cells

Chronic treatments consisting of D-serine alone didnot change the number of spontaneously active cells

Fig. 3. Effect of MK 801 (0.5 mg/kg, i.v.) on the mean firing ac-tivity (A) and mean burst activity (B) of VTA DA neurons in controlrats and in rats acutely treated with haloperidol (1 mg/kg, i.p.) andolanzapine (10 mg/kg, i.p.). The same neurons were recorded in thecomplete sequence. The number shown in the top of each pair ofhistograms represents the number of neurons tested. The figureshows that both firing and burst activity are significantly enhancedby MK 801 in control rats and in rats pretreated with haloperidol.On the other hand, no significant effects of MK 801 on burst activitywere observed in olanzapine pretreated rats. ***P < 0.001, **P < 0.01,*P < 0.04, compared with respective controls, paired Student’s t-test.



encountered per descent into the VTA relative to con-trol (control group: 2.02 6 0.21 cell/track, n ¼ 38; D-serine group: 1.5 6 0.23 cell/track, n ¼ 24). Whenhaloperidol alone was chronically administered with adose of 1 mg/kg/day, there was a significant reductionin the number of spontaneously active cells by amean value of 42% compared with controls (haloperi-dol: 1.18 6 0.162; n ¼ 49; P < 0.002, as comparedwith control, Student’s t-test). A further significantdecrease is observed when haloperidol was given incombination with D-serine; the number of spontane-ously active cells decreasing to a value 51% lowerthan the control levels (combination treatment: 1.006 0.127 cell/track; n ¼ 65; P < 0.00001, unpaired Stu-dent’s t-test, when compared with controls). However,there were no significant differences between thecombination treatment and the haloperidol-only treat-ment groups (P ¼ 0.37, Student’s t-test). Mean firing

and burst activity were not significantly different inthe different treatment groups (not shown).

DISCUSSION

The data presented here confirms previous observa-tions that the NMDA antagonist MK 801 increasedthe firing rate and burst activity of midbrain dopami-nergic neurons (French and Ceci, 1990). This effecthas also been demonstrated with other noncompeti-tive open channel blockers, e.g., PCP and ketamine(French and Ceci, 1990; French et al., 1993), in con-trast to the lack of effects observed with competitiveantagonists (Connelly and Shepard, 1997; Frenchet al., 1993). The increase in firing and burst activitymay well be related to the psychotomimetic effect ofMK 801. Behavioral studies have shown that theresponses produced by the systemic administration of

Fig. 4. Effect of 2 doses of MK 801 (0.5 mg/kg, i.v.) on the meanfiring activity (A) and mean burst activity (B) of VTA DA neuronsin control rats and in rats chronically treated with haloperidol (1mg/kg/day, 21 days), haloperidol/D-serine combination (1/300 mg/kg/day, 21 days), olanzapine (10 mg/kg/day, 21 days), and D-serine only(300 mg/kg/day, 21 days). The number shown in the top of eachgroup of histograms represents the number of neurons tested. For

each different treatment group, the same neurons were recorded inthe complete sequence. The figure shows that both firing and burstactivity are significantly enhanced in both control and D-serinetreated rats, but not in rats treated with the different neurolepticregimen. **P < 0.01, *P < 0.05, compared with respective controls,Newman–Keuls test after significant repeated measure ANOVA.



MK 801 closely resemble those produced by the localapplication of a low dose of MK 801 directly into theVTA, suggesting that the VTA is the primary site ofaction of MK 801 (Narayanan et al., 1996). In paral-lel, neurochemical studies have found that the stimu-lation of the release of DA in the nucleus accumbensinduced by systemic administration of MK 801 occursin a manner that is directly dependent on the impulseactivity of VTA DA neurons (Mathe et al., 1999).

The circuitry involved in the stimulatory effectsseen after systemic administration of MK 801 is notfully understood. Locally, there is probably a loss ofinhibitory control from VTA GABAergic interneuronsthat are tonically stimulated by ambient glutamateacting on NMDA receptors. In support of this arefindings that the systemic administration of MK 801reduces the spontaneous activity of nondopaminergiccells in the VTA, which are believed to be GABAergicneurons (Steffensen et al., 1998; Stobbs et al., 2004;Zhang et al., 1993).

