CONTRIBUTION OF OPIOID AND STRESS SYSTEMS TO IMPULSIVE
ACTION IN RODENT MODELS
by
Megan Kelly Mahoney
A thesis submitted to the Department of Psychology
In conformity with the requirements for
the degree of Doctor of Philosophy
Queen’s University
Kingston, Ontario, Canada
(January, 2015)
Copyright © Megan Kelly Mahoney, 2015
i
Abstract
Impulsive action, the inability to withhold responding, is a primary symptom of many psychiatric
disorders, including drug addiction. Impulsive action is mediated through a complex interplay of
many different neurotransmitters. The purpose of the experiments presented in this thesis was to
further elucidate the role of opioid activity and environmental influences (e.g., stress) on
impulsive action. Chapters 2 – 4 extend previous findings showing mu and delta opioid receptor
(M/DOR) involvement in impulsive action. In all experiments, animals were trained to withhold
responding for sucrose until a discriminative stimulus (cue light) was presented. In Chapter 2,
rats were tested in a Variable delay response inhibition (RI) task in which the pre-response (i.e.,
premature) phase ranged from 1-60 s; thus, animals could not predict the length of the premature
phase. Following training, rats were tested for impulsive action following injections of the DOR
agonist, SNC80, or the MOR agonist, morphine. SNC80 dose-dependently increased premature
responding in the Variable delay task, whereas none of the morphine doses had any effect.
SNC80-induced increases in premature responding were blocked by a highly specific DOR
antagonist (naltrindole). Morphine had no effect on responding in the Variable delay task, so in
Chapter 3 we tested the effects of morphine in two Fixed delay tasks (4- and 60-s). Morphine
dose-dependently increased premature responding in both the Fixed 4s and 60s versions of the RI
task, an effect blocked by MOR antagonism. In Chapter 4, we assessed impulsive action in mice
conditionally lacking MORs or DORs located on GABAergic forebrain neurons (cKO), and in
constitutive MOR or DOR knockout mice (KO). Surprisingly, neither cKO nor KO mice differed
in impulsive responding compared to control mice. Finally, in Chapter 5, we examined whether
the pharmacological stressor, yohimbine, increases impulsive action. Yohimbine dosedependently increased impulsive action; however antagonism of stress (CRF and glucocorticoid
ii
receptors), noradrenergic, dopaminergic, or opioidergic systems did not alter this response.
Collectively, these studies demonstrate an important role for both opioid systems and stress in
impulsive action, but highlight the importance of methodological differences when interpreting
pharmacological and genetic behavioural studies.
iii
Acknowledgements
First and foremost, I want to express my utmost gratitude to my supervisor, Cella
Olmstead. Cella is an exceptional supervisor. I would not have had such a fulfilling experience
nor would I be the researcher or person I am now had it not been for her supervisory role. I
aspire to be half as great a mentor as she is.
Secondly, I want to thank all members of the Olmstead lab, both past and present. Scott
Hayton was an amazing friend and mentor during the initial phases of my PhD, and set a work
and sarcasm standard that I have always aimed to achieve. Justin Giannoccaro, Amanda Maracle,
Sylvia Magrys, and Edmund Ong have also been helpful colleagues, and more importantly, they
have become life-long friends. I’m also fortunate to have worked with some wonderful
undergraduate students. Mason Silveira and John Barnes in particular were excellent students
and delightful people who greatly contributed to my projects, and sanity.
Roland Dupras, Rick Eves, Justin Siu, and Chris Degen – the guys in the shop – were so
helpful during the many technical errors that happened during my thesis. I am particularly
grateful to Roland, who, on multiple occasions, offered to come in during holidays and
vacations, if I needed him. These guys really made things run smoothly, and were always
friendly in doing so.
I would like to greatly thank Brigitte Kieffer for the opportunity to conduct research in
her lab at the IGBMC in Strasbourg, France. Additionally, thanks to many members of the
Kieffer lab who not only taught me technical skills, but also made going to work enjoyable.
Notably, Katia Befort, Paul Chu Sin Chung, Pauline Charbogne, Lauren Faget, Dominique
Filliol, Olivier Gardon, Claire Gaveriaux-Ruff, Dominique Massotte, and Audrey Matifas all
added to my wonderful French experience. I’d like to thank Katia Befort in particular, who acted
iv
as a great mentor and friend both while I was in France, and ever since. In addition to my lab
mates, I’d like to acknowledge Felicie Hentzgen and Yannick Lafue who were exceptionally
welcoming friends when I first arrived to France, and ever since. All of these people made living
in Strasbourg, France one of my most memorable life experiences to date, and I’m so fortunate to
have met such exceptional people.
And to my closest friends – Jessica Livingston, Claire Rowsell, Livia Chyurlia, Maaike
Hum, Alex Mattioli, and Sarah Simpson – you guys mean the world to me, and I appreciate all
the support you’ve given me throughout the years. I could recount endless memories, but I’ll
save that for another time!
Finally, I give my greatest thanks to my parents and family, who are always so positive
and comforting. My family, including Evelyn, Julie, Kent and Marie, and their partners and
children have been such a positive support for me throughout my education – from start to finish.
And to my parents, who not only provided me with financial support throughout my education,
but more importantly, emotional support. I am particularly appreciative of them for always
giving me autonomy to make my own path. Thanks to them for teaching me to preserver and for
helping me become the person I am today. I am so happy to dedicate my dissertation to them.
v
Table of Contents
Abstract ........................................................................................................................................ i
Acknowledgements .................................................................................................................... iii
Table of Contents ........................................................................................................................ v
List of Tables ............................................................................................................................... x
List of Figures and Illustrations ................................................................................................. xi
List of Abbreviations ................................................................................................................ xiii
General Introduction ...................................................................................................... 1
1.1 Overview ............................................................................................................................... 1
1.2 Impulsivity ............................................................................................................................ 1
Impulsive choice ............................................................................................................. 2
Impulsive action ............................................................................................................. 2
Impulsivity as a risk factor for psychiatric disorders ..................................................... 4
Neurocircuitry of impulsive action ................................................................................. 6
1.3 Opioid system ...................................................................................................................... 11
Cre-LoxP recombination strategy for constitutive and conditional gene knockout ..... 13
M/DOR constitutive and conditional knockout mouse construction............................ 15
Opioid receptor involvement in impulsivity................................................................. 17
1.4 Stress system ....................................................................................................................... 18
Stress as a risk factor of drug addiction ........................................................................ 20
Stress-impulsivity interactions ..................................................................................... 21
1.5 Objectives ............................................................................................................................ 22
Effects of Delta Opioid Receptor Activation on a Response Inhibition Task in Rats . 23
2.1 Abstract ............................................................................................................................... 23
vi
2.2 Introduction ......................................................................................................................... 24
2.3 Materials and methods ........................................................................................................ 26
Subjects ......................................................................................................................... 26
Apparatus ...................................................................................................................... 26
Drugs ............................................................................................................................ 27
Response inhibition training ......................................................................................... 27
Statistical analyses ........................................................................................................ 32
2.4 5 Results .............................................................................................................................. 33
Response inhibition training ......................................................................................... 33
Effects of morphine on accuracy .................................................................................. 34
Effects of SCN80 on accuracy...................................................................................... 35
Effects of morphine and SNC80 on locomotor activity ............................................... 40
2.5 Discussion ........................................................................................................................... 41
Increased Impulsive Action in Rats: Effects of Morphine in a Short and Long FixedDelay Response Inhibition Task ................................................................................................... 46
3.1 Abstract ............................................................................................................................... 46
3.2 Introduction ......................................................................................................................... 47
3.3 Materials and methods ........................................................................................................ 49
Subjects ......................................................................................................................... 49
Apparatus ...................................................................................................................... 49
Drug Administration ..................................................................................................... 49
Behavioral procedures .................................................................................................. 50
Data and statistical analyses ......................................................................................... 52
vii
3.4 Results ................................................................................................................................. 54
Baseline response inhibition training ........................................................................... 54
Effect of morphine on premature responding in Fixed 4-s and 60-s conditions .......... 54
Effect of naloxone on morphine-induced premature responding ................................. 57
Effect of morphine on the proportion of premature responses ..................................... 61
Effect of naloxone on morphine-induced increases in the proportion of premature
responses ................................................................................................................................ 64
3.5 Discussion ........................................................................................................................... 66
Lack of an Impulsivity Phenotype in Constitutive or Conditional Mu or Delta Opioid
Receptor Knockout Mice .............................................................................................................. 71
4.1 Abstract ............................................................................................................................... 71
4.2 Introduction ......................................................................................................................... 71
4.3 Materials and methods ........................................................................................................ 73
Subjects ......................................................................................................................... 73
Apparatus ...................................................................................................................... 74
Statistical analyses ........................................................................................................ 77
4.4 Results ................................................................................................................................. 78
Training ........................................................................................................................ 78
Test ............................................................................................................................... 83
4.5 Discussion ........................................................................................................................... 99
Acute Pharmacological Stress Increases Impulsive Action in a Rodent Model ........ 103
5.1 Abstract ............................................................................................................................. 103
5.2 Introduction ....................................................................................................................... 103
viii
5.3 Materials and methods ...................................................................................................... 106
Subjects ....................................................................................................................... 106
Apparatus .................................................................................................................... 106
Drug administration .................................................................................................... 106
Behavioural procedures .............................................................................................. 107
Data and statistical analyses ....................................................................................... 109
5.4 Results ............................................................................................................................... 110
Baseline response inhibition training and drug-free test sessions .............................. 110
Effect of yohimbine on impulsive action ................................................................... 111
Effect of antalarmin on yohimbine-induced impulsive action ................................... 113
Effect of mifepristone on yohimbine-induced impulsive action ................................ 115
Effect of prazosin on yohimbine-induced impulsive action ....................................... 116
Effect of guanfacine on yohimbine-induced impulsive action ................................... 117
Effect of SCH 39166 on yohimbine-induced impulsive action .................................. 118
Effect of naloxone on yohimbine-induced impulsive action ...................................... 119
5.5 Discussion ......................................................................................................................... 124
General Discussion ..................................................................................................... 128
6.1 Summary of findings ......................................................................................................... 128
Opioid receptor involvement in impulsive action ...................................................... 128
Stress involvement in impulsive action ...................................................................... 130
6.2 Neuropharmacological mechanisms ................................................................................. 131
Opioid-induced impulsive action................................................................................ 131
Stress-induced impulsive action ................................................................................. 133
ix
Cognitive mechanims of opioid- and stress-induced impulsive action ...................... 134
6.3 Theoretical considerations: parsing impulsive action from other behavioural measures . 136
Risky decision making................................................................................................ 137
Locomotor reactivity .................................................................................................. 138
Reversal learning ........................................................................................................ 139
Anxiety-like behaviour ............................................................................................... 140
Pavlovian reward cue respond .................................................................................... 140
Vulnerability to addiction: unique contribution of behavioural endophenotypes ...... 141
6.4 Conclusions and clinical relevance ................................................................................... 144
References ............................................................................................................................... 146
Appendix I Responding in drug-free yohimbine test sessions ............................................... 171
Appendix II ANOVA summary table for drug-free yohimbine test sessions........................... 172
Appendix III Effects of antagonists on yohimbine-induced responding during drug-free test
sessions .................................................................................................................................... 173
Appendix IV Effects of antagonists on yohimbine-induced premature response latencies in
drug-free test sessions ............................................................................................................. 174
Appendix V ANOVA summary table for drug-free test sessions for antagonists on yohimbineinduced responding ................................................................................................................. 175
x
List of Tables
List of Figures and Illustrations ................................................................................................. xi
List of Abbreviations ................................................................................................................. xiii
Table 1.1 Role of different neurotransmitter systems involved in impulsive action in rodent
models. ............................................................................................................................................ 9
Table 1.2 Classification of opioid receptor endogenous ligands ................................................. 12
Table 2.1 Effects of SNC80 on omissions in the RI task ............................................................. 38
Table 2.2. Effects of SCN80 on latencies to a premature response in the RI task. ...................... 39
Table 2.3. Effect of SCN80 on latencies to a correct response in the RI task. ............................ 39
Table 3.1. Effects of morphine on behavioral measures in the RI task........................................ 56
Table 3.2. Effect of naloxone on morphine-induced behavioral changes in the RI task ............. 60
Table 3.3. Effects of naloxone on behavioral measures in the RI task ........................................ 61
Table 4.1. Proportion of mice that reached criterion during operant training .............................. 77
Table 4.2. Summary of ANOVA results on measures of the Fixed 4 s Task in Delta Mice ....... 85
Table 4.3. Summary of ANOVA results on measures of the VT 1-8 s Task in Delta Mice ........ 90
Table 4.4. Summary of ANOVA results on measures of the Fixed 4 s Task in Mu Mice........... 95
Table 5.1. Effects of yohimbine on measures in the RI task ...................................................... 112
Table 5.2. Effects of antagonists on yohimbine-induced responding ........................................ 121
Table 5.3. Effects of antagonists on yohimbine-induced responding ........................................ 123
Table 6.1. Role of behavioural phenotypes to drug take during different phases of drug
addiction ...................................................................................................................................... 143
Appendix I Responding in the RI task of yohimbine on drug-free test sessions ....................... 171
Appendix II ANOVA summary table for drug-free test sessions of yohimbine......................... 172
Appendix III Effects of antagonists on yohimbine-induced responding on measures of the RI
task on drug-free test sessions..................................................................................................... 173
Appendix IV Effects of antagonists yohimbine-induced responding on premature response
latency in the RI task on drug-free test sessions ......................................................................... 174
Appendix V ANOVA summary table for drug-free test sessions on yohimbine-induced
responding on measures of the RI task ....................................................................................... 175
xi
List of Figures and Illustrations
Figure 1.1 Simplified schematic of the Cre-LoxP genetic recombination strategy ..................... 15
Figure 1.2 Sagittal view of a representative mouse brain illustrating M/DOR receptor a) gene
expression and b) ligand binding in M/DOR cKO mice............................................................... 17
Figure 1.3 Schematic of hypothalamic-pituitary-adrenal axis and extra-hypothamalamic brain
regions involved in the stress response. ........................................................................................ 19
Figure 2.1 Response inhibition task ............................................................................................. 30
Figure 2.2 Effects of morphine on motor impulsivity ................................................................. 35
Figure 2.3 Effects of SNC80 on motor impulsivity ..................................................................... 37
Figure 2.4 Effects of SNC80 on locomotor responses ................................................................. 40
Figure 3.1 Effects of morphine on impulsive action .................................................................... 55
Figure 3.2 Effects of naloxone on morphine-induced increases in impulsive action .................. 58
Figure 3.3 Effects of morphine on the proportion of responses during the premature phase ...... 63
Figure 3.4 Effects of naloxone on morphine-induced increases in proportion of responses during
the premature phase ...................................................................................................................... 65
Figure 4.1 Schematic of the progression of each trial in the mouse response inhibition task ..... 76
Figure 4.2 Number of sessions to criterion for each of the training phases for delta mice. ........ 80
Figure 4.3 Number of sessions to criterion for each of the training phases for delta mice ......... 81
Figure 4.4 Number of sessions to criterion for each of the training phases for mu mice ............ 84
Figure 4.5 Efficiency in delta mice .............................................................................................. 84
Figure 4.6 Mean a) inter-trial interval, b) premature, and c) correct response nose pokes for delta
mice ............................................................................................................................................... 86
Figure 4.7 Mean a) reward presentations, b) percent reward omissions, and c) trial omissions for
delta mice ...................................................................................................................................... 87
Figure 4.8 Mean a) premature response latency (s), b) correct response latency (s), and c) nonreinforced nose pokes for delta mice ............................................................................................ 88
Figure 4.9 Efficiency in delta mice .............................................................................................. 89
Figure 4.10 Mean a) inter-trial interval, b) premature, and c) correct response nose pokes for
delta mice ...................................................................................................................................... 91
xii
Figure 4.11 Mean a) reward presentations, b) percent reward omissions, and c) trial omissions
for delta mice ................................................................................................................................ 92
Figure 4.12 Mean a) premature response latency (s), b) correct response latency (s), and c) nonreinforced nose pokes for delta mice ............................................................................................ 93
Figure 4.13 Efficiency in mu mice............................................................................................... 94
Figure 4.14 Mean a) inter-trial interval, b) premature, and c) correct response nose pokes for mu
mice ............................................................................................................................................... 96
Figure 4.15 Mean a) reward presentations, b) percent reward omissions, and c) trial omissions
for mu mice ................................................................................................................................... 97
Figure 4.16 Mean a) premature response latency (s), b) correct response latency (s), and c) nonreinforced nose pokes for mu mice ............................................................................................... 98
Figure 5.1 Effects of yohimbine on impulsive action in the response inhibition task ............... 111
Figure 5.2 Effects of antalarmin on yohimbine-induced impulsive action ................................ 114
Figure 5.3 Effects of mifepristone on yohimbine-induced impulsive action ............................. 116
Figure 5.4 Effects of prazosin on yohimbine-induced impulsive action ................................... 117
Figure 5.5 Effects of guanfacine on yohimbine-induced impulsive action ............................... 118
Figure 5.6 Effects of SCH 39166 on yohimbine-induced impulsive action .............................. 119
Figure 5.7 Effects of naloxone on yohimbine-induced impulsive action .................................. 120
Figure 6.1 Schematic of opioid receptor distribution in impulsive action-related regions ........ 133
xiii
List of Abbreviations
5-CSRTT
5-choice serial reaction time task
5-HT
Serotonin
ACC
Anterior cingulate cortex
Ant
Antalarmin
ADHD
Attention-deficit hyperactivity disorder
ANOVA
Analysis of variance
BLA
Basolateral amygdala
BNST
Bed nucleus of the stria terminalis
cAMP
Cyclic adenosine monophoshphate
cKO
Conditional knockout
Cre
Cyclic recombinase
CRE
Cis-regulatory element
CREB
Cyclic adenosine monophoshphate response element binding
CRF
Continuous reinforcement
CRF
Corticotropin releasing factor
CRF1
Corticotropin releasing factor receptor 1
CMV
Cytomegalovirus
CPu
Caudate putamen
D1
Dopamine receptor subtype 1
D1/5
Dopamine receptor subtype 1/5
D2/3
Dopamine receptor subtype 2/3
DA
Dopamine
DOR
Delta opioid receptor
DS
Dorsal striatum
Fl
Flox
GABA
Gamma-aminobutyric acid
GABAA
Gamma-aminobutyric acid receptor subtype A
GC
Glucocorticoid
GPCR
G-protein coupled receptor
xiv
Gua
Guanfacine
HI
High impulsive
HI-Act
High impulsive action
HI-Ch
High impulsive choice
HP
Hippocampus
HPA
Hypothalamic-pituitary-adrenal
IL
Infralimbic
ITI
Inter-trial interval
LI-Act
Low impulsive action
LI-Ch
Low impulsive choice
KO
Knockout
KOR
Kappa opioid
Mif
Mifepristone
Mor
Morphine
MOR
Mu opioid receptor
mPFC
Medial prefrontal cortex
MSN
Medium spiny neuron
mGluR
Metabotropic glutamate receptor
Nal
Naloxone
NRL
Non-reinforced lever presses
NTI
Naltrindole
NA
Noradrenalin
NAcb
Nucleus accumbens
NAcbC
Nucleus accumbens core
NAcbS
Nucleus accumbens shell
NMDA
N-methyl-D-aspartate
nNOS
Nitric oxide synthase
OFC
Orbitofrontal cortex
PL
Prelimbic
PLd
Dorsal prelimbic
PFC
Prefrontal cortex
xv
Pra
Prazosin
PVN
Paraventricular nucleus of the hypothalamus
RI
Response inhibition
SED
Standard error of the differences of the means
SSRT
Stop-signal reaction time task
Veh
Vehicle
vmPFC
Ventromedial prefrontal cortex
VT
Variable-time
VTA
Ventral tegmental area
1
General Introduction
1.1 Overview
The following introduction summarizes impulsivity, including a description of
impulsivity subtypes, the role of impulsivity in psychiatric disorders, and the neurocircuitry that
mediates impulsive action. In the second portion of the introduction, I will describe opioid and
stress system involvement in impulsivity. I will briefly describe these findings with respect to all
psychiatric disorders, but will then focus on impulsivity’s role in drug addiction, as it is highly
implicated in drug-taking behaviours.
1.2 Impulsivity
Impulsivity is a multifaceted construct that includes several distinct features. Impulsivity
is defined as a preference for immediate over delayed reward, an inability to withhold
responding, an inability to think about the future consequences of actions, and increased
engagement in risky or novelty-seeking behaviours. These factors, or characteristics of
impulsivity, appear to be dissociable in that the behavioural and neurobiological underpinnings
of each do not necessarily correlate (Evenden 1999; Robinson et al. 2009; Broos et al. 2012b;
Richards et al. 2013; Jenstch et al. 2014). Of these factors, preference for immediate over
delayed reward (impulsive choice) and an inability to withhold responding (impulsive action)
have been most extensively examined in both clinical and preclinical populations as they are
most implicated in psychiatric disorders, specifically substance abuse (Olmstead 2006; Pattij and
De Vries 2013; Jentsch and Pennington 2014). Broadly speaking, impulsive choice relates more
2
to a dysfunctional reward system that is associated with aversion for delayed reward whereas
impulsive action is conceptualized as impairments in inhibitory control (Dalley et al. 2011).
Impulsive choice
In measures of impulsive choice, subjects (humans and non-human) must choose between
two options that are presented concurrently. One example is a delay discounting task in which
subjects are provided with two alternatives: a lesser reward presented immediately (e.g., money,
drug, sucrose pellet), or a larger reward presented following one of several temporal delays (e.g.,
1 day versus 6 months in humans, and 0 s versus 60 s in rats) (Cardinal et al. 2001; Kirby and
Petry 2004; Pothuizen et al. 2005). Models of impulsive choice measure how the subjective
value of the delayed reward decreases as delays increase: high impulsive subjects tend to devalue
the distal reward option. Impulsive choice is often assessed by an indifference point value; the
point at which immediate and delayed reward options are chosen equally. In other words, high
impulsive individuals, both humans and non-humans, have lower indifference points.
Impulsive action
Impulsive action (referred to as motor impulsivity and response inhibition in Chapter 2)
refers to the inability to withhold responding, and can be further divided into cancellation and
withholding subtypes. Both subtypes of impulsive action encompass inhibitory control of a
behaviour; however, cancellation and withholding subtypes involve the inability to stop an
already-initiated response and withhold (i.e., prevent) a response from occurring, respectively. In
other words, they differ in when the subject (human or animal) should inhibit responding. In
paradigms measuring cancellation impulsivity, subjects must inhibit an already initiated response
upon presentation of an infrequent “stop signal” (i.e., a discriminative stimulus). In the stop-
3
signal reaction-time task (SSRT), rodents are trained to make rapid responses between two levers
presented concurrently (Logan and Cowan 1984). In this task, approximately 20% of trials are
stop trials, in which animals must initiate the same response as go trials (i.e., press the first
lever), but they must inhibit responding on the second lever after stop signal presentation (i.e.,
stop the completion of the go response). The stop signal (i.e., sound) is presented at a variable
time (in milliseconds) after the rat presses the first lever. As it is impossible to calculate the
reaction time of an inhibited response, these are estimated as probabilities using validated race
model calculations based on reaction time during go trials (Logan 1994). Longer reaction times
on stop signal trials are indicative of diminished inhibitory control in the SSRT.
Withholding paradigms require subjects to withhold a response until the presentation of a
specific discriminative stimulus, or for a certain delay. Unlike cancellation paradigms of
impulsive action, subjects must not initiate a response. The go/no go task requires subjects to
make (go trial) or inhibit (no/go trial) a response on discrete trials, depending on the presented
discriminative stimuli; increased impulsive action is measured as a greater proportion of no/go
trial responses (Newman et al. 1985). A more commonly used measure of this subtype is the 5choice serial reaction time task (5-CSRTT), which measures impulsive action, visuo-spatial
attention, and perseverative responding (Robbins 2002). This complex test requires months of
training and is cognitively demanding. To circumvent these limitations, we developed a model of
impulsive action, the response inhibition (RI) task (Hayton et al. 2012). In this task, animals
must withhold responding for either a fixed or variable delay (premature phase), with the
difference being that animals can not predict the length of the premature phase in the variable
delay task. The RI task requires less training than other impulsive action paradigms, and allows
us to examine the role of predictability of the premature delay length on impulsive responding.
4
We previously validated the RI task with amphetamine, a DA agonist, known to increase
impulsive action on other measures (Cole and Robbins 1987; Hayton et al. 2012).
