Further
Characterization of Quinpirole-Elicited Yawning
As a Model of
Dopamine
D3 Receptor Activation in Male and Female
Monkeys
Susan E. Martelle, Susan H. Nader, Paul W.
Czoty, William S. John, Angela N. Duke,
Pradeep K. Garg, Sudha Garg, Amy H. Newman
and Michael A. Nader
Departments of Physiology
and Pharmacology
and Radiology, Wake Forest
School of Medicine, Winston-Salem,
NC
and the Intramural Research
Program, National Institute on Drug Abuse,
Baltimore, MD
Abstract
The dopamine (DA) D3 receptor (DRD3) has
been associated with impulsivity, pathological
gambling and drug addiction making it a
potential target for pharmacotherapy
development. Positron emission tomography (PET)
studies using the DRD3-preferring radioligand
[11C]-(+)-propyl-hexahydro-naphtho-oxazin
([11C]PHNO) have shown higher binding
potentials in drug abusers compared to control
subjects.
Preclinical studies have examined DRD3
receptor activation using the DA agonist
quinpirole and the unconditioned behavior of
yawning. However, the relationship
between quinpirole-elicited yawning and
DRD3 receptor availability has not been
determined.
In Experiment 1, 10 drug-naive male rhesus
monkeys were scanned with [11C]PHNO and
the ability of quinpirole (0.01-0.3 mg/kg, i.m.)
to elicit yawning was examined. Significant
relationships between DRD3 receptor availability
and quinpirole-induced yawns were noted in
several brain regions. Experiment 2 replicated
earlier findings that a history of cocaine
self-administration did not affect
quinpirole-induced yawning and extended this to
examine monkeys with a history of
methamphetamine (MA) self-administration and
found that monkeys with experience
self-administering MA showed greater potency and
significantly higher quinpirole-elicited yawning
compared to controls.
Finally, quinpirole-elicited yawning was
studied in drug-naive female monkeys and
compared to drug-naive male. Sex differences
were noted, with quinpirole being more potent
and significantly eliciting more yawns in males
compared to females.
Taken together these findings support the
use of quinpirole-elicited yawning as a
behavioral tool for examining DRD3 activation in
monkeys and that both drug history and sex may
influence individual sensitivity to the
behavioral effects of DRD3 compounds
Psychostimulant drug use is a global
problem. In 2011, the United Nations Office on
Drugs and Crime estimated 14 to 21 million
people worldwide used cocaine at least once in
2009, with 20% of users in the United States
meeting the criteria of drug dependence
(Degenhardt and Hall, 2012).
Despite the array of negative health and
societal consequences, a successful
pharmacotherapy for psychostimulant addiction
has remained elusive (Newman et al., 2005,
2012). There is evidence that dopamine (DA) D3
receptors (DRD3), a subtype of the D2-like
family of DA receptors, mediate many of the
effects of psychostimulants associated with high
abuse potential (Heidbreder and Newman, 2010;
Heidbreder, 2012), including the role of
conditioned stimuli (Achat-Mendes et al., 2010;
Neisewander et al., 2004; Orio et al., 2010; Yan
et al., 2013), discriminative stimulus effects
(Achat-Mendes et al., 2010; Collins and Woods,
2009; Martelle et al., 2007), and cue
conditioning (Le Foll et al., 2002).
In addition, DRD3 partial agonists and
antagonists have been shown to reduce
self-administration of methamphetamine (MA) in a
rodent model of compulsive drug intake (Orio et
al., 2010). Postmortem studies indicate higher
DRD3 densities in cocaine overdose victims
compared to age-matched controls (Staley and
Mash, 1996; Segal et al., 1997), supporting a
role for this receptor subtype in drug
addiction. Finally, recent brain imaging studies
in humans using the positron emission tomography
(PET) DRD3-preferring ligand
[11C]-(+)-propyl-hexahydro-naphtho-oxazin
([11C]PHNO) revealed that DRD3
availability was higher in MA polydrug users
(Boileau et al., 2012) and cocaine-dependent
individuals (Payer et al., 2014) compared to
age-matched controls; an outcome opposite to the
decreases in D2-like receptor availability
following chronic stimulant exposure (e.g.,
Volkow et al., 1990, 2001; Nader et al.,
2006).