However, MK 801 not only increased firing activitybut also burst firing. Bursting activity of dopaminer-gic cells is an index of dopaminergic neuronal activa-tion associated with increased in DA release in termi-nal areas (see Moore et al., 1999, for review). Burstactivity is thought to be predominantly driven by acti-vation of specific excitatory receptors on DA cells,including principally the different ionotropic glutama-tergic receptor subtypes that tonically activate DAneurons (Chergui et al., 1993), though muscariniccholinergic receptors (Gronier and Rasmussen, 1998),and norepinephrine alpha one receptors (Grenhoffand Svensson, 1993) may also participate.

In the present study we demonstrated that 5-HT2Areceptor stimulation may be accounted for increasingburst activity following the acute administration ofMK 801. When the selective 5-HT2A antagonistM 100,907 was administered by i.v. bolus (400 mg/kg)it produced no effect on baseline firing of dopaminer-gic cells. However, when applied after MK 801, it was

Fig. 5. A: Example of firing rate histogram showing theresponse of a VTA DA neuron from a rat subchronically treatedwith the combination haloperidol/D-serine (1/300 mg/kg/day) to thesubsequent administration of two doses of the NMDA antagonistMK 801 (0.5 mg/kg, i.v.) followed by two dose of D-serine (200 mg/kg, i.v.). The dopaminergic identity of this neuron was confirmed bythe marked inhibitory effect induced by a high dose of the DA ago-nist apomorphine. B: Representative burst train of the same neuron

corresponding to 1 min recording obtained in baseline condition andafter the i.v. administration of the second dose of MK 801 (1 mg/kg).C: Interspike interval histograms of the same neuron representedabove assessed before (100–200 s) and after (400–500 s) the admin-istration of MK 801. Despite MK 801 has a modest inhibitory effecton the firing activity, it substantially reduces the burst activity inthis particular neuron.



able to reverse, at least partially, the increase inburst activity produced by the NMDA antagonist.However, some DA cells (4/9) acutely pretreated withthe atypical antipsychotic olanzapine had still theirburst activity stimulated by MK 801, despite this an-tipsychotic is a strong 5-HT2 antagonist. In addition,neither M 100,907, nor acute olanzapine, did reverseor prevent the activation of firing induced by MK 801.These data suggests that central 5-HT2A receptoractivation is only one of the different components ofthe neuronal circuitry that drives the activation ofDA cells following MK 801 administration. Neverthe-less, elevation of burst activity, restrained by 5-HT2Aantagonists, may be associated with some of the be-havioral effects of the NMDA antagonists. Interest-ingly, previous behavioral studies have found thatsome of the effects produced by acute administrationof noncompetitive NMDA antagonists, such as psycho-motor stimulation, prepulse inhibition deficit, ‘‘impulsetype’’ behavior, social withdrawal, were abolished (orprevented) by the administration of M 100,907 (Bakshyand Geyer, 1995; Carlsson, 1995; Corbett et al., 1995;O’Neill & Shaw, 1999; Varty et al., 1999) or by atypi-cal neuroleptics such as clozapine or olanzapine(Ninan and Kulkarni, 1999, Corbett 1995). In keepingwith these behavioral studies, M 100,907 has beenfound to prevent the stimulatory effect of MK 801or PCP on DA release in the nucleus accumbens(Schmidt and Fadayel, 1996). However, no positiveclinical data on schizophrenic patients treated withsole 5-HT2A antagonists have been published so far.

