Journal of Neuroscience Research 51:583592 (1998)

Non-NMDA Excitatory Amino Acid Receptors

in the Ventral Tegmental Area Mediate
Systemic Dizocilpine (MK-801) Induced
Hyperlocomotion and Dopamine Release
in the Nucleus Accumbens
Jan M. Mathe, George G. Nomikos, Bjorn Schilstrom, and Torgny H. Svensson*
Department of Physiology and Pharmacology, Section of Neuropsychopharmacology,
Karolinska Institutet, Stockholm, Sweden

This study investigated the putative role of nonNMDA excitatory amino acid (EAA) receptors in the
ventral tegmental area (VTA) for the increase in
dopamine (DA) release in the nucleus accumbens
(NAC) and behavioral stimulation induced by systemically administered dizocilpine (MK-801). Microdialysis was utilized in freely moving rats implanted with
probes in the VTA and NAC. Dialysates from the NAC
were analyzed with high-performance liquid chromatography for DA and its metabolites. The VTA was
perfused with the AMPA and kainate receptor antagonist CNQX (0.3 or 1 mM) or vehicle. Forty min after
onset of CNQX or vehicle perfusion of the VTA,
MK-801 (0.1 mg/kg) was injected subcutaneously.
Subsequently, typical MK-801 induced behaviors were
also assessed in the same animals by direct observation. MK-801 induced hyperlocomotion was associated with a 50% increase of DA levels in NAC
dialysates. Both the MK-801 evoked hyperlocomotion
and DA release in the NAC was antagonized by CNQX
perfusion of the VTA in a concentration-dependent
manner. None of the other rated MK-801 evoked
behaviors, e.g. head weaving or sniffing, were affected
by CNQX perfusion of the VTA. By itself the CNQX
or vehicle perfusion of the VTA alone did not affect DA
levels in NAC or any of the rated behaviors. These
results indicate that MK-801 induced hyperlocomotion and DA release in the NAC are largely elicited
within the VTA via activation of non-NMDA EAA
receptors, tentatively caused by increased EAA release. Thus, the locomotor stimulation induced by
psychotomimetic NMDA receptor antagonists may
not only reflect impaired NMDA receptor function,
but also enhanced AMPA and/or kainate receptor
activation in brain, e.g., in the VTA. In view of their
capacity to largely antagonize the behavioral stimulation induced by psychotomimetic drugs, such as MK-801,

r 1998 Wiley-Liss, Inc.

AMPA, and/or kainate receptor antagonists may possess antipsychotic efficacy. J. Neurosci. Res. 51:583
592, 1998. r 1998 Wiley-Liss, Inc.
Key words: CNQX; MK-801; locomotion; ventral
tegmental area; dopamine
INTRODUCTION
A drug-related psychosis remarkably similar to
schizophrenia induced by phencyclidine (PCP) has been
described in numerous studies since its introduction as a
general anesthetic in the mid-1950s. Unlike the amphetamine-induced paranoid psychosis, the PCP-induced psychosis may exhibit both positive symptoms of schizophrenia, such as auditory hallucinations and paranoid delusions,
as well as negative symptoms, i.e., flattening of affect,
and impaired attention and motivation. The PCP-induced
psychosis also frequently incorporates the formal thought
disorder and neuropsychological deficits that commonly
are associated with schizophrenia (see Javitt and Zukin,
1991).
When administered in psychotomimetic doses, PCP
acts primarily as a non-competitive antagonist of N-methyl-

Contract grant sponsor: Swedish Medical Research Council; Contract

grant numbers: 4747, 11026; Contract grant sponsor: Karolinska
Institutet; Contract grant sponsor: Fredrik och Ingrid Thuring Stiftelse;
Contract grant sponsor: Swedish Society for Medical Research;
Contract grant sponsor: AB LEOs i Helsingborg Stiftelse for Forskning; Contract grant sponsor: Janssen Pharmaceutica, Beerse; Contract
grant sponsor: The Council for Tobacco Research, USA, Inc.; Contract
grant sponsor: Astra Arcus AB, Sodertalje.
*Correspondence to: Torgny H. Svensson, Department of Physiology
and Pharmacology, Section of Neuropsychopharmacology, Karolinska
Institutet, S-171 77 Stockholm, Sweden. E-mail: torgny.svensson@fyfa.ki.se
Received 27 August 1997; Revised 22 October 1997; Accepted 27
October 1997

584

Mathe et al.

D-aspartate receptors (see Javitt and Zukin, 1991). Other

non-competitive NMDA receptor antagonists, such as
ketamine and memantine, as well as competitive NMDA
receptor antagonists, have also been reported to possess
psychotomimetic properties, which appear well correlated to their affinity for NMDA receptors (see Kornhuber
and Weller, 1995). Thus, NMDA receptor antagonism
seems primarily to underlie the psychotomimetic properties of PCP-like drugs.
When administered systemically to rodents, noncompetitive NMDA receptor antagonists, including PCP
and the approximately ten-fold more potent and selective
NMDAantagonist dizocilpine [MK-801, (1)-5-methyl-10,11dihydroxy-5H-dibenzo-(a,d)cyclohepten-5,10-imine) maleate, see Wong et al., 1986], evoke characteristic behavioral effects such as hyperlocomotion, head weaving,
sniffing, and, in high doses, ataxia (Clineschmidt et al.,
1982; Loscher et al., 1992). This behavioral syndrome in
rodents has been proposed as an animal model of
schizophrenia (Contreras et al., 1987) and, accordingly,
major components of this syndrome can be antagonized
by catecholamine depletion or by administration of
dopamine (DA) receptor antagonists (Fessler et al., 1980;
Clineschmidt et al., 1982; Criswell et al., 1993; Willins et
al., 1993). In addition, the behavioral effects of systemic
administration of PCP and MK-801 have been reported to
be antagonized by local, bilateral microinjections of the
g-amino-butyric acid (GABA)B receptor agonist baclofen
into the ventral tegmental area (VTA; Narayanan et al.,
1996), which inhibits DA neuronal activity (Grace and
Bunney, 1980; Seutin et al., 1994). Thus, activation of the
mesolimbic and mesocortical DA systems and, specifically, the DA cell body region within the VTA appears to
play a crucial role in the behavioral stimulation induced
by PCP-like drugs.
In accordance with the above findings, previous
electrophysiological experiments demonstrate that both
PCP and MK-801 evoke essentially identical effects on
VTA DA neuronal activity (French, 1986; Pawlowski et
al., 1990; Rouillard et al., 1990; Zhang et al., 1992;
Murase et al., 1993). Our subsequent experiments have
revealed that the drugs induce a high-frequency firing
pattern in DA cells, which largely project subcortically,
e.g., to the nucleus accumbens (NAC), but reduce burst
firing in DA neurons projecting primarily to prefrontal
cortex (PFC; Pawlowski et al., 1990; Murase et al., 1993).
Recent microdialysis studies in freely moving animals
have demonstrated that MK-801 evokes a long-lasting
increase in DA output within the terminal regions of the
mesocorticolimbic and the nigrostriatal dopamine systems (Wolf et al., 1993; Wedzony et al., 1993; Mathe et
al., 1996a; Miller and Abercrombie, 1996).
The aforementioned stimulation of mesolimbic DA
activity by potent NMDA receptor antagonists is, how-

