Molecular Psychiatry (2005) 10, 877–883

& 2005 Nature Publishing Group All rights reserved 1359-4184/05 $30.00
www.nature.com/mp

ORIGINAL RESEARCH ARTICLE

Dopamine receptor contribution to the action of PCP,

LSD and ketamine psychotomimetics
P Seeman1,2, F Ko1 and T Tallerico2
1
Department of Pharmacology, University of Toronto, Toronto, Ontario, Canada; 2Department of Psychiatry, University of
Toronto, Toronto, Ontario, Canada

Although phencyclidine and ketamine are used to model a hypoglutamate theory of

schizophrenia, their selectivity for NMDA receptors has been questioned. To determine the
affinities of phencyclidine, ketamine, dizocilpine and LSD for the functional high-affinity state
of the dopamine D2 receptor, D2High, their dissociation constants (Ki) were obtained on
[3H]domperidone binding to human cloned dopamine D2 receptors. Phencyclidine had a high
affinity for D2High with a Ki of 2.7 nM, in contrast to its low affinity for the NMDA receptor, with a
Ki of 313 nM, as labeled by [3H]dizocilpine on rat striatal tissue. Ketamine also had a high
affinity for D2High with a Ki of 55 nM, an affinity higher than its 3100 nM Ki for the NMDA sites.
Dizocilpine had a Ki of 0.3 nM at D2High, but a Kd of 1.8 nM at the NMDA receptor. LSD had a Ki of
2 nM at D2High. Because the psychotomimetics had higher potency at D2High than at the NMDA
site, the psychotomimetic action of these drugs must have a major contribution from D2
agonism. Because these drugs have a combined action on both dopamine receptors and
NMDA receptors, these drugs, when given in vivo, test a combined hyperdopamine and
hypoglutamate theory of psychosis.
Molecular Psychiatry (2005) 10, 877–883. doi:10.1038/sj.mp.4001682; published online 26 April 2005
Keywords: dopamine receptor; phencyclidine; domperidone; ketamine; NMDA receptors;
psychotomimetics