Impulsivity as a risk factor for psychiatric disorders
At extreme levels, impulsive behaviour is maladaptive: it may be risky, premature, and
inappropriate. Several psychiatric disorders are associated with extreme degrees of impulsivity
which have negative consequences to the individual, and are associated with undesirable and
poor clinical outcomes (Dom et al. 2006; Heyman and Gibb 2006; von Diemen et al. 2008;
American Psychiatric Association 2013). Impairments of impulsivity are a primary feature of
several disorders including attention deficit hyperactivity disorder (ADHD) (Solanto et al. 2001;
Castellanos and Tannock 2002; Urcelay and Dalley 2012), problem gambling (van Holst et al.
2010; Brevers et al. 2012; Goudriaan et al. 2014), eating disorders (Claes et al. 2005; Cassin and
von Ranson 2005), borderline personality disorder (Bornovalova et al. 2005), and drug abuse and
addiction (Verdejo-García et al. 2008; Jentsch and Pennington 2014). Of these disorders, drug
abuse and addiction has received significant attention, in part because of the particularly severe
social and financial outcomes and high prevalence of this disease. As a result, determining
predisposing vulnerability factors and respective neurobiological mechanisms that increase ones
risk to addiction is particularly important. High levels of impulsivity is one such risk factor that
predicts all aspects of drug addiction from initiation of drug use to heightened degrees of craving
during withdrawal (De Wit 2009; Tziortzis et al. 2011). For example, in non-dependent
populations, impulsivity predicts the degree of future drug use in non-using adolescents (Nigg et
al. 2006; von Diemen et al. 2008), as well as severity of current drug use in adults (Kollins
2003). In drug dependent populations, it predicts craving, relapse, and poly drug use (Charney et
al. 2010; Joos et al. 2013).
5
Studies with current and previously drug-dependent populations are difficult to interpret
because the relationship between impulsivity and drug use is bidirectional, with chronic drug use
further increasing impulsivity (Jentsch and Taylor 1999; Perry and Carroll 2008; De Wit 2009).
To address this problem, animal models are used in which subjects are tested on impulsive
responding prior to drug access in order to assess pre-existing, trait-like levels of impulsive
responding. For instance, rats are categorized as high (HI-Act) or low (LI-Act) on impulsive
action based on premature responding in the 5-CSRTT (Dalley et al. 2007), and high (HI-Ch) or
low (LI-Ch) on impulsive choice based on indifference points on the delay discounting task
(Diergaarde et al. 2008). This type of experimental design is beneficial in that neurobiological
differences between the low and high impulsivity groups may represent an underlying biological
vulnerability marker of drug addiction. For instance, HI-Act rats are more likely to escalate their
cocaine or nicotine intake (Dalley et al. 2007; Belin et al. 2008; Diergaarde et al. 2008; Besson et
al. 2013), seek cocaine despite negative consequences (Belin et al. 2008), and reinstate cocaine
seeking following abstinence (Koob and Le Moal 2008; Economidou et al. 2009; Pattij and De
Vries 2013); these three behaviours are defining features of drug addiction that can be assessed
using animal models (American Psychiatric Association 2013). Similarly, HI-Ch subjects also
display escalated cocaine (but not nicotine) self-administration (Perry et al. 2005; Perry et al.
2008; Diergaarde et al. 2008; Anker et al. 2009), are resistant to cocaine extinction (Broos et al.
2012a), and show greater reinstatement for cocaine and nicotine (Diergaarde et al. 2008; Broos et
al. 2012b).
In a series of experiments, Dalley, Belin and colleagues showed that HI-Act animals have
a greater propensity to transition from stable to compulsive intake (Dalley et al. 2007; Belin et al.
2008), even though they do not necessarily differ on their propensity to initially (first few self-
6
administration sessions) acquire drug self-administration or maintain drug intake (Dalley et al.
2007; Belin et al. 2008; Perry et al. 2008; Besson et al. 2013). These findings suggests that
impulsivity does not necessarily affect initial drug intake, but rather, once drug is on board, there
is an interaction between impulsivity and drug-action in high impulsive, but not low impulsive,
animals. These findings highlight the importance of a high impulsivity endophenotype as being a
crucial factor that makes some individuals more vulnerable to drugs of abuse by directly
interacting with drug action.
Neurocircuitry of impulsive action
Impulsivity is regulated by a cortical-striatal-hippocampal network, and consequently
involves several neurotransmitter systems within these circuits. A complete discussion of the
literature on the circuitry of impulsive action is out of the scope of this discussion. Instead, I
present an overview of the neurocircuitry within these main brain regions associated with
impulsive action and their respective neurotransmitter systems (Table 1.1).
1.2.4.1 Striatum
The striatum, composed of the nucleus accumbens (NAcb) core (C) and shell (S), and
caudate putamen (i.e., dorsal striatum), is crucial for regulating impulsive action, with lesions to
this area increasing impulsivity in rodents (Rogers et al. 2001; Cardinal et al. 2001; Pothuizen et
al. 2005). More specifically, impulsive responding is mediated by dopamine (DA)-receptor
containing GABAergic medium spiny neurons of the NAcb (Pattij and Vanderschuren 2008;
Dalley et al. 2011; Caprioli et al. 2014); these neurons receive glutamatergic and dopaminergic
input from the prefrontal cortex (PFC) and ventral tegmental area (VTA) (Kalivas et al. 2005;
Sesack and Grace 2010).
7
Striatal dopaminergic receptors are essential for regulating impulsivity. For instance, HIAct rats (described above) have reduced DA D2/3 autoreceptor binding in NAcbS (Dalley et al.
2007; Caprioli et al. 2014), while in healthy humans, reduced D2/3 receptor functioning in the
VTA and substantia nigra (SN) predicts increased impulsivity (Buckholtz et al. 2010). In the
latter study, these authors went on to show that D2/3 binding in the VTA and SN is negatively
associated with increased striatal DA release (Buckholtz et al. 2010). Therefore, dopaminergic
activity in the NAcbC increases impulsivity, likely through both D1 and D2/3 activation (Besson
et al. 2010; Wiskerke et al. 2011a; Economidou et al. 2012; Jupp et al. 2013b; Moreno et al.
2013). Data are more equivocal with respect to DA-NAcbC involvement, with some data
showing no role (Economidou et al. 2012; Moreno et al. 2013), and others showing increased
impulsivity following D2/3 antagonism (Besson et al. 2010). In the latter study, researchers
argued that the NAcbC and NAcbS have opposing roles in regulating impulsivity; a fact that was
indirectly supported by the lack of a systemic D2/3 antagonist effect (Besson et al. 2010).
Moreover, even though the NAcb is made up predominantly of GABAergic medium
spiny neurons, little research has examined their role in impulsive action directly. It was recently
shown that knocking down glutamate decarboxylase (GAD65/67; catalyst of GABA) gene
expression in the NAcbC increases premature responding, suggesting that GABAergic activity
within this region may also play a role in impulsive action (Caprioli et al. 2014).
1.2.4.2 Extra-striatal regions
The PFC is involved in top-down inhibitory control of behaviour, including impulsivity.
The PFC is composed of two major neuronal populations: glutamatergic pyramidal projection
neurons and GABAergic interneurons. Glutamatergic projections from the PFC act to regulate
striatal dopaminergic activity (Sesack and Grace 2010). Indeed, increased premature responding
8
following medial prefrontal cortex (mPFC) lesions or GABAA agonism can be attenuated by D2/3
receptor antagonism in the NAcb (Pezze et al. 2009; Pezze et al. 2014).
Within the PFC, reduced noradrenergic and glutamatergic activity increases impulsivity
(Chamberlain et al. 2006; Chamberlain and Robbins 2013). For instance, intra-PFC infusions of
atomoxetine, a selective NA reuptake inhibitor, improves impulse control in rodents (Bari et al.
2011), although the exact mechanisms are unknown. Additionally, NMDA-R antagonism within
the PFC increases impulsive action through increased presynaptic metabotropic glutamate 2/3
receptor (mGluR2/3) activity (Pozzi et al. 2011).
Serotonin (5-HT) has its main effect on impulsive action within the PFC (Dalley et al.
2008; Puig and Gulledge 2011). Broadly, increasing prefrontal activity via stimulation of 5-HT1a
receptors (inhibitory receptors) increases premature responding (Harrison et al. 1997; Dalley et
al. 2002a; Winstanley et al. 2006b), as does central 5-HT depletion and 5-HT2a (excitatory
receptors) antagonism (Winstanley et al. 2004). Therefore, 5-HT overstimulation or insufficient
receptor stimulation within the PFC may both increase impulsive action (Dalley et al. 2002b;
Dalley et al. 2008). These effects may in part be driven by 5-HT – DA interactions, as 5-HT
depletion attenuates amphetamine-induced increases in impulsive responding (Harrison et al.
1997; Dalley et al. 2002a; Dalley et al. 2008). In fact, some authors have argued that serotonin’s
effects on impulsive action are indirect, and are a consequence of its modulation of striatal
dopamine (Harrison et al. 1997).
The hippocampus has also more recently been implicated in impulsive action. Although
the exact mechanisms merit further research, bilateral lesions to the ventral (but not dorsal)
hippocampus increase impulsive action (Chudasama et al. 2003). Additionally, disconnecting the
vPFC-ventral hippocampus pathway increases impulsive action to the same degree as lesions to
9
the vPFC, suggesting the importance of this pathway in regulating impulsive action (Chudasama
et al. 2012).
Table 1.1 Role of different neurotransmitter systems involved in impulsive action in rodent
models.
Neurotransmitter
System
Pharmacological
Agent
Brain Region
Result
Reference
Dopamine
DA agonism (e.g.,
reuptake inhibition)
Amphetamine
Methylphenidate
Methylphenidate
Methylphenidate
Methylphenidate
Systemic
NAcb Core
NAcb Shell
PL mPFC
IL mPFC
↑
↑
=
=
=
(Pattij et al. 2007b)
(Economidou et al. 2012)
(Economidou et al. 2012)
(Economidou et al. 2012)
(Economidou et al. 2012)
D1 agonism
SKF 38393
SKF 38393
NAcb
OFC
↑
=
(Pezze et al. 2007)
(Winstanley et al. 2010)
D1 antagonism
SCH 23390
SCH 23390
SCH 23390
SCH 23390
NAcb Core
NAcb Shell
OFC
DS
↑
↑
↓
↓
(Pattij et al. 2007b)
(Pattij et al. 2007b)
(Winstanley et al. 2010)
(Agnoli and Carli 2011)
D2/3 agonism
Quinpirole
Quinpirole
Quinpirole
Quinpirole
Nacb
NAcb Core
NAcb Shell
OFC
↑
↑
=
=
(Pezze et al. 2007)
(Moreno et al. 2013)
(Moreno et al. 2013)
(Winstanley et al. 2010)
D2/3 antagonism
Eticlopride
Nafadotride
Eticlopride
NAcb Core
NAcb Core
NAcb Shell
=
↑
=
(Pattij et al. 2007b)
(Besson et al. 2010)
(Pattij et al. 2007b)
Nafadotride
NAcb Shell
↓
(Besson et al. 2010)
SR141716A
SR141716A
O-2050
WIN55,212-2
D9-THC
Systemic
Systemic
Systemic
Systemic
Systemic
↓
↓
↓
=
=
(Pattij et al. 2007a)
(Wiskerke et al. 2011b)
(Wiskerke et al. 2011b)
(Pattij et al. 2007a)
(Wiskerke et al. 2011b)
Cannabinoid
CB1 antagonism
CB1 agonism
10
Noradrenalin
NA agonism (e.g,
reuptake inhibition)
Atomoxetine
Atomoxetine
Atomoxetine
Atomoxetine
Systemic
Systemic
NAcb Core
NAcc Shell
↓
↓
=
↓
(Fernando et al. 2012)
(Robinson 2012)
(Economidou et al. 2012)
(Economidou et al. 2012)
α2A-adrenoceptor
Agonism
Guanfacine
Systemic
↓
(Fernando et al. 2012)
α2A-adrenoceptor
antagonism
Atipamizole
Atipamizole
Systemic
Systemic
↑
↑
(Sirviö et al. 1994)
(Koskinen et al., 2003
β-adrenoceptor
agonism
Clenbuterol
Systemic
↓
(Pattij et al. 2012)
β-adrenoceptor
antagonist
Propanolol
Systemic
↓
(Milstein et al. 2010)
5-HT agonism (e.g.,
re-uptake inhibition)
Citalopram
Systemic
↓
(Baarendse and
Vanderschuren 2012)
5-HT2A antagonism
M100907
M100907
M100907
M100907
M100907
Systemic
NAcb
mPFC
IL mPFC
PL mPFC
DMS
↓
↓
↓
=
=
=
(Winstanley et al. 2003)
(Robinson et al. 2008)
(Winstanley et al. 2003)
(Robinson et al. 2008)
(Robinson et al. 2008)
(Agnoli and Carli 2012)
5-HT2C agonism
Ro60-0175
DMS
=
(Agnoli and Carli 2012)
5-HT2C antagonism
SB242084
SB242084
SB242084
NAcb
IL mPFC
PL mPFC
↑
=
=
(Robinson et al. 2008)
(Robinson et al. 2008)
(Robinson et al. 2008)
NMDA antagonism
MK 801
(R)-CPP
(R)-CPP
Systemic
IL mPFC
mPFC
↑
↑
↑
(Paine et al. 2007)
(Murphy et al. 2012)
(Pozzi et al. 2011)
NMDA2B antagonist
Ro63-1908
Systemic
↓
(Burton and Fletcher 2012)
mGluR2/3 antagonism
LY341495
Systemic
=
(Semenova and Markou
2007)
Serotonin
Glutamate
11
GABA*
GABAA agonism
Muscimol
Muscimol
Muscimol
mPFC
mPFC
IL mPFC
↑
↑
↑
(Paine et al. 2011)
(Pezze et al. 2014)
(Murphy et al. 2012)
GABAA antagonism
Picrotoxin
Bicuculline
mPFC
mPFC
=
=
(Pezze et al. 2014)
(Paine et al. 2011)
GABAB agonism
Baclofen
IL mPFC
↑
(Murphy et al. 2012)
↓ and ↑ denote decreases and increases in impulsive action (premature responding), respectively;
= denotes no effect. *Pharmacological agonism or antagonism of GABAA receptors are
synonymous with disinhibition and hyperactivation, respectively.
5-HT, serotonin; ACC, anterior cingulate cortex; CB1, cannabinoid receptor 1; D1, dopamine
receptor 1; D2/3, dopamine 2/3 receptor; DA, dopamine; DMS; dorsomedial striatum; DS; dorsal
striatum; GABA, gamma-aminobutyric acid; IL, infralimbic; mPFC, medial prefrontal cortex;
mGluR; metabotropic glutamate receptor; NA, noradrenalin; NAcb, nucleus accumbens; NMDA,
N-methyl-D-aspartate; OFC, orbitofrontal cortex; PL, prelimbic; PLd, dorsal prelimbic; vmPFC,
ventromedial prefrontal cortex
1.3 Opioid system
Opioids are ubiquitous within the central nervous system, and consist of three
heterogeneous receptor subtypes: mu, delta, and kappa (Martin et al. 1976; Lord et al. 1977).
Opioids are notably implicated in pain-, reward-, and addiction-related processes (Le Merrer et
al. 2009; Pradhan et al. 2011; Chu Sin Chung and Kieffer 2013). Opioid receptors are
preferentially coupled to inhibitory Gαi/o-protein coupled receptors (GPCRs) with a 7transmembrane domain. Opioid receptor binding has effects on several downstream systems, but
ultimately result in inhibition of neuron activation via reduced neurotransmitter release or cell
firing (Pathan and Williams 2012; Atwood et al. 2014; Klenowski et al. 2014). Naturally
occurring ligands include the enkephalins, endorphins, dynorphins, and endomorphins, which
bind to different receptor subtypes (Table 1.2) and are distributed broadly in the brain, including
in regions associated with impulsivity (Figure 1.2) (Le Merrer et al. 2009).
12
Mu opioid receptors (MORs) are found pre- and post-synaptically, predominantly on
GABAergic neurons (Jiang and North 1992; Martin et al. 1997; Chieng and Williams 1998), in
areas associated with impulsivity-related disorders. In the VTA, NAcb, dorsal striatum,
hippocampus, and ventral pallidum, activation of MORs located on presynaptic GABAergic
interneurons causes hyperpolarization resulting in disinhibition of postsynaptic neurons (e.g.,
GABAergic medium spiny neurons; MSNs) (Austin and Kalivas 1990; Johnson and North 1992;
Chieng and Williams 1998; Kupchik et al. 2014; Atwood et al. 2014). In the NAcb, MORs are
also located presynaptically on glutamatergic terminals that synapse onto GABAergic MSNs,
and postsynaptically on GABAergic MSNs (Jiang and North 1992; Martin et al. 1997).
Delta opioid receptors (DORs) are predominantly expressed in cholinergic interneurons in
the striatum (Le Moine et al. 1994; Wang et al. 2003). In the hippocampus, DORs are
presynaptically located on GABAergic interneurons that synapse onto glutamatergic principal
cells; activation of these DORs results in increased GABAergic neuron firing (Svoboda et al.
1999; Rezaï et al. 2012; Erbs et al. 2012; Piskorowski and Chevaleyre 2013).
Table 1.2 Classification of opioid receptor endogenous ligands
Opioid Receptor Subtype
Endogenous Ligand
Mu (μ)
β-endorphin (not selective)
Enkephalin (not selective)
Endomorphin-1
Endomorphin-2
Delta (δ)
β-endorphin (not selective)
Enkephalin (not selective)
Kappa (κ)
Dynorphin A
Dynorphin B
α-neoendorphin
13
Cre-LoxP recombination strategy for constitutive and conditional gene knockout
In the mid-1990s, genes coding for mu (oprm1), delta (oprd1), and kappa (oprk1)
receptors (M/D/KORs) were identified in the human and rodent genome and were constitutively
and conditionally knocked out in mice thereafter (Befort et al. 1994; Le Moine et al. 1994;
Simonin et al. 1995; Charbogne et al. 2014). Both constitutive knockout (KO) (i.e., full gene
deletion) and conditional knockout (cKO) (i.e., gene knockout in selective tissue, brain regions,
etc) can be achieved using Cre-LoxP recombination strategies. Cre-LoxP recombination is a sitespecific genetic strategy used to manipulate DNA sequences. It involves cyclic recombinase
(Cre) and LoxP manipulation, both derived from the P1 bacteriophage. Cre is an enzyme that
catalyses DNA recombination between two of its 34 base-pair (bp) DNA recognition sites,
referred to as LoxP sites. Therefore, the DNA sequence (i.e., an exon of a gene) flanked by the
LoxP sites is recombined, resulting in insertion, deletion, translocation or inversion of specific
DNA sequences. The outcome is a gene knockout or knockin, depending on the specific
methodology.
Selectivity of Cre-LoxP systems is derived from the promoter chosen. A promoter is the
region of DNA that initiates transcription of a particular gene, and is located upstream (3’ region)
from the gene sequence. Cytomegalovirus (CMV) is a commonly used promoter that is highly
non-specific, and will be expressed in all cells. Therefore, if CMV is used as a promotor, Cre
will be expressed in all cell types early in embryonic development.
Promoters can also be tissue specific, cell specific, developmentally specific or
responsive to exogenous compounds, allowing for a broad range of applications. The benefit of
this type of genetic approach is temporal (i.e., trigger is at a specific point in development or
experimental phase) and/or spatial (i.e., only specific cell types or brain regions are targeted)
14
specificity. Cre-LoxP is an efficient strategy, as it can be used in any cellular environment and on
any type of DNA, and in conjunction with homologous recombination-based gene targeting
(Lakso et al. 1992; Orban et al. 1992).
To produce a desired modified mouse line, mice expressing Cre are bred with those
expressing the LoxP sites (Figure 1.1). The gene coding for Cre produces the enzyme that then
cleaves the exon being flanked by the LoxP sites. Importantly, the exon being flanked should
have functional relevance for the gene being transcribed: if the methodological aim is a gene
knockout, excision of the flanked exon should render the gene non-functional. To accomplish a
Cre-LoxP system, the Cre gene or the LoxP sequence of DNA is implanted into DNA of a mouse
blastocyst. These two mouse lines are then bred (F0 generation) to produce an F1 generation. F1
mice will be chimeras: the gene will be excised in some, but not all cells. F1 generation mice are
then cross bred with wild-type mice (F2), creating heterozygous mice (they will have one copy of
the knocked out gene in all cells). These heterozygous offspring are then interbred (F3-onward),
creating some animals that carry no functional copy of the original unaltered gene (homozygous
for the gene of interest; KO).
15
Figure 1.1 Simplified schematic of the Cre-LoxP genetic recombination strategy. Mice
expressing the Cre gene under the cytomegalovirus (CMV) promotor control is bred with
animals expressing LoxP sites flanking Exon X (a). In the presence of the Cre gene, Cre
recombinase is synthesized, which excises the region flanked by the LoxP sites, rendering Exon
X inactive and resulting in gene X being non-functional (b). In this example, LoxP sites are
placed in a trans orientation resulting in Cre recombinase excising the flanked region of DNA.
M/DOR constitutive and conditional knockout mouse construction
Mice with a floxed delta (Oprd1fl/fl) or mu (Oprm1fl/fl) opioid receptor gene were
constructed by flanking exon 2 (delta) or exon 2 and 3 (mu) of the gene with LoxP sites
(Gaveriaux-Ruff et al. 2011; Weibel et al. 2013). These floxed mice show intact DOR and MOR
gene expression and binding (Gaveriaux-Ruff et al. 2011; Weibel et al. 2013). Constitutive MOR
(CMV-Oprm1-/-) and DOR (CMV-Oprm1-/-) KO mice were produced by breeding Oprm1fl/fl or
Oprd1fl/fl mice with CMV-Cre mice, respectively. In these mice, opioid receptor activation is not
detectable in the brain, suggestive of effective gene deletion (Gaveriaux-Ruff et al. 2011; Weibel
et al. 2013).
Recently, mice lacking MORs or DORs in GABAergic forebrain neurons were produced
by breeding Oprm1fl/fl and Oprd1fl/fl mice with Dlx5/6-Cre mice that express Cre recombinase in
16
Dlx5/6 positive neurons. In mammals, there are six Dlx homeobox genes (Stock et al. 1996).
Generally, homeobox genes are involved in the regulation of anatomical development in
vertebrates. Specifically, Dlx1, Dlx2, Dlx5, and Dlx6 genes have distinct roles in the different
steps of differentiation, migration and survival of forebrain progenitors (Anderson et al. 1997;
Liu et al. 1997; Cobos et al. 2005; Cobos et al. 2007). In particular, the Dlx5/Dlx6 cluster (i.e.,
two genes that share the same endogenous promoter, and are transcribed together) is expressed in
differentiating and migrating forebrain GABAergic neurons during embryonic development
(Zerucha et al. 2000; Stühmer et al. 2002). Therefore, when Dlx5/6-Cre (Monory et al. 2006)
animals are crossed with Oprm1fl/fl or Oprd1fl/fl, MORs or DOR genes will be selectively excised
from GABAergic forebrain neurons, producing cKO M/DOR mice. Since M/DORs are primarily
located on GABAergic neurons in the forebrain (Charbogne et al. unpublished data; Chu Sin
Chung et al. unpublished data), these conditional animals lack the vast majority of M/DORs from
forebrain regions. Specifically in both Dlx5/6-Oprd1-/- (DOR cKO) and Dlx5/6-Oprm1-/- (MOR
cKO) mice, respectively, Oprd1 or Oprm1 is not expressed in the caudate putamen (CPu), and
NAcb, partially expressed in the prefrontal cortex (PFC) and hippocampus, and fully expressed
in the ventral tegmental area (Figure 1.2) (Charbogne et al. unpublished data; Chu Sin Chung et
al. unpublished data).
17
Figure 1.2 Sagittal view of a representative mouse brain illustrating M/DOR receptor gene
expression (a) and ligand binding (b) in M/DOR cKO mice in regions associated with impulsive
action. ↓↓ reflects abolished mRNA expression; ↓ reflects decreased mRNA expression (a) or
binding (b); = no change. HP: Hippocampus; CPu: Caudate putamen; VTA: Ventral tegmental
area; PFC: Prefrontal cortex.
Opioid receptor involvement in impulsivity
Initial evidence for opioid involvement in impulsivity came from studies showing that
chronic opiate abusers are more impulsive than non-users (Madden et al. 1997; Baldacchino et
al. 2014). Additionally, human PET data suggest a possible association, with cognitive
impulsivity predicting greater MOR concentrations in several brain regions associated with
impulsivity, including the orbitofrontal cortex (OFC), anterior cingulate cortex (ACC), and
basolateral amygdala (BLA) (Love et al. 2009). To date, it has been shown in rats that MOR
18
activation (via morphine) increases both impulsive choice, measured on a delay discounting task,
and impulsive action in the 5-CSRTT (Pattij et al. 2009; Wiskerke et al. 2011a). In addition,
animals lacking MORs are less impulsive than wild-type controls (Olmstead et al. 2009).