ever, somewhat at variance with the wide body of

evidence which unequivocally demonstrates that glutamate and various excitatory amino acid (EAA) receptor
agonists also augment central DA activity (Scarnati and
Pacitti, 1982; Chergui et al., 1993; Kalivas et al., 1989;
Schilstrom et al., 1997). Moreover, microiontophoretic
application of PCP-like NMDA receptor antagonists
reportedly does not affect the neuronal activity of midbrain DA cells (French, 1986; Rouillard et al., 1990;
Zhang et al., 1992). Therefore, the stimulant action of
PCP-like NMDA receptor antagonists on central DA
systems is likely to be mediated via indirect mechanisms
(Freeman and Bunney, 1984; Mathe et al., 1996a,b;
Svensson et al., 1997).
Utilizing single unit recording techniques in vivo,
we previously observed a high-frequency firing pattern in
VTA DA neurons evoked by systemic PCP or MK-801,
which, when analyzed utilizing the criteria of Grace and
Bunney (1984), was interpreted as continuously bursting
(Pawlowski et al., 1990; Murase et al., 1993). This firing
pattern was, however, largely devoid of typical post-burst
pauses in DA neuronal activity and the characteristic,
progressive reduction of spike amplitude during each
burst, as we have previously observed e.g. after local
application of NMDA onto DA cells (Chergui et al.,
1993). Instead, the firing pattern was quite similar to the
high-frequency non-bursting firing pattern observed after
local application of quisqualate or kainate, a-amino-3hydroxy-5-methyl-4-isoxazole proprionic acid (AMPA)
and kainate receptor agonists, onto the DA neurons
(Mathe et al., 1996b; Svensson et al., 1997).
In view of the above findings, we hypothesized that
blockade of NMDA receptors by PCP-like drugs may,
indirectly, lead to a relative hyperfunction of AMPA
and/or kainate receptors within the VTA, an effect that
might cause both stimulation of DA release in the NAC,
as well as the associated hyperlocomotion evoked by
systemically administered PCP-like NMDA receptor antagonists. In order to address this question, we have here
utilized a dual-probe microdialysis technique in freely
moving animals, with local perfusion with the AMPA and
kainate receptor antagonist 6-cyano-7-nitroquinoxaline2,3-dione (CNQX) of the VTA and simultaneous measurement of extracellular DA levels in the NAC during
systemic administration of MK-801. In addition, typical
MK-801 induced behavioral effects, e.g., hyperlocomotion and head weaving, were also assessed in the same
animals by means of direct observation.
METHODS
Animals
Male Bkl:WR rats weighing between 250350 g
were used (Bantin and Kingman Universal AB). Animals

MK-801 Effects Caused by AMPA Receptors in VTA

arrived at least 1 week before use and were housed five

per cage under standard laboratory conditions. They were
maintained on a 12 hr light:dark cycle with lights on at
06:00 and had access to R34 rat chow and water ad
libitum. Only experimentally naive rats were used. All
animals were handled for at least 5 min per day during the
last 3 days before experiments.
Surgery and Microdialysis Experiments
Rats were anesthetized with sodium pentobarbital,
60 mg/kg, injected intraperitoneally. They were subsequently mounted in a stereotaxic apparatus and their body
temperature was maintained at 37C with a thermostatically controlled heating pad. In each animal, one vertical
probe of concentric type was stereotaxically implanted
into the NAC, AP 1 1.6 mm, ML 21.4 mm, and DV
28.2 mm relative to bregma. In addition, a second
vertical microdialysis probe was stereotaxically implanted within the ipsilateral VTA, AP 1 3.0 mm, ML
20.6 mm, and DV 28.5 mm relative to lambda. Dialysis
occurred through a 2.25 mm semipermeable membrane
for the NAC probe, and through a 1.0 mm membrane for
the VTA probe (AN69 Hospal; copolymer of acrylonitrile
and sodium methallyl sulphonate, i.d. 5 0.24 mm, 40,000
Da molecular weight cutoff).
After surgery, the animals were housed individually
in plexiglass cages (32 3 35 3 50 cm) and given ad
libitum access to food and water. All experiments commenced approximately 48 hr after surgery in awake,
freely moving animals during the mid-day. After four
consecutive stable baseline dialysates were obtained
(,10% variation), typically about 4 hr after commencement of perfusion, rats were perfused with vehicle or
CNQX through the VTA probe (cf. below). Forty minutes
later, rats were subcutaneously (s.c.) injected with MK801 (0.1 mg/kg) or saline. Treatment groups consisted of
57 rats each.
Microdialysis was performed using automated online sampling (Nomikos et al., 1989, 1994). The dialysis
probe was perfused with a solution containing 147 mM
sodium chloride, 3.0 mM potassium chloride, 1.3 mM
calcium chloride, 1.0 mM magnesium chloride, and 1.0
mM sodium phosphate, pH 7.4, at a rate of 2.5 l/min set
by a microinfusion pump (Univentor 684). The perfusate
was loaded directly into the sample loop of the injector
(Valco Instruments, Houston, TX). Samples were automatically injected into the analytical system every 20
min. An IBM compatible computer controlled the loading
and injection modes of the injector, through use of
Turbochrom software (Perkin Elmer, Oak Brook, IL).
Upon completion of the experiments, rats were killed by
an overdose of sodium pentobarbital and the brains were
removed and stored in 5% formaldehyde and 25%
sucrose. Each brain was sectioned on a microtome (50