While the clinical anti-dopaminergic actions of anti-
psychotic drugs are compatible with the hyperdopa-
mine hypothesis of psychosis and schizophrenia,1,2
the psychoses caused by glutamate antagonists such
as phencyclidine or ketamine suggest a hypo-gluta-
mate component in psychosis.3–5 However, phency-
clidine also lowers plasma prolactin6 and elicits
rotation,7 suggesting a direct or indirect dopamine-
mimetic action of phencyclidine.
Although phencyclidine and ketamine are not
selective for glutamate NMDA receptors,8 the precise
affinities of these drugs for dopamine D2 receptors
need to be clarified in order to determine their
dopaminergic and non-dopaminergic components of
action. More specifically, although phencyclidine had
a dissociation constant of 37 000 nM at the D2
receptor in rat striatal homogenate,8 phencyclidine
had a dissociation constant of 1.3 nM for the func-
tional high-affinity site of the cloned D2 receptor, or
the D2High receptor.9
This apparent discrepancy may be resolved by the
recent finding that dopamine itself has a dissociation
constant of 3000 nM when competing vs [3H]raclo-
Correspondence: Dr P Seeman, MD, PhD, Departments of
Pharmacology and Psychiatry, Medical Science Building, Room
4344, University of Toronto, 1 King’s College Circle, Toronto,
Ontario, Canada M5S 1A8. E-mail: philip.seeman@utoronto.ca
Received 23 November 2004; revised 4 March 2005; accepted 21
March 2005; published online 26 April 2005
pride at striatal D2 receptors, but has a dissociation
constant of 1.5 nM at the D2High receptor when the
link between dopamine D1 and D2 receptors is
blocked by the D1 antagonist SCH23390.10 The link
between D1 and D2 receptors arises from several
sources, including the colocalization of dopamine D1
and D2 receptors in at least 50% of the medium spiny
neurons in the striatum, the cooperation and mutual
potentiation of D1 and D2 agonists on various
behaviors, and the biochemical conversion of D2
receptors from their functional high-affinity state,
D2High, into their low-affinity state, D2Low, in the
presence of a D1 antagonist.10–13 This D1–D2 link has
been detected in vitro in human and rat striata,10,11
and in tissue culture cells transfected with both D1
and D2,12 but has not yet been demonstrated in
humans by means of positron emission tomography
and [11C]raclopride.
However, in order to compare the potencies of
psychotomimetics at dopamine D2 receptors with
their potencies at NMDA receptors, it would be better
to avoid using SCH23390. This is because SCH23390
can also affect the potency of drugs at the NMDA
receptor, especially since it is known that D1 and
NMDA receptors interact.14–16
Although the blockade of D1 (by SCH23390)
unmasks the functional D2High states,11 the addition
of SCH23390 is only needed when using [3H]raclo-
pride to measure D2 receptors. [3H]Domperidone,
however, is a selective ligand for D2 that readily
 Dopamine receptors, phencyclidine and ketamine
P Seeman et al
878
detects D2High receptors without the addition of
SCH23390.17
Therefore, in order to obtain the potencies of
phencyclidine, ketamine, dizocilpine and LSD at the
functional D2High receptor, the present study used
[3H]domperidone to label D2 receptors. The results
indicate that the potencies of the psychotomimetics
were greater at D2High than at the NMDA receptor.
Materials and methods
We used the human cloned dopamine D2Long receptor
(in Chinese hamster ovary (CHO) cells18) . The cells
were harvested by gently scraping the cells off the
bottom of the Petri dish, centrifuged, resuspended in
phosphate-buffered saline (0.9% NaCl), recentrifuged
and the pellet frozen at
701. When used, the frozen
pellet was thawed, and the cells suspended at 200 mg
protein/ml. The suspension was homogenized for
5 s (Brinkmann Polytron, setting 5 (maximum on
scale was 10)), without any further washing or
centrifugation.
Frozen rat brains were purchased (Pel-Freez,
Rogers, AR, USA), partly thawed and the striata
removed. To prepare the tissues for the drug/[3H]
ligand competition-type experiments, the striatal
tissues were blotted and weighed; buffer was added
(50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 5 mM KCl,
1.5 mM CaCl2, 4 mM MgCl2, with a final concentration
of 120 mM NaCl added later in the final incubation
tubes) to yield 4 mg tissue per ml. The suspension
was homogenized with a glass-Teflon homogenizer
(10 up–down strokes of a piston rotating at 500 rpm).
The homogenate was not washed, centrifuged or
preincubated because previous work found that these
procedures resulted in a loss of 30–60% of dopamine
D2 receptors.19
[3H]Dizocilpine (or [3H]( þ )MK801; 17 Ci/mmol;
3 nM final concentration in the incubation tubes)
was purchased from PerkinElmer Life Sciences Inc.
(Boston, MA, USA). [3H]Domperidone was custom
synthesized as [phenyl-3H(N)]domperidone (42 Ci/
mmol) by PerkinElmer Life Sciences Inc., and used
at a final concentration of 1.2–1.8 nM.
The competition-type experiment between a drug
and the [3H]ligand for binding at the receptors was
carried out as follows. Each incubation tube
(12  75 mm, glass) received, in the following order,
0.5 ml buffer (with a final incubate concentration of
either 10 mM or, more usually, 120 mM NaCl), 0.25 ml
[3H]ligand and 0.25 ml of tissue homogenate. The
tubes, containing a total volume of 1 ml, were
incubated for 2 h at room temperature (201C), after
which the incubates were filtered, using a 12-well cell
harvester (Titertek, Skatron, Lier, Norway) and buffer-
presoaked glass fiber filter mats (No. 7034, Skatron,
Sterling, VA, USA). After filtering the incubate, the
filter mat was rinsed with buffer for 15 s (7.5 ml
buffer). The filters were pushed out and placed in
scintillation minivials (Packard Instruments, Chicago,
IL, USA). The minivials received 4 ml each of
Molecular Psychiatry
scintillant (Ready Solve, Beckman Co., CA, USA),
and were monitored 6 h later for tritium in a Beckman
L5000 scintillation spectrometer at 55% efficiency.
The nonspecific binding of the [3H]ligand to the
dopamine D2 receptors was defined as that which
occurred in the presence of 10 mM S-sulpiride
(Ravizza, Milan). The dissociation constants (Kd) of
[3H]domperidone for the dopamine D2 receptors
(0.42 nM on the cloned D2 receptor and 0.47 nM on
the rat striatal tissue17) were obtained by the above
method, using a range of 12 final concentrations of
0.1–20 nM [3H]domperidone. The dissociation con-
stant for [3H]dizocilpine was 1.8 nM (rat striatum),
using 100 mM phencyclidine to define nonspecific
binding. The Kd value was calculated by nonlinear
regression analysis of the [3H]ligand saturation
curve.19 The competition data were analyzed as
previously described;20 the program provided two
statistical criteria to judge whether a two-site fit was
better than a one-site fit, or whether a three-site fit
was better than a two-site fit.
The drug-induced incorporation of [35S]GTP-g-S
(1250 Ci/mmol; final concentration 0.2–0.3 nM)
was carried out using the procedure previously
described.8
Results
Phencyclidine
As noted above, both [3H]raclopride and [3H]domper-
idone selectively label D2 receptors, but [3H]domper-
idone is more sensitive than [3H]raclopride to the
competitive action of dopamine and other drugs and
readily allows the detection of D2High sites.17 There-
fore, using the more readily displaceable [3H]domper-
idone, phencyclidine had a dissociation constant at
D2High of 2.770.8 nM (n ¼ 9) (Figure 1, top) for the
human cloned D2 receptor and 4.370.8 nM for the rat
striatal tissue (Figure 1, middle). These Ki values were
essentially unchanged when phencyclidine was
tested in the presence of 100 nM SCH23390 to
occlude other receptors such as dopamine D1 recep-
tors and serotonin-2 receptors that may interact with
the D2 receptor.10
In contrast, the phencyclidine Ki for inhibiting
the binding of [3H]dizocilpine to rat striatal homo-
genate was 313748 nM (n ¼ 9) (Figure 1, bottom).
The addition of 2 mM glutamate did not change
the phencyclidine Ki values.
In order to determine the functional concentration
range for phencyclidine, the effect of phencyclidine
on the incorporation of [35S]GTP-g-S was tested on
the human cloned D2 receptors. Phencyclidine at
concentrations 2 nM or higher stimulated the incor-
poration of [35S]GTP-g-S, with an EC50% (relative
effective concentration for 50% incorporation) of
90 nM, as shown in Figure 2.
Ketamine
The data for ketamine were qualitatively similar
to those for phencyclidine. For example, using