Administration of MK 801 may at one site in thebrain increase the release of serotonin, which in turnactivates excitatory 5-HT2A receptors and, eitherdirectly or indirectly, activates the burst activity ofDA cells. It can be envisioned that the circuitryinvolved arises from the dorsal raphe nucleus (DRN)serotonergic neurons that innervate the frontal cor-tex. In the DRN, MK-801 has been found to dramati-cally increase the firing activity of serotonergic neu-rons (Lejeune et al., 1994), and this may produce anincrease of serotonin release within the PFC(Kashiwa et al., 1995). Pyramidal cortical neuronspossess a high density of excitatory 5-HT2A receptors(Cornea-Hebert et al., 2002; Amargos-Bosch et al.,2004). In addition, it has recently been found that thelocal application of DOI in the prefrontal cortex canproduce an increase in firing and burst activity ofVTA dopaminergic cells (Bortholozzi et al., 2005).There is also evidence that MK 801 activates a gluta-mate excitatory input to the VTA as infusion ofAMPA or kainate antagonists into the VTA preventsthe increase in DA release in the accumbens (Matheet al., 1998). However, there is no indication that thisexcitatory drive comes from the prefrontal cortex.

The results of this study show that D-serine byitself is not capable of reversing any of the effects of

MK 801. As MK 801 is a noncompetitive antagonist,increasing the concentration of the endogenous ago-nist D-serine may not be able to overcome the inhibi-tory effect of the channel blocker. However, otherinvestigators (Javitt et al., 2004) have found thatsome of the behavioral or neurochemical effects elic-ited by PCP can be prevented by high doses of gly-cine, or by glycine uptake inhibitors, e.g., NFPS.

Demonstrated here for the first time is the observa-tion that chronically treating animals with the typicalantipsychotic haloperidol, as well as with the atypicalantipsychotic olanzapine, prevents the activation offiring and burst activity of DA cells induced by MK801. Therefore, during a chronic treatment with neu-roleptics, the neuronal circuitry induced by MK 801and producing the activation of dopaminergic neuro-nal activity has become less effective. Both typicaland atypical neuroleptic chronic treatments, as wellas the combination therapy, may act via a commonmechanism by altering the glutamatergic excitatorycontrol over DA cell activity. It is possible that thisaction contributes to attenuate the manifestation ofpositive symptoms, on which all types of neuroleptics,when administered chronically, have an effect. Thisstands in contrast to results obtained when thesedrugs were administered acutely.

It cannot be ruled out that neuroleptic treatmentscan exert a direct influence on the NMDA receptor(Bennett and Gronier, 2005), or alter the distributionof the different NMDA receptors involved in theaction of MK 801 on DA cells (Schmitt et al., 2003;Spurney et al., 1999).

These different types of antipsychotic treatmentsmay also act by dramatically altering the modulatoryaction of the GABAergic interneurons that controlmidbrain DA neurons firing. Some GABA neuronshave their firing activity markedly stimulated bylocally applied DA and dramatically reduced byNMDA antagonists (Steffensen et al., 1998; Stobbset al., 2004). Consequently, the permanent blockadeof DA receptors during a chronic treatment with neu-roleptics may have induced an important reduction inthe firing activity of these GABA neurons. In addi-tion, the process of depolarization/inactivation ofsome DA neurons, which is known to occur duringchronic typical and atypical antipsychotic treatments(Chiodo and Bunney, 1983; Skarsfeldt, 1995) and dur-ing the combination haloperidol/D-serine treatment(our study), would have produced a significant deple-tion in the local level of DA, which may furtherreduce the basal activity of the GABA interneuronsand altered their capacity to regulate the firing activ-ity of midbrain DA neurons. Interestingly, it was inthe two groups of animals chronically treated withhaloperidol, which is a more potent DA receptor an-tagonist than olanzapine, where the neurons appearto be the more insensitive to MK 801. Further studies



are needed to better characterize the chronic effectsof antipsychotic treatment on the activity of VTAinterneurons and their possible modulatory role onthe stimulatory effects of MK 801.

After the chronic treatment with the combined re-gime of haloperidol and D-serine, the stimulatingeffects of MK 801 were also abolished. Additionally, itseemed that more neurons (�20% more) than therewere in the other treatment groups responded to MK801 by an inhibition of their burst activity. Whetherthis stronger inhibitory effect can reflect a better clin-ical response toward some of the symptoms of schizo-phrenia is an interesting question to address.