Even less research has examined the role of DORs on impulsivity; with the exception of
the finding that DOR KO mice exhibit increased impulsive action compared to wild-type
controls (Olmstead et al. 2009). Additional, indirect evidence for DOR involvement in
impulsivity comes from the fact that DORs are highly expressed in regions involved in
impulsivity, including the VTA and striatum (Le Merrer et al. 2009; Charbogne et al. 2014).
1.4 Stress system
Physiological responses in the stress system help organisms adapt to acute, day to day life
stressors. In a normally functioning system, acute stress activates the hypothalamic-pituitaryadrenal (HPA) axis, which results in a cascade of physiological responses. Specifically,
corticosterone releasing factor (CRF), released from the paraventricular nucleus (PVN) of the
hypothalamus, increases adrenocorticotropic hormone (ACTH) secretion from the anterior
pituitary. In turn, the adrenal glands release glucocorticoids (GC) (e.g., corticosterone; CORT).
In addition, endogenous opioids (enkephalins and endorphins) are released in response to stress,
and diminish the stress response (Drolet et al. 2001). Glucocorticoids, cortisol in humans and
corticosterone in rodents, target both the HPA axis (pituitary and PVN) and extra-hypothalamic
stress regions, notably the extended amygdala, hippocampus, and mPFC (Figure 1.3). Broadly,
following acute stress, hippocampus activity inhibits the HPA axis; in contrast extended
amygdala activation increases HPA activity (Herman and James 2012). The mPFC, a region
particularly influential in modulating the effects of acute stress, is responsible for both stressrelated activation and attenuation of the HPA axis by infralimbic and prelimbic PFC regions,
19
respectively (Herman and James 2012). Overall, these responses promote beneficial adaptations
to acute stress and allows the organisms to cope with the stressor.
Figure 1.3 Schematic of hypothalamic-pituitary-adrenal (HPA) axis and extra-hypothamalamic
brain regions involved in the acute stress response. In response to a stressor, corticotropin
releasing factor (CRF) is released from the paraventricular nucleus (PVN) of the hypothalamus,
which in turn stimulates adrenocorticotropic hormone (ACTH) release from the anterior pituitary
gland. Glucocorticoids (e.g., cortisol) then feedback to the HPA (PVN and pituitary), extended
amygdala (amygdala and bed nucleus of the solitary track; BNST), hippocampus (HP), and
medial prefrontal cortex (PFC). + represents increased activation following stress; - represents
decreased activity following stress.
In contrast to acute stress responses, prolonged or extreme stress can cause long-term
upregulation of HPA activity, in part via extra-hypothalamic neurocircuitry feeding back to the
PVN. Upregulation of HPA activity results in several changes in the stress response, including
sensitization of the HPA axis to acute stressors, increased basal glucocorticoid secretion,
enhanced CRF1 and GC receptor activity, and neuroplastic changes within extra-hypothalamic
brain regions (Herman and James 2012). Importantly, following acute stress, HPA activity
returns to basal, homeostatic functioning; conversely, prolonged or extreme stress results in longterm upregulation of HPA activity, resulting in continued HPA hyperactivity.
20
Stress as a risk factor of drug addiction
Notably, regions involved in the stress response (mPFC, hippocampus, and extended
amygdala) are also highly implicated in drug-related behaviour, suggesting a common
mechanism of action. Both acute stress and altered HPA functioning from chronic stress can
precipitate or exacerbate psychiatric disorders, including drug addiction (Goeders 1997; Piazza
and Le Moal 1998; Sarnyai et al. 2001). More specifically, stressful life events (both day to day
and long-lasting) contribute to increased drug use at all stages of drug addiction, from initiation
to drug use to craving and relapse (Kreek et al. 2005; Koob and Kreek 2007; Koob and Le Moal
2008; Sinha 2008; Haass-Koffler and Bartlett 2012; Schank et al. 2012). Individuals who report
high levels of perceived stress also report longer periods of cocaine use, compared to low-stress
controls (Karlsgodt et al. 2003). Furthermore, perceived stressful events can trigger relapse, even
after long periods of abstinence (Sinha et al. 1999; Stine et al. 2002; Sinha et al. 2006; Fox et al.
2007; Koob 2008; Li et al. 2010; Greenwald et al. 2013), and acute pharmacological stress (i.e.,
yohimbine) increases opioid seeking and craving in drug dependent patients (Stine et al. 2002;
Umhau et al. 2011; Greenwald et al. 2013; Moran-Santa Maria et al. 2014). Animal research
confirms that acute (one session) and chronic (repeated sessions) stress contribute to all stages of
drug-taking behaviour, from initiation to escalation of drug intake to relapse (Shaham et al. 1996;
Goeders 1997; Goeders 2002; Lu et al. 2003; Koob and Volkow 2010; Shalev et al. 2010).
Mechanistically, stress differentially alters drug intake across the phases of addiction via both
HPA (e.g., changes in neuroendocrine systems) and extra-hypothalamic activity.
It appears that stress circuitry activates both the rewarding pathways implicated in drugtaking behavior and the aversive aspects of withdrawal (Koob 2009). More specifically, during
early stages of intake, stress exacerbates drug-induced dopaminergic activity in the mesolimbic
21
pathway (Saal et al. 2003; Ungless et al. 2010). Following chronic use, interactions between drug
and HPA activity result in HPA dysregulation via inhibition of peripheral GC feedback in the
hypothalamus and pituitary gland, and recruitment of extra-hypothalamic sites (Kreek and Koob
1998; McEwen 2008; Herman and James 2012). For instance, drug-induced CRF1 and GC
receptor up- and down-regulation in various stress-related brain regions contribute to the
transition from casual drug use to dependence (Ambroggi et al. 2009; Koob and Volkow 2010;
Gilpin 2012; Vendruscolo et al. 2012). In this state, CRF and noradrenergic activity, particularly
within the extended amygdala, mediate the aversive effects of acute withdrawal (Koob and
Volkow 2010).
Stress-impulsivity interactions
Both stress and impulsivity have been clearly defined as environmental and personality
factors that precipitate, promote, and exacerbate drug addiction; however, they are often viewed
as independent predictors of this maladaptive behavior. Few clinical or pre-clinical studies have
examined, directly, the relationship between these two factors, and the mechanisms that may
mediate this interaction in a drug-naïve state are not well understood. Some evidence suggests
that adverse life events are correlated with impulsivity in cocaine-dependent individuals (Hayaki
et al. 2005), whereas other evidence does not support a relationship between total life stress
scores and either delay discounting or trait impulsivity in cocaine dependent populations (Ross et
al. 2013). These discordant findings likely reflect different measures of impulsivity (delay
discounting versus impulsivity questionnaires), life stress questionnaires, and clinical
populations studied.
More direct physiological markers of HPA activity support a stress-impulsivity
interaction: basal cortisol levels predict trait like impulsivity in both individuals with borderline
22
personality disorders and healthy controls (Almeida et al. 2010). Additionally, pharmacological
stress (e.g., yohimbine) increases impulsivity in healthy individuals (Swann et al. 2005), and
increases impulsive action in the 5-CSRTT in rats (Sun et al. 2010). In the latter study,
yohimbine-induced increases in premature responding were attenuated by blocking cyclic
adenosine monophoshphate (cAMP) response element binding (CREB) activity in the OFC (Sun
et al. 2010). Since CREB is a transcription factor implicated in the stress response (Briand and
Blendy 2011), these authors argued that yohimbine’s effects on impulsive responding were
stress-response specific, rather than due to general noradrenergic activation (Sun et al. 2010).
1.5 Objectives
The purpose of the experiments presented in the following chapters is to further elucidate
the mechanisms of impulsive action, with respect to the pharmacological underpinnings and
environmental influences (e.g., stress) that mediate this behaviour. In the first three experiments,
we aimed to extend findings showing an involvement of the opioid system in impulsive action.
To accomplish this, pharmacological manipulations were implemented in Chapters 2 and 3; in
Chapter 4, we specifically examined the role of MORs or DORs located on GABAergic
forebrain neurons on impulsive action by testing mice lacking these receptors. In rodent
experiments, we used morphine as a MOR agonist because more selective agonists do not readily
pass the blood brain barrier. Additionally, because morphine is already available as a
pharmacotherapy for pain management, its use in the study of impulse control disorders has
more clinical relevance than other MOR agonists. Finally, in Chapter 5, we examined whether
stress, via pharmacological intervention, increases impulsivity, and further examined possible
neurotransmitter system involvement in this interaction.
23
Effects of Delta Opioid Receptor Activation on a Response Inhibition Task in
Rats
As published in: Befort K, Mahoney MK, Chow C, Hayton SJ, Kieffer BL, Olmstead MC
(2011). Effects of delta opioid receptors activation on a response inhibition task in rats.
Psychopharmacology, 214, 967–976.
2.1 Abstract
Rationale: Response inhibition, a primary symptom of many psychiatric disorders, is mediated
through a complex neuropharmacological network that involves dopamine, serotonin, glutamate,
noradrenaline, and cannabinoid mechanisms. Recently we identified an opioidergic contribution
to response inhibition by showing that deletion of mu or delta opioid receptors in mice alters
motor impulsivity. Objectives: We investigated this phenomenon further by testing whether
pharmacological activation of opioid receptors disrupts the ability to inhibit a motor response.
Methods: Long-Evans rats were trained to withhold a lever-pressing response for sucrose until a
discriminative stimulus (lever light) was presented. The delay to the discriminative stimulus (1 to
60 s) was varied so animals could not predict, on any given trial, the length of the pre-response
phase. Motor impulsivity was assessed as the inability to inhibit lever pressing prior to the
discriminative stimulus. Rats were tested following an injection of the mu opioid receptor
agonist morphine (0, 0.5, 1, 2, 4, 6, 8 or 10 mg/kg) or the delta receptor agonist SNC80 (0, 2.5, 5,
or 10 mg/kg). Results: The highest dose of SNC80 increased premature responses and locomotor
activity, but had no effect on the speed of responding or non-reinforced lever presses. The
SNC80-induced decrease in accuracy was blocked by the delta opioid receptor antagonist
naltrindole. Morphine had no effect on any behavioural measure. Conclusions: These findings
24
point to a more complicated role for opioid receptors in disinhibition than would have been
predicted from previous opioid receptor deletion studies. Possible sources of these discrepant
results are discussed.
2.2 Introduction
Impulsivity is a primary pathology in many psychiatric disorders including attention
deficit hyperactivity disorder (ADHD), substance abuse, pathological gambling, binge eating,
and antisocial personality disorder (American Psychiatric Association 2013). The prevalence of
impulsivity across different patient populations is not surprising as the construct encompasses
several behavioral and cognitive processes (Evenden 1999). Acting before thinking, opting for
immediate gratification in lieu of delayed rewards, making a decision prior to integrating all
relevant information, and the inability to withhold a motor response are all examples of
impulsive responding. The latter process, commonly referred to as response inhibition or motor
impulsivity, is the common trait in disorders of impulse control and, hence, the subject of intense
scientific investigation (for reviews see (Pattij and Vanderschuren 2008; Hayton and Olmstead
2009).
Response inhibition is mediated through a complex neuropharmacological network that
involves dopamine (DA), serotonin (5-HT), glutamate, noradrenaline (NA), and cannabinoid
mechanisms (Pattij and Vanderschuren 2008; Hayton and Olmstead 2009). The role of the opioid
system in this process has not been studied extensively despite evidence that heroin abusers show
deficits in motor impulsivity (Verdejo et al. 2005), and that opioid antagonists enhance control of
motor responses (Mitchell et al. 2007) and may be an effective treatment for impulse-control
disorders (Kim 1998). Recently, we identified an opioidergic contribution to response inhibition
in a signaled nose poke task by showing that mice lacking the mu opioid receptors exhibit lower
25
levels of motor impulsivity, whereas those lacking delta opioid receptors exhibit higher levels
(Olmstead et al. 2009). A role for mu opioid receptors in response inhibition was confirmed in
rats using a 5-choice serial reaction time task (5-CSRTT) (Pattij et al. 2009).
Our finding that delta opioid receptors play a role in motor impulsivity is particularly
intriguing as this was the first evidence for a contribution of these receptors to any aspect of
impulsivity. Moreover, delta opioid receptors make an important contribution to emotional
processing in that reductions in delta opioid receptor signaling increase anxiety and depressivelike symptoms (Filliol et al. 2000; Perrine et al. 2008) whereas receptor activation has the
opposite effect (Broom et al. 2002b; Perrine et al. 2006; Jutkiewicz 2006). Disruptions in
emotional processing and impulsivity often co-occur in clinical populations, suggesting that
alterations in delta opioid receptor function may be a common underlying mechanism in some
psychiatric conditions. Finally, if deletion of delta opioid receptors increases motor impulsivity,
activation of these receptors would be expected to decrease motor impulsivity, pointing to a
potential new therapeutic intervention for response inhibition deficits.
The purpose of the present study was to examine, in greater depth, the role of mu and
delta opioid receptors in motor impulsivity. To do so, we tested the effects of morphine and
SNC80 (mu and delta opioid receptor agonists respectively) in a response inhibition task (Hayton
et al. 2010) that mimics the signaled nose poke task in mice (Bowers and Wehner 2001). Rats are
required to withhold responding for a sucrose reward until a discriminative stimulus (lever light)
is presented. In the current study, the delay to the discriminative stimulus was varied so that
animals could not predict, on any given trial, the length of the pre-response phase. This ensured
that optimal performance was not influenced by the ability to accurately time intervals. A further
advantage of this task is that animals were tested at short and long intervals in the same session,
26
and that both increases and decreases in task accuracy could be observed across sessions. Motor
impulsivity was assessed as the number of responses that occurred prior to the presentation of the
discriminative stimulus, modeling the failure to inhibit an inappropriate verbal or motor response
in humans.
2.3 Materials and methods
Subjects
Male Long-Evans rats (Charles River, PQ, Canada) weighing 250-350 g at the start of the
experiments were pair-housed on a reverse light-dark cycle in standard polycarbonate cages.
Behavioral testing was conducted during the dark cycle. Initially, rats had free access to food
(Lab Diet; PMI Nutrition International, Inc.) and water. Beginning three days prior to
behavioural training, rats were food restricted to 90 min of free access per day such that they
gained 10-15 g per week. Animal care was conducted in accordance with the guidelines provided
by the Canadian Council on Animal Care, and the experiments were approved by the Queen’s
University Animal Care Committee.
Apparatus
Training and testing were conducted in operant conditioning chambers (26.5 × 22.0 ×
20.0 cm), each housed in a sound-attenuating chamber (constructed in house). Each box was
fitted with two retractable levers positioned equidistantly on one wall. Levers were 4 cm wide, 5
cm from the mesh floor, and 13 cm apart. Sucrose pellets (45 mg; Bio-Serv, N.J., USA) were
delivered to a food magazine via a pellet dispenser (Med Associates, Vt., USA). The food
magazine was located between the two levers and was equipped with infrared sensors to detect
food collection. One small dim light was situated 4 cm above each lever and the magazine. An
27
indirect houselight could illuminate the chamber. An IBM-type computer in an adjacent room
controlled the equipment and was used for data collection (software written in-house using
ECBASIC).
Drugs
Morphine sulfate (BDH Pharmaceuticals, Toronto ON) and naloxone (TOCRIS
Biosciences, USA) were dissolved in 0.9% physiological saline. Morphine was tested at doses of
0, 0.5, 1, 2, 4, 6, 8 and 10 mg/kg. The opioid receptor antagonist, naloxone, was tested at a dose
of 1 mg/kg. SNC80 ((+)-4-[(α-R*)-α-((2S*,5S*)-4-allyl-2,5-dimethyl-1-piperazinyl)-3methoxybenzyl]-N,N-diethylbenzamide) (Tocris Biosciences, USA) was dissolved in 8% 1 M
HCl solution to yield doses of 0, 2.5, 5, and 10 mg/kg. The selective delta opioid receptor
antagonist, naltrindole (NTI) (Tocris Biosciences, USA), was dissolved in 25% DMSO to yield a
dose of 5 mg/kg, a dose which blocks SNC80-induced behavioural changes (Spina et al. 1998).
Morphine, naloxone and SNC80 were injected s.c. 30 min prior to behavioral testing. Doses were
administered in an ascending or descending order, counterbalanced across animals. Drug doses
were separated by 24 hr (1-lever task) or 48 hr (2-lever task). In the latter, animals were tested
drug-free on the intervening days. NTI was injected s.c. 15 min prior to the SNC80 injection,
based on previous evidence of effective blockade of SNC80 behavioral effects at this interval
(Broom et al. 2002a).
Response inhibition training
Rats were magazine trained for 2 days, receiving 20 sucrose pellets on a random time
(RT) 90-s schedule. Levers were withdrawn and lever lights (discriminative stimuli) were turned
28
off. The house-light was illuminated until a food pellet was dispensed. At this point, the
magazine light turned on and the house light turned off for 1 s and the next trial commenced.
Rats were then trained to lever press for food on a continuous reinforcement (CRF)
schedule. In order to minimize cognitive demands, such as discrimination, only one lever was
inserted into the chamber with the lever assignment (left versus right) counterbalanced across
animals. The house light and discriminative stimulus (lever light) were turned on throughout
these sessions, except during delivery of the reward, which was signaled by illumination of the
magazine light (1 s). There was no time limit for animals to respond on each trial; training
continued until a minimum of 80 pellets were earned in a 60-min session for 2 consecutive days.
Following CRF training, the response phase of the task was shortened such that rats had
10 s to lever press following the lever insertion. The house light and discriminative stimulus
were turned on during this 10-s period. If 10 s elapsed, the lever was retracted for a 10-s intertrial interval (ITI) and the trial was scored as an omission. Both the house light and
discriminative stimulus were turned off during the ITI. Training continued until animals reached
a criterion of fewer than 20% omitted trials for four consecutive sessions. The purpose of these
sessions is to train animals to respond within a 10-s window before they learn to inhibit
responding during the premature phase. Over 90% of the animals reached criterion in 2 days,
with the remaining completing this stage of training in 4 days.
In the next stage of training, each trial commenced with a 10-s ITI in which the lever was
retracted, the houselight was off. At the end of the ITI, the lever was inserted and the houselight
was illuminated but the discriminative stimulus was delayed for a variable period. Lever presses
prior to the discriminative stimulus (i.e., premature responses) reinstated the ITI with no reward.
Responses in the presence of the discriminative stimulus (10-s limit) were reinforced with a
29
sucrose pellet and initiated the next trial. Animals received 5 training sessions at each of the
following delays: 1, 4, 15, and 60 s with 100 trials per session. Subjects then progressed to the
full response inhibition task. In this stage, each session consisted of 20 trials at delay intervals of
1, 4, 15, 30, and 60 s (100 trials in total), that were presented in a random order. As previously,
the lever was retracted and both house light and discriminative stimulus were turned off during
the 10-s ITI. The house light turned on and the lever was inserted at the beginning of the
premature phase (variable interval). The response phase was signaled by the presentation of the
discriminative stimulus. A lever press during the response phase turned off the house light and
discriminative stimulus, retracted the lever, turned on the magazine light (1 s), delivered a food
pellet, and initiated the next trial. Failure to respond during the 10-s response phase was counted
as an omission and the next trial commenced. See Figure 2.1a for details.
30
Figure 2.1 Response inhibition task. (a) Visual schematic of the progression of trials in operant
boxes used in the response inhibition task. Each trial starts with a 10-s inter-trial-interval (ITI).
The houselight is turned on and a lever is inserted at the beginning of the subsequent premature
phase (1, 4, 15, 30 or 60 s); responses during the premature phase reinstate the ITI. At the end of
the premature phase, the houselight is turned off and a discriminative stimulus (light) is
presented indicating that lever presses will be reinforced with a sucrose pellet. Failure to respond
during the 10-s correct phase reinstates the ITI with no sucrose pellet. Arrows indicate possible
outcomes of each phase of the task. (b) Motor impulsivity, measures as accuracy in the response
inhibition task was stable across 14 days of baseline testing in the 1-lever (left) and 2-lever
(right) versions of the task. Accuracy increased as the length of the premature phase decreased in
both tasks indicating that rats had more difficulty inhibiting a motor response as the delay to the
discriminative stimulus increased.
Animals were tested for 14 days prior to drug administration with the last three days of
testing constituting baseline. Accuracy was assessed as the percentage of trials in which animals
successfully inhibited a lever pressing response prior to the discriminative stimulus. This
dependent measure was calculated as (correct responses/(premature + correct responses))*100.
The number of omissions and the latencies to premature and correct responses were recorded for
31
each session. Latencies were calculated as the time from lever insertion (premature response) or
discriminative stimulus presentation (correct response) to the first lever press.
In a subsequent set of experiments, a second lever was inserted into the chamber on the
same schedule as the first lever. This manipulation allowed us to test whether significant drug
effects were due to non-specific motor activation. We also tested animals, drug-free, on the days
between each drug dose in order to confirm that responding returned to baseline levels between
drug testing. Responses on this second, non-reinforced, lever were recorded but had no
programmed consequences. Non-reinforced responses could occur repeatedly within each trial
and the available time to respond increased with the delay interval. To account for these
differences, non-reinforced responses were analyzed using response rate (lever presses/s) as the
dependent variable.
Effects of morphine on accuracy
Three separate groups of rats were tested in the response inhibition task following an
injection of the mu opioid receptor agonist, morphine. The first group (n=10) was tested with one
lever using a low dose range (0, 0.5, 1, 2, and 4 mg/kg) and the second (n=8) at a higher dose
range (0, 6, 8, and 10 mg/kg). A third group (n=8) was tested in the 2-lever paradigm following
an injection of 0, 2, 4, or 8 mg/kg morphine (twice at each dose). A separate group of animals
(n=8) was tested in the 2-lever paradigm following an injection of naloxone (0 or 1 mg/kg).
Effects of SNC80 on accuracy
The effect of the delta opioid receptor agonist, SNC80, was tested using three separate
groups of rats. The first (n=10) were tested in the 1-lever paradigm following an injection of
SNC80 (0, 2.5, 5, and 10 mg/kg). At the end of drug testing, all animals underwent one drug-free
test in the response inhibition paradigm. The highest dose of SNC80 produced a significant
32
increase in impulsivity so this dose was used in a second study (n=16). Eight rats were tested
following an injection of SNC80 (10 mg/kg), the remaining 8 animals were tested following an
injection of NTI (5 mg/kg) plus SNC80. Following these tests, the second group was re-divided
into two and tested following an injection of NTI or vehicle (HCl + DMSO). The third group of
animals (n=16) was tested in the 2-lever paradigm following an injection of SNC80 (0 or 10
mg/kg).
Effects of morphine and SNC80 on locomotor activity
Locomotor activity was measured in activity boxes (50× 40×40 cm) with horizontal
movements monitored via a grid of infrared beams (beam breaks as the dependent measure).
Separate groups (n=8) were tested twice, once following an injection of SNC80 (10 mg/kg) or
morphine (2 mg/kg) and once following a vehicle injection (order counterbalanced within the
group). In each session, rats were placed in the activity cages for 1 h (basal exploration), injected
with saline (1 ml/kg s.c.), and returned to the cages for 1 h. Finally, rats were injected with drug
or vehicle and activity was recorded for 1 h.
Statistical analyses
Data were analyzed using analysis of variance (ANOVA), computed with the Statistical
Program for the Social Sciences (SPSS; V.18.0). Degrees of freedom for repeated measures were
adjusted using the Greenhouse-Geisser correction if assumptions of sphericity were violated
(Greenhouse and Geisser 1959). When significant main effects were observed in betweensubjects analyses, post-hoc tests were performed using a Bonferroni correction. Within-subjects
significant main effects were further analyzed with post-hoc paired t-tests with an  correction
for the number of comparisons made. The dependent measures in the response inhibition task
were accuracy, omissions, rate of responding on the non-reinforced lever (2 lever task only) and
33
latencies to either premature or correct responses. In order to examine whether accuracy was
stable across testing and between different groups of animals, data from the baseline sessions
were analyzed using a 3-way ANOVA with session (baseline 1-14) and delay interval (1, 4, 15,
30 and 60 s) as within-subjects factors, and group (3 morphine and 3 SNC80) as a betweensubjects factor. Data from drug testing sessions were analyzed with dose and delay as withinsubjects factors and order of injection (ascending versus descending) as repeated-measures
factors, with one exception noted below. Including injection order as a factor in the ANOVA
controls for the effect of repeated drug exposure on behavioral measures, and makes the analysis
more conservative. Horizontal beam breaks was the dependent measure in the locomotor test and
data were analyzed using group (vehicle vs SNC80) and order of injection as between-subjects
factors and time (10 min bins) as a within-subjects factor. The session was divided into baseline,
saline and drug injection, which were analyzed separately.