Figure 1 Top: Using 1 nM [3H]domperidone to label
human cloned dopamine D2 receptors, phencyclidine had
a Ki of 2.770.8 nM (n ¼ 9) at D2High. This high-affinity state
was eliminated by the presence of 200 mM guanilylimido-
diphosphate (open circles). Total binding was 3000 dpm/
filter; nonspecific binding in the presence of 10 mM S-
sulpiride was 500 dpm/filter. Middle: Using 1.4–1.75 nM
[3H]domperidone, phencyclidine had a Ki of 4.370.8 nM
(n ¼ 4) at D2High. Using 100 nM SCH23390 to occlude other
receptors (such as dopamine D1 and serotonin-2) did not
significantly alter the Ki values. Bottom: Using 3.3 nM
[3H]dizocilpine to label NMDA receptors in rat striatal
homogenate, phencyclidine only recognized low-affinity
binding sites with a Ki of 313748 nM (n ¼ 9).
Dopamine receptors, phencyclidine and ketamine
P Seeman et al
879
Figure 2 Phencyclidine, ketamine, dizocilpine and dopa-
mine stimulated the incorporation of 0.3 nM [35S]GTP-g-S in
the human cloned D2Long receptors in CHO cells. Vertical
bars indicate SE (n ¼ 3). The final concentration of GDP was
10 mM. Maximum incorporation of 100% corresponded to
6000 cpm/filter. Nonspecific binding in the absence of drug
was 2500 cpm/filter.
[3H]domperidone on the cloned D2 receptor, ketamine
had a Ki value of 55712 nM (n ¼ 3) at D2High (Figure 3,
top). In the presence of 200 mM guanilylimidodipho-
sphate, the high-affinity site (Ki of 55 nM) was
eliminated, consistent with the property of a G
protein-linked receptor. Ketamine stimulated the
incorporation of [35S]GTP-g-S into the cloned D2
receptor at concentrations above 20 nM, with 50%
incorporation at 110 nM (Figure 2).
At the NMDA receptor, however, ketamine was less
potent than at the D2 receptor, having a Ki value of
31007300 nM (n ¼ 6) vs the binding of [3H]dizocil-
pine (Figure 3, bottom) to rat striatal tissue. The
addition of 2 mM glutamate had no effect on the
ketamine Ki values.
Dizocilpine
The data for dizocilpine were also qualitatively
similar to those for phencyclidine and ketamine, but
dizocilpine was much more potent. Dizocilpine
inhibited the binding of [3H]domperidone to the
cloned D2High receptor with a Ki of 0.370.08 nM
(n ¼ 5) (Figure 4, top). This high-affinity site was
abolished in the presence of 200 mM guanilylimido-
diphosphate (Figure 4, top). Matching its high
potency to inhibit [3H]domperidone binding in the
1–10 nM range, dizocilpine also stimulated the in-
corporation of [35S]GTP-g-S into the cloned D2
receptor at concentrations between 1 and 20 nM, with
50% incorporation at 8 nM (Figure 2).
In order to confirm that dizocilpine was indeed
recognizing a high-affinity site for D2, as suggested by
the observation that dizocilpine inhibited the binding
of [3H]domperidone at dizocilpine concentrations of
Molecular Psychiatry
 Dopamine receptors, phencyclidine and ketamine
P Seeman et al