Our results are in agreement with previous literaturedata showing that the administration of haloperidolreduces the number of spontaneously active DA cells(Chiodo and Bunney, 1983; Grace, 1992). We found thatthe chronic coadministration of D-serine, which had nosignificant effects when administered on its own, withhaloperidol is unable to potentiate significantly theeffect of haloperidol on DA cell activity. Combinationtreatment with typical neuroleptics and D-serine hasbeen shown to significantly improve both the positiveand negative symptoms, as well as cognitive perform-ance of schizophrenics when compared with the mono-therapy with antipsychotics (Tsai et al., 1998). By com-parison, the effects of D-serine on the positive symptomswere modest (17% improvement), in the same range oforder as it further decreased the number of cell/track,though this was not significant in our study. Overall,our study does not formally demonstrate that D-serine,whether administered alone or in combination withhaloperidol, has a significant effect on DA neuronal ac-tivity that could explain its beneficial therapeuticeffects. Finally, it is worth to mention that the lack ofresponse of DA neurons to MK 801 was seen on the DAcells that were apparently not affected by the process ofdepolarisation/inactivation induced by the differentchronic neuroleptic treatments (�50% of the DA neu-rons). These cells may represent a group of neuronswith some other common physiological characteristicsthat it would be particularly interesting to identify.

CONCLUSIONS

The present study shows that the chronic, but notthe acute, administration of antipsychotics sup-presses, or even occasionally reverses, the activatoryresponse of midbrain DA neurones to systemicallyadministered MK-801, suggesting that the function ofthe NMDA receptors that regulate the activity of DAneurons is affected by the chronic administration ofneuroleptics. In addition, we have also demonstratedthat the circuitry involved in the stimulatory effectsof MK 801 requires the participation of 5-HT2Areceptors to elicit burst activation. The latter maycontribute to the behavioral effects of the NMDA

antagonists. Finally, our results also indicate thatadding D-serine to the typical neuroleptic haloperidoldoes not seem to strengthen significantly the modula-tory effect of haloperidol on midbrain DA cells.

REFERENCES

Amargos-Bosch M, Bortolozzi A, Puig MV, Serrats J, Adell A, CeladaP, Toth M, Mengod G, Artigas F. 2004. Co-expression and in vivointeraction of serotonin1A and serotonin2A receptors in pyramidalneurons of prefrontal cortex. Cereb Cortex 14:281–299.

Bakshi VP, Geyer MA. 1995. Antagonism of phencyclidine-induceddeficits in prepulse inhibition by the putative atypical antipsychoticolanzapine. Psychopharmacology (Berl) 122:198–201.

Bennett S, Gronier B. 2005. Modulation of striatal dopamine releasein vitro by agonists of the glycineB site of NMDA receptors; interac-tion with antipsychotics. Eur J Pharmacol 527:52–59.

Bortolozzi A, Diaz-Mataix L, Scorza MC, Celada P, Artigas F. 2005.The activation of 5-HT receptors in prefrontal cortex enhances dopa-minergic activity. J Neurochem 95:1597–1607.

Carlsson ML. 1995. The selective 5-HT2A receptor antagonist MDL100,907 counteracts the psychomotor stimulation ensuing manipula-tions with monoaminergic, glutamatergic or muscarinic neurotrans-mission in the mouse–implications for psychosis. J Neural TransmGen Sect 100:225–237.

Chergui K, Charlety PJ, Akaoka H, Saunier CF, Brunet JL, Buda M,Svensson TH, Chouvet G. 1993. Tonic activation of NMDA receptorscauses spontaneous burst discharge of rat midbrain dopamine neu-rons in vivo. Eur J Neurosci 5:137–144.

Chiodo LA, Bunney BS. 1983. Typical and atypical neuroleptics: Dif-ferential effects of chronic administration on the activity of A9 andA10 midbrain dopaminergic neurons. J Neurosci 3:1607–1619.

Connelly ST, Shepard PD. 1997. Competitive NMDA receptor antago-nists differentially affect dopamine cell firing pattern. Synapse25:234–242.