585

m), stained with neutral red, and examined for probe

placement. Only rats with probes verified to be located
within both the nucleus accumbens and ventral tegmental
area, according to the anatomical atlas of Paxinos and
Watson (1986), were included in this study.
Biochemical Assay
Concentrations of DA, its metabolites dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA)
were determined by high-performance liquid chromatography (HPLC) with electrochemical detection as described previously (Nomikos et al., 1994). The detection
limit of this assay is approximately 1 fmol for DA and
DOPAC, and 3 fmol for HVA. Separation of DA and the
acid metabolites was achieved by reverse-phase liquid
chromatography (column: 150 3 4.6 mm, Nucleosil 5
m, C18) with a mobile phase consisting of 0.055 M
sodium acetate with 0.1 mM octanesulfonic acid, 0.01
mM Na2EDTA, and 5% methanol, pH 3.8, adjusted with
glacial acetic acid. The mobile phase was delivered by an
HPLC pump (LKB; 2150) at 0.8 ml/min. Chromatograms
were recorded by an IBM compatible computer as well as
by a two-pen chart recorder (Kipp and Zonen).
Behavioral Experiments
Typically, rats begin to display MK-801 evoked
behaviors at about 10 min after subcutaneous administration of this compound (unpublished observations). Thus,
10 min after systemic (MK-801) or local (CNQX) drug
administration and throughout the microdialysis experiments, the behavior of the above animals was continuously observed to ascertain any overt behavioral changes.
Characteristic MK-801 induced behaviors (Clineschmidt
et al., 1982; Loscher et al., 1992) were rated by two
independent observers and were measured by the time
during which the rats exhibited the following behaviors:
locomotor activity, ipsi- and contra-lateral turning, sniffing, head weaving, and ataxia. Head weaving was determined as repetitive side-to-side, left-to-right swaying
movements of the head and upper torso. Ataxia was
defined as tottering of the hindquarters, abduction and
dragging of hindlimbs, flat body posture, and loss of
balance (Loscher et al., 1992). All behavioral monitoring
was conducted during the light phase of the daily cycle
between 12:00 and 17:00.
Drug Treatment
(1)-MK-801 hydrogen maleate [(1)-5-methyl10,11-dihydroxy-5H-dibenzo-(a,d)cyclohepten-5,10-imine hydrogen maleate, Research Biochemicals, Inc.,
Natick, MA] was dissolved in 0.9% sodium chloride and
administered at a dose of 0.1 mg/kg. Saline solution refers

586

Mathe et al.

to a physiological 0.9% sodium chloride aqueous solution. MK-801 and saline were administered subcutaneously (s.c.) in a volume of 1 ml/kg.
CNQX (6-cyano-7-nitroquinoxaline-2,3-dione, Research Biochemicals, Inc., Natick, MA) was dissolved to
a concentration of 100 mM in pure dimethylsulfoxide
(DMSO). Immediately before use, this CNQX stock was
diluted with perfusion solution (cf. above) to a concentration of 1 mM or 300 M CNQX. Subsequently, CNQX
was administered by reverse microdialysis perfusion
through the VTA probe. These concentrations were
chosen considering the approximately 0.5% in vitro
recovery of this probe type for similar compounds
(Marshall et al., 1997), which would thus deliver about
510 M to the surrounding brain tissue, a concentration
which corresponds to functionally relevant concentrations that block AMPA and kainate, but not NMDA
evoked responses in brain slice preparations (Mercuri et
al., 1996). In addition, we have recently confirmed the
pharmacological and functional specificity of CNQX
administered by virtually identical reverse microdialysis
techniques (Schilstrom et al., 1997).
Vehicle perfusion refers to perfusion solution plus
1% DMSO, corresponding to the highest concentration of
DMSO in CNQX-treated animals. Vehicle was administered identically to CNQX, i.e., by reverse microdialysis
perfusion via the VTA probe.
Data Analysis
Microdialysis data were analyzed utilizing Turbochrom software (Perkin Elmer, Oak Brook, IL). For
graphical representation of microdialysis data, the average of four baseline samples immediately preceding drug
injection was defined as 100%. All subsequent measurements were transformed to a mean percentage of baseline
values for each subsequent 20 min sampling period (Fig.
1). For calculation of overall output of DA, DOPAC, and
HVA, the mean percentage values over 140 min after drug
administration were calculated (Fig. 2). Behavioral data
are presented as raw time duration values of locomotor
activity, turning, head weaving, sniffing and ataxia (Fig.
3).
Statistical evaluation was performed using the Statistica software suite (StatSoft, Inc., Tulsa, OK). For
evaluation of temporal changes in dialysate levels of DA
(Fig. 1), a one- and two-way (treatment 3 time) analysis
of variance (ANOVA) with repeated measures was used,
followed by the post-hoc Newman-Keuls test for multiple
comparisons. Overall microdialysate levels of DA,
DOPAC and HVA were analyzed using two-way ANOVA
followed by the Newman-Keuls test (Fig. 2). Behavioral
data were also analyzed by one-way ANOVA and the
post-hoc Newman-Keuls test (Fig. 3). In all statistical
tests, a P value ,0.05 was considered significant.