880

Figure 4 Schematic showing that dizocilpine recognized

the high-affinity state of the dopamine D2 receptor, D2High.
Using 1.3 nM [3H]domperidone to label human cloned
dopamine D2 receptors, dizocilpine had a Ki of
0.370.08 nM (n ¼ 5) at D2High. This high-affinity state was
eliminated in the presence of 200 mM guanilylimidodi-
phosphate. Total binding was B9000 dpm/filter; nonspecific
binding in the presence of 10 mM S-sulpiride was
B5000 dpm/filter.

Figure 3 Top: Using [3H]domperidone to label cloned D2

receptors, ketamine recognized a high-affinity site with a Ki
value of 55712 nM (n ¼ 3). Bottom: Using [3H]dizocilpine to
label NMDA receptors in rat striatal homogenate, ketamine
had a Ki of 31007300 nM (n ¼ 6). Nonspecific binding was
defined by 100 mM phencyclidine.

3–100 nM (Figure 4), it was necessary to do the

converse, namely, to compete dopaminergic drugs vs
[3H]dizocilpine. This was carried out by competing a
standard D2 antagonist, haloperidol, vs 0.8 nM
[3H]dizocilpine on CHO cells containing transfected
D2Long receptors. The data in Table 2 show that
dopaminergic drugs, two antagonists and two ago-
nists, inhibited the binding of [3H]dizocilpine to the
D2Long receptors with the appropriate rank order
characteristic of dopamine D2 receptor agonists,
namely that n-propyl-norapomorphine is more potent Figure 5 Top: Using 1.4–1.7 nM [3H]domperidone, LSD-25
than apomorphine. had a Ki of 270.3 nM for human cloned dopamine D2
The Kd for [3H]dizocilpine was 1.870.4 nM for the receptors. Bottom: LSD-25 had no effect on the binding of
3.2 nM [3H]dizocilpine. Nonspecific binding was defined by
NMDA sites in the striatal tissue. The presence of
100 mM phencyclidine.
2 mM glutamate had no effect on the dizocilpine
dissociation constants.

LSD-25 stimulated incorporation was 80% of that stimulated

Using [3H]domperidone, LSD-25 inhibited the bind-
ing at D2High with a Ki value of 270.3 nM (n ¼ 3)
(Figure 5, top) on human cloned D2 receptors. LSD-25
(up to 10 000 nM) had no effect on the binding of
[3H]dizocilpine (Figure 5, bottom). LSD-25 stimulated
the incorporation of [35S]GTP-g-S into the cloned D2
receptor at concentrations between 10 and 200 nM,
with an LSD-25 EC50% of 40 nM (data not shown in
Figure 2, because the data points overlap those for
phencyclidine). The total absolute amount of LSD-
by dopamine.
Discussion
The data show that phencyclidine, ketamine, dizo-
cilpine and LSD all have high affinities for the high-
affinity state of dopamine D2 receptors, D2High with
dissociation constants of B3, 55, 0.3 and 2 nM,
respectively (Table 1).