2.4 5 Results
Response inhibition training
ANOVA was used to examine whether accuracy (the dependent measure in the response
inhibition task) was stable between groups and across testing. This analysis revealed no
differences across baseline sessions [F(13,611) = 1.99, p = 0.07] and no difference in accuracy
between the experiments [F(5,47) = 1.13, p = 0.36] (Figure 2.1b). The significant main effect of
delay, confirmed that accuracy declined (i.e., animals made more premature responses) as the
delay interval increased [F(4,188) = 204.33, p < 0.01]. Paired post-hoc t-tests revealed that
accuracy scores were significantly different at each interval (all ps < 0.01 with  correction for 4
comparisons). None of the interactions reached statistical significance. Furthermore, an ANOVA
34
comparing accuracy during baseline sessions (averaged across the last 3 sessions) with accuracy
during the final drug-free test session in the first SNC80 group confirmed that repeated testing
had no effect on this measure [F(1,8) = 0.1, p = 0.77]. Additionally, we examined whether
responding was stable across drug testing by comparing accuracy scores during baseline sessions
with accuracy across drug washout days (i.e., between drug testing in the 2-lever task). ANOVA
revealed a main effect of delay for both morphine [F(4,16) = 14.79, p < 0.01] and SNC80
[F(4,56) = 36.11, p < 0.01] experiments, but no main effects of session or injection order and no
significant interactions.
Effects of morphine on accuracy
Figure 2.2 shows that morphine had no effect on accuracy in the response inhibition task.
The lower (Figure 2.2a) and higher (Figure 2.2b) dose ranges yielded non-significant effects of
dose and injection, but statistically significant effects of delay: [F(4,5) = 33.31, p < 0.01] and
[F(4, 24) = 68.50, p < 0.01]. None of the interactions reached statistical significance. In the 2lever task (Figure 2.2c), two subjects were dropped from the analysis due to malfunctioning
equipment on one of the test days. Because both subjects were in the ascending dose condition,
order of injection was dropped as a factor in this analysis. Nonetheless, the data were consistent
with results from the 1-lever task: main effect of delay [F(4,20) = 21.63, p < 0.01] but no effect
of dose [F(3,15) = 0.23, p = 0.88]. Neither delay nor drug dose had an effect on the rate of nonreinforced lever presses (Figure 2.2d).
In all three groups of animals, morphine had no effect on omissions or on latencies to a
premature or correct response (all ps > 0.05). In addition, naloxone had no effect on any
dependent measure in the response inhibition task (data not shown).
35
Figure 2.2 Effects of morphine on motor impulsivity in the response inhibition task. Neither low
(a) nor high (b) doses of morphine altered accuracy in the response inhibition task, suggesting no
change in motor impulsivity. Morphine also had no effect on accuracy in a 2-lever version of the
task (c) and did not affect the rate of non-reinforced lever presses (d). Accuracy was calculated
as (correct responses/(premature + correct responses))*100. SED = standard error of differences
of means; SED is calculated as sqrt(2*MSerror/n) where n is the harmonic mean of the sample
sizes and MSerror is the overall interaction term.
Effects of SCN80 on accuracy
In contrast to the lack of morphine effects, SNC80 produced a dose-dependent decrease
in accuracy (Figure 2.3a), manifested as a significant main effect of dose [F(3,24) = 8.47, p <
0.01]. The significant main effect of delay [F(4,32) = 104.21, p < 0.01] confirmed that accuracy
declined across delay intervals and the delay x dose interaction [F(12,96) = 7.41, p < 0.01]
indicated that this effect differed across dose. Simple effects revealed a significant effect of drug
at the 15, 30, and 60 s delay (ps < 0.05 with  correction for 5 comparisons). Post-hoc paired t-
36
tests revealed that at 15, 30 and 60 s all doses of SNC80 significantly decreased accuracy
compared to vehicle (ps < 0.05 with  correction for 3 comparisons.
Figure 2.3b shows that the effect of SNC80 on accuracy was blocked by prior
administration of NTI. Statistical analysis confirmed a main effect of delay [F(4,112) = 61.29, p
< 0.01] and drug group [F(3,28) = 5.32, p < 0.01], and a significant delay x drug group
interaction [F(12,112) = 2.06, p < 0.05]. Simple effects revealed a significant effect of drug at
the 4, 15, 30, and 60 s delays (ps < 0.05 with  correction for 5 comparisons). Post-hoc tests
revealed no differences between NTI and vehicle at any of the delays, but SNC80 + NTI
significantly increased accuracy at the 4, 15, 30 and 60 s delay compared to SNC80 alone (ps <
0.05 with  correction for 2 comparisons).
The decrease in accuracy following an SNC80 injection was confirmed in the 2-lever task
(Figure 2.3c) by a main effect of dose [F(1,14) = 5.58, p < 0.05], delay [F(4,56) = 43.45, p <
0.01] and a delay x dose interaction [F(4,56) = 3.31, p < 0.05]. As previously described, post-hoc
paired t-tests revealed that accuracy declined with increases in the delay interval with a
significant difference at 60 s but not at the other intervals (ps < 0.05 with  correction for 5
comparisons). Despite these changes, SNC80 had no effect on the rate of non-reinforced lever
presses (Figure 2.3d) or on omissions in the 1- or 2-lever tasks. See Table 2.1 for details.
37
Figure 2.3 Effects of SNC80 on motor impulsivity in the response inhibition task. The highest
dose of SNC80 decreased accuracy (i.e., increased impulsivity) in the response inhibition task
(a), an effect that was blocked (b) by the selective delta opioid receptor antagonist naltrindole
(NTI). SNC80 also decreased accuracy in a 2-lever version of the task (c), but did not affect the
rate of non-reinforced lever presses (d). Accuracy was calculated as (correct
responses/(premature + correct responses))*100. SED = standard error of differences of means;
SED is calculated as sqrt(2*MSerror/n) where n is the harmonic mean of the sample sizes and
MSerror is the overall interaction term. a, c, and d: Vehicle = 8% 1 M HCl; b: Vehicle = 8% 1
M HCl plus 25% DMSO
38
Table 2.1 Effects of SNC80 on omissions in the response inhibition task
Group 1
Group 2
Group 3
Delay
Vehicle
10 mg/kg
Vehicle
10 mg/kg
Vehicle
10 mg/kg
1s
0.15 + 0.11
0.20 + 0.20
0+0
0+0
0.56  0.30
0.56  0.22
4s
0.25 + 0.25
0.15 + 0.08
0+0
0+0
0.44  0.22
0.38  0.31
15 s
0.50 + 0.30
0.15 + 0.08
0+0
0.12 + 0.09
0.38  0.22
0.19  0.10
30 s
0.30 + 0.17
0.10 + 0.07
0+0
0.12 + 0.07
0.56  0.29
0.31  0.15
60 s
0.25 + 0.13
0.05 + 0.05
0+0
0.75 + 0.10
0.06  0.06
0.19  0.10
Values represent the mean (+SEM) number of omissions at each delay interval following an
injection of vehicle or SNC80 (10 mg/kg) in the 1-lever (Groups 1 and 2) and 2-lever (Group 3)
versions of the task. Vehicle Group 1 and 3 = 8% 1 M HCl solution; Vehicle Group 2 = 8% 1 M
HCl plus 25% DMSO.
Table 2.2 and Table 2.3 show latencies to premature and correct responses, respectively,
following injections of 0 or 10 mg/kg SNC80 (the latter being the dose which decreased
accuracy). SNC80 had no effect on the latency to premature or correct responses, although
latencies to premature responses increased with the delay interval in the 1-lever task [F(4,5) =
96.35, p < 0.01] and [F(4,5) = 3.14, p < 0.05]. In the 2-lever task, there was a main effect of drug
[F(3,28) = 3.94, p < 0.05] and delay [F(4,112) = 88.24, p < 0.01] on latencies for premature, but
not correct, responses. Similarly, the delay x group interaction was statistically significant for
premature [F(12,112) = 4.01, p < 0.05], but not correct, responses. Simple effects of latencies to
premature responding revealed a significant effect of drug at the 30 and 60 s delays (ps < 0.05
with α correction for 5 comparisons). Post hoc t-test showed no group differences at the 30 s
delay; however the NTI group had longer latencies than the SNC80 group at the 60 s delay (ps <
0.05 with α correction for 4 comparisons).
39
Table 2.2. Effects of SCN80 on latencies to a premature response in the response inhibition
task.
Group 1
Group 2
Group 3
Delay
Vehicle
10 mg/kg
Vehicle
10 mg/kg
Vehicle
10 mg/kg
1s
0.23 + 0.05
0.26 + 0.07
0+0
0.82 + 0.03
0.02 + 0.01
0.02 + 0.01
4s
1.51 + 0.20
1.72 + 0.20
1.60 + 0.27
2.22 + 0.21
0.26 + 0.10
0.32 + 0.10
15 s
3.59 + 0.04
5.47 + 0.57
3.05 + 0.52
5.52 + 0.39
1.35 + 0.44
1.91 + 0.45
30 s
9.15 + 1.71
11.8 + 1.39
10.67 + 1.09
8.19 + 0.97
5.25 + 1.30
5.36 + 1.12
60 s
23.32 + 1.72
19.59 + 1.01
24.77 + 2.35
21.21 + 1.53
13.29 + 1.87 15.64 + 1.09
Values represent the mean (+SEM) latency (s) to responding prematurely at each delay interval
following an injection of vehicle or SNC80 (10 mg/kg) in the 1-lever (Groups 1 and 2) and 2lever (Group 3) versions of the task. Vehicle Group 1 and 3 = 8% 1 M HCl solution; Vehicle
Group 2 = 8% 1 M HCl plus 25% DMSO.
Table 2.3. Effect of SCN80 on latencies to a correct response in the response inhibition task.
Group 1
Group 2
Group 3
Delay
Vehicle
10 mg/kg
Vehicle
10 mg/kg
Vehicle
10 mg/kg
1s
1.82 + 0.22
1.64 + 0.16
1.26 + 0.15
1.34 + 0.18
1.42  0.21
1.78  0.25
4s
1.64 + 0.21
1.39 + 0.16
1.08 + 0.12
1.52 + 0.15
1.31  0.14
1.39  0.15
15 s
1.88 + 0.21
1.82 + 0.19
1.38 + 0.12
1.19 + 0.25
1.17  0.25
1.18  0.26
30 s
1.82 + 0.15
1.47 + 0.15
1.45 + 0.14
1.36 + 0.30
0.88  0.15
0.84  0.23
60 s
1.91 + 0.18
1.71 + 0.17
0.99 + 0.17
0.58 + 0.17
0.73  0.12
0.49  0.15
Values represent the mean (+SEM) latency (s) to respond correctly at each delay interval
following an injection of vehicle or SNC80 (10 mg/kg) in the 1-lever (Groups 1 and 2) and 2lever (Group 3) versions of the task. Vehicle Group 1 and 3 = 8% 1 M HCl solution; Vehicle
Group 2 = 8% 1 M HCl plus 25% DMSO.
40
Effects of morphine and SNC80 on locomotor activity
As shown in Figure 2.4, As shown in Fig. 4, activity levels declined in all groups across
the habituation phase of testing [SNC80: F(5,30)= 56.26, p<0.01; morphine: F(5,30)=40.79,
p<0.01]. Activity continued to decline following a saline injection [SNC80: F (5,30)=3.58,
p=0.01; morphine: F(5,30)=14.09, p<0.01]; both drugs significantly increased activity levels
[SNC80: F (1,6)=257.53, p<0.01; morphine: F(1,6)=9.99, p<0.05], which declined across the
next 2 h [SNC80: F(11,66)= 3.44, p=0.01; morphine: F(11,66)=3.12, p<0.01].
Figure 2.4 Effects of SNC80 and morphine on locomotor responses. Data points represent beam
breaks per 10-min bin during habituation (60 min), following a saline injection (120 min), and
following an injection of 10 mg/kg SNC80 (top) or 2 mg/kg morphine (bottom) (120 min).
Arrows indicate the time of saline (left) and drug or vehicle (right) injections.
41
2.5 Discussion
The primary, novel finding in this study is that administration of a delta opioid receptor
agonist (SNC80) to rats decreases accuracy in a response inhibition task. Pretreatment with the
selective antagonist, NTI, reversed the disinhibitory effect of SNC80, confirming a key role for
delta opioid receptors in this effect. The fact that activation of delta opioid receptors increased
premature responses appears to contradict our original report (Olmstead et al. 2009) that genetic
deletion of delta opioid receptors had similar effects in mice. Gene deletion and behavioral
pharmacology studies frequently yield discrepant results (Filliol et al. 2000; Gingrich and Hen
2001; Zhang et al. 1999), a difference that is often attributed to developmental issues in
genetically modified animals and/or poor specificity of pharmacological tools (Kieffer and
Gaveriaux-Ruff 2002; Portugal and Gould 2008). In addition to species and experimental
manipulation, the apparent contradiction between our previous and current findings could be
explained by a differential contribution of opioid mechanisms to task performance. This seems
likely given the behavioral requirements of the two paradigms. In the signaled nose poke and
response inhibition tasks, animals are trained with a set versus a random ITI, respectively. In the
former, a 20-s ITI is followed by a variable premature phase (1–8s), but unlike the response
inhibition task, the onset of the premature phase is not signaled. Moreover, mice may nose poke
at any point during a trial, in contrast to the response inhibition task in which the lever is
retracted during the ITI. Finally, although there is no evidence that inhibition of a nose poke and
lever pressing response invoke different neural systems, this remains a possible source of
discrepant findings. In sum, any or all of these task differences may have accounted for the
differential effects of gene knockout versus drug administration on these two measures of motor
impulsivity.
42
The fact that both deletion and activation of delta opioid receptors increase premature
responses may also be explained by task differences in the length of the premature phase. In the
signaled nose poke task, mice are required to withhold motor responses for 1–8 s, whereas rats
must inhibit responding in the response inhibition task for 1–60 s. SNC80 only produced an
effect on accuracy at the longest interval, although differences were emerging at the 15- and 30-s
intervals. There is no direct evidence that inhibition of a motor process at long and short intervals
involves distinct neural systems, although neuropharmacological systems mediating inhibition in
tasks with long and short delays (e.g., differential reinforcement of low rate responding and fivechoice serial reaction time tasks) are dissociable (Hayton and Olmstead 2009). Thus, one
hypothesis is that decreases in delta opioid receptor signaling disrupt the ability to inhibit a
motor response at short intervals, whereas increases in receptor signaling disrupt motor
inhibition at long intervals. This hypothesis remains speculative because knockout mice were not
tested at long intervals (Olmstead et al. 2009). Moreover, even though SNC80 had no effect at
short delays, accuracy was close to ceiling levels in these trials so it may not have been possible
to observe behavioral changes. The idea that delta opioid receptors make different contributions
to response inhibition at long and short intervals, therefore, requires further study but does not
detract from the principle finding that delta opioid receptors play an important role in this test of
motor impulsivity. The contribution of delta opioid receptors in specific brain regions could be
addressed in future studies using conditional gene knockout technology, an extremely powerful
approach to elucidate the function of single genes (Gaveriaux-Ruff and Kieffer 2007).
The contribution of delta opioid receptors to accuracy in the response inhibition task is
not likely due to non-selective increases in behavioral responding because SNC80 did not
increase latencies to either premature or correct responses in this task. In addition, the dose that
43
increased premature responses did not increase the rate of non-reinforced lever pressing. That is,
SNC80 had similar effects on accuracy when rats had the opportunity to respond on one or two
levers. At the same time, SNC80 increases locomotor activity (present findings) (Spina et al.
1998), which may have produced subtle changes in responding during task performance. The
stimulant effects of delta opioid receptor activation had no detectable effect on accuracy in the
response inhibition task, suggesting that the two effects may be dissociable. This fits with
evidence that the effect of delta opioid receptor deletion on premature responding in a signaled
poke task in mice is independent of changes in locomotor activity (Olmstead et al. 2009). Nor is
it obvious how increases in premature responses could be explained by SNC80-induced
alterations in anxiety or depressive-like symptoms (Russell 2007). Thus, in addition to the wellestablished role in emotional processing (Filliol et al 2000; Nieto et al. 2005), our findings
suggest that delta opioid receptors are involved in cognitive processes such as response
inhibition.
In contrast, our finding do not support a direct role for mu opioid receptors in
impulsivity, consistent with the evidence that morphine has no effect in the stop signal reaction
time task (Pattij et al. 2009). However, our results appear to contradict the evidence that
morphine increases action impulsivity in rats (Pattij et al. 2009) and that deletion of mu opioid
receptors in mice decreases motor impulsivity (Olmstead et al. 2009). It is easier to reconcile
discrepancies between mice genetic and rat pharmacological studies (Kieffer and Gaveriaux-Ruff
2002), than to explain why morphine had different effects on premature responses in one species.
It is unlikely that inconsistencies between the two studies are due to route of drug administration
(i.p. versus s.c.) or to drug dose. Indeed, we extended the dose range of morphine that we tested
(from 4 to 10 mg/kg) after the Pattij et al. data were published because we questioned whether
44
we had missed the effective dose. The most plausible explanation for the differential effects of
morphine on premature responses is that unique cognitive processes are engaged in the two tasks
(Hayton and Olmstead 2009). A primary distinction between the five-choice and response
inhibition paradigms is the use of a fixed versus a variable delay to the discriminative stimulus
indicating reward availability: Rats in the Pattij et al study learned to inhibit a motor response for
a set period (5 s) whereas rats in our study could not predict the length of this premature phase
(1–60 s). Morphine, therefore, may have disrupted the ability to accurately time intervals in the
Pattij study, a process that would not be engaged (and thus not disrupted) in the response
inhibition task. This would also explain why morphine has no effect in the stop signal reaction
time, as this paradigm uses random, rather than fixed, delays to a discriminative stimulus.
The difficulty in establishing how opioid systems contribute to motor impulsivity is also
apparent in human research. There are no published reports examining opiate-induced changes in
motor impulsivity in humans, and the evidence for inhibitory deficits in opiate addicts is
equivocal (Clark et al. 2006; Forman et al. 2004; Passetti et al. 2008; Verdejo-Garcia et al. 2008;
Verdejo-Garcia et al. 2007). Addicts in these studies are tested drug free, with the assumption
that basal functioning of the opioidergic system is altered due to chronic opiate use. Our finding
that neither NTI nor naloxone affected performance in the response inhibition task suggests that
endogenous opioid tone is not involved in motor impulsivity. Endogenous opioid tone (Kieffer
and Gaveriaux-Ruff 2002) refers to changes in basal peptides levels, often produced in responses
to physiological challenge (e.g., hunger). Release of endogenous peptides may be measured
using microdialysis (Torregrossa et al. 2006) or inferred by antagonist-induced changes in
behavior. Our suggestion that the release of endogenous opioid peptides does not play a major
role in motor impulsivity fits with previous evidence (Pattij et al., 2009), although the effects
45
may be different in animals chronically treated with opiates. Under those conditions, morphine
and other drugs of abuse may produce robust and consistent effects on response inhibition,
paralleling the disinhibited behaviors that characterize addiction (Everitt et al. 2008; Olmstead
2006; Perry and Carroll 2008). Our study focused on one type of impulsivity (response inhibition
or motor impulsivity), but opioid systems contribute to other types of impulsive behavior. For
example, morphine increases impulsive decision making in three different rodent tasks (Everitt et
al. 2008; Kieres et al. 2004; Pitts and McKinney 2005) and opiate addicts display deficits in
cognitive (Bickel and Marsch 2001; Kirby et al. 1999; Verdejo-Garcia et al. 2007) as well as
reflection (Clark et al. 2006) impulsivity. This adds to existing evidence that 5-HT, DA, NA,
glutamate, and cannabinoid systems contribute to impulsive responding (Pattij and
Vanderschuren 2008). It should not be surprising that multiple neural systems are implicated in
impulsivity because this is a poorly defined cognitive function that involves a number of
different processes. It is unlikely that any of these function in isolation, so a full characterization
of neural mechanism that mediates impulsivity must be built around interactions between
neuropharmacological systems. Only then can progress in the development of therapies for
disorders of impulse control be advanced.
46
Increased Impulsive Action in Rats: Effects of Morphine in a Short and Long
Fixed-Delay Response Inhibition Task
As published in: Mahoney MK, Silveira, MM, and Olmstead, MC. (2013). Increased impulsive
action in rats: effects of morphine in a short and long fixed-delay response inhibition task.
Psychopharmacology, 4, 569-77.
3.1 Abstract
Rationale: Impulsive action is mediated through several neurochemical systems although it is
not clear which role each of these play in the inability to withhold inappropriate responses.
Manipulations of the opioid system alter impulsive action in rodents, although the effects are not
consistent across tasks. Previously, we speculated that these discrepancies reflect differences in
the cognitive mechanisms that control responding in each task. Objectives: We investigated
whether the effect of morphine, a mu opioid receptor (MOR) agonist, on impulsive action
depends on the ability to time the interval during which subjects must inhibit a response.
Methods: Male Long–Evans rats were trained in a response inhibition (RI) task to withhold
responding for sucrose during a 4 or 60 s delay; impulsive action was assessed as increased
responding during the delay. Rats were tested following an injection of morphine (0, 1, 3, 6
mg/kg). In a subsequent experiment, the effects of morphine (6 mg/kg) plus the MOR antagonist
naloxone (0, .3, 1, 3 mg/kg) were investigated. Results: Morphine increased impulsive action,
but had different effects in the two conditions: the drug increased the proportion of premature
responses as the 4-s interval progressed and produced a general increase in responding across the
60-s interval. Naloxone blocked all morphine-induced effects. Conclusions: The finding that
47
morphine increases impulsive action in a fixed-delay RI task contrasts with our previous
evidence showing no effect in the same task with a variable delay. Thus, MORs disrupt
impulsive action only when rats can predict the delay to respond.
3.2 Introduction
Impulsivity includes several distinct processes that can be dissociated neurochemically,
anatomically, and behaviourally (Evenden 1999; Dalley et al. 2011). The broadest division is
between cognitive and motor impulsivity, defined as an inability to delay gratification and an
inability to inhibit responding, respectively (Olmstead 2006). The latter, recently termed
impulsive action (Winstanley et al. 2006a), is manifested in a number of psychiatric disorders
including substance abuse, pathological gambling, attention deficit hyperactivity disorder, binge
eating, and antisocial personality disorder (American Psychiatric Association 2013).
Several brain regions and neurotransmitters are implicated in impulsive action (Pattij and
Vanderschuren 2008; Fletcher et al. 2011), which may reflect an interplay of cognitive processes
that control responding in different tasks (Hayton and Olmstead 2009). Recently, we uncovered
an opioidergic contribution to this process by showing that mu opioid receptor (MOR) deletion
in mice decreases impulsive action in a signaled nose poke task (Olmstead et al. 2009). In line
with our findings, morphine injections, which activate MORs, increase impulsive action in a 5choice serial reaction time task (5-CSRTT), an effect that is blocked by the MOR antagonist,
naloxone (Pattij et al. 2009). Both findings help to explain clinical reports that opioid antagonists
enhance control of motor responses (Mitchell et al. 2007) and may be an effective treatment for
impulse-control disorders (Kim 1998).
Given this evidence, we were surprised to observe no effect of morphine on impulsive
action in a response inhibition (RI) task (Befort et al. 2011). In this paradigm, rats withhold lever
48
pressing for sucrose until a signal is presented: responses prior to the signal are a measure of
impulsive action. Delays to the signal presentation are varied from 1-60 s, such that rats cannot
predict the length of the premature phase on any given trial. We speculated that distinct cognitive
mechanisms, mediated by dissociable neurochemical systems, would be engaged when animals
can and can not time the interval during which they must inhibit a response. In other words,
opioid systems may contribute to impulsive action in one task condition (fixed delay) but not
another (variable delay). Indeed, amphetamine has opposite effects on impulsive action in the RI
task with fixed and variable delays: the drug decreases impulsive action in a variable delay task,
but increases impulsive action in a fixed delay task (Hayton et al. 2012).
To test this possibility, we examined the effects of morphine in the RI task (Hayton et al.
2010) using a fixed delay, as a comparison to our previous results using a variable delay (Befort
et al. 2011). With the exception of the premature delay duration (variable versus fixed), the two
tasks were identical in terms training, discriminative stimuli signaling reward availability, and
session length. Separate groups of rats were trained to withhold responding for a short (4 s) or
long (60 s) delay until responding was stable across sessions. Testing was conducted following
an injection of morphine and, in a subsequent experiment, naloxone plus morphine. We also
examined whether morphine alters timing mechanisms by analyzing the proportion of premature
responses across trials with short and long delays. A similar analysis revealed unique patterns of
time-dependent errors under different task conditions following amphetamine administration
(Hayton et al. 2012).