The experiments were conducted with the in accordance with the guidelines of, and with the consent of, the
local ethical committee (Stockholms Norra och Sodra
Forsoksdjursetiska Kommitteer).

RESULTS
Microdialysis Experiments
There were no significant differences in mean
values of basal DA and its metabolites among the
treatment groups. The overall mean 6 standard error of
the mean (S.E.M.) of basal values of DA, DOPAC and
HVA was 2.17 6 0.22, 740 6 59, and 534 6 51 fmol/min
from the NAC (n534), respectively.
CNQX perfusion of the VTA followed by subcutaneous saline injection (1 ml/kg) did not affect dialysate
concentrations of DA (Fig. 1A), DOPAC, or HVA, nor did
vehicle perfusion of the VTA and saline injection (n55 in
all three groups; data not shown).
Vehicle perfusion of the VTA and systemic MK-801
administration (0.1 mg/kg s.c.) evoked a significant
increase in DA levels in dialysates, (142 6 12% overall,
P,0.05, n57, Figs. 1B,C, 2). MK-801 of 0.1 mg/kg
induced a progressive elevation of dialysate DA levels,
reaching statistical significance 60 min after administration. Dialysate DA levels remained significantly elevated
throughout the experiment, reaching a maximum of
154 6 17% during the 80 min sampling period (P,0.001,
Fig. 1B,C). Dialysate DOPAC and HVA concentrations
increased in response to vehicle and MK-801 (0.1 mg/kg,
131 6 5%, and 110 6 2% overall, P values ,0.001 and
,0.05, respectively; Fig. 2). Generally, DOPAC and HVA
concentrations increased in a time course which parallelled that of dopamine, albeit with few samples delay
(data not shown).
CNQX perfusion of the VTA at a concentration of
300 M partially antagonized systemic MK-801 (0.1
mg/kg, s.c.) evoked DA release during the early period
after MK-801 administration (60 to 100 min) but also
significantly reduced overall DA levels (P values ,0.05
0.01, n56, Figs. 1B, 2). Also, increases in DA levels did
not attain statistical significance after MK-801 injection
(maximum increase 131 6 5%). Furthermore, systemic
MK-801 induced elevations in dialysate levels of DOPAC
were significantly antagonized by local perfusion in the
VTA with CNQX (P,0.05, Fig. 2).
CNQX perfusion of the VTA at a concentration of 1
mM completely abolished MK-801 induced increases in
microdialysate concentrations of DA (P values ,0.05
0.01, n56; Figs. 1C, 2, respectively). In addition, MK801 evoked increases in DOPAC and HVA were significantly antagonized by local perfusion with CNQX in the
VTA (P values ,0.010.001, Fig. 2).

MK-801 Effects Caused by AMPA Receptors in VTA

Fig. 1. Effects of CNQX (300 M and 1 mM) perfusion of the

ventral tegmental area (VTA) alone (A), MK-801 (0.1 mg/kg,
s.c.) during CNQX (300 M) or vehicle perfusion of the VTA
(B), and MK-801 during CNQX (1 mM) or vehicle perfusion of
VTA (C) on extracellular concentrations of dopamine in the
nucleus accumbens, as assessed by microdialysis in freely
moving rats. Data are presented as mean (6 S.E.M., n55-8)
percent changes from basal values before drug injection. Gray
bars indicate VTA perfusion period. Arrow heads indicate
injection of saline or MK-801 (B and C). **P,0.01 and
***P,0.001, vehicle and MK-801 as compared to last baseline
sample. 1P,.05, 11P,.01, and 111P,0.001, vehicle and
MK-801 compared to CNQX and MK-801.

587

Behavioral Experiments
In all rats observed, no spontaneous locomotor
activity or behavioral signs were evident before injection
of MK-801. All rats had been in their microdialysis cages
for at least 4 hr prior to drug administration and were thus
habituated to their environment. In general, before the
commencement of drug administration, rats were either
asleep or resting in their cages.
Vehicle or CNQX perfusion of the VTA or subcutaneous injection of 0.9% sodium chloride (1 ml/kg) did not
cause any visible changes in rat behavior. Typically,
animals resumed resting within 2 min after being lifted
from their cages and injected with saline (data not
shown).
MK-801 administration during vehicle perfusion of
the VTA induced typical behavioral effects as previously
reported (Clineschmidt et al., 1982; Loscher et al., 1992;
Mathe et al., 1996a) commencing approximately 10 min
after the MK-801 injection (Fig. 3). These effects included a long-lasting locomotor activation with a duration of 84.0 6 8.3 min (ten out of ten rats). In addition,
MK-801 administration evoked pronounced head weaving (50.0 6 6.1 min, 10/10 rats), which generally
preceded the MK-801 induced locomotor activity, but
also subsided more quickly. Some ipsi- and contra-lateral
turning and sniffing was observed in the rats that received
MK-801 (30 6 1.4 min, in 7/10 animals). Only a limited
ataxia was observed at the dose of MK-801 used (10 6
4.5 min, 4/10 animals).
In rats that were perfused with CNQX via the VTA
probe the duration of MK-801 induced locomotion was
significantly reduced from 84.0 6 8.3 min (vehicle and
MK-801, cf. above) by both concentrations of the antagonist, 300 M and 1 mM, to 20 6 10 min (3/6 rats
displaying locomotion) and 13 6 6 min (3/8 rats showing
locomotion), respectively (Fig. 3, both P values ,0.001).
However, the MK-801 induced turning behavior was not
affected by intra-VTA perfusion of CNQX 300 M (Fig.
3). Since there were no animals which exhibited only
ipsi- or contra-lateral turning, these data were combined.
CNQX perfusion of the VTA did not significantly affect
any of the other measured MK-801 induced behaviors,
i.e., head weaving, sniffing, or ataxia (Fig. 3).
DISCUSSION
The major finding of this study is the demonstration
that local perfusion of the VTA with CNQX potently
antagonizes both the evoked DA release in the NAC by
systemic MK-801 as well as the associated locomotor
stimulation. Thus, these results strongly support previous
behavioral experiments in the rat, indicating that the
locomotor stimulation by low doses of non-competitive
NMDA receptor antagonists, such as PCP or MK-801, is
dependent on DA in brain and is elicited within the VTA

588

Mathe et al.