Molecular Psychiatry

Table 1 Psychotomimetic Ki values (nM)
D2High
Clone (present)
Striatum (present)
Phencyclidine 2.770.8
4.370.8a
Ketamine 55712
Dizocilpine 0.370.08
LSD 270.3
a
Rat striatum.
Table 2 Dopaminergic drug Ki values using [3H]dizocilpine
on dopamine D2Long receptors in CHO cells
Dopaminergic drug Ki (7SE) (nM)
Antagonists
Haloperidol 0.3170.1
Domperidone 2879
Agonists
N-propyl-norapomorphine-(
) 1174
Apomorphine 5607100
SE for triplicate measurements.
Ki ¼ IC50/(1 þ C*/Kd), where IC50 is the concentration for
50% inhibition of [3H]dizocilpine binding to D2Long recep-
tors, C* is the concentration of [3H]dizocilpine and Kd is the
dissociation constant of [3H]dizocilpine at D2High (Figure 4).
The present data are relevant to problems asso-
ciated with testing the hypoglutamate theory of
psychosis, because it has been proposed that dopa-
mine abnormalities in psychosis are secondary to a
primary hypofunction in glutamate neurotransmis-
sion. The hypoglutamate theory is supported by the
fact that phencyclidine and ketamine are NMDA
antagonists and elicit psychotic symptoms.3,4
The present findings illustrate, however, that
phencyclidine and ketamine also act on the high-
affinity state of dopamine D2 receptors with potent
dissociation constants of 3 and 55 nM, respectively.
These values compare with 313 and 3100 nM for
phencyclidine and ketamine at NMDA receptors,
respectively.
Moreover, the biphasic pattern of inhibition of
[3H]domperidone binding by phencyclidine and
ketamine is compatible with agonist action at the
dopamine D2High state of the D2 receptors, considering
that virtually all dopamine agonists have this bipha-
sic pattern, in contrast to all dopamine antagonists,
which reveal a monophasic pattern of ligand inhibi-
tion.13,17,20
In addition, because receptor agonists accelerate
the exchange of receptor-bound GDP for GTP, the
stimulation of [35S]GTP-g-S incorporation is a useful
Dopamine receptors, phencyclidine and ketamine
P Seeman et al
881
NMDA
Cortex34 Clone34
313748 196 124
31007300 3150 2270
(Kd) 1.870.4 3 3
410 000 Not done
index of agonist action. Phencyclidine and ketamine
also increased the incorporation of [35S]GTP-g-S into
cloned D2-containing cells8 (Figure 2).
[3H]Domperidone is the optimal ligand for detect-
ing the high-affinity state of dopamine D2 receptors,
because [3H]spiperone and [3H]raclopride in physio-
logical saline (B120 mM NaCl) primarily reveal the
low-affinity state of D2 receptors.17 In fact, the
dissociation constant of 1.5 nM for dopamine to
inhibit the binding of [3H]domperidone at D2High is
identical to the dissociation constant of [3H]dopamine
at the dopamine D2 receptor;17 the dopamine dis-
sociation constants to inhibit [3H]spiperone or [3H]
raclopride at D2High are invariably higher (B20–200 nM)
and do not correspond to the dissociation constant of
[3H]dopamine at the dopamine D2High receptor.17
Although early experiments did not reveal the
D1–D2 link in any rat striatum or in every human
striatum,11 recent work, using a competition method,
has consistently revealed the D1–D2 link in rat and
human striata.10 This recent work showed that the
block of D1 receptors unveiled the presence of a high-
affinity state for D2 in a dopamine/[3H]raclopride
competition-type experiment, with a dopamine dis-
sociation constant of 1.5 nM. Such a high-affinity
state for D2 is not normally detected in striata, despite
the fact that the high-affinity state of the dopamine D2
receptor is the physiologically functional state of the
receptor in the anterior pituitary gland.13 In fact,
previous work found it difficult to detect high-affinity
sites for dopamine D2 receptors in rat or human
striatum in the presence of physiological concentra-
tions of sodium ions, because these high-affinity sites
were generally less than 5% of the population in
120 mM NaCl.21 Thus, the determination of agonist
dissociation constants at the high-affinity site of the
D2 receptor was generally carried out in the absence
of NaCl.20
Considering that all three psychotomimetics, phen-
cyclidine, ketamine and LSD, act on dopamine D2
and other receptors (such as serotonin receptors for
LSD), these compounds are not sufficiently selective
for testing theories of psychotomimetic action, as had
been thought.22 LSD is a known direct-acting partial
agonist at dopamine D2 receptors, lowering the
release of prolactin and stimulating rotation.23,24 The
Molecular Psychiatry
 Dopamine receptors, phencyclidine and ketamine
P Seeman et al
882
present data that LSD stimulated the incorporation of
[35S]GTP-g-S less than dopamine are consistent with
the partial agonist action of LSD reported by Giaco-
melli et al.23 Phencyclidine also stimulates dopamine
receptors by a direct or indirect action, lowering the
release of prolactin25 and causing rotation.6,7
Although prolactin release from hemi-pituitary
glands in vitro was not directly inhibited by 1 mM
phencyclidine,25 lower concentrations of phencycli-
dine need to be tested to determine whether phency-
clidine can directly alter the release of prolactin
release in vitro. This is because Pampillo et al26 found
that L-glutamate stimulates prolactin release via