49
3.3 Materials and methods
Subjects
Male Long-Evans rats (Charles River, PQ, Canada), weighing 250-350 g at arrival, were
pair-housed on a reverse light-dark cycle in standard polycarbonate cages. Behavioral testing
was conducted during the dark cycle. Three days prior to training, rats were food restricted to 90
min of free access to regular chow (Lab Diet: PMI Nutritional International, Inc) per day such
that they gained 10-15 g per week. Animal care was conducted in accordance with the guidelines
provided by the Canadian Council on Animal Care, and the experiments were approved by the
Queen’s University Animal Care Committee.
Apparatus
Training and testing were conducted in operant conditioning chambers (26.5 × 22.0 ×
20.0 cm), each housed in a sound-attenuating chamber (constructed in-house). Each chamber
was fitted with two retractable levers, 4 cm wide, 5 cm from the mesh floor, and 13 cm apart.
Sucrose pellets (45 mg; Bio-Serv, NJ, USA) were delivered to a food magazine via a pellet
dispenser (Med Associates, Vt., USA). The food magazine was located between the two levers
and was equipped with infrared sensors to detect food collection. Lights were situated 4 cm
above both levers and the magazine. A house light was situated outside of the operant chamber.
An IBM-type computer in an adjacent room controlled the equipment and was used for data
collection (software written in-house using ECBASIC).
Drug Administration
Morphine sulfate (BDH Pharmaceuticals, Toronto ON) and naloxone (Tocris
Biosciences, MO, USA) were dissolved in 0.9 % physiological saline. Morphine was tested at
50
doses of 0, 1, 3 and 6 mg/kg s.c., with each subject receiving two injections at each dose. The
opioid receptor antagonist, naloxone, was tested once at doses of 0, .3, 1, and 3 mg/kg. In
combined-drug studies, naloxone doses were tested with either 0 or 6 mg/kg morphine. Naloxone
was injected s.c. 30 min prior to behavioral testing, whereas morphine was injected s.c. 20 min
prior to testing. All drug doses were administered in an ascending or descending order,
counterbalanced across animals. Drug testing was separated by 48 h and animals were tested
drug-free on the intervening days.
Behavioral procedures
Response inhibition training took place over a series of sequential phases. In the first, rats
were magazine trained for one day, receiving 20 sucrose pellets on a random time (RT) 90-s
schedule. The house light was illuminated until a food pellet was dispensed. At this point, the
magazine light turned on, the house light turned off for 1 s, and the next trial commenced.
Rats were then trained to lever press for food on a continuous reinforcement (CRF) schedule.
Two levers were inserted into the chamber, with reinforced lever assignment counterbalanced
across animals (left versus right). The house light and discriminative stimulus (lever light) were
turned on throughout these sessions (response phase), except during delivery of the reward,
which was signaled by illumination of the magazine light (1 s). Training continued until a
minimum 80 pellets was earned in a 60-min session for two consecutive sessions.
Following CRF training, the response phase of the task was shortened such that rats had
10 s to lever press following lever insertion. The house light and discriminative stimulus were
turned on during this 10-s period. If 10 s elapsed, the levers were retracted for a 10-s inter-trial
interval (ITI) and the trial was scored as an omission. Both the house light and discriminative
stimulus were turned off during the ITI. Training continued until animals reached a criterion of
51
fewer than 20% omitted trials for two consecutive sessions. Over 90% of the animals reached
criterion in 2 sessions, with the remaining completing this stage of training in 4 sessions.
In the next stage of training, each trial commenced with a 10-s ITI in which the levers were
retracted and the house light was off. At the end of the ITI, the levers were inserted and the
house light was illuminated but the discriminative stimulus was delayed for a variable period.
Lever presses prior to the discriminative stimulus (i.e., premature responses) reinstated the ITI
with no reward. Responses in the presence of the discriminative stimulus (10-s limit) were
reinforced with a sucrose pellet and initiated the next trial. Animals received five training
sessions at 4-s and15-s delays (both 4-s and 60-s conditions), and at a 60-s delay (60-s
condition). All training sessions consisted of 100 trials.
Subjects then progressed to the full RI task described previously (Hayton et al. 2010;
Befort et al. 2011; Hayton et al. 2012). Each session consisted of 100 trials at a Fixed 4-s delay,
or 50 trials at a Fixed 60-s delay, so that the session duration was consistent in both conditions.
To achieve stable responding, animals were tested in the full RI task for 14 sessions (Fixed 4-s
condition; n = 16 for morphine-only experiment, n = 16 for combined drug experiment) or 21
sessions (Fixed-60-s condition; n = 16 for morphine-only experiment, n = 16 for combined drug
experiment) prior to drug administration, with the last three sessions constituting a baseline.
Briefly, following a 10-s ITI, each trial progressed to a premature response phase,
signaled by both levers being inserted and illumination of the house light. A single lever press on
the reinforced lever during the premature phase (4 s or 60 s) initiated a new trial, and was
recorded as a premature response. If subjects did not lever press during the premature phase, the
trial progressed to a correct response phase, during which the discriminative stimulus was
illuminated, and the house light was extinguished. Lever pressing on the reinforced lever during
52
this phase was reinforced by a sucrose pellet, and initiated the next trial. Failure to respond
during the 10-s response phase was counted as an omission and initiated the next trial.
Data and statistical analyses
Premature responding was measured as the percentage of trials in which animals did not
inhibit a lever pressing response prior to the discriminative stimulus. This dependent measure
was calculated as (premature responses/(premature + correct responses))*100. Proportion of
premature responses reflected the likelihood of making a premature response during a set period;
this measure was analyzed using four 1-s bins and twelve 5-s bins for the 4-s and 60-s
conditions, respectively (Hayton et al. 2011). Proportion of responses was calculated as the
number of premature responses during a bin, divided by the total number of trials in which rats
responded (premature + correct responses). For example if a subject made two premature
responses (100 trials/session) in the 0-1 s bin of the 4-s condition, the proportion of responses
during this time bin would be 2/100 = 0.02. If the same subject made 5 premature responses in
the 1-2 s bin, the proportion would be 5/100 = 0.05. The number of omissions and the latencies
to premature and correct responses were recorded for each session. Latencies were calculated as
the time from lever insertion (premature response) or discriminative stimulus presentation
(correct response) to the first lever press. Responses on the non-reinforced lever were recorded
but had no programmed consequences, and were calculated as the rate of responding per minute.
Data were analyzed using analysis of variance (ANOVA), computed with the Statistical
Package for the Social Sciences (SPSS; V.19.0). Degrees of freedom for repeated measures were
adjusted using the Greenhouse-Geisser correction if assumptions of sphericity were violated
(Greenhouse and Geisser 1959). When significant main effects were observed for betweensubjects analyses, post-hoc tests were performed using a Bonferroni correction. Within-subjects
53
significant main effects were further analyzed with post-hoc paired t-tests with an  correction
for the number of comparisons made. The dependent measures in the response inhibition
condition were percent premature responses, premature response proportion, omissions, rate of
non-reinforced responding, and latencies to either premature or correct responses. In order to
address the fact that the variance of the omission data was scaled by dose, we performed square
root transformations to omission data in all experiments, to adjust the evident positive skew and
to reduce scaling. Data from each session were excluded from analysis if the number of
omissions were greater than 3 standard deviations from the mean of each dose group.
Consequently, four animals were excluded from each of the morphine 4-s and 60-s conditions
and one animal was excluded from each of the combined drug (naloxone plus morphine)
conditions. In order to examine whether percent premature responding was stable across testing
and between different groups of rats, data from the baseline sessions were analyzed using a 2way ANOVA with session (last 3 sessions) as a within-subjects factor, and group (morphine
alone and combined-drug groups) as a between-subjects factor. Baseline calculations of response
proportion were made by pooling the final seven sessions, and the effects of morphine on this
measure were calculated by pooling data from the two sessions at each dose. Drug testing data
were analyzed with dose as a within-subjects factor and dosing order (ascending versus
descending) as a between-subjects factor. In addition, injection number (first or second injection
of the same dose) was included as a within-subjects factor in the morphine-only testing analyses.
54
3.4 Results
Baseline response inhibition training
There was no significant difference in the percent of premature responses across the last
three baseline sessions and no session x experiment interaction for this measure. The significant
main effect of condition [F(3,50)= 4.04, p = 0.012] was due to higher premature responding in
the Fixed 60-s than in the Fixed 4-s condition (all ps < 0.05). Additionally, a comparison of
premature responding during baseline sessions (last 3 days pooled) and across morphine or
combined drug experiment washout days revealed no significant main effects or interactions for
either the 4-s or 60-s conditions.
Effect of morphine on premature responding in Fixed 4-s and 60-s conditions
As shown in Figure 3.1, morphine produced a dose-dependent increase in percent
premature responses in the Fixed 4-s and 60-s conditions [F(3,30) = 4.60, p = 0.027] and
[F(3,30) = 10.46, p = 0.001]. Post hoc t-tests of the significant main effect of dose in each
condition showed that both 3 mg/kg and 6 mg/kg of morphine significantly increased percent
premature responding compared to vehicle injections (all ps < 0.05). In neither the Fixed 4-s nor
60-s condition, were there any significant main effects of dose order, or injection, or interactions
between these two factors.
55
Figure 3.1 Effects of morphine on impulsive action in the response inhibition task.
The data represent percentage of premature responses following an injection of morphine in the
Fixed 4-s (n = 12) (a) and Fixed 60-s (n = 12) (b) conditions, calculated as (premature
responses/(premature + correct responses))*100. Data are expressed as means ( SEM).
* p < 0.05 versus vehicle (0 mg/kg).
56
Table 3.1 shows the effects of morphine on premature and correct latencies, nonreinforced lever pressing and omissions. Morphine did not alter the latency to respond during the
premature or correct phase or non-reinforced lever pressing for either the 4-s or 60-s conditions.
In both the 4-s or 60-s conditions, dose had a significant effect on omissions [F(3,33) = 29.12, p
= 0.001] and [F(3,39) = 12.02, p = 0.01] with post hoc t-tests revealing that 6 mg/kg morphine
increased omissions compared to vehicle in both conditions (all ps < 0.05). There was no
significant effect of, or interaction with, dose order or injection number for any of the variables
measured in the morphine experiments (all ps > 0.05).
Table 3.1. Effects of morphine on behavioral measures in the response inhibition task
Fixed 4-s condition
Dose
(mg/kg)
Premature
Latency (s)
Correct Latency
(s)
NRL presses
(min)
Omissions (%)
Vehicle
3.12  0.08
1.54  0.17
7.80  2.82
0.13 ± 0.10
1.0
3.03  0.12
1.44  0.17
7.80  2.94
0.38 ± 0.18
3.0
2.88  0.13
1.61  0.22
5.94  2.16
0.85 ± 0.24
6.0
2.72  0.14
2.10  0.18
7.20  3.60
2.72* ± 0.46
Fixed 60-s condition
Dose
(mg/kg)
Premature
Latency (s)
Correct Latency
(s)
NRL presses
(min)
Omissions (%)
Vehicle
31.093.46
1.750.22
1.380.60
0.46±0.20
1.0
33.812.13
1.350.15
0.900.33
0.81±0.30
3.0
33.272.66
2.160.35
0.500.22
1.80±0.39
6.0
29.512.56
2.390.27
0.190.084
3.45*±0.73
Values represent mean (SEM) premature and correct response latencies, number of nonreinforced lever (NRL) presses per min, and percent omissions for the Fixed 4-s (n = 12) and
Fixed 60-s (n = 12) conditions following varying doses of morphine. Note: Represented
omission data are square root transformations. * p < 0.05, compared to vehicle.
57
Effect of naloxone on morphine-induced premature responding
As shown in Figure 3.2, naloxone significantly reduced morphine-induced increases in
percent premature responses in both the Fixed 4-s (Figure 3.2a) and 60-s (Figure 3.2b)
conditions, [F(4,52) = 17.72, p = 0.001] and [F(4,52) = 48.39, p = 0.001]. Post hoc t-tests of the
dose main effect showed that all doses of naloxone blocked morphine-induced increases in
premature responding for both conditions (all ps < 0.05). There was no significant main effect of
order and no dose x dose order interaction for either conditions. When tested alone, naloxone did
not alter percent premature responses in either the 4-s or 60-s conditions nor was there a
significant main effect of dose order or dose x order interaction in either condition (all ps >
0.05).
58
Figure 3.2 Effects of naloxone on morphine-induced increases in impulsive action in the
response inhibition task. The data represent percentage of premature responding following
injections of naloxone (Nal) + morphine (Mor) in the Fixed 4-s (n = 15) (a) and Fixed 60-s (n =
15) (b) conditions, calculated as (premature responses/(premature + correct responses))*100.
Data are expressed as means (SEM). * p < 0.05 versus vehicle (0 mg/kg), # p < 0.05 compared
to morphine-vehicle.
59
Table 3.2 shows the effects of combined naloxone-morphine injections on premature and
correct latencies, non-reinforced lever pressing and omissions for both 4-s and 60-s conditions.
The combined drug injection did not alter the latency to respond during the premature phase for
either condition; however, there was a main effect of drug on latency to respond during the
correct phase in both the 4-s and 60-s conditions: [F(4,52) = 8.90, p = 0.01] and [F(4,52) = 9.08,
p = 0.01]. Post hoc t-tests revealed that 1 mg/kg naloxone plus 6 mg/kg morphine significantly
decreased correct response latency in the 4-s condition, whereas vehicle-morphine significantly
increased correct response latency in the 60-s condition. Non-reinforced lever pressing did not
differ across the combined naloxone-morphine doses for either condition. The combined drug
increased the number of omissions in both the 4-s [F(4,52) = 6.33, p = 0.01] and 60-s condition
[F(4,52) = 47.26, p = 0.001]. Post hoc t-test tests revealed a significantly higher number of
omissions in the 4-s condition following vehicle-morphine compared to morphine and 0.3 mg/kg
naloxone, while significantly higher number of omissions in the 60-s condition were observed
following vehicle-vehicle compared to morphine-vehicle (all ps < 0.05). There was no significant
effect of, or interaction with, dose order for any of the variables measured in the combined drug
experiments (all ps > 0.05).
60
Table 3.2. Effect of naloxone on morphine-induced behavioral changes in the response
inhibition task
Fixed 4-s condition
Premature
Latency (s)
Correct
Latency (s)
NRL presses
(min)
Omissions (%)
Veh-Veh
3.12  0.20
2.38  0.27
13.20  3.06
1.16 ± 0.51
Mor-Veh
2.52  0.19
2.69  0.26
4.98  1.20
2.75 ± 0.60
Mor-Nal 0.3
3.02  0.15
1.91  0.22
22.20  7.80
0.34* ± 0.16
Mor-Nal 1.0
3.09  0.12
1.72**  0.19
15.60  7.80
0.49 ± 0.17
Mor-Nal 3.0
2.75  0.28
1.85  0.24
18.60  7.20
0.93 ± 0.37
Dose (mg/kg)
Fixed 60-s condition
Premature
Latency (s)
Correct
Latency (s)
NRL presses
(min)
Omissions (%)
Veh-Veh
34.13  3.37
1.81  0.17
0.35  0.084
0.85 ± 0.32
Mor-Veh
31.24  1.71
2.95**  0.23
0.132  0.049
6.99** ± 0.91
Mor-Nal 0.3
29.58  2.25
1.94  0.17
0.84  0.45
1.09 ± 0.31
Mor-Nal 1.0
34.50  2.74
1.68  0.14
0.58  0.20
0.29 ± 0.15
Mor-Nal 3.0
34.13  3.37
1.81  0.17
0.35  0.084
0.62 ± 0.20
Dose (mg/kg)
Values represent mean (SEM) premature and correct response latencies, number of nonreinforced lever (NRL) presses per min, and percent omissions for the Fixed 4-s (n = 15) and
Fixed 60-s (n = 15) conditions following varying doses of naloxone (Nal) + 6 mg/kg morphine
(Mor). Note: Represented omission data are square root transformations. * p < 0.05 compared to
vehicle + 6 mg/kg morphine (Mor-Veh), ** p < 0.05, compared to vehicle-vehicle (0 mg/kg).
61
Table 3.3 shows the effects of naloxone injections on premature and correct latencies,
non-reinforced lever pressing and omissions for both 4-s and 60-s conditions. In the 60-s
condition, there was a significant main effect of correct responding latency [F(3,39) = 4.16, p =
0.012], with post hoc tests revealing that 1.0 mg/kg naloxone significantly increased the correct
response latency compared to saline. There were no significant main effects or interactions of
naloxone dose on any of the other variables analyzed (all ps > 0.05).
Table 3.3. Effects of naloxone on behavioral measures in the response inhibition task
Fixed 4-s condition
Dose
(mg/kg)
Premature
Latency (s)
Correct
Latency (s)
NRL presses
(min)
Omissions (%)
Vehicle
2.95  0.70
2.45  1.50
19.29  5.70
0.85  0.40
Nal 0.3
3.36  0.39
2.34  1.38
22.88  6.02
0.25  0.15
Nal 1.0
2.89  0.93
2.34  1.47
19.88  5.26
0.19  0.13
Nal 3.0
3.30  0.39
2.17  1.33
20.67  5.33
0.26  0.14
Fixed 60-s condition
Dose
(mg/kg)
Premature
Latency (s)
Correct
Latency (s)
NRL presses
(min)
Omissions (%)
Vehicle
28.95  3.25
1.55  0.14
0.29  0.067
0.42  0.16
Nal 0.3
57.06  24.78
1.64  0.17
0.3  0.078
0.46  0.16
Nal 1.0
31.14  3.57
1.91*  0.16
0.34  0.079
0.47  0.15
Nal 3.0
32.32  2.48
1.71  0.17
0.48  0.13
0.33  0.14
Values represent mean (SEM) premature and correct response latencies, number of nonreinforced lever (NRL) presses per min, and percent omissions for the Fixed 4-s (n = 15) and
Fixed 60-s (n = 15) conditions following varying doses of naloxone (Nal). Note: Represented
omission data are square root transformations. * p < 0.05, compared to vehicle-vehicle (0
mg/kg).
Effect of morphine on the proportion of premature responses
62
As shown in Figure 3.3, morphine increased the proportion of premature responses in the
Fixed 4-s condition (four, 1-s bins) [F(3,39) = 6.15, p = 0.01], which also increased for all doses
across time [F(3,39) = 48.56, p = 0.001]. Simple main effects of the significant dose x time
interaction [F(9,117) = 3.80, p = 0.01] revealed that 1 m/kg morphine significantly increased
proportion of responses in bin 3 (2-3s), and bin 4 (3-4s) mg/kg; both 3 mg/kg and 6 mg/kg
morphine significantly increased the proportion of responses in bin 2 (1-2 s), bin 3 (2-3s), and
bin 4 (3-4s) (all ps < 0.05; Figure 3.3a). Similarly, morphine dose dependently increased
premature response proportion in the Fixed 60-s condition [F(3,39) = 11.06, p = 0.001], with
post hoc t-tests showing that both 3 mg/kg and 6 mg/kg morphine increased the likelihood of
making a premature response (Figure 3.3b). In contrast to the 4-s condition, there was no effect
of time and no dose x time interaction in the Fixed 60-s condition (ps > 0.05). Furthermore, in
the 60-s condition, dose had no effect on response proportion in the first or last 5 s of the delay
(1-s bins) and there was no effect of time or dose x time interaction. Finally, there was no main
effect of dose order and no dose x dose order interaction for either conditions (all ps > 0.05).
63
Figure 3.3 Effects of morphine on the proportion of responses during the premature phase of the
Fixed 4-s (n = 12) (a) and Fixed 60-s (n = 12) (b) conditions of the response inhibition task.
Response proportion is displayed in 1-s bins for the Fixed 4-s condition, and 5-s bins for the
Fixed 60-s condition. See text for description of response proportion calculation. SED standard
error of differences of means; SED is calculated as sqrt(2 × MSerror/n) where n is the harmonic
mean of the sample sizes and MSerror is the overall interaction term. * p < 0.05, versus vehicle
(Veh); # p < 0.05 versus vehicle (Veh + Veh) in the Fixed 60-s condition.
64
Effect of naloxone on morphine-induced increases in the proportion of premature
responses
Figure 3.4 shows the significant main effects of naloxone dose [F(4,52) = 9.61, p =
0.001] and time [F(4,39) = 15.46, p = 0.001] on a morphine-induced change in the proportion of
premature responses in the Fixed 4-s condition (four 1-s bins). Simple main effects of the
significant dose x time interaction [F(12,156) = 1.83, p = 0.047] revealed that vehicle-morphine
significantly increased response proportion compared to vehicle-vehicle injections in bin 3 (2-3s)
and bin 4 (3-4s). Furthermore, 0.3 mg/kg naloxone blocked morphine-induced increases in bin 3
(2-3s); 1.0 and 3.0 mg/kg morphine blocked morphine-induced increases in bins 3 (2-3s) and 4
(3-4s) (all ps < 0.05; Figure 3.4a). In the Fixed 60-s condition, the main effect of dose [F(4,32) =
21.01, p = 0.000], and the dose x time interaction [F(44,352) = 2.52, p = 0.04] were significant.
In contrast, the main effect of time [F(11,88) = 1.22, p = 0.32] was not statistically significant.
Simple main effects of the dose x time interaction revealed that morphine-induced increases in
premature response proportion was blocked by: 0.3 mg/kg naloxone in bin 2 (5-9s); 1.0 mg/kg
naloxone in bin 1 (0-4s) and bin 2 (5-9s); 3.0 mg/kg naloxone in bin 1 (0-4s), bin 2 (4-9s), bin 3
(10-14 s) , and bin 4 (15-19s) (all ps < 0.05; Figure 3.4b).When only the first or last 5 s (1-s bins)
of the 60-s condition was analyzed, there was a main effect of dose for the first 5 s [F(4,32) =
5.50, p = 0.021] but not the last 5 s [F(4,32) = 0.58, p = 0.62]. There was no main effect of dose
order and no dose x dose order interaction for either conditions (all ps > 0.05).
65
Figure 3.4 Effects of naloxone on morphine-induced increases in proportion of responses during
the premature phase of the Fixed 4-s (n = 15) (a) and Fixed 60-s (n = 15) (b) conditions of the
response inhibition task. Response proportion is displayed in 1-s bins for the Fixed 4-s condition,
and 5-s bins for the Fixed 60-s condition. See text for description of response proportion
calculation. SED standard error of differences of means; SED is calculated as sqrt(2 ×
MSerror/n) where n is the harmonic mean of the sample sizes and MSerror is the overall
interaction term. * p < 0.05, versus vehicle (Veh); # p < 0.05 versus vehicle-vehicle (Veh + Veh)
in the Fixed 60-s condition.
66
3.5 Discussion
Morphine dose-dependently increased impulsive action in a fixed-delay version of the RI
task, regardless of whether the delay was short (4 s) or long (60 s). These results are in stark
contrast to our previous report that morphine has no effect on impulsive action in the same task
with a variable delay (Befort et al. 2011). The latter study used a wide range of morphine doses
over three separate experiments, so we are confident that the drug does not alter responding in
the RI task when the duration of the premature phase is unpredictable. Moreover, both
amphetamine (Hayton et al. 2012) and the delta opioid receptor agonist, SNC80 (Befort et al.
2011) increase impulsive action in a variable-delay RI task (Hayton et al. 2012) confirming that
this task variant is not insensitive to drug effects. Comparing the results of the fixed and variable
delay versions of the RI task suggests that morphine alters premature responding only when
subjects are able to predict the duration of the premature phase. This fits with evidence that
morphine increases premature responding in the 5-CSRTT that uses a set delay, but has no effect
on reaction times in the SSRT, which has a variable delay (Pattij et al. 2009). Recently, we
reported a similar phenomenon with amphetamine: differential effects on premature responding
in the RI task depending on whether the premature delay was fixed (predictable) or variable
(unpredictable) (Hayton et al. 2012). Amphetamine increased impulsive action in the first
condition and decreased it in the second.
Although morphine increased impulsive action at both short and long fixed delays, it
produced a different pattern of time-dependent errors in the two conditions. As with
amphetamine (Hayton et al. 2012) morphine enhanced behavioral patterns which characterize
responding in each condition. More specifically, the proportion of premature responses in the 4-s
condition increased as the interval progressed. In contrast, the proportion of premature responses
67
in the 60-s condition was consistent across the delay phase. Morphine increased both tendencies,
leading to elevated responding near the end of the 4-s interval, but a stable increase in
responding across the 60-s interval. The differential effects of morphine on the proportion of
responses in short- and long-delay conditions suggest that the drug may be increasing impulsive
action through distinct processes. One possibility is that morphine disrupts the ability to time
intervals, a mechanism that appears to be important in the 4-s, but not the 60-s, version of the RI
task: increased proportion of responses across time in the former, but not the latter, indicates that
animals are anticipating the termination of the short-delay interval. This type of increased
responding across time also characterizes long delay intervals of DRL schedules (e.g. 72 s;
(Kramer and Rilling 1970; McClure et al. 1997)) but, unlike the RI task, these do not include a
signal at the end of the interval. We speculated previously (Hayton et al. 2012) that animals
responding in the RI task with long delays rely on the discriminative stimulus as a cue for reward
availability, resulting in consistent rates of responding throughout the premature phase.