Fig. 2. Effects of vehicle or CNQX (300M or 1 mM)

perfusion of the ventral tegmental area after saline (1 ml/kg,
s.c.) or MK-801 (0.1 mg/kg or 0.3 mg/kg, s.c.) administration,
on extracellular concentrations of dopamine (DA), dihydroxyphenylacetic acid (DOPAC), and homovanillic acid (HVA) in
the nucleus accumbens. Data are presented as mean (1 S.E.M.,

n55-8) percent changes over 140 min (overall effect) during

CNQX (300 M or 1 mM) or vehicle perfusion following
injection of saline or MK-801. 1P,.05, 11P,.01, and
111P,.001, vehicle and MK-801 compared to vehicle and
saline. *P,.05, **P,0.01, and ***P,0.001, vehicle and
MK-801 compared to CNQX plus MK-801, as shown.

(cf. Introduction). In addition, our experiments demonstrate that the evoked DA output in the NAC caused by
systemic MK-801 is, indeed, initiated within the VTA, the
site for nerve impulse generation in the mesolimbic DA
system. Consequently, the augmented DA output in the
NAC is in all probability related to, and dependent on, the
previously, electrophysiologically demonstrated hyperactivity of VTA DA neurons induced by systemic MK-801.
Although a direct stimulant effect of the drug on the DA
cells could be excluded (cf. Introduction), an indirect
action within the VTA is indicated, since neither lesions
of medial PFC, nor NAC, nor forebrain hemitransection
affect the stimulatory action of systemically administered
non-competitive NMDA antagonists on VTA DA neuro-

nal activity in chloral hydrate anesthetized rats (Ceci and

French, 1989; Rouillard et al., 1990; Zhang et al., 1992).
The mechanism within the VTA by which systemic
MK-801 evokes DA release in the NAC in this study
appears to be critically dependent upon increased activation of AMPA and/or kainate receptors in the VTA, since
local antagonism of AMPA and kainate receptors in the
VTA with CNQX completely blocked MK-801 evoked
DA release in the NAC. This view is in consonance with
previous studies, which demonstrate that local application of AMPA, quisqualate, or kainate in the somatodentritic region of the DA neurons potently stimulates DA
neuronal activity and release (Kalivas et al., 1989;
Chergui et al., 1993; Schilstrom et al., 1997). Such an

MK-801 Effects Caused by AMPA Receptors in VTA

589

Fig. 3. Effects of vehicle or CNQX (300 M or 1 mM) perfusion of the ventral tegmental area
(VTA) after administration of MK-801 (0.1 mg/kg, s.c.) on typical MK-801 induced behaviors,
i.e., locomotor activity, turning (ipsi- and contra-lateral), head weaving, sniffing, and ataxia,
rated during the above microdialysis experiments. Data are presented as mean (1 S.E.M.,
n56-10) time values for the entire 200 min measurement period (overall effect). ***P,0.001,
vehicle plus MK-801 compared to CNQX and MK-801, as indicated.

AMPA and/or kainate receptor mediated mechanism is

also in line with the preferential stimulatory effect of
systemic MK-801 on VTA DA cells localized within the
paranigral nucleus, which largely provides the DA projection to subcortical areas, e.g., the NAC (Murase et al.,
1993). These findings are in consonance with the reported
regional differences in the EAA receptor mediated regulation of mesolimbic vs. mesocortical DA systems (Kalivas
et al., 1989): Although the DA projection to the PFC
predominantly appears modulated by the NMDA subtype
of EAA receptors, the DA projection to the NAC, in
contrast, seems predominantly modulated by AMPA and
kainate receptors. The pharmacological and functional
specificity of the AMPA and kainate receptor blockade of,
and absence of NMDA receptor antagonism by CNQX
when locally perfused with these concentrations through
reverse microdialysis in the VTA has been recently
confirmed (Schilstrom et al., 1997). The reported increase
in DA release after systemic administration of MK-801 is
also well in agreement with respect to magnitude of
increase and time course, as reported in previous studies
(Wolf et al., 1993; Mathe et al., 1996a). Thus, the use of a

dual-probe compared to a single probe microdialysis

technique did not affect the present results with respect to
these parameters.
The demonstrated AMPA and/or kainate receptordependent increase in DA levels in NAC evoked by
systemic MK-801 suggests indirectly that MK-801 may
increase EAA release in the VTA. We have, in fact,
recently found that systemic MK-801 elevates dialysate
levels of glutamate and aspartate in the VTA (Svensson et
al., 1997). However, owing to very large intra- and
inter-animal variations in EAA efflux, these changes did
not reach statistical significance, in spite of long probe
washout periods, extensive, careful animal handling and
habituation. Indeed, a recent study suggests that systemic
administration of the NMDA receptor antagonist ketamine may also elevate glutamate levels in the prefrontal
cortex (Moghaddam et al., 1997). In addition, local
administration of various EAA receptor antagonists,
either in the hippocampus or striatum, has been shown to
enhance release of glutamate and aspartate in these
structures, as assessed by microdialysis (Liu et al., 1995),
thus supporting a central EAA-releasing action of PCP-

590

Mathe et al.