ionotropic glutamate receptors, but inhibits prolactin
release via metabotropic glutamate receptors. It will
be essential, therefore, to test the direct in vitro action
of phencyclidine on the release of prolactin from cells
in culture, using various methods to separate phen-
cyclidine’s actions on both the glutamate receptors
and the dopamine D2 receptors. The analogue of
phencyclidine, ketamine, directly reduces the basal
release of prolactin.27
Because the present data are entirely in vitro, they
should not be confused or admixed with separate
clinical data. Nevertheless, the present in vitro data
are at least consistent with the clinical data on this
topic. That is, the clinical agonist actions of phency-
clidine and LSD in eliciting psychotic symptoms are
selectively treated in the hospital emergency room by
haloperidol and other dopamine D2-selective antago-
nists,28,29 supporting the idea that dopamine D2
receptors are the primary target for eliciting psychotic
symptoms. Although it has been claimed that psy-
chotic symptoms elicited by ketamine are not blocked
by haloperidol,3 Giannini et al30 have found that 5 mg
of intramuscular haloperidol caused significant re-
ductions in psychotic symptoms.31
It is important to note that the serum concentration
of phencyclidine that elicits psychotomimetic action
is between 10 and 50 nM.32 This is precisely the active
range of phencyclidine that directly acts on D2High
(Figure 1) and stimulates D2 (Figure 2). In contrast,
the NMDA receptor has phencyclidine Ki values of
97 nM34 and 124 nM35 on rat brain membranes, and
196 nM35 on the human cloned NMDA receptor (Table
1).
A similar situation exists for ketamine, where the
psychotomimetic concentration in the serum is of the
order of 100–500 nM,8 a value that matches the
ketamine action (100–500 nM) at D2High (Figure 3,
top), but not that at the NMDA receptor, which has a
ketamine Ki of 3100 nM for striatal tissue (Figure 3,
bottom), 760–2270 nM for brain membranes33,34 and
3150 nM34 for cloned NMDA receptors (Table 1).
The data in Figure 4 show that dizocilpine
recognized a high-affinity site for the D2 receptor at
dizocilpine concentrations of 3–100 nM when com-
peting vs the D2 ligand, [3H]domperidone. As men-
tioned in Results, it was necessary to confirm the
existence of this high-affinity site by using dopami-
nergic drugs to compete vs [3H]dizocilpine. This was
Molecular Psychiatry
carried out for the data in Table 2, which shows that
standard D2 agonists (NPA or N-propyl-norapor-
phine-(
), and apomorphine) and antagonists (halo-
peridol and domperidone) inhibited the binding of
[3H]dizocilpine to D2Long in the cloned CHO cells.
While it is known that drug Ki values generally
depend on the [3H]ligand used,17 nevertheless, the Ki
value for the standard D2 antagonist, haloperidol, for
example, inhibited the binding of [3H]dizocilpine at a
haloperidol Ki of 0.31 nM. This value of 0.31 nM
closely matches the [3H]haloperidol Kd of 0.4 nM and
the haloperidol Ki value of 0.7 nM when using
[3H]raclopride on cloned D2Long receptors.35 More
important than the absolute Ki values for NPA and
apomorphine in Table 2 is the observation that the
rank order of the agonists NPA and apomorphine
was characteristic for typical dopamine D2 re-
ceptors, namely that NPA was more potent than
apomorphine.
As discussed by Javitt and Zukin,32 phencyclidine
elicits a variety of behaviors in animals, including
unique discrimination of phencyclidine, hyperactiv-
ity, stereotypy and tranquilization, the latter effect
being especially pronounced in monkeys. While it is
not possible to attribute each of these behaviors to
single neurotransmitter systems, the pharmacology of
phencyclidine-induced hyperlocomotion helps to
clarify the role of dopamine. In particular, the work
of Ögren and Goldstein36 has shown that three
dopamine antagonists (haloperidol, raclopride and
remoxipride) blocked locomotion elicited by low and
high doses of phencyclidine. Because these three
dopamine antagonists have very low affinity for the
NMDA receptor,37 with raclopride and remoxipride
being highly selective for dopamine D2 receptors,38,39
dopamine must be an important component in the
phencyclidine hypoglutamate model of schizophre-
nia.
In conclusion, with phencyclidine and ketamine
having considerable potency at D2High receptors, the
use of these compounds as provoking agents in
animals or humans actually tests a combined dopa-
mine/glutamate theory of psychosis.
Acknowledgements
We thank Dr H-C Guan for excellent technical
assistance. Françoise Ko was supported by the
University of Toronto Adel Sedra Scholarship and
an Ontario Graduate Studentship. This research was
supported by the Ontario Mental Health Foundation
(Regular and Special Initiatives grants), NARSAD (the
National Alliance for Research on Schizophrenia and
Depression, to PS and FL), the CIHR (Canadian
Institutes of Health Research, to FL and PS), the
CPRF (Canadian Psychiatric Research Foundation, to
FL and PS), NIDA (the National Institute on Drug
Abuse), the SMRI (Stanley Medical Research Insti-
tute, to PS) and by donations from Dr Karolina Jus and
the Medland and O’Rorke families.