The present findings, combined with our previous results, indicate that amphetamine and
morphine produce similar patterns of responding in the RI task. Opiates and psychostimulants,
therefore, may alter impulsive action via a common mechanism. The most likely possibility is
that opioid effects are mediated through dopaminergic systems as opiates stimulate dopamine
release (Di Chiara and Imperato 1988; Spanagel et al. 1990) in areas now known to be involved
in impulsive action (Pattij and Vanderschuren 2008; Eagle and Baunez 2010) and both naloxone
and CTAP (MOR antagonists) block amphetamine-induced impulsive action (Wiskerke et al.
2011a). The reverse, whether morphine-induced increases in impulsive action can be blocked by
dopamine antagonism, is unknown. On the other hand, it is equally possible that morphine and
68
amphetamine alter different cognitive mechanisms, both of which are manifested as disinhibited
responding in only one version of the RI task.
The mechanism by which morphine increases impulsive action may involve alterations in
time perception, at least in the Fixed 4-s condition when premature responses are elevated at the
end of the delay interval. This effect could involve an interaction with DA systems as
psychostimulants alter the perception of time (Meck 1983; Fowler et al. 2009), whereas there is
no evidence for similar changes following opiate administration. In the 60-s condition, morphine
may increase premature responding by enhancing the perceptual salience of the predictive
reward stimuli (e.g., lever insertion and house light illumination), analogous to increased
prepulse inhibition following morphine administration in humans (Quednow et al. 2008). This
may not be observed in the variable delay task where lever insertion and house light illumination
do not allow for predictions of delay length. Again, this process likely involves DA-opioid
interactions as MOR agonists attenuate DA-induced impairments in prepulse inhibition (Ukai
and Okuda 2003) and naloxone blocks amphetamine-induced impairments in prepulse inhibition
(Swerdlow et al. 1991). Finally, despite the similarities and interactions between
pharmacological systems, we cannot rule out the possibility that the increased impulsive action
following morphine and amphetamine administration are mediated by distinct mechanisms.
It is unlikely that the effects of morphine on impulsive action in our study were due to
non-selective increases in behavioural responding because the highest dose of morphine
significantly increased omissions as well as correct response latencies in the Fixed 60-s
conditions. In other words, a general slowing of responses with increasing drug doses would be
more likely to produce decreases, not increases, in premature responding. Moreover, we
observed significant increases in premature responding following the 3 mg/kg dose of morphine
69
that did not affect omissions or response latencies. Despite the effect on other behavioral
measures, we chose to test the effects of naloxone with the highest does of morphine (6 mg/kg)
because initial experiments with the short delay task did not reveal changes to latency or
omissions, and a complementary study (Pattij et al. 2009) reported effects of morphine on
impulsive action only at this dose. Given that all doses of naloxone blocked behavioral changes
induced by 6 mg/kg morphine, there is no reason to believe that antagonist actions would be
ineffective at the lower (i.e. 3 mg/kg) dose. Most importantly, the blockade of morphine-induced
increases in impulsive action by naloxone confirms an involvement of MORs in this effect.
Interestingly, naloxone alone did not alter impulsive action, even though MOR knockout
mice are more impulsive (Olmstead et al. 2009). This demonstrates differences between agonistinduced MOR signalling and MOR expression. That is, the expression of MORs, even the
absence of evoked activation, has an effect on impulsive action, possibly through basal tone or
through mu receptor interactions with other systems. The removal of such basal activity was only
achieved using genetic approaches to abolish MOR expression.
We showed that pharmacological activation of MORs increases impulsive action at both
short and long delays of a fixed duration. In contrast, we previously showed that morphine has
no effect on impulsive responding in a variable delay task (Befort et al. 2011). These findings
may provide an explanation for discrepant findings: Morphine alters impulsive action when the
delay is predictable, but not unpredictable. We further showed distinct patterns of premature
responding in the short (4 s) and long (60 s) conditions, which were exacerbated by morphine.
The fact that MOR activation increases impulsive action is in line with our previous findings
showing that MOR knockout mice are less impulsive (Olmstead et al. 2009), and other
pharmacological studies showing that MOR agonists increase premature responding (Pattij et al.
70
2009; Wiskerke et al. 2011a). We have previously shown that SNC80, a selective delta opioid
receptor agonist, increases impulsive action; paradoxically, delta knockout mice also show
increased premature responding. The fact that these two manipulations produce similar
behavioral effects may reflect an interaction with emotional processing, as DORs have a
prominent role in this function (Filliol et al. 2000). Overall, these findings show a role for the
opioid system in behaviors that impact impulsive action. This knowledge may prove to be
fruitful in the treatment of impulse control disorders.
71
Lack of an Impulsivity Phenotype in Constitutive or Conditional Mu or Delta
Opioid Receptor Knockout Mice
4.1 Abstract
Impulsive action, the inability to withhold behavioural responding, is mediated through many
neurotransmitter systems. We previously identified a role for opioid receptors by showing that
constitutive mu and delta opioid receptor (M/DOR) gene knockout mice show decreased and
increased impulsive action, respectively. The current study extended this investigation by
examining whether conditional knockout mice lacking M/DORs in GABAergic forebrain
neurons exhibit changes in impulsive action. Male and female mice were trained in a response
inhibition task to withhold responding for sucrose during a Fixed (4 s) or Variable (1-8 s) delay;
impulsive action was assessed as increased responding during the delay. Neither constitutive nor
conditional M/DOR mice expressed an impulsivity endophenotype. The findings highlight the
importance of genetic background and genetic construction on an impulsivity phenotype.
4.2 Introduction
Impulsivity is a multi-faceted construct characterized by the inability to inhibit
responding, acting without considering consequences, and preference for immediate over
delayed rewards. High impulsivity is associated with several psychiatric disorders, most notably
drug addiction, attention deficit hyperactivity disorder, problem gambling, and eating disorders
such as anorexia and bulimia nervosa (American Psychiatric Association 2013). As such,
improving our understanding of the mechanisms of impulsivity has potential clinical relevance.
72
Several neurotransmitter systems regulate impulsivity, with the monoamine system,
namely dopamine, serotonin, noradrenalin, being best documented (Chamberlain and Robbins
2013; Bari and Robbins 2013; Jentsch and Pennington 2014; Jenstch et al. 2014). More recently,
other systems including GABA and opioids (Jupp et al. 2013a; Hayes et al. 2014) have also been
linked to impulsivity. For example, we and others showed that pharmacological activation of
both mu and delta opioid receptors (M/DORs) increases impulsive action, defined as the inability
to withhold responding (Pattij et al. 2009; Befort et al. 2011; Wiskerke et al. 2011a; Mahoney et
al. 2013). Furthermore, constitutive (i.e., full) knockout (KO) mice lacking MORs are less
impulsive, whereas mice lacking delta opioid receptors (DORs) are more impulsive (Olmstead et
al. 2009). Together, these data suggest that M/DORs are involved in impulsive action, but it is
not clear which brain network mediates this response. Recent converging evidence suggests that
increased impulsivity may be due to impaired interactions within the cortical-striatalhippocampal network (Basar et al. 2010; Chudasama et al. 2012; Jupp et al. 2013b). Therefore,
M/DOR activity in this network may alter impulsive action. In line with this idea, MOR
activation in the NAcbS increases impulsive responding (Wiskerke et al. 2011a), lending support
to striatal opioid involvement in impulsive action. To date, few studies have examined whether
opioid activity in the striatum or other brain regions contributes to impulsive action.
To elucidate the role of M/DORs within the forebrain on impulsive responding, we
assessed this behaviour in mice conditionally lacking either MORs or DORs in GABAergic
forebrain neurons (Charbogne et al. unpublished data; Chu Sin Chung et al. 2014). M/DOR
expression and binding is greatly reduced in the striatum and hippocampus of these mice. Based
on this pattern of M/DOR expression and previous findings in constitutive KO animals
(Olmstead et al 2009), we hypothesized that M/DOR cKOs would show improved and impaired
73
impulsive action compared to flox controls, respectively. To test these hypotheses, M/DOR cKO,
M/DOR KO and flox control mu and delta mice were tested in a modified version of a response
inhibition task (Olmstead et al. 2009; Hayton et al. 2010).
4.3 Materials and methods
Subjects
Delta and mu conditional knockout mice, as well as constitutive KO mice and their flox
controls, were bred in-house and genotyped using a piece of tail or digit at 2 weeks of age. Both
delta flox controls (Oprd1fl/fl) and cKO (Dlx5/6-Oprd1-/-) mice were C57Bl6/J 66%, 129SVPas
18%, and unknown14%; both mu flox controls (Oprm1fl/fl) and cKO (Dlx5/6-Oprm1fl/fl) mice
were Bl/6J 53 %,129/SvPas 44% and unknown 3%. Both delta KO (CMV-Oprd1-/-) and mu KO
(CMV-Oprm1-/-) mice were C57Bl6/J 75% and 129/SVPas 25%. All mice were bred for more
than four generations. A detailed description of the construction of these lines and their
genotyping is provided elsewhere (Charbogne et al. unpublished data; Chu Sin Chung et al.2014;
Gaveriaux-Ruff et al., 2011; Monory et al., 2006). All animals were housed in a temperature
(21± 2°C) and humidity (45±5%) controlled environment on 12 h dark-light cycle (lights on at
07:00). Food was available ad libitum throughout the experiment; access to water was restricted
to 2 h per day beginning 2–4 h after testing. Behavioral testing commenced at 9-23 (Delta 4 s
Task), 9-22 (Delta VT 1-8 s task) and 10-17 (Mu 4 s Task) weeks of age and finished at 23-41
(Delta 4 s task), 24-42 (Delta VT 1-8 s Task) and 25-32 (Mu 4 s Task) weeks of age. All
experiments and animal care were conducted in accordance with the European Communities
Council directive of 24 November 1986 (86/609/EEC) and the experiments were approved by
the Comité régional d’éthique en matière d’experimentation animale de Strasbourg, CREMEAS.
74
Apparatus
Mice were trained in the mouse version of the response inhibition task as described
previously (Olmstead et al. 2009; Hayton et al. 2010; Befort et al. 2011; Mahoney et al. 2013)
with minor modifications (Figure 4.1). One day prior to magazine training, and following 21 h of
water deprivation, a sucrose solution (25% w/v) was presented in the home cage for one hour.
Beginning the next day, mice underwent daily operant training sessions. Training started with
magazine training (Phase 1) during which liquid sucrose presentation (dipper elevation) occurred
at random intervals of 15-45 s. During sucrose presentation, the house light turned off, the
magazine light was illuminated for 5 s, and the mouse had 20 s to enter the magazine. The liquid
dipper remained elevated for 5 s following magazine entry; if the mouse did not enter the
magazine, the dipper retracted from the magazine, and a new trial commenced. The program
ended after 30 sucrose presentations. Magazine entries were recorded and used as an indicator
that mice had learned the location of the sucrose reward. Magazine training continued until mice
received a minimum of 20 sucrose presentations in one session.
Mice were then trained to nose poke for sucrose on a fixed-ratio (FR1) schedule of
reinforcement (Phase 2); nose poke hole (left or right) was counterbalanced across mice. The
discriminative stimulus (i.e., visual cue light in the nose poke hole) and house light were
illuminated throughout these sessions, except during sucrose presentation. Following a nose
poke, the magazine light illuminated for 5 s and the dipper remained elevated for 20 s. Following
magazine entry, the mice had 5 s to drink from the dipper. If the mouse did not respond during
the 20-s response window, the dipper retracted from the magazine, and a new trial commenced.
A session ended following either 40 min or 25 sucrose presentations. Training continued until
mice reached a criterion of 25 sucrose presentations in one session. Following FR1 training,
75
response requirements were increased to an FR3 schedule (Phase 3) and the response window
(dipper elevation) was limited to 5 s. A session ended following either 40 min or 25 sucrose
presentations. Training continued until mice reached a criterion of 25 sucrose presentations in
one session.
In the next stage, each trial commenced with a 10-s inter-trial interval (ITI) (Phase 4).
Following the ITI, the ‘response phase’ commenced with the house light and nose poke light
illuminating for 10 s. A nose poke response (FR1) resulted in magazine light illumination (5 s),
5-s dipper elevation, and the house light and nose poke lights turning off. Following magazine
entry, the mice had 5 s to drink from the dipper. If the mouse did not respond during the
premature or response phase, the dipper retracted from the magazine, and a new trial
commenced. A session ended following 51 entries into the ITI or 40 min. Training continued
until mice reached a criterion of 25 sucrose presentations for two consecutive sessions.
Mice then progressed to either the Fixed or Variable-delay nose poke task for 12 sessions
(Figure 4.1). The Fixed 4 s task was identical to Phase 4 training described above, with the
exception of the addition of a premature phase. Following a 10-s ITI, a 4 s ‘premature phase’
commenced. The Variable delay task was identical to Phase 4 training with the exception of a
lengthened ITI (20 s), and the addition of a variable time (VT) 1-8 s ‘premature phase’. In both
tasks, there were no discernable differences from the ITI, except that a nose poke during the
premature phase was counted as a premature response, and initiated the next trial. If the mouse
withheld responding during the premature phase, the 5-s response phase commenced, which had
the same outcomes as Phase 4 training. A session ended following 51 entries.
If mice did not reach criterion for training, they were not tested in the response inhibition
task (see Table 4.1 for a summary of excluded animals). Final sample sizes were: delta Fixed 4 s
76
task (nOprd1fl/fl = 28; n Dlx5/6-Oprd1-/- = 18; nCMV-Oprm1-/- = 9); delta VT 1-8 s task (nOprd1fl/fl = 14; n
Dlx5/6-Oprd1-/- =
Oprm1-/-
17; nCMV-Oprm1-/- = 6); mu Fixed 4 s task (nOprm1fl/fl = 10; n Dlx5/6-Oprm1-/-= 9; nCMV-
= 15).
Figure 4.1 Schematic of the progression of a representative trial in the mouse response inhibition
task. In the Fixed 4 s task, a 10-s ITI was followed by a 4 s premature phase. In the Variable
delay task, following a 20-s ITI, rats progressed to a variable-time (VT) 1-8 s premature phase.
In both tasks, responding during the premature phase was counted as a premature response and
initiated a new trial. If mice withheld responding during the premature phase, the correct phase
commenced, signalled by nose poke light and house light illumination. Responding during the
correct response phase initiated the reward phase, and a sucrose solution reward was presented
for 5 s. If mice did not respond during the premature or correct phase, a trial omission was
counted, and a new trial commenced; a reward omission occurred if the mice made a correct
response, but did not enter the magazine during reward presentation.
77
Table 4.1 Proportion of mice that reached criterion during operant training
Experimental Group
N reaching RI Task/
N Trained
Proportion Completing
Operant Training
Male
Female
Male
Female
Fixed 4 s Delta Mice
Oprd1fl/fl
Dlx5/6-Oprd1-/CMV-Oprd1-/-
15/17
10/11
2/14
13/14
8/9
7/10
0.88
0.90
0.14
0.93
0.89
0.70
VT 1-8 s Delta Mice
Oprd1fl/fl
Dlx5/6-Oprd1-/CMV-Oprd1-/-
8/10
11/11
2/6
6/8
6/7
4/9
0.80
1.0
0.33
0.75
0.86
0.44
Fixed 4 s Mu Mice
Oprm1fl/fl
Dlx5/6-Oprm1-/CMV-Oprm1-/-
5/6
6/8
8//8
5/5
3/5
7/8
0.83
0.75
1.0
1.0
0.6
0.88
Abbreviations: N, sample size; RI, Response inhibition; VT, variable-time. Proportion completed
was calculated by dividing the number of animals that completed operant training (per group) by
the total number of animals (per group) trained.
Statistical analyses
Efficiency was measured as the proportion of trials in which animals inhibited a nose
poke response prior to the discriminative stimulus presentation. This dependent measure was
calculated as (correct responses/(premature + correct + ITI responses)). The number of trials and
reward omissions, latencies to premature and correct responses, and non-reinforced nose pokes
were recorded for each session. Trial omissions occurred if the mouse did not respond during the
premature or response phase; reward omissions occurred if the mouse made a correct nose poke
response, but did not enter the magazine during sucrose presentation. Reward omissions were
analyzed as a percentage, calculated as reward omissions/total responses. Latencies were
calculated as the time from the start of the premature phase (premature response) or response
78
phase (correct response) to the first nose poke. Finally, responses on the non-reinforced nose
poke hole were recorded but had no programmed consequences.
Data were analyzed using mixed model analysis of variances (ANOVAs), computed with
the Statistical Package for the Social Sciences (SPSS; V.20.0). Training data (sessions to
criterion for each training phase) were analyzed using a between-subjects ANOVA with
genotype as the between subjects factor. Test data were analyzed using mixed model ANOVAs
with genotype as the between subjects factor, and testing session as the within subjects factor.
There was no effect of sex on any of the measures analyzed for either mu or delta mice, so this
factor was collapsed in all reported results. If assumptions of sphericity were violated, degrees of
freedom for repeated measures were adjusted using the Greenhouse-Geisser correction
(Greenhouse and Geisser 1959). Significant interactions were followed up with simple main
effects, which were further analyzed with post-hoc paired t-tests with an alpha correction for the
number of comparisons made.
4.4 Results
Training
Fixed 4 s Delta Mice
All data for training criteria were positively skewed, and were transformed using a square
root transformation. As shown in Figure 4.2, KO (CMV-Oprd1-/-) mice required a significantly
greater total number of sessions to reach training criterion compared to both flox (Oprd1fl/fl) and
cKO mice (Dlx5/6-Oprd1-/-) [F(2,54) = 6.67, p = 0.003, η2 = 0.20], which did not differ. More
specifically, genotype did not alter number of sessions to magazine training criterion [F(2,53) =
3.16, p = 0.051]; however during FR1 training, KO mice required significantly more sessions
than flox and cKO mice [F(2,50) = 8.43, p = 0.001, η2 = 0.26]. During FR3 training, KO mice
79
did not differ from flox, but did require significantly more of sessions to reach criterion
compared to cKO mice [F(2,58) = 3.62, p = 0.033, η2 = 0.12]. Finally, there was no genotype
difference for 10-s ITI training.
VT 1-8 s Delta Mice
Data for sessions to magazine, FR1, and FR3 training criterion were positively skewed,
so these data were transformed using a square root transformation. As shown in Figure 4.3,
overall, cKO (Dlx5/6-Oprd1-/-) mice required significantly fewer sessions to criterion than flox
(Oprd1fl/fl) mice; however KO (CMV-Oprd1-/-) did not differ from flox mice [F(2,25) = 4.04, p =
0.03, η2 = 0.26]. More specifically, there were no genotype differences in magazine training
sessions [F(2, 37) = 1.13, p = 0.34], but during FR1 training flox and cKO mice required
significantly fewer sessions than KO mice [F(2,43) = 7.80, p = 0.001, η2 = 0.28] to reach
training criterion. During FR3 training, cKO mice reached criterion in fewer sessions than flox
mice [F(2,33) = 4.62, p = 0.017, η2 = 0.23]. Finally, mice did not differ on criterion to 10-s ITI
training [F(2,29) =1.66, p = 0.21].
Fixed 4 s Mu Mice
Data for sessions to magazine training criterion were positively skewed, so these data
were transformed using a square root transformation. As shown in Figure 4.4, mice did not differ
on total sessions to training criterion [F(2,34) = 1.53, p = 0.24]. Although flox (Oprm1fl/fl) mice
required significantly more magazine training sessions than both cKO (Dlx5/6-Oprm1-/-) and KO
(CMV-Oprm1-/-) mice [F(2,52) = 6.99, p = 0.002, η2 = 0.22], there were no genotype difference
for the number of training sessions to reach criterion for FR1 [F(2,41) = 1.782, p = 0.18], FR3
[F(2,31) = 2.43, p = 0.11], or 10-s ITI [F(2,29) = 1.03, p = 0.37].
80
Figure 4.2 Mean (± SEM) number of sessions to criterion for each of the training phases for
delta Fixed 4 s flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and constitutive
knockout (CMV-Oprd1-/-) mice. Data represent square root transformations for magazine
training (Mag Train), FR1, and FR3 training. * p < 0.05, compared to Oprd1fl/fl; # p < 0.05,
compared to Dlx5/6-Oprd1-/-.
81
Figure 4.3 Mean (± SEM) number of sessions to criterion for each of the training phases for
delta VT 1-8 s flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and constitutive
knockout (CMV-Oprd1-/-) mice. Data represent square root transformations for magazine
training (Mag Train), FR1, and FR3 training. * p < 0.05, compared to Oprd1fl/fl; # p < 0.05,
compared to Dlx5/6-Oprd1-/-.
82
Figure 4.4 Mean (± SEM) number of sessions to criterion for each of the training phases for mu
4 s flox (Oprm1fl/fl), conditional knockout (Dlx5/6-Oprm1-/-), and constitutive knockout (CMVOprm1-/-) mice. Data represent square root transformations for magazine training (Mag Train),
FR1, and FR3 training. * p < 0.05, compared to Oprm1fl/fl.
83
Test
Fixed 4 s Delta Mice
Overall, efficiency increased for all mice across the 12 test sessions [F(11, 583) = 21.95,
p < 0.001, partial η2 = 0.29] (Figure 4.5). Although there was no significant main effect of
genotype [F(2, 53) = 1.99, p = 0.15], there was a significant genotype x session interaction
[F(22, 583) = 2.21, p = 0.005, partial η2 = 0.07] on efficiency. Follow-up simple main effects
showed that cKO (Dlx5/6-Oprd1-/-) mice had significantly higher efficiency than flox mice for
sessions 1-3. Results for all other behavioural measures are presented in Table 4.2, and depicted
in Figures 4.6-4.8.
VT 1-8 s Delta Mice
There was no main effect of genotype [F(2, 34) = 1.43, p = 0.25], and no interaction
between genotype and session [F(22,374) = 0.54, p = 0.90] on efficiency across the 12 test
sessions (Figure 4.9). Furthermore, there was no main effect of session on this dependent
measure [F(11, 374) = 1.15, p = 0.32]. Results for all other behavioural measures are presented
in Table 4.3, and depicted in Figures 4.10-4.12.
Fixed 4 s Mu Mice
Overall, efficiency increased for all mice across the 12 test sessions [F(11, 341) = 7.47, p
< 0.001, partial η2 = 0.19]; however, there was no significant main effect of genotype [F(2, 31) =
0.15, p = 0.86, and no genotype x session interaction [F(22, 341) = 0.88, p = 0.62] (Figure 4.13).
Results for all other behavioural measures are summarized in Table 4.4 and Figures 4.14-4.16.
84
Figure 4.5 Efficiency in delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and
constitutive knockout (CMV-Oprd1-/-) mice in a Fixed 4 s premature delay response inhibition
(RI) task across 12 testing sessions. Efficiency was calculated as total correct nose pokes/(correct
+ premature + ITI nose pokes). Data represent mean (± SEM). * p < 0.05, Dlx5/6-Oprd1-/compared to Oprd1fl/fl.
85
Table 4.2 Summary of ANOVA results for measures of the Fixed 4 s Task in Delta Mice
Behavioural Measure
Inter-trial Interval Nose Pokes
Genotype
Session
Genotype x Session
Premature Response Nose Pokes
Genotype
Session
Genotype x Session
Correct Response Nose Pokes
Genotype
Session
Genotype x Session
Rewards
Genotype
Session
Genotype x Session
Reward Omissions (%)
Genotype
Session
Genotype x Session
Trial omissions
Genotype
Session
Genotype x Session
Premature Response Latency
Genotype
Session
Genotype x Session
Correct Response Latency
Genotype
Session
Genotype x Session
Non-reinforced Nose Pokes
Genotype
Session
Genotype x Session
* denotes a significant effect, p < 0.05
df
F
p
Partial η2
2, 53
11, 583
22, 583
0.39
12.45
1.89
0.68
< 0.001*
0.03*
0.19
0.066
2, 53
11, 583
22, 583
0.66
6.02
1.54
0.52
< 0.001*
0.056
0.10
-
2, 53
11, 583
22, 583
6.21
12.71
1.34
0.004*
< 0.001*
0.19
0.19
0.19
-
2, 53
11, 583
22, 583
6.50
10.56
1.24
0.003*
0.001*
0.21
0.20
0.17
-
2, 53
11, 583
22, 583
7.77
2.11
1.74
0.001*
0.035*
0.039*
0.23
0.038
0.061
2, 53
11, 583
22, 583
6.22
5.42
0.99
0.004*
< 0.001*
0.47
0.19
0.093
-
2, 53
11, 583
22, 583
0.078
1.18
1.73
0.93
0.31
0.043
-
2, 53
11, 583
22, 583
0.30
0.73
3.37
0.74
0.61
< 0.001*
0.11
2, 53
11, 583
22, 583
3.34
0.50
0.614
0.043*
0.91
0.92
0.11
-
86
Figure 4.6 Mean number (± SEM) of inter-trial interval (ITI) (a), premature (b), and correct
response nose pokes (c) for delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and
constitutive knockout (CMV-Oprd1-/-) mice in a Fixed 4 s premature delay response inhibition
(RI) task across 12 testing sessions. * p < 0.05, compared to Oprd1fl/fl.