like drugs. However, none of these studies have, as yet,

unequivocally demonstrated that the glutamate measured
was released from neurons and was not related to, e.g.,
metabolic processes in brain. In fact, mounting recent
evidence indicates that synaptic EAA release is tightly
regulated and compartmentalized and, thus, that extracellular EAA levels, as assessed by microdialysis, may
largely reflect changes in processes that are not necessarily related to neurotransmission (see Westerink, 1995;
Timmerman and Westerink, 1997). Nevertheless, the
most likely explanation for the increased mesolimbic DA
neuronal activity caused by systemic MK-801 is, clearly,
enhanced EAA release, thus resulting in a relative hyperactivation of AMPA and kainate receptors in the VTA.
Indirect, electrophysiological evidence suggests that
local GABAergic mechanisms in the VTA may contribute
to the activation of DA neuronal activity observed after
systemic administration of MK-801 or PCP. Systemic
administration of these NMDA antagonists was found to
inhibit the neuronal activity of local, presumed GABA
containing interneurons and might thereby release DA
neurons from tonic inhibition, although no further evidence has been presented in support of this hypothesis
(Ceci and French, 1989; Zhang et al., 1993). However,
utilizing this logic, further EAA receptor antagonism by
simultaneous, local application of CNQX should therefore be predicted to disinhibit DA cells even further. In
contrast, local perfusion with CNQX (1 mM) in the VTA
completely blocked systemic MK-801 evoked DA release. Albeit these experiments cannot determine whether
this effect is due to a direct action of MK-801 on EAA
neurons modulating DA function, or if MK-801 inhibits
NMDA receptors on GABAergic cells, thereby indirectly
regulating the activity of glutamatergic neurons involved
in the control of DA cell activity, our data suggest that
EAA-mediated processes represent the major mechanism
by which MK-801 activates mesolimbic DA activity.
Our data clearly indicate that mechanisms located at
the DA cell body level within the VTA are largely
involved also in the behavioral stimulation caused by low
doses (i.e., ,0.25 mg/kg) of MK-801, in particular its
locomotor stimulant action, albeit without induction of
pronounced ataxia, which has been reported to contribute
to severe locomotor incoordination in rats given doses of
MK-801 higher than 0.3 mg/kg (Criswell at al., 1993).
The potent antagonistic action of CNQX, locally administered in the VTA, on systemic MK-801 evoked locomotor
activity may well explain the mechanism by which
intra-VTA microinjections of MK-801 mimic the behavioral effects of systemic injections of this drug (cf.
introduction). In fact, recent studies indicate that systemic
administration of the AMPA and kainate receptor antagonist GYKI 52466 can antagonize MK-801 evoked locomotor stimulation, as well as MK-801 induced increases in

DA turnover in the NAC (Bubser at al., 1995). Our results

indicate the possibility that this antagonistic action of
GYKI 52466 on MK-801 evoked locomotor activity may
be executed at the VTA level. Our data demonstrate a
correlation between MK-801 induced DA release in the
NAC and the stimulation of locomotion evoked by this
compound. Consequently, our data lend further support to
the hypothesis that DA-dependent mechanisms play a
causative role in eliciting hyperlocomotion observed after
systemic administration of MK-801 within this dose
range.
Interestingly, although local application of CNQX
in the VTA significantly antagonized MK-801 evoked
locomotor activity, it did not block MK-801 induced head
weaving. A recent finding indicates that head weaving,
unlike locomotor activity, is not primarily associated with
DA, but rather with 5-HT in brain, and specifically, with
5-HT1A receptor activation (Loscher et al., 1992).
The present results show that even unilateral blockade of AMPA and kainate receptors in the VTA can
significantly antagonize systemic MK-801 evoked hyperlocomotion, albeit without a concomitant increase in
ipsilateral turning as might be expected after unilateral
reduction of central DAergic activity (Ungerstedt, 1971).
However, our data do not reveal whether local, unilateral
VTA perfusion with CNQX also affects DA release
evoked by systemic MK-801 in the contralateral NAC. In
view of the adjacency of the left and right side VTA
nuclei, a unilateral microdialysis probe of 0.3 mm
diameter, such as the ones used in these experiments,
placed close to the midline (0.20.7 mm) may allow
CNQX to diffuse also into the contralateral VTA and
thereby significantly antagonize AMPA and kainate receptors on both right and left side VTA nuclei. In fact, it is
generally assumed that microdialysis probes sample a
volume that probably extends several hundred micrometers radially into the neural tissue (Westerink and Justice,
1991). Conversely, it has recently been shown that
infusion of [3H]-nicotine in the striatum via reverse
microdialysis results in a spreading of 80% of the
measured radioactivity to within 1 mm of the microdialysis probe (Toth et al., 1992). Conceivably, a similarly
lipophilic drug, such as CNQX, infused via reverse
microdialysis in any brain structure probably diffuses
over a similar volume around the probe. Thus, under the
present experimental conditions, unilateral infusion of
CNQX in the VTA may affect AMPA and kainate
receptors on the contralateral side as well. In turn, CNQX
might thus reduce MK-801 evoked release in both the
right and left side NAC, and consequently reduce associated locomotor activity in general, rather than to evoke
circling behavior.
When administered to schizophrenic patients, PCP
has been reported to induce psychotic relapse in a manner

MK-801 Effects Caused by AMPA Receptors in VTA

characteristic to the symptoms of the individual patients,

i.e., PCP did not evoke additional schizophrenic symptoms but rather exacerbated the pre-existing symptomatology of the patients. These findings suggest that schizophrenia and PCP-like NMDA receptor antagonists, e.g.,
MK-801, may share common mechanisms in the generation of psychotic symptoms. In addition, several lines of
evidence have implicated reduced glutamatergic activity
in brain in schizophrenia, specifically involving reduced
NMDA receptor activity (see Javitt and Zukin, 1991).
Since both local and systemic application of a potent
AMPA and kainate receptor antagonist in the VTA
alleviated several behavioral consequences of drug induced NMDA receptor hypofunction, a commonly used
pharmacological model of schizophrenia, AMPA and/or
kainate receptor antagonists may possess antipsychotic
efficacy. Importantly, the AMPA and kainate receptor
antagonist CNQX reduced MK-801 evoked but not basal
DA release. In addition, perfusion of the VTA with
CNQX attenuated behavioral stimulation induced by
MK-801 but did not affect spontaneous behaviors when
given alone. Recent studies of animal models of schizophrenia suggest an increased sensitivity of central DA
systems to environmental or pharmacological challenge
(Deutch et al., 1990; Jaskiw et al., 1990). Consequently,
AMPA and/or kainate receptor antagonists might specifically suppress the augmentation of schizophrenic symptoms frequently seen during conditions of environmental
stress.
In summary, the behavioral effects of psychotomimetic NMDA antagonists may not only involve blockade
of NMDA receptors but may also reflect a relatively
increased activation of non-NMDA excitatory amino acid
receptors, e.g., in the VTA. Therefore, a novel approach
to alleviate not only the behavioral effects of psychotomimetic NMDA receptor antagonists, but also psychotic
symptoms in schizophrenia, may thus involve blockade
of AMPA and/or kainate receptors.