References
1 Van Rossum JM. The significance of dopamine-receptor blockade
for the action of neuroleptic drugs. In: Brill H, Cole J, Deniker P,
Hippius H, Bradley PB (eds). Neuropsychopharmacology, Proceed-
ings of the 5th Collegium Internationale Neuropsychopharmacolo-
gicum. Excerpta Medica Foundation: Amsterdam, 1967, pp 321–329.
2 Seeman P. Dopamine receptors and the dopamine hypothesis of
schizophrenia. Synapse 1987; 1: 133–152.
3 Tamminga CA, Holcom HH, Gao X-M, Lahti AC. Glutamate
pharmacology and the treatment of schizophrenia: current status
and future directions. Int Clin Psychopharmacol 1995; 10(Suppl
3): 29–37.
4 Jentsch JD, Roth RH. The neuropsychopharmacology of phency-
clidine: from NMDA receptor hypofunction to the dopamine
hypothesis of schizophrenia. Neuropsychopharmacology 1999; 20:
201–225.
5 Goff DC, Coyle JT. The emerging role of glutamate in the
pathophysiology and treatment of schizophrenia. Am J Psychiat
2001; 158: 1367–1377.
6 Lozovsky D, Saller CF, Bayorh MA, Chiueh CC, Rice KC, Burke Jr
TR et al. Effects of phencyclidine on rat prolactin, dopamine
receptor and locomotor activity. Life Sci 1983; 32: 2725–2731.
7 Mele A, Wozniak KM, Hall FS, Pert A. The role of striatal
dopaminergic mechanisms in rotational behavior induced by
phencyclidine and phencyclidine-like drugs. Psychopharmacol-
ogy (Berl) 1998; 135: 107–118.
8 Kapur S, Seeman P. NMDA receptor antagonists ketamine and PCP
have direct effects on the dopamine D2 and serotonin 5-HT2
receptors—implications for models of schizophrenia. Mol Psychiat
2002; 7: 837–844.
9 Seeman P. Comment on ‘Diverse psychotomimetics act through a
common signaling pathway’. Science 2004; 305: 180.
10 Seeman P, Tallerico T. Link between dopamine D1 and D2 receptors
in rat and human striatal tissues. Synapse 2003; 47: 250–254.
11 Seeman P, Niznik HB, Guan HC, Booth G, Ulpian C. Link between
D1 and D2 dopamine receptors is reduced in schizophrenia and
Huntington diseased brain. Proc Natl Acad Sci USA 1989; 86:
10156–10160.
12 Seeman P, Sunahara RK, Niznik HB. Receptor–receptor link in
membranes revealed by ligand competition: example for dopa-
mine D1 and D2 receptors. Synapse 1994; 17: 62–64.
13 George SR, Watanabe M, Di Paolo T, Falardeau P, Labrie F, Seeman P.
The functional state of the dopamine receptor in the anterior pituitary
is in the high-affinity form. Endocrinology 1985; 117: 690–697.
14 Lee FJ, Xue S, Pei L, Vukusic B, Chery N, Wang Y et al. Dual re-
gulation of NMDA receptor functions by direct protein–protein in-
teractions with the dopamine D1 receptor. Cell 2002; 111: 219–230.
15 Pei L, Lee FJ, Moszczynska A, Vukusic B, Liu F. Regulation of
dopamine D1 receptor function by physical interaction with the
NMDA receptors. J Neurosci 2004; 24: 1149–1158.
16 Salter MW. D1 and NMDA receptors hook up: expanding on an
emerging theme. Trends Neurosci 2003; 26: 235–237.
17 Seeman P, Tallerico T, Ko F. Dopamine displaces [3H]domperidone
from high-affinity sites of the dopamine D2 receptor, but not
[3H]raclopride or [3H]spiperone in isotonic medium. Implications
for human positron emission tomography. Synapse 2003; 49:
209–215.
18 Liu IS, George SR, Seeman P. The human dopamine D2(Longer)
receptor has a high-affinity state and inhibits adenylyl cyclase.
Mol Brain Res 2000; 77: 281–284.
19 Seeman P, Ulpian C, Wreggett, KA, Wells J. Dopamine receptor
parameters detected by 3H-spiperone depend on tissue concentra-
tion: analysis and examples. J Neurochem 1984; 43: 221–235.
Dopamine receptors, phencyclidine and ketamine
P Seeman et al