A behavioral assay that has been frequently
used to characterize DRD3 agonists is the
unconditioned behavior of drug-elicited yawning
using the D3/D2 receptor agonist quinpirole in
rodents (e.g., Collins et al., 2005, 2007;
Baladi and France, 2009) and nonhuman primates
(Martelle et al., 2007; Hamilton et al., 2010;
Blaylock et al., 2011).
The quinpirole-induced yawning dose-response
curve is an inverted-U shaped function of dose.
Collins, Woods and colleagues concluded that the
ascending limb of the quinpirole dose-response
curve is mediated by DRD3, while the descending
limb is primarily D2 receptor (DRD2)-mediated
and coincides with an onset of hypothermia
(Collins et al., 2005, 2007). Despite the
differences noted in D2-like and DRD3
availability following cocaine exposure,
quinpirole-elicited yawning was not different in
cocaine-naïve vs. cocaine-experienced
monkeys (Blaylock et al., 2011). However, while
DRD3 partial agonists do not elicit yawning when
administered to cocaine-naïve monkeys
(Martelle et al., 2007), they do elicit yawning
in monkeys with an extensive cocaine history
(Blaylock et al., 2011).
This finding suggests that drug-elicited
yawning may be an effective baseline on which to
assess functional consequences of chronic
cocaine exposure. In the present study, we
utilized quinpirole-elicited yawning to examine
the relationship between behavioral sensitivity
and measures of DRD3 receptor availability using
PET imaging with [11C]PHNO in
drug-naïve male rhesus monkeys. Secondly,
we extended previous findings that a cocaine
history does not influence sensitivity to
quinpirole-elicited yawning (Blaylock et al.,
2011; Collins et al., 2011) to monkeys with an
extensive MA self-administration history.
Finally, recent data investigating DA D2/D3
receptor availability in cocaine abuse (Nader et
al., 2012) suggested possible sex differences in
the relationship between DA receptor function
and cocaine abuse. Thus, we examined the effects
of quinpirole-elicited yawning in a group of
drugnaïve female rhesus monkeys and
compared them to drug-naïve male rhesus
monkeys.
DISCUSSION
The purpose of the present study was
three-fold. The first objective was to ascertain
the relationship between DRD3 binding potentials
and the behavioral effects of the D3-preferring
agonist quinpirole in monkeys. PET imaging using
the D3-preferring radioligand [11C]PHNO
revealed a significant correlation between DRD3
binding potential and quinpirole-induced yawning
in several regions of the brain including the
caudate nucleus, putamen, globus pallidus,
ventral pallidum and hippocampus.
The second objective was to test whether a
history of MA self-administration differentially
affected sensitivity to the unconditioned
behavioral effects of quinpirole. Using a
cumulative quinpirole dosing procedure,
differences between MAexperienced monkeys and
drug-naïve monkeys were noted, with the MA
group showing greater sensitivity to quinpirole.
We also replicated an earlier study (Blaylock et
al., 2011) showing no differences in
quinpirole-elicited yawning between
drug-naïve and cocaine-experienced
monkeys.
Lastly, the comparison of drug-naïve
male and female monkeys revealed sex
differences, with quinpirole showing greater
efficacy and greater potency in eliciting
yawning in males compared to females. These
findings suggest that the unconditioned behavior
of quinpirole-elicited yawning reflects DRD3
function in monkeys and that both sex and drug
history are determinants of individual
sensitivity. Receptor autoradiography studies
have indicated similar DRD3 receptor
distribution in nonhuman primates compared to
humans (Levant, 1998; Morisette et al., 1998;
Sun et al., 2012).
In the present study, the highest uptake and
binding potentials were in the caudate nucleus,
putamen and ventral striatum. Previous studies
in humans have found a similar pattern of
distribution of PHNO binding, with highest
binding potentials in the putamen, globus
pallidus and substantia nigra (Searle et al.,
2010; Boileau et al., 2012, 2013; Gallezot et
al., 2014). Tziortzi and colleagues (2011)
utilized the DRD3 antagonist GSK598809 to better
determine the contribution of DRD3 and DRD2 to
the PHNO signal in various brain regions.