ACKNOWLEDGMENTS
The excellent technical assistance of Mrs. Anna
Malmerfelt, Mrs. Annika Olsson, and Mr. Martin Svensson is gratefully acknowledged.

REFERENCES
Bubser M, Tzschentke T, Hauber W (1995): Behavioural and neurochemical interactions of the AMPA antagonist GYKI 52466 and
the non-competitive NMDA antagonist dizocilpine in rats. J
Neural Transm 101:115126.
Ceci A, French ED (1989): Phencyclidine-induced activation of ventral
tegmental A10 dopamine neurons is differentially affected by
lesions of the nucleus accumbens and medial prefrontal cortex.
Life Sci 45:637646.

591

Chergui K, Charlety PJ , Akaoka H, Saunier CF, Brunet JL, Buda M,

Svensson TH, Chouvet G (1993): Tonic activation of NMDA
receptors causes spontaneous burst discharge of rat midbrain
dopamine neurons in vivo. Eur J Neurosci 5:137144.
Clineschmidt BV, Martin GE, Bunting PR, Papp NL (1982): Central
sympathomimetic activity of (1)-5-methyl-10,11-dihydro-5Hdibenzo[a,d]cyclohepten-5,10-imine (MK-801), a substance with
potent anticonvulsant, central sympathomimetic, and apparent
anxiolytic properties. Drug Dev Res 2:135145.
Contreras PC, Monahan JB, Lanthorn TH, Pullan LM, DiMaggio DA,
Handelmann GE, Gray NM, ODonohue TL (1987): Phencyclidine: Physiological actions, interactions with excitatory amino
acids and endogenous ligands. Mol Neurobiol 1:191211.
Criswell HE, Johnson KB, Mueller RA, Breese GR (1993): Evidence
for involvement of brain dopamine and other mechanisms in the
behavioral action of the N-methyl-D-aspartic acid antagonist
MK-801 in control and 6-hydroxydopamine-lesioned rats. J
Pharmacol Exp Ther 265:10011010.
Deutch AY, Clark WA, Roth RH (1990): Prefrontal cortical dopamine
depletion enhances the responsiveness of mesolimbic dopamine
neruons to stress. Brain Res 521:311315.
Fessler RG, Sturgeon RD, Meltzer HY (1980): Effects of phencyclidine
and methylphenidate on d-amphetamine-induced behaviors in
reserpine pretreated rats. Pharmacol Biochem Behav 13:835
842.
French ED (1986): Effects of phencyclidine on ventral tegmental A10
dopamine neurons in the rat. Neuropharmacology 25:241248.
Freeman AS, Bunney BS (1984): The effects of phencyclidine and
n-allylnormetazocine on midbrain dopamine neuronal activity.
Eur J Pharmacol 104:287293.
Grace AA, Bunney BS (1980): Effects of baclofen on nigral dopaminergic cell activity following acute and chronic haloperidol
treatment. Brain Res Bull 5:537543.
Grace AA, Bunney BS (1984): The control of the firing pattern in nigral
dopamine neurons: Burst firing. J Neurosci 4:28772890.
Jaskiw GE, Karoum F, Freed WJ, Phillips I, Kleinman JE, Weinberger
DR (1990): Effect of ibotenic acid lesions of the medial
prefrontal cortex on amphetamine-induced locomotion and
regional brain catecholamine concentrations in the rat. Brain
Res 534:263272.
Javitt DC, Zukin SR (1991): Recent advances in the phencyclidine
model of schizophrenia. Am J Psychiatry 148:13011308.
Kalivas PW, Duffy P, Barrow J (1989): Regulation of the mesocorticolimbic dopamine system by glutamic acid receptor subtypes. J
Pharmacol Exp Ther 251:378387.
Kornhuber JM, Weller M (1995): Predicting psychotomimetic properties of PCP-like NMDA receptor antagonists, In R. Fog,
J.Gerlach, R. Hemmingsen (eds): Schizophrenia, Alfred Benzon Symposium. Copenhagen: Munksgaard 38:314325.
Liu J, Moghaddam B (1995): Regulation of glutamate efflux by
excitatory amino acid receptors: Evidence for tonic inhibitory
and phasic excitatory regulation. J Pharmacol Exp Ther 274:
12091215.
Loscher WR, Honack D (1992): The behavioural effects of MK-801 in
rats: involvement of dopaminergic, serotonergic and noradrenergic systems. Eur J Pharmacol 215:199208.
Marshall DL, Redfern PH, Wonnacott S (1997): Presynaptic nicotinic
modulation of dopamine release in the three ascending pathways studied by in vivo microdialysis: Comparison of naive and
chronic nicotine-treated rats. J Neurochem 68:15111519.
Mathe JM, Nomikos GG, Hildebrand BE, Hertel P, Svensson TH
(1996a): Prazosin inhibits MK-801 induced hyperlocomotion
and dopamine release in the nucleus accumbens. Eur J Pharmacol 309:111.
Mathe JM, Schilstrom B, Nomikos GG, Svensson TH (1996b):

592

Mathe et al.