Corbett R, Camacho F, Woods AT, Kerman LL, Fishkin RJ, Brooks K,Dunn RW. 1995. Antipsychotic agents antagonize non-competitiveN-methyl-D-aspartate antagonist-induced behaviors. Psychopharma-cology (Berl) 120:67–74.

Cornea-Hebert V, Riad M, Wu C, Singh SK, Descarries L. 2002. Simi-lar ultrastructural distribution of the 5-HT2A serotonin receptorand microtubule-associated protein MAP1A in cortical dendrites ofadult rat. Neuroscience 113:23–35.

French ED, Ceci A. 1990. Non-competitive N-methyl-D-aspartateantagonists are potent activators of ventral tegmental A10 dopa-mine neurons. Neurosci Lett 119:159–162.

French ED, Mura A, Wang T. 1993. MK-801, phencyclidine (PCP),and PCP-like drugs increase burst firing in rat A10 dopamine neu-rons: Comparison to competitive NMDA antagonists. Synapse13:108–116.

Grace AA. 1992. The depolarization block hypothesis of neurolepticaction: Implications for the etiology and treatment of schizophrenia.J Neural Transm Suppl 36:91–131.

Grace AA, Bunney BS. 1984a. The control of firing pattern in nigraldopamine neurons: Single spike firing. J Neurosci 4:2866–2876.

Grace AA, Bunney BS. 1984b. The control of firing pattern in nigraldopamine neurons: Burst firing. J Neurosci 4:2877–2890.

Grenhoff J, Svensson TH. 1993. Prazosin modulates the firing patternof dopamine neurons in rat ventral tegmental area. Eur J Pharma-col 233:79–84.

Gronier B, Rasmussen K. 1998. Activation of midbrain presumed do-paminergic neurons by muscarinic cholinergic receptors: An in vivoelectrophysiological study in the rat. Br J Pharmacol 124:455–464.

Imperato A, Scrocco MG, Bacchi S, Angelucci L. 1990. NMDA recep-tors and in vivo dopamine release in the nucleus accumbens andcaudatus. Eur J Pharmacol 187:555–556.

Javitt DC. 1999. Treatment of negative and cognitive symptoms. CurrPsychiatry Rep 1:25–30.

Javitt DC, Zukin SR. 1991. Recent advances in the phencyclidinemodel of schizophrenia. Am J Psychiatry 148:1301–1308.

Javitt DC, Balla A, Burch S, Suckow R, Xie S, Sershen H. 2004.Reversal of phencyclidine-induced dopaminergic dysregulation by

N-methyl-D-aspartate receptor/glycine-site agonists. Neuropsycho-pharmacology 29:300–307.

Kashiwa A, Nishikawa T, Nishijima K, Umino A, Takahashi K. 1995.Dizocilpine (MK-801) elicits a tetrodotoxin-sensitive increase inextracellular release of dopamine in rat medial frontal cortex. 1995.Neurochem Int 26:269–279.

Lane HY, Chang YC, Liu YC, Chiu CC, Tsai GE. 2005. Sarcosine orD-serine add-on treatment for acute exacerbation of schizophrenia:



A randomized, double-blind, placebo-controlled study. Arch Gen Psy-chiatry 62:1196–1204.

Lejeune F, Gobert A, Rivet JM, Millan MJ. 1994. Blockade of trans-mission at NMDA receptors facilitates the electrical and syntheticactivity of ascending serotoninergic neurons. Brain Res 656:427–431.

Lindsley CW, Shipe WD, Wolkenberg SE, Theberge CR, Williams DLJr, Sur C, Kinney GG. 2006. Progress towards validating theNMDA receptor hypofunction hypothesis of schizophrenia. Curr TopMed Chem 6:771–785.

Mathe JM, Nomikos GG, Schilstrom B, Svensson TH. 1998. Non-NMDA excitatory amino acid receptors in the ventral tegmentalarea mediate systemic dizocilpine (MK-801) induced hyperlocomo-tion and dopamine release in the nucleus accumbens. J NeurosciRes 51:583–592.