These investigators found that 100% of PHNO
signal in the substantia nigra and hypothalamus
and the majority of the signal in the ventral
pallidum and globus pallidus were DRD3 mediated.
In contrast, only 20% of the signal in the
ventral striatum was attributed to DRD3 whereas
100% of the signal in the caudate nucleus was a
result of DRD2 binding (Tziortzi et al.,
2011).
Since the ascending limb of the
quinpirole-elicited yawning dose-response curve
is thought to be DRD3 mediated (Collins et al.,
2005, 2007), the results of the mixed model
which showed approximately 5 greater yawns for
every 1-unit increase in binding potential in
the globus pallidus, supports this
conclusion.
Furthermore, in the present study, yawning
was negatively correlated with binding in the
caudate nucleus and putamen, consistent with
predominant DRD2-mediated signal associated with
[11C]PHNO binding potential and
pharmacological studies suggesting that the
descending limb of the quinpirole-elicited
yawning dose-response curve is DRD2 mediated
(Collins et al., 2005, 2007). Recent PET studies
of chronic cocaine (Payer et al., 2014) and
poly-drug MA (Boileau et al., 2012) users
reported that both drugs were associated with
elevated levels of DRD3 availability in the
substantia nigra. Furthermore, the use of
cocaine and MA has previously been shown to be
associated with decreased D2-like receptor
availability (Volkow et al., 1990, 2001; Nader
et al., 2006; Payer et al., 2014).
In the present study, quinpirole-elicited
yawning was used to assess group differences in
DRD3 activity in monkeys with a cocaine or MA
history and drug-naïve male monkeys. In
drug-naïve monkeys, the quinpirole
dose-response curve was an inverted U-shaped
function of dose, as has been reported
previously in monkeys (Martelle et al., 2007;
Hamilton et al., 2010; Blaylock et al., 2011).
Similarly, quinpirole-elicited yawning was not
different between drug-naïve and
cocaine-exposed monkeys (Blaylock et al., 2011),
although there was substantial
individual-subject variability in the quinpirole
dose-response curves. The reasons for these
individual differences are not yet determined,
but may be due to differences in cocaine
intake.
Consistent with the present findings, a
recent study found substantial individual
subject differences in the reinforcing effects
of DRD3 agonists, quinpirole, pramipexole, and
ropinirole (Koffarnus et al., 2012), which may
be due to differences in cocaine
selfadministration histories.
Additional studies are needed to better
understand how different cocaine exposures
affect DRD3 compounds. In contrast to the lack
of differences between drug-naïve and
cocaine-exposed monkeys, MA-exposed monkeys were
more sensitive to the behavioral effects of
quinpirole compared to control monkeys. Changes
in sensitivity on the ascending limb of the
quinpirole dose-response curve have been
hypothesized to be due to upregulation of DRD3
and/or downregulation of DRD2 (Collins et al.,
2011); PET and receptor autoradiography studies
support this mechanism following long-term MA
exposure (Kuczenski et al., 2009; Boileau et
al., 2012).
It also remains possible that changes in DA
transporter (DAT) and subsequent changes in
extracellular DA may interact with quinpirole's
effects at DRD3 and DRD2. Consistent with this
hypothesis is that MA exposure has been shown to
decrease DAT (Chu et al., 2008; Groman et al.,
2013) while chronic cocaine exposure has been
reported to increase DAT density (Letchworth et
al., 2001). Irrespective of the mechanisms, the
present findings suggest that despite both
stimulants resulting in increases in
[11C]PHNO binding (Boileau et al., 2012;
Payer et al., 2014), the behavioral effects of
DRD3 compounds may be different in subjects with
a MA history compared to a cocaine history.