87
Figure 4.7 Mean number (± SEM) of reward presentations (a), percent reward omissions (b),
and trial omissions (c) for delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and
constitutive knockout (CMV-Oprd1-/-) mice in a Fixed 4 s premature delay response inhibition
(RI) task across 12 testing sessions. * p < 0.05, compared to Oprd1fl/fl; # p < 0.05, compared to
Dlx5/6-Oprd1-/-.
88
Figure 4.8 Mean number (± SEM) of premature response latency (s) (a), correct response
latency (s) (b), and non-reinforced nose pokes (c) for delta flox (Oprd1fl/fl), conditional knockout
(Dlx5/6-Oprd1-/-), and constitutive knockout (CMV-Oprd1-/-) mice in a Fixed 4 s premature
delay response inhibition (RI) task across 12 testing sessions. * p < 0.05, compared to Oprd1fl/fl;
# p < 0.05, compared to Dlx5/6-Oprd1-/-; + p < 0.05 compared to CMV-Oprd1-/-.
89
Figure 4.9 Efficiency in delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and
constitutive knockout (CMV-Oprd1-/-) mice in a VT 1-8 s premature delay response inhibition
(RI) task across 12 testing sessions. Efficiency was calculated as total correct nose pokes/(correct
+ premature + ITI nose pokes). Data represent mean (± SEM). * p < 0.05, Dlx5/6-Oprd1-/compared to Oprd1fl/fl.
90
Table 4.3 Summary of ANOVA results for measures of the VT 1-8 s Task in Delta Mice
Behavioural Measure
Inter-trial Interval Nose Pokes
Genotype
Session
Genotype x Session
Premature Response Nose Pokes
Genotype
Session
Genotype x Session
Correct Response Nose Pokes
Genotype
Session
Genotype x Session
Rewards
Genotype
Session
Genotype x Session
Reward Omissions (%)
Genotype
Session
Genotype x Session
Trial Omissions
Genotype
Session
Genotype x Session
Premature Response Latency
Genotype
Session
Genotype x Session
Correct Response Latency
Genotype
Session
Genotype x Session
Non-reinforced Nose Pokes
Genotype
Session
Genotype x Session
* denotes a significant effect, p < 0.05
df
F
p
Partial η2
2, 32
11, 352
22, 352
0.03
1.01
0.85
0.97
0.44
0.61
-
2, 34
11, 374
22, 374
0.54
0.54
0.80
0.59
0.88
0.73
-
2, 34
11, 374
22, 374
6.08
1.30
1.10
0.006*
0.25
0.36
0.26
-
2, 34
11, 374
22, 374
6.42
1.29
0.90
0.004*
0.26
0.56
0.27
2, 33
11, 363
22, 363
3.39
1.17
1.00
0.046*
0.32
0.46
0.17
-
2, 34
11, 374
22, 374
4.62
1.11
1.13
0.017*
0.36
0.33
0.21
-
2, 34
11, 374
22, 374
0.015
0.80
0.65
0.99
0.64
0.72
-
2, 34
11, 374
22, 374
0.12
0.79
0.99
0.89
0.61
0.46
-
2, 31
11, 341
22, 341
0.08
0.67
1.46
0.92
0.69
0.14
-
91
Figure 4.10 Mean number (± SEM) of inter-trial interval (ITI) (a), premature (b) and correct
response nose pokes (c) for delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and
constitutive knockout (CMV-Oprd1-/-) mice in a VT 1-8 s premature delay response inhibition
(RI) task across 12 testing sessions. * p < 0.05, compared to Oprd1fl/fl; + p < 0.05, compared to
CMV-Oprd1-/-.
92
Figure 4.11 Mean number (± SEM) of reward presentations (a), percent reward omissions (b),
and trial omissions (c) for delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6-Oprd1-/-), and
constitutive knockout (CMV-Oprd1-/-) mice in a VT 1-8 s premature delay response inhibition
(RI) task across 12 testing sessions. * p < 0.05, compared to Oprd1fl/fl; + p < 0.05, compared to
CMV-Oprd1-/-.
93
Figure 4.12 Mean (± SEM) premature response latency (s) (a), correct response latency (s) (b),
and non-reinforced nose pokes (c) for delta flox (Oprd1fl/fl), conditional knockout (Dlx5/6Oprd1-/-), and constitutive knockout (CMV-Oprd1-/-) mice in a VT 1-8 s premature delay
response inhibition (RI) task across 12 testing sessions.
94
Figure 4.13 Efficiency in mu flox (Oprm1fl/fl), conditional knockout (Dlx5/6-Oprm1-/-), and
constitutive knockout (CMV-Oprm1-/-) mice in a Fixed 4 s premature delay response inhibition
(RI) task across 12 testing sessions. Efficiency was calculated as total correct nose pokes/(correct
+ premature + ITI nose pokes). Data represent mean (± SEM).
95
Table 4.4 Summary of ANOVA results for measures of the Fixed 4 s Task in Mu Mice
Behavioural Measure
Inter-trial Interval Nose Pokes
Genotype
Session
Genotype x Session
Premature Response Nose Pokes
Genotype
Session
Genotype x Session
Correct Response Nose Pokes
Genotype
Session
Genotype x Session
Rewards
Genotype
Session
Genotype x Session
Reward Omissions (%)
Genotype
Session
Genotype x Session
Trial omissions
Genotype
Session
Genotype x Session
Premature Response Latency
Genotype
Session
Genotype x Session
Correct Response Latency
Genotype
Session
Genotype x Session
Non-reinforced Nose Pokes
Genotype
Session
Genotype x Session
* denotes a significant effect, p < 0.05
df
F
p
Partial η2
2, 31
11, 341
22, 341
1.60
7.47
1.10
0.22
< 0.001*
0.36
0.19
-
2, 31
11, 341
22, 341
2.37
5.35
1.15
0.11
< 0.001*
0.32
0.15
-
2, 31
11, 341
22, 341
12.93
1.71
1.46
< 0.001*
0.12
0.14
0.46
-
2, 31
11, 341
22, 341
8.51
1.41
1.52
0.001*
0.21
0.12
0.35
-
2, 30
11, 330
22, 330
0.77
0.56
0.84
0.47
0.73
0.60
-
2, 30
11, 330
22, 330
12.29
1.00
1.42
<0.001*
0.42
0.17
0.45
-
2, 31
11, 341
22, 341
0.31
1.20
1.81
0.74
0.31
0.045*
0.11
2, 31
11, 341
22, 341
5.55
1.03
0.61
0.009*
0.41
0.85
0.26
-
2, 31
11, 341
22, 341
0.59
0.90
1.47
0.56
0.50
0.15
-
96
Figure 4.14 Mean number (± SEM) of inter-trial interval (ITI) (a), premature (b), and correct
response nose pokes (c) for mu flox (Oprm1fl/fl), conditional knockout (Dlx5/6-Oprm1-/-), and
constitutive knockout (CMV-Oprm1-/-) mice in a Fixed 4 s premature delay response inhibition
(RI) task across 12 testing sessions. * p < 0.05, compared to Oprm1fl/fl.
97
Figure 4.15 Mean number (± SEM) of reward presentations (a), percent reward omissions (b),
and trial omissions (c) for mu flox (Oprm1fl/fl), conditional knockout (Dlx5/6-Oprm1-/-), and
constitutive knockout (CMV-Oprm1-/-) mice in a Fixed 4 s premature delay response inhibition
(RI) task across 12 testing sessions. * p < 0.05, compared to Oprm1fl/fl.
98
Figure 4.16 Mean (± SEM) premature response latency (s) (a), correct response latency (s) (b),
and non-reinforced nose pokes (c) for mu flox (Oprm1fl/fl), conditional knockout (Dlx5/6-Oprm1
-/), and constitutive knockout (CMV-Oprm1-/-) mice in a Fixed 4 s premature delay response
inhibition (RI) task across 12 testing sessions.
99
4.5 Discussion
Overall, there were no genotype differences in either constitutive DOR or MOR KO mice
in impulsive action in either a Fixed or Variable delay impulsive action task. Male, and to a
lesser extent, female DOR KO mice (CMV-Oprd1-/-) required more sessions to reach criterion
compared to flox mice, and fewer reached training criterion at all. In contrast, MOR mice did not
differ from flox mice in terms of number of sessions to training criterion or dropout rate,
suggesting no impairments in associative learning across the mu genotypes.
These data were contrary to our hypothesis, as we previously showed that DOR KO mice
show increased impulsivity, whereas MOR KO mice show decreased impulsivity (Olmstead et
al. 2009); neither group exhibited learning impairments in this test. These contrasting results
across studies may be attributed to mouse strain or genetic background. There are several
possibilities for how strain and genetic background could account for phenotypic differences
across studies. Basal differences in behaviour across different strains of mice have been wellestablished for fear conditioning (Smith et al. 2007) and spatial memory (Koopmans et al. 2003),
such that Bl6 mice show the most consistent learning overall, and 129SvPas strains and hybrids
with 129SvPas show inconsistent endophenotypes across tasks in terms of learning and
performance (Bothe et al. 2004; Smith et al. 2007). Basal strain differences fit with our previous
findings that wild-type controls with a Bl6-129SvPas hybrid background are less impulsive than
mice backcrossed onto a Bl6 background (Olmstead et al. 2009). Mice in the current study are
more similar to Bl6 background, which may explain the lack of constitutive KO or cKO
endophenotype on impulsive action in either MOR or DOR mice. In support of this idea, delta
CMV KO mice (as used in this study) do not show an anxiety endophenotype (Chu Sin Chung et
100
al. unpublished data), which was previously reported in 129SvPas hybrid mice (Filliol et al.,
2000).
Furthermore, it is possible that there is an interaction between strain and genetic
manipulation on impulsive responding. For example, nitric oxide synthase (nNOS) KO mice on a
129S4/SvJae-C57BL/6J background show increased aggressive behavior; however, after only
five generations of backcrossing to the non-aggressive C57BL/6J strain, nNOS mice no longer
show an aggressive endophenotype (Le Roy et al. 2000). These results illustrate an important
interplay between basal levels of aggressive behaviour and genetic modification.
It is further possible that simply backcrossing mice onto different mouse strains altered
an impulsive endophenotype. More specifically, backcrossing onto different mouse strains (as
opposed to the strain in which the genetic modification was originally produced) can result in cosegregation of the modified gene with other neighbouring genes (Crusio 2004). In other words, a
neighbouring gene’s normal expression is modified as a result of the modification to the gene of
interest. As a consequence of this co-segregation, endophenotypes may be falsely attributed to
the genetic modification, rather than the co-segregating genes (Crusio 2004). For instance, even
though the mice used in the present studies were backcrossed onto C57 strains, they still retain a
portion of the 129 genome (Bolivar et al. 2001; Cook et al. 2002; Bothe et al. 2004; Crusio
2004), which may alter behavioural responding.
Not only can strains express differing basal endophenotypes, but genetic background and
strain may also alter phenotypic penetrance or expressivity (Doetschman 2008). These terms
refer to the proportion and degree to which an endophenotype is expressed within a genotype,
respectively. Thus, the same genetic manipulation may produce different endophenotypes, which
is in part dependent on strain. The fact that penetrance and expressivity are often more variable
101
on mixed/hybrid backgrounds (Doetschman 2008) fits with possibility that genetic background
alters impulsive responding.
Genetic construction may also explain differences across studies. MOR (CMV-Oprm1-/-)
and DOR (CMV-Oprd1-/-) KO mice used in the current study were produced using a Cre-LoxP
system in which floxed mice were bred with CMV-Cre mice (Gaveriaux-Ruff et al. 2011;
Weibel et al. 2013). Specifically, exon 2 of the Oprd1 gene was floxed (and therefore excised) in
CMV-Oprd1-/- mice, whereas exon 2 and 3 of the Oprm1 gene were floxed in CMV-Oprm1-/mice. Conversely, M/DOR KO mice used previously were produced by replacing the initiation
codon (exon 1) of Oprd1 and Oprm1 with a neo resistance cassette, respectively (Matthes et al.
1996; Filliol et al. 2000; Olmstead et al. 2009). It is possible that there are functional
consequences to excising different exons; excising exon 2 and 3 may allow for partial, residual
M/DOR receptor functionality (e.g., partial ligand binding or downstream consequences) (Nagy
2000), which may be sufficient for regulating impulsive behaviour. If this is true, we would not
observe an impulsivity endophenotype in our animals. It should be noted that minimal or absent
mRNA expression and ligand binding is reported for all of these KO mice, suggesting lack of
receptor functionality (Matthes et al. 1996; Filliol et al. 2000; Gaveriaux-Ruff et al. 2011;
Weibel et al. 2013).
In addition to a lack of endophenotype in either M/DOR mice, we observed a high
dropout rate and increased sessions to criterion in the male delta KO mice. It may be that
acquisition and retention of reinforcement contingencies, and/or cue-related associations were
impaired in these mice, as DORs are involved in learning (Klenowski et al. 2014). Specifically,
KO DOR mice exhibit spatial learning and object recognition impairments (Le Merrer et al.
2013), and deficits in cue-related associations (Le Merrer et al. 2012). It should be noted,
102
however, that Olmstead et al 2009 did not report learning deficits in the delta KO mice, lending
further support to phenotypic differences (rather than learning impairments) as a result of strain
or genetic construction.
In conclusion, we did not observe an impulsivity endophenotype in either the M/DOR
constitutive or conditional KO mice; however, these findings do not necessarily imply that the
opioid system is not involved in impulsive action. It is possible that our lack of endophenotype
was due to strain or genetic construction effects, which needs to be clarified in order to elucidate
opioid involvement in impulsivity in these genetically constructed mice. Overall, the role of
opioidergic receptors in striatal and other brain regions brain on impulsivity has yet to be
resolved.
103
Acute Pharmacological Stress Increases Impulsive Action in a Rodent Model
5.1 Abstract
Rationale. Both impulsivity and stress precipitate, promote, and exacerbate many psychiatric
disorders, however how these two factors may interact is poorly understood. Objectives. We
examined the mechanism by which yohimbine, an α2 adrenoceptor antagonist and
pharmacological stressor, increases impulsive action (premature responding) in a response
inhibition task. Methods. Male Long-Evans rats were tested on impulsive action following
various doses of yohimbine (0 - 5.0 mg/kg). In subsequent experiments, yohimbine (2.5 mg/kg)induced impulsive responding was tested following pre-treatment with CRF1 antagonist,
antalarmin (0 - 20 mg/kg), glucocorticoid receptor antagonist, mifepristone (0 - 30 mg/kg), α1
adrenoceptor antagonist, prazosin (0 - 2 mg/kg), α2 adrenoceptor agonist, guanfacine (0 - 0.5
mg/kg), D1/5 receptor antagonist, SCH 39166 (0 - 0.0625 mg/kg), or mu opioid receptor
antagonist, naloxone (0 – 2 mg/kg). Results. Yohimbine dose dependently increased premature
responding; however none of the agents tested blocked this effect, nor did they affect any of the
other behavioural measures assessed. Conclusions: These findings suggest that yohimbineinduced impulsive action may be mediated by synergistic influence of multiple neurotransmitter
systems.
5.2 Introduction
Impulsivity is a multi-faceted construct that encompasses several unique characteristics
including a preference for immediate reward (impulsive choice), acting without thinking, and an
inability to withhold behavioural responses (impulsive action). A high impulsivity
104
endophenotype can precipitate, promote, and exacerbate psychiatric disorders, including
attention deficit hyperactivity disorder (ADHD) and substance abuse disorders (American
Psychiatric Association 2013). Stress is also a risk factor for many of the same psychiatric
disorders, and it may be that stress interacts with impulsivity to alter behaviors associated with
drug addiction. This could explain why adverse life events predict impulsivity in cocainedependent individuals (Hayaki et al. 2005), and self-reported impulsivity mediates the positive
relationship between stress and alcohol intake (Hamilton et al. 2013).
Although it is not clear how stress and impulsivity interact, it likely involves activation of
physiological stress responses. For instance, basal cortisol levels predict trait-like impulsivity,
both in individuals with borderline personality disorders and in healthy controls (Almeida et al.
2010). Similarly, the pharmacological stressor, yohimbine, increases impulsivity in humans, nonhuman primates, and rodents (Ma et al. 2003; Swann et al. 2005; Sun et al. 2010). Yohimbineinduced impulsivity may occur through one or more pharmacological systems. Yohimbine
elevates noradrenergic activity through presynaptic α2a noradrenergic autoreceptor activation
and, consequentially, increases arousal, cortisol levels, and anxiety in healthy participants and
rodents (Johnston and File 1989; Stine et al. 2002). Yohimbine’s behavioural effects on drug
taking behaviour are mediated by several systems, including stress-related hypothalamicpituitary-adrenal axis activation (e.g., through corticotropin releasing factor (CRF) and
glucocorticoid (GC) receptor activity), and an action on noradrenergic (NA) and dopaminergic
(DA) systems (Johnston and File 1989; Millan et al. 2000). CRF1 and GC receptor antagonism
block yohimbine-induced reinstatement of alcohol seeking (Marinelli et al. 2007; Simms et al.
2012). Similarly, antagonizing noradrenergic activity, through presynaptic α2-adrenoceptor
agonism and post-synaptic α1-adrenoceptor antagonism, blocks yohimbine-induced
105
reinstatement of drug (alcohol) seeking in rodents, and stress-induced craving in humans (Lê et
al. 2009; Lê et al. 2011; Fox et al. 2012). In addition, yohimbine’s direct action on DA D1/5
receptors mediates yohimbine-induced cocaine reinstatement (Brown et al. 2012). Finally,
although yohimbine does not directly act on the opioid system, opioid antagonism decreases
yohimbine-induced action (Nielsen et al. 2012), which may be reflective of increased HPA
activity.
The aim of the present work was to validate yohimbine-induced increases in impulsive
action that have been observed in the 5-choice serial reaction time task (Sun et al. 2010), and to
determine whether stress, adrenergic, dopaminergic, or opioidergic systems are involved in this
effect. Specifically, in a series of experiments, we assessed impulsive action in rats following
acute, systemic doses of yohimbine. In separate groups of animals, we tested whether antalarmin
(CRF1 antagonist), mifipristone (GC receptor antagonist), guanfacine (presynaptic α2
adrenoceptor agonist), prazosin (post-synaptic α1 adrenoceptors antagonist), SCH 39166 (D1/5
receptor antagonist), or naloxone (MOR antagonist) block yohimbine-induced increases in
impulsive action. The role of systemic antalarmin and mifepristone on impulsive action have yet
to be tested; conversely, guanfacine, but not prazosin, attenuates impulsive action in rodents
(Franowicz et al. 2002; Koskinen et al. 2003; Bari et al. 2009; Milstein et al. 2010; Fernando et
al. 2012; Roychowdhury et al. 2012; Abela and Chudasama 2014). Additionally, D1/5 antagonism
reduces impulsive action (van Gaalen et al. 2006; Pattij et al. 2007b). Since these systems are
implicated in impulsive action, it is possible that they mediate stress-induced impulsive action.
106
5.3 Materials and methods
Subjects
Male Long–Evans rats (N = 60 Charles River, PQ) weighing 200–250 g at arrival were
pair-housed on a reverse light–dark cycle with testing conducted during the dark cycle. Three
days prior to training, rats were food restricted to 120 min of free access (Lab Diet: PMI
Nutritional International, Inc) per day; water was freely available. Rats were monitored daily to
ensure body weight did not fall below 85% of their normal free-feeding weight. Animal care was
conducted in accordance with Canadian Council on Animal Care guidelines, and the experiments
were approved by the Queen’s University Animal Care Committee.
Apparatus
Training and testing were conducted in operant boxes (26.5×22.0×20.0 cm), each housed
in a sound-attenuating chamber (constructed in house). Each box was fitted with two retractable
levers and a house light. Sucrose pellets (45 mg; Bio-Serv, NJ, USA) were delivered to a food
magazine, located between the levers, via a pellet dispenser (Med Associates, VT, USA). Lights
were situated 4 cm above each lever and the magazine. An IBM-type computer controlled the
equipment and was used for data collection (software written in-house using ECBASIC or Med
Associates software; Med Associates, VT, USA).
Drug administration
Yohimbine hydrochloride (Tocris Biosciences, USA), an α2 adrenoceptor antagonist, was
tested at doses of 0, 0.625, 1.25, 2.5, and 5 mg/kg. In combined-drug experiments, doses of 0 and
2.5 mg/kg yohimbine were tested. Antalarmin hydrochloride (Cayman Chemical, USA), a CRF1
antagonist, was tested at doses of 0, 5, 10, 15, and 20 mg/kg. Mifepristone (RU-486; Sigma-
107
Aldrich, USA), a glucocorticoid receptor antagonist, was tested at doses of 0, 7.5, 15, and 30
mg/kg. Prazosin hydrochloride (Sigma-Aldrich, USA), an α1 adrenoceptor antagonist, was
tested at doses of 0, 0.5, 1, and 2 mg/kg. Guanfacine hydrochloride (Sigma-Aldrich, USA), an
α2 adrenoceptor agonist, was tested at doses of 0.125, 0.25, and 0.5 mg/kg. SCH 39166
hydrobromide (Tocris Biosciences, USA), a high affinity D1/5 receptor antagonist, was tested at
doses of 0.0156 and 0.0625 mg/kg. Naloxone hydrochloride (Tocris Biosciences, USA), a mu
opioid receptor antagonist, was tested at doses of 0.5, 1.0, and 2.0 mg/kg. Yohimbine was
dissolved in sterile water, prazosin, guanfacine, SCH 39166, and naloxone were dissolved in
0.9% saline, mifepristone was suspended in 100% propylene glycol, and antalarmin was
dissolved in 10% Tween 80 dissolved in saline. With the exception of the 2.0 mg/kg dose of
prazosin and the 30 mg/kg dose of mifepristone, which were administered in a volume of 2
ml/kg, all drugs were administered in a volume of 1 ml/kg. In all cases, yohimbine was injected
10 min before testing. In combination drug studies, antalarmin, prazosin, guanfacine, SCH
39166, and naloxone were injected 30 min prior to testing, mifepristone was injected 45 min
prior to testing. All injections were i.p.
Behavioural procedures
Training. Response inhibition training occurred over several phases. Rats were first
magazine trained for 1 day, receiving 20 sucrose pellets on a random time 90-s schedule. The
house light remained illuminated until a food pellet was dispensed, at which point the house light
turned off, the magazine light illuminated for 1 s, and the next trial commenced.
Rats were then trained to lever press for food on a continuous reinforcement (CRF)
schedule. Two levers were inserted into the chamber, with reinforced lever assignment
counterbalanced across animals (left versus right). The house light and discriminative stimulus
108
(lever light) were illuminated throughout these sessions, except during delivery of the reward,
which was signalled by the magazine light (1 s). There was no time limit for responding on each
trial; training continued until a minimum of 80 pellets were earned in a 60-min session for two
consecutive sessions. Following CRF training, the response phase (house light and
discriminative stimulus on) was shortened to 10 s following lever insertion. If 10 s elapsed, the
levers retracted for a 10-s inter-trial interval (ITI), and the trial was omitted. Training continued
until animals reached a criterion of fewer than 20% omitted trials for two consecutive sessions.
Over 90% of the animals reached criterion in 2 sessions, with the remaining completing this
stage of training in 4 sessions.
In the next stage, each trial commenced with a 10-s ITI in which both levers were
retracted and the house light was off. Following the ITI, the levers inserted and the house light
illuminated, but the discriminative stimulus remained off for a variable delay period (see below).
A lever press on the reinforced lever prior to the discriminative stimulus (i.e., premature
response) initiated the next trial, and no reinforcement was delivered. Responses during
discriminative stimulus (10-s window) presentation were reinforced with a sucrose pellet (correct
response) and initiated the next trial. Non-reinforced lever presses and omissions (failure to
respond) were recorded but had no programmed consequences. Rats received five training
sessions at each of the following delays: 1, 4, 15, and 60 s with 100 trials per session.