Mathe JM, Nomikos GG, Blakeman KH, Svensson TH. 1999. Differ-ential actions of dizocilpine (MK-801) on the mesolimbic and meso-cortical dopamine systems: Role of neuronal activity. Neuropharma-cology 38:121–128.

Millan MJ. 2002. N-methyl-D-aspartate receptor-coupled glycineBreceptors in the pathogenesis and treatment of schizophrenia: Acritical review. Curr Drug Targets CNS Neurol Disord 1:191–213.

Moore H, West AR, Grace AA. 1999. The regulation of forebrain dopa-mine transmission: Relevance to the pathophysiology and psychopa-thology of schizophrenia. Biol Psychiatry 46:40–55.

Narayanan S, Willins D, Dalia A, Wallace L, Uretsky N. 1996. Role ofdopaminergic mechanisms in the stimulatory effects of MK-801injected into the ventral tegmental area and the nucleus accum-bens. Pharmacol Biochem Behav 54:565–73.

Ninan I, Kulkarni SK. 1999. Preferential inhibition of dizocilpine-induced hyperlocomotion by olanzapine. Eur J Pharmacol 368:1–7.

O’Neill MF, Shaw G. 1999. Comparison of dopamine receptor antago-nists on hyperlocomotion induced by cocaine, amphetamine, MK-801 and the dopamine D1 agonist C-APB in mice. Psychopharma-cology 145:237–250.

Paxinos G, Watson C. 1998. The rat brain in stereotaxic coordinates,4th ed. NewYork: Academic Press.

Schmitt A, Zink M, Muller B, May B, Herb A, Jatzko A, Braus DF,Henn FA. 2003. Effects of long-term antipsychotic treatment onNMDA receptor binding and gene expression of subunits. Neuro-chem Res 28:235–241.

Schmidt CJ, Fadayel GM. 1996. Regional effects of MK-801 on dopa-mine release: Effects of competitive NMDA or 5-HT2A receptorblockade. J Pharmacol Exp Ther 277:1541–1549.

Skarsfeldt T. 1995. Differential effects of repeated administration ofnovel antipsychotic drugs on the activity of midbrain dopamine neu-rons in the rat. Eur J Pharmacol 281:289–294.

Spurney CF, Baca SM, Murray AM, Jaskiw GE, Kleinman JE, HydeTM. 1999. Differential effects of haloperidol and clozapine on iono-tropic glutamate receptors in rats. Synapse 34:266–276.

Steffensen SC, Svingos AL, Pickel VM, Henriksen SJ. 1998. Electro-physiological characterization of GABAergic neurons in the ventraltegmental area. J Neurosci 18:8003–8015.

Stobbs SH, Ohran AJ, Lassen MB, Allison DW, Brown JE, SteffensenSC. 2004. Ethanol suppression of ventral tegmental area GABAneuron electrical transmission involves N-methyl-D-aspartate recep-tors. J Pharmacol Exp Ther 311:282–289.

Tsai G, Yang P, Chung LC, Lange N, Coyle JT. 1998. D-serine addedto antipsychotics for the treatment of schizophrenia. Biol Psychiatry44:1081–1089.

Varty GB, Bakshi VP, Geyer MA. 1999. M100907, a serotonin 5-HT2Areceptor antagonist and putative antipsychotic, blocks dizocilpine-induced prepulse inhibition deficits in Sprague-Dawley and Wistarrats. Neuropsychopharmacology 20:311–321.

Yan QS, Reith ME, Jobe PC, Dailey JW. 1997. Dizocilpine (MK 801)increases not only dopamine but also serotonin and norepinephrinetransmissions in the nucleus accumbens as measured by microdial-ysis in freely moving rats. Brain Res 765:149–158.

Zhang JL, Chiodo LA, Freeman AS. 1993. Effects of phencyclidine,MK-801 and 1,3-di(2-tolyl)guanidine on non-dopaminergic midbrainneurons. Eur J Pharmacol 230:371–374.



Effects of antipsychotic treatments and D-serine supplementation on the electrophysiological...

Documents

Transcript of Effects of antipsychotic treatments and D-serine supplementation on the electrophysiological...