Non-NMDA receptors in the ventral tegmental area mediate

hyperlocomotion and dopamine release in the n. accumbens
induced by systemic MK-801. (Abstract) Proceedings of the
Eighth International Catecholamine Symposium, Pacific Grove,
CA, USA.
Mercuri NB, Grillner P, Bernardi G (1996): N-methyl-d-aspartate
receptors mediate a slow excitatory postsynaptic potential in the
rat midbrain dopaminergic neurons. Neuroscience 74:785792.
Miller DW, Abercrombie ED (1996): Effects of MK-801 on spontaneous and amphetamine-stimulated dopamine release in striatum
measured with in vivo microdialysis in awake rats. Brain Res
Bull 40:5762.
Moghaddam B, Adams B, Verma A, Daly D (1997): Activation of
glutamatergic neurotransmission by ketamine: A novel step in
the pathway from NMDA receptor blockade to dopaminergic
and cognitive disruptions associated with the prefrontal cortex. J
Neurosci 17:29212927.
Murase S, Mathe JM, Grenhoff J, Svensson TH (1993): Effects of
dizocilpine (MK-801) on rat midbrain dopamine cell activity:
Differential actions related to anatomical localization. J Neural
Transm (Gen Sect) 91:1325.
Narayanan S, Willins D, Dalia A, Wallace L, Urtesky NJ (1996): Role
of dopaminergic mechanisms in the stimulatory effects of
MK-801 injected into the ventral tegmental area and the nucleus
accumbens. Pharmacol Biochem Behav 54:565573.
Nomikos GG, Damsma G, Wenkstern D, Fibiger HC (1989): Acute
effects of buproprion on extracellular dopamine concentrations
in rat striatum and nucleus accumbens studied by in vivo
microdialysis. Neuropsychopharmacology 2:273279.
Nomikos GG, Iurlo M, Andersson JL, Kimura K, Svensson TH (1994):
Systemic administration of amperozide, a new atypical antipsychotic drug, preferentially increases dopamine release in the rat
medial prefrontal cortex. Psychopharmacology 115:147156.
Pawlowski L, Mathe JM, Svensson TH (1990): Phencyclidine activates
rat A10 dopamine neurons but reduces burst activity and causes
regularization. Acta Physiol Scand 139:529530.
Paxinos G, Watson C (1986): The Rat Brain in Stereotaxic Coordinates, 2nd Ed. London: Academic Press.
Rouillard C, Chiodo LA, Freeman AS (1990): The effects of the
phencyclidine analogs BTCP and TCP on nigrostriatal dopamine neuronal activity. Eur J Pharmacol 182:227235.
Scarnati E, Pacitti C (1982): Neuronal responses to iontophoretically
applied dopamine, glutamate and GABA of identified dopaminergic cells in the rat substantia nigra after kainate acid-induced
destruction of the striatum. Exp Brain Res 46:377382.
Schilstrom B, Nomikos GG, Nisell M, Hertel P, Svensson TH (1998):

NMDA receptor antagonism in the ventral tegmental area

diminishes the systemic nicotine-induced dopamine release in
the nucleus accumbens. Neuroscience 82:781789.
Seutin V, Johnsson SW, North RA (1994): Effect of dopamine and
baclofen on n-methyl-d-aspartate-induced burst firing in rat
ventral tegmental neurons. Neuroscience 58:201206.
Svensson TH, Mathe JM, Nomikos GG, Schilstrom B, Marcus M,
Fagerquist M (1997): Interactions between catecholamines and
serotonin: relevance to the pharmacology of schizophrenia. In
DS Goldstein, G Eisenhofer, R McCarty (Eds.) Advances in
Pharmacology. Academic Press, pp. 814818.
Timmerman W, Westerink BHC (1997): Brain microdialysis of GABA
and glutamate: What does it signify?. Synapse 27:242261.
Toth E, Sershen H, Hashim A, Vizi ES, Lajtha A (1992): Effects of
nicotine on levels of extracellular amino acids in regions of the
rat brain in vivo. Neuropharmacology 32:827832.
Ungerstedt U (1971): Stiatal dopamine release after amphetamine or
nerve degeneration revealed by rotational behavior. Acta Physiol
Scand 82:4968.
Wedzony K, Klimek V, Golembiowska K (1993): MK-801 elevates the
etracellular concentration of dopamine in the rat prefrontal
cortex and increases the density of striatal dopamine D1
receptors. Brain Res 622:325329.
Westerink BHC, Justice JB (1991): Microdialysis compared with other
in vivo release models, In: Robinson TE, Justice JB (eds):
Microdialysis in the Neurosciences. Elsevier Science Publishers B.V., pp. 2343.
Westerink BHC (1995): Brain microdialyis and its application for the
study of animal behaviour. Behav Brain Res 70:103124.
Willins DL, Nayaranan S, Wallace LJ, Uretsky NJ (1993): The role of
dopamine and AMPA/kainate receptors in the nucleus accumbens in the hypermotility response to MK-801. Pharmacol
Biochem Behav 46:881887.
Wolf ME, White FJ, Hu XT (1993): Behavioral sensitization to
MK-801 (dizocilpine): Neurochemical and electrophysiological
correlates in the mesoaccumbens dopamine system. Behav
Pharmacol 4:429442.
Wong EHF, Kemp JA, Priestley T, Knight AR, Woodruff GN, Iversen
LL (1986): The anticonvulsant MK-801 is a potent N-methyl-Daspartate antagonist. Proc Natl Acad Sci USA 83:71047108.
Zhang J, Chiodo LA, Freeman AS (1992): Electrophysiological effects
of MK-801 on rat nigrostriatal and mesoaccumbal dopaminergic neurons. Brain Res 590:153163.
Zhang J, Chiodo LA, Freeman AS (1993): Effects of phencyclidine,
MK-801 and 1,3-di(2-tolyl)guanidine on non-dopaminergic
midbrain neurons. Eur J Pharmacol 230:371374.