883
20 Seeman P, Watanabe M, Grigoriadis D, Tedesco JL, George SR,
Svensson U et al. Dopamine D2 receptor binding sites for agonists.
A tetrahedral model. Mol Pharmacol 1985; 28: 391–399.
21 Grigoriadis D, Seeman P. Complete conversion of brain D2dopa-
mine receptors from the high- to the low-affinity state for
dopamine agonists, using sodium ions and guanine nucleotide.
J Neurochem 1985; 44: 1925–1935.
22 Svenningsen P, Nomikos GG, Greengard P. Response to comment
on ‘Diverse psychotomimetics act through a common signaling
pathway’. Science 2004; 305: 180.
23 Giacomelli S, Pamery M, Romanelli R, Cheng CY, Silvestrini B.
Lysergic acid diethylamide (LSD) is a partial agonist of D2
dopaminergic receptors and it potentiates dopamine-mediated
prolactin secretion in lactotrophs in vitro. Life Sci 1998; 63:
215–222.
24 Meltzer HY, Fessler RG, Simonovic M, Doherty J, Fang VS.
Lysergic acid diethylamide: evidence for stimulation of pituitary
dopamine receptors. Psychopharmacology (Berl) 1977; 54: 39–44.
25 Meltzer HY, Simonovic M, Gudelsky GA. Phencyclidine-induced
inhibition of rat prolactin secretion: increased portal blood
dopamine. Eur J Pharmacol 1985; 110: 143–146.
26 Pampillo M, Theas S, Duvilanski B, Seilicovich A, Lasaga M.
Effect of ionotropic and metabotropic glutamate agonists and D-
aspartate on prolactin release from anterior pituitary cells. Exp
Clin Endocrinol Diabetes 2002; 110: 138–144.
27 Login IS. Direct stimulation of pituitary release by glutamate. Life
Sci 1990; 47: 2269–2275.
28 Giannini AJ, Eighan MS, Loiselle RH, Giannini MC. Comparison of
haloperidol and chlorpromazine in the treatment of phencyclidine
psychosis. J Clin Pharmacol 1984; 24: 202–204.
29 Giannini AJ, Nageotte C, Loiselle RH, Malone DA, Price WA.
Comparison of chlorpromazine, haloperidol and pimozide in the
treatment of phencyclidine psychosis: DA-2 receptor specificity.
J Toxicol Clin Toxicol 1984–85; 22: 573–579.
30 Giannini AJ, Underwood NA, Condon M. Acute ketamine
intoxication treated by haloperidol: a preliminary study. Am J
Ther 2000; 7: 389–391.
31 Krystal JH, D’Souza DC, Karper LP, Bennett A, Abi-Dargham A,
Abi-Saab D et al. Interactive effects of subanesthetic ketamine and
haloperidol in healthy humans. Psychopharmacology (Berl) 1999;
145: 193–204.
32 Javitt DC, Zukin SR. Recent advances in the phencyclidine model
of schizophrenia. Am J Psychiat 1991; 148: 1301–1308.
33 Murray F, Kennedy J, Hutson PH, Elliot J, Huscroft I, Mohnen K et
al. Modulation of [3H]MK-801 binding to NMDA receptors in vivo
and in vitro. Eur J Pharmacol 2000; 137: 263–270.
34 Grimwood S, Le Bourdelles BL, Atack JR, Barton C, Cockett W,
Cook SM et al. Generation and characterization of stable cell lines
expressing recombinant human N-methyl-D-aspartate receptor
subtypes. J Neurochem 1996; 66: 2239–2247.
35 Seeman P. Atypical antipsychotics: mechanism of action. Can J
Psychiat 2002; 47: 27–38.
36 Ögren SO, Goldstein M. Phencyclidine- and dizocilpine-induced
hyperlocomotion are differentially mediated. Neuropsychophar-
macology 1994; 11: 167–177.
37 Reynolds IJ, Miller RJ. Tricyclic antidepressants block N-methyl-D-
aspartate receptors: similarities to the action of zinc. Br J
Pharmacol 1988; 95: 95–102.
38 Köhler C, Hall H, gren S-O, Gawell L. Specific in vitro and in vivo
binding of 3H-raclopride. Biochem Pharmacol 1985; 34: 2251–
2259.
39 Hall H, Sällemark M, Jerning E. Effects of remoxipride and some
related new substituted salicylamides on rat brain receptors. Acta
pharmacol toxicol 1986; 58: 61–70.
Molecular Psychiatry