Rats then progressed to the full response inhibition (RI) task described previously
(Hayton et al. 2010; Befort et al. 2011; Mahoney et al. 2013). The full RI task was identical to
the previous stage of training, with the exception that each session consisted of 20 trials at delay
intervals of 1, 4, 15, 30, and 60 s (100 trials in total) that were presented in a random order.
109
Drug testing. Rats were tested for impulsive responding 14 sessions prior to drug
injections with the last 3 sessions constituting baseline. In one cohort, animals were tested
following varying doses of yohimbine (n = 12). A separate cohort of animals was tested
following pre-treatment of antalarmin (n = 16). From the antalarmin cohort, 8 animals were
tested following 0, 10, and 20 mg/kg doses (high dose group); the 10 mg/kg, but not the 20
mg/kg, dose reduced yohimbine-induced premature responding. To explain these findings, an
additional 8 animals were tested following pre-treatment with 0, 5, and 15 mg/kg antalarmin
(low dose group). One animal was removed from the mifepristone cohort (n = 15) because of
illness prior to drug administration. One group of animals (n = 16) was used to test prazosin,
guanfacine, and SCH 39166. Animals were first tested with either prazosin first (n = 8) or
guanfacine first (n = 8), and then tested on the other noradrenergic drug. Since neither drug
induced any behavioural effects on premature responding, baseline was re-established for 7
sessions, and then animals (n = 16) were tested following pretreatment with SCH 39166. Again,
since there were no effects with SCH 39166, baseline was re-established for 7 sessions, and then
animals (n = 16) were tested following pretreatment with naloxone. In all cases, rats received
doses according to a partial Latin square design and were given each dose once. Animals were
tested with drug-free test sessions on intervening days.
Data and statistical analyses
The dependent measures of the RI task were premature responses, total number of nonreinforced responses, omissions, and latencies to either premature or correct responses.
Premature responding was defined as the percentage of trials in which rats made a premature
response (lever press prior to presentation of the discriminative stimulus). This dependent
measure was calculated as (premature responses/(premature + correct responses)) × 100.
110
Latencies were calculated as the time from lever insertion (premature response) or discriminative
stimulus presentation (correct response) to the first lever press.
Data were analyzed using analysis of variance (ANOVA), computed with the Statistical
Program for the Social Sciences (SPSS; V.22.0). Degrees of freedom for repeated measures were
adjusted using the Greenhouse–Geisser correction if assumptions of sphericity were violated
(Greenhouse and Geisser 1959). When significant main effects were observed in betweensubjects analyses, post hoc tests were performed using a Bonferroni correction. Within-subjects
significant main effects were analyzed further with post hoc paired t-tests with an α correction
for the number of comparisons made. Significant interactions between repeated measures were
analyzed by simple main effects. In order to examine whether premature responding was stable
across testing and between different experimental cohort of rats, data from baseline sessions
were analyzed using a three-way ANOVA with session (last three sessions) and delay interval (1,
4, 15, 30, and 60 s) as within-subjects factors, and group (yohimbine alone and combination-drug
groups) as a between-subjects factor. Drug testing data and drug-free test sessions were analyzed
with dose as the within subjects factor for omission and non-reinforced variables (totaled across
delay) and, dose and delay as within-subjects factors for premature responding and latency
variables. For all experiments, premature latencies for the 1 and 4 s delay were not included in
analyses due to minimal responding at these delays.
5.4 Results
Baseline response inhibition training and drug-free test sessions
Premature responding was stable across the last three baseline sessions [F(2,140) = 1.76,
p = 0.18], and across experimental cohorts [F(8,140) = 1.24, p = 0.28]. Additionally, yohimbine
111
did not alter either total premature responding on drug-free washout test sessions [F(4,44) =
1.28, p = 0.29], or across delay [F(16,176) = 1.19, p = 0.33] (Appendix I); nor did it affect any
of the other variables measured (all ps > 0.05) (Appendix I and II). Finally, none of the
combination drug experiments altered premature responding or any of the other variables
measured on drug-free test sessions (ps > 0.05) (Appendix III-V).
Effect of yohimbine on impulsive action
Figure 5.1, yohimbine dose-dependently increased percent premature responding
[F(4,44) = 7.43, p = 0.001], with 1.25, 2.5, and 5.0 mg/kg doses all producing significant
increases (ps < 0.05). Additionally, all yohimbine doses significantly increased premature
responding at the 60-s delay. At the 30-s delay 1.25, 2.5 and 5 mg/kg increased premature
responding, and at the 15-s delay, 2.5 and 5.0 mg/kg increased premature responding (dose x
delay interaction: [F(16,176) = 3.65, p = 0.004]).
Figure 5.1 Effects of yohimbine on impulsive action in the response inhibition task. The data
represent percentage of premature responses at each delay (a), or combined across delays (b),
calculated as (premature responses/(premature + correct responses))*100. Data are expressed as
means ( SEM).  p < 0.05 versus vehicle (0 mg/kg) at the same delay; * p < 0.05 versus
vehicle.
112
Table 5.1 shows that yohimbine did not alter total number of omissions [F(4,44) = 0.38, p
= 0.72], non-reinforced responses [F(4,44) = 1.81, p = 0.18], correct response latency [F(4,44) =
1.48, p = 0.22], or premature response latency [F(4,32) = 0.68, p = 0.61], nor were there any
significant interactions of dose x delay for correct or premature response latencies (ps > 0.05).
Table 5.1 Effects of yohimbine on measures in the response inhibition task
Behavioural Measure
Dose (mg/kg)
Omissions
Non-reinforced
Lever Presses
Correct Response Latency
(s)
Vehicle
0.625
1.25
2.5
5.0
1.50 ± 0.98
0.83 ± 0.67
0.68 ± 0.31
0.68 ± 0.43
1.67 ± 1.14
203.72 ± 87.53
206.52 ± 82.05
282.42 ± 89.60
293.00 ± 96.68
161.75 ± 46.56
1.74 ± 0.20
1.65 ± 0.32
1.59 ± 0.13
1.68 ± 0.16
1.96 ± 0.22
Premature Response Latency (s)
Dose (mg/kg)
Vehicle
0.625
1.25
2.5
5.0
Delay (s)
15
30
60
6.00 ± 0.89
6.80 ± 0.75
6.10 ± 0.76
6.22 ± 0.56
5.92 ± 0.79
12.60 ± 1.71
14.17 ± 1.60
10.32 ± 0.66
12.48 ± 1.62
11.54 ± 1.42
21.30 ± 3.38
24.80 ± 2.95
22.90 ± 2.49
22.24 ± 2.94
17.52 ± 2.26
Values represent mean (SEM) total omissions, total non-reinforced responses, and correct
response latency collapsed across delay, and premature responses across delay. Premature delay
latencies for the 1 and 4 s delay were not analyzed due to minimal responding at these delays.
None of the yohimbine doses significantly differed from Vehicle (0 mg/kg).
113
Effect of antalarmin on yohimbine-induced impulsive action
As shown in Figure 5.2, antalarmin did not reliably block yohimbine-induced premature
responding at either the high (0 – 20 mg/kg) or low (0 – 15 mg/kg) dose range. For the high dose
range, the main effect of dose [F(3,21) = 6.55, p = 0.03] and delay [F(4,28) = 15.22, p < 0.001]
were significant, but not the dose x delay interaction [F(12,84) = 2.42, p = 0.08]. The significant
dose effect was due to increased premature responding following the vehicle-yohimbine dose. In
addition, the 10 mg/kg, but not the 20 mg/kg, dose significantly decreased premature responding
(Figure 5.2a,b). Because of these inconsistent dose effects, pre-treatment with 5 mg/kg and 15
mg/kg doses were tested in a second cohort of animals. For the low dose range, the main effect
of dose [F(3, 21) = 13.71, p < 0.001], delay [F(4, 28) = 48.67, p < 0.001], and the dose x delay
interaction [F(12, 84) = 4.77, p = 0.003] were all significant; however this was due to increased
premature responding following the vehicle-yohimbine dose at the 4, 15, 30, and 60-s delays
compared to the vehicle-vehicle dose (ps < .05). In contrast, neither the 5 or 15 mg/kg doses of
antalarmin blocked the yohimbine-induced increase in premature responding (ps > 0.05) (Figure
5.2c,d).
114
Figure 5.2 Effects of the corticotropin releasing factor receptor 1 antagonist, antalarmin (ANT),
on yohimbine (2.5 mg/kg)-induced impulsive action in the response inhibition task. All drugs
were administered i.p, and doses are listed as mg/kg. The data represent percentage of premature
responses at each delay (a,c), or combined across delays (b,d), calculated as (premature
responses/(premature + correct responses))*100. Data are expressed as means ( SEM).  p <
0.05 versus vehicle-vehicle (0-0 mg/kg) at the same delay; * p < 0.05 versus vehicle-vehicle. + p
< 0.05 versus vehicle-yohimbine.
As shown in Table 5.2 and Table 5.3, none of the antalarmin-yohimbine doses in the high
dose range (0-20 mg/kg) altered omissions [F(3,21) = 1.24 , p = 0.31], non-reinforced
responding [F(3,21) = 0.26, p = 0.79], correct response latency [F(3,21) = 2.49, p = 0.09], or
premature response latency [F(3,21) = 1.69, p = 0.23]. Neither premature nor correct response
latency variables had significant dose x delay interactions (ps > 0.05).
115
Similarly, as shown in Table 5.2 and Table 5.3, none of the antalarmin-yohimbine doses
in the low dose range (0-15 mg/kg) altered omissions [F(3,21) = 0.80, p = 0.46], non-reinforced
responding [F(3,21) = 0.45, p = 0.61], or correct response latency [F(3,21) = 0.53, p = 0.67],
although the vehicle-vehicle dose caused significantly longer premature response latencies at the
60-s delay compared to all other doses [F(6,42) = 5.16, p < 0.001]. The correct response latency
variable did not have a significant dose x delay interaction (ps > 0.05).
Effect of mifepristone on yohimbine-induced impulsive action
As shown in Figure 5.3, mifepristone did not block yohimbine-induced premature
responding. The main effect of dose [F(4, 56) = 8.60, p < 0.001], delay [F(4, 56) = 71.73, p <
0.001] and dose x delay interaction [F(16, 224) = 4.62, p < 0.001] were all significant; however
this was due to increased premature responding following the vehicle-yohimbine dose at the 4,
15, 30, and 60-s delays compared to the vehicle-vehicle dose (ps < 0.05). In contrast, none of the
other mifepristone-yohimbine doses differed from the vehicle-yohimbine dose (ps > 0.05).
As shown in Table 5.2 and Table 5.3, none of the mifepristone-yohimbine doses altered
omissions [F(4,52) = 0.15, p = 0.96], non-reinforced responding [F(4,52) = 1.43, p = 0.26], or
correct response latency [F(4,56) = 1.78, p = 0.15]. The main effect of dose on premature
response latency [F(4,56) = 3.02, p = 0.03] was due to decreased latencies following the vehicleyohimbine, 7.5 mg/kg mifepristone-yohimbine, and 15 mg/kg mifepristone-yohimbine doses (ps
< 0.05). Neither premature nor correct response latency variables had significant dose x delay
interactions (ps > 0.05).
116
Figure 5.3 Effects of the glucocorticoid receptor antagonist, mifepristone (MIF), on yohimbine
(2.5 mg/kg)-induced impulsive action in the response inhibition task. All drugs were
administered i.p, and are in mg/kg. The data represent percentage of premature responses at each
delay (a), or combined across delays (b), calculated as (premature responses/(premature +
correct responses))*100. Data are expressed as means ( SEM).  p < 0.05 versus vehiclevehicle (0-0 mg/kg) at the same delay; * p < 0.05 versus vehicle-vehicle.
Effect of prazosin on yohimbine-induced impulsive action
Figure 5.4 shows that prazosin did not block yohimbine-induced premature responding.
The main effect of dose [F(4,60) = 8.45, p < 0.001], delay [F(4,60) = 56.32, p < 0.001], and
dose x delay interaction [F(16,240) = 3.04, p = 0.007] were all significant. The vehicleyohimbine dose significantly increased premature responding compared to the vehicle-vehicle
dose at the 15, 30, and 60-s delays (ps < 0.05). In contrast, none of the other prazosin-yohimbine
combination doses significantly differed from the vehicle-yohimbine dose (ps > 0.05).
As shown in Table 5.2 and Table 5.3, none of the prazosin-yohimbine doses altered
omissions [F(4,60) = 0.94, p = 0.38], non-reinforced responding [F(4,56) = 1.08, p = 0.36], or
correct response latency [F(4,60) = 0.58, p = 0.63]. The significant main effect of dose on
premature response latency [F(4,60) = 7.42, p < 0.001] was the result of decreased latencies
117
following all combination drug doses compared to vehicle-vehicle (ps < 0.05). Neither premature
nor correct response latency variables had significant dose x delay interactions (ps > 0.05).
Figure 5.4 Effects of the α1 adrenoceptor antagonist, prazosin (PRA), on yohimbine (2.5
mg/kg)-induced impulsive action in the response inhibition task. All drugs were administered i.p,
and are in mg/kg. The data represent percentage of premature responses at each delay (a), or
combined across delays (b), calculated as (premature responses/(premature + correct
responses))*100. Data are expressed as means ( SEM).  p < 0.05 versus vehicle-vehicle (0-0
mg/kg) at the same delay; * p < 0.05 versus vehicle-vehicle.
Effect of guanfacine on yohimbine-induced impulsive action
Figure 5.5 shows that guanfacine did not block yohimbine-induced premature
responding. Although the dose x delay interaction was not significant [F(16, 56) = 1.35, p =
0.28], both the main effect of dose [F(4, 28) = 3.99, p < 0.011] and delay [F(4, 28) = 36.86, p <
0.001] were. The vehicle-yohimbine dose significantly increased premature responding
compared to the vehicle-vehicle dose (ps < 0.05). In contrast, none of the other guanfacineyohimbine combination doses significantly differed from the vehicle-yohimbine dose (ps >
0.05).
118
As shown in Table 5.2 and Table 5.3, none of the guanfacine-yohimbine doses altered
any of the variables measured: omissions [F(4, 28) = 1.03, p = 0.35], non-reinforced responding
[F(4,28) = 2.00, p = 0.18], correct response latency [F(4,28) = 0.48, p = 0.62], or premature
response latency [F(4,28) = 1.93, p = 0.16]. Neither premature nor correct response latency
variables had significant dose x delay interactions (ps > 0.05).
Figure 5.5 Effects of the α2 adrenergic agonist, guanfacine (GUA), on yohimbine (2.5 mg/kg)induced impulsive action in the response inhibition task. All drugs were administered i.p, and are
in mg/kg. The data represent percentage of premature responses at each delay (a), or combined
across delays (b), calculated as (premature responses/(premature + correct responses))*100. Data
are expressed as means ( SEM). * p < 0.05 versus vehicle-vehicle (0-0 mg/kg).
Effect of SCH 39166 on yohimbine-induced impulsive action
Figure 5.6 shows the effect of SCH 39166 on yohimbine-induced premature responding.
Although the dose x delay interaction was not significant [F(12, 180) = 1.77, p = 0.06], both the
main effect of dose [F(3, 45) = 9.56, p < 0.01] and delay [F(4, 60) = 35.46, p < 0.001] were. The
vehicle-yohimbine dose significantly increased premature responding compared to the vehiclevehicle dose at the 15, 30, and 60-s delays (ps < 0.05). In contrast, none of the other naloxone-
119
yohimbine combination doses significantly differed from the vehicle-yohimbine dose (ps >
0.05).
Table 5.2 and Table 5.3 show that SCH 39166-yohimbine doses did not alter any of the
variables measured: omissions [F(3, 45) = 1.56, p = 0.21], non-reinforced responding [F(3,39) =
2.71, p = 0.06], correct response latency [F(3,45) = 1.06, p = 0.34], or premature response
latency [F(3,45) = 0.049, p = 0.97]. Neither premature nor correct response latency variables had
significant dose x delay interactions (ps > 0.05).
Figure 5.6 Effects of the D1/5 receptor antagonist, SCH 39166 (SCH), on yohimbine (2.5
mg/kg)-induced impulsive action in the response inhibition task. All drugs were administered i.p,
and are in mg/kg. The data represent percentage of premature responses at each delay (a), or
combined across delays (b), calculated as (premature responses/(premature + correct
responses))*100. Data are expressed as means ( SEM). * p < 0.05 versus vehicle-vehicle (0
mg/kg).
Effect of naloxone on yohimbine-induced impulsive action
Figure 5.7 shows that naloxone did not block yohimbine-induced premature responding.
The main effect of dose [F(4, 56) = 10.81, p < 0.01], delay [F(4, 56) = 51.58, p < 0.001], and
dose x delay interaction [F(15, 224) = 3.17, p = 0.07] were all significant. The vehicle-
120
yohimbine dose significantly increased premature responding compared to the vehicle-vehicle
dose at the 4, 15, 30, and 60-s delays (ps < 0.05). In contrast, none of the other naloxoneyohimbine combination doses significantly differed from the vehicle-yohimbine dose (ps >
0.05). The vehicle-yohimbine dose significantly increased premature responding compared to the
vehicle-vehicle dose at the 15, 30, and 60-s delays (ps < 0.05). In contrast, none of the other
naloxone-yohimbine combination doses significantly differed from the vehicle-yohimbine dose
(ps > 0.05).
Table 5.2 and Table 5.3 show that naloxone-yohimbine doses did not alter any of the
variables measured: omissions [F(4,60) = 1.11, p = 0.33], non-reinforced responding [F(4,56) =
0.76, p = 0.43], premature response latency [F(4,60) = 2.67, p = 0.061] or correct response
latency [F(4,56) = 2.72, p = 0.11]. Neither premature nor correct response latency variables had
significant dose x delay interactions (ps > 0.05).
Figure 5.7 Effects of the mu opioid receptor antagonist, naloxone, on yohimbine (2.5 mg/kg)induced impulsive action in the response inhibition task. All drugs were administered i.p, and are
in mg/kg. The data represent percentage of premature responses at each delay (a), or combined
across delays (b), calculated as (premature responses/(premature + correct responses))*100. Data
are expressed as means ( SEM).  p < 0.05 versus vehicle-vehicle (0-0 mg/kg) at the same
delay; * p < 0.05 versus vehicle-vehicle.
121
Table 5.2 Effects of corticotropin releasing factor, glucocorticoid, adrenergic, and
dopaminergic agents on yohimbine-induced responding on measures of the response
inhibition task
Dose
(mg/kg)
Omissions
Non-reinforced
Lever Presses
Correct
Response
Latency (s)
Veh – Veh
Veh – Yoh
Ant 10 – Yoh
Ant 20 – Yoh
3.25 ± 2.06
0.75 ± 0.41
1.13 ± 0.52
1.63 ± 0.60
149.38 ± 70.67
183.13 ± 51.33
183.25 ± 65.47
159.00 ± 41.75
1.91 ± 0.26
1.50 ± 0.21
1.64 ± 0.21
1.82 ± 0.25
Veh – Veh
Veh – Yoh
Ant 5 – Yoh
Ant 15 – Yoh
1.75 ± 1.03
0.25 ± 0.25
1.63 ± 1.49
0.50 ± 0.27
66.13 ± 24.35
89.13 ± 46.56
97.75 ± 54.79
99.13 ± 48.95
1.43 ± 0.17
1.17 ± 0.10
1.42 ± 0.34
1.26 ± 0.18
Mifepristone
(GC ant)
Veh – Veh
Veh – Yoh
Mif 7.5 –Yoh
Mif 15 – Yoh
Mif 30 – Yoh
1.90 ± 1.30
1.14 ± 0.61
1.79 ± 1.03
1.21 ± 1.07
1.21 ± 0.37
40.24 ± 14.59
81.20 ± 26.35
66.20 ± 20.82
59.71 ± 15.69
95.80 ± 35.42
1.82 ± 0.19
1.65 ± 0.18
1.70 ± 0.17
1.73 ± 0.17
2.01 ± 0.18
Prazosin
(α1 adrenergic ant)
Veh – Veh
Veh – Yoh
Pra 0.5 – Yoh
Pra 1 – Yoh
Pra 2 – Yoh
3.75 ± 2.45
1.81 ± 0.77
0.81 ± 0.45
1.69 ± 0.83
1.44 ± 1.05
110.07 ± 33.52
149.00 ± 38.79
111.07 ± 29.09
113.40 ± 30.77
98.87 ± 27.39
1.82 ± 0.12
1.69 ± 0.13
1.85 ± 0.17
1.91 ± 0.17
1.70 ± 0.12
Guanfacine
(α2 adrenergic ago)
Veh – Veh
Veh – Yoh
Gua 0.125 – Yoh
Gua 0.25 – Yoh
Gua 0.5 – Yoh
0.63 ± 0.38
3.50 ± 2.94
2.63 ± 1.59
1.00 ± 0.63
0.75 ± 0.49
122.13
142.13
175.25
196.50
120.38
± 39.95
± 45.44
± 59.40
± 76.15
± 39.29
1.71 ± 0.25
1.91 ± 0.25
1.88 ± 0.17
1.65 ± 0.21
1.79 ± 0.24
SCH 39166
(D1/5 ant)
Veh – Veh
Veh – Yoh
SCH 0.015 – Yoh
SCH 0.0625 – Yoh
3.38 ± 1.34
1.31 ± 0.74
1.81 ± 0.67
1.44 ± 0.74
113.57 ± 27.53
118.43 ± 21.95
108.71 ± 26.90
85.43 ± 19.01
1.93 ± 0.16
1.54 ± 0.12
1.99 ± 0.35
1.80 ± 0.13
Naloxone
(MOR ant)
Veh – Veh
Veh – Yoh
Nal 0.5 – Yoh
Nal 1.0 – Yoh
Nal 2.0 – Yoh
2.38 ± 0.98
1.88 ± 0.95
1.44 ± 0.67
3.06 ± 1.56
0.56 ± 0.26
119.47 ± 41.57
151.00 ± 50.53
138.33 ± 38.11
85.53 ± 23.07
115.40 ± 30.74
1.87 ± 0.21
1.51 ± 0.13
1.33 ± 0.12
1.69 ± 0.21
1.28 ± 0.08
Drug
Antalarmin
(CRF1 ant)
Values represent mean (SEM) total omissions, total non-reinforced lever pressing, and correct
response latency collapsed across delay for combined-drug experiments. The yohimbine dose
122
was 0 or 2.5 mg/kg; other drug doses are in mg/kg. None of the drug combinations significantly
differed from their respective Veh-Veh (0 – 0 mg/kg) or Veh-Yoh (0 – 2.5 mg/kg) doses.
Ago, agonist; ant, antagonist; Ant, antalarmin; CRF1, corticotropin releasing factor receptor 1;
D1/5, dopamine 1/5 receptor family; GC, glucocorticoid receptor; Gua, guanfacine, Mif,
mifepristone; MOR, mu opioid receptor; Nal, naloxone; Pra, prazosin; SCH, SCH 39166
123
Table 5.3 Effects of corticotropin releasing factor, glucocorticoid, adrenergic, and
dopaminergic agents on yohimbine-induced responding on premature response latency in
the response inhibition task
Premature Delay (s)
Dose
(mg/kg)
15
30
60
Veh – Veh
Veh – Yoh
Ant 10 – Yoh
Ant 20 – Yoh
5.97 ± 0.90
6.67 ± 1.23
6.53 ± 0.60
5.44 ± 0.77
12.15 ± 2.32
11.46 ± 1.35
10.34 ± 1.66
12.52 ± 1.62
25.54 ± 6.12
19.41 ± 3.57
26.04 ± 4.86
19.56 ± 3.34
Veh – Veh
Veh – Yoh
Ant 5 – Yoh
Ant 15 – Yoh
7.15 ± 0.96
5.56 ± 0.84
5.19 ± 0.85
5.43 ± 0.95
11.95 ± 1.64
10.11 ± 1.50
7.94 ± 1.46
10.35 ± 1.47
29.21 ± 4.23*
15.89 ± 2.27*
15.06 ± 3.11*
13.06 ± 3.03*
Mifepristone
(GC ant)
Veh – Veh
Veh – Yoh
Mif 7.5 –Yoh
Mif 15 – Yoh
Mif 30 – Yoh
7.32 ± 1.00
5.18 ± 0.66
6.48 ± 0.64
5.84 ± 0.58
5.90 ± 0.53
11.79 ± 1.45
10.56 ± 0.87
10.04 ± 1.21
12.05 ± 1.59
8.67 ± 1.24
28.46 ± 1.53
19.74 ± 2.54
12.05 ± 1.59
19.67 ± 2.32
24.37 ± 2.83
Prazosin
(α1 adrenergic ant