There is accumulating evidence of sex
differences in cocaine abuse (O'Brien and
Anthony, 2005), including greater vulnerability
in initiating drug use, progressing to
dependence faster, and more adverse physical,
mental, and social consequences of abuse in
women compared to men (Zilberman et al., 2003;
Greenfield et al., 2010). Several animal models
using female subjects have supported these
observations (e.g., Lynch and Carroll, 2002;
Mello et al., 2007; Mello, 2010; Nader et al.,
2012).
As it relates to DRD3 function, sex
differences have not been reported. In the
present study, we found significant differences
between male and female monkeys at lower
quinpirole doses, with females being less
sensitive to the behavioral effects of
quinpirole. It is possible that the entire
quinpirole dose-response curve for females is
shifted to the right of males and that if higher
doses of quinpirole were tested, even greater
sex differences would have been observed.
However, it is important to note that
quinpirole-induced hypothermia was not different
in drug-naïve females and males suggesting
that baseline DRD2 availability is not
different.
PET data using DRD2-like radiotracers in
monkeys have confirmed no significant
differences between male and female monkeys
(Hamilton et al., 2010). Although
DRD2-availability may not be different at
baseline, our data suggest that sex differences
in pharmacodynamics may be influencing the
behavioral results. Consistent with this
hypothesis, an investigation of healthy human
subjects reported that drug-elicited DA release
in the ventral striatum, caudate nucleus and
putamen was greater in male subjects compared to
females (Munro et al., 2012).
Additionally, a recent study showed that
female smokers have significantly higher D2-like
receptor availability than male smokers (Brown
et al., 2012), further supporting the growing
evidence of gender differences in dopaminergic
system dynamics (Becker, 1999; Dreher et al.,
2007; Festa et al., 2010; Hedges et al., 2010).
On the other hand, the quinpiroleelicited
yawning dose-response curve has been shown to be
malleable to changes in food content and body
weight (Baladi and France, 2009). Although males
and females in this study ate the same
high-protein diet, the heaviest female weighed
less than the lightest male.
Whether changes in diet would similarly
affect the quinpirole dose-response curve in
males and females remains to be investigated.
There is evidence supporting the use of DRD3
antagonists and partial agonists for the
treatment of cocaine and MA abuse as well as
accompanying symptoms such as cognitive
dysfunction (Newman et al., 2012; Mugnaini et
al., 2013; Nakajima et al., 2013). Considering
female rhesus monkeys in the current study were
less sensitive to the behavioral effects of
quinpirole, it is likely that drug potency will
be a critical variable in future studies
investigating DRD3 compounds as a treatment
option.
As there are known gender differences in
drug pharmacokinetics and pharmacodynamics (see
Gandhi et al., 2004 for review), and also that
the fields of Neuroscience and Pharmacology
publish predominately male subject-based studies
(Beery and Zucker, 2011), the present findings
support the consideration of sex differences as
a critical variable in the development of
treatment strategies for drug abuse (Zilberman
et al., 2003; see Becker and Hu, 2008). Finally,
this and other studies (Bouileou et al., 2013;
Payer et al., 2014) have associated DRD3
availability and a history of psychostimulant
abuse but have yet to establish the nature of
this relationship. Future studies will need to
focus on whether high or low DRD3 availability
is a risk factor in psychostimulant abuse or a
result of exposure to the drug. There are some
limitations to the present study.
The relationship between [11C]PHNO
binding potential and quinpirole-elicited
yawning is not consistent with DRD3 and DRD2
signal in every brain region examined. For
example, monkeys experienced fewer yawns with
increasing DRD3 binding potential in the ventral
pallidum. This relationship is not consistent
with the high DRD3 signal reported in that brain
region (Tziortzi et al., 2011), suggesting
possible alternative mechanism.
Past studies have implicated specific brain
areas in dopamine agonistelicited yawning
including the nigrostriatal pathway,
hypothalamus, and the paraventricular nucleus
(Dourish et al., 1985; Melis et al., 1987), thus
it is expected that yawning does not necessarily
have to correlate with DRD3 and DRD2 signal in
every brain region. Furthermore, yawning is not
limited to dopaminergic modulation and several
other neurotransmitters and neurohormones are
involved (Argiolas and Melis, 1998) creating a
complex neural circuit (Sanna et al., 2012) that
requires additional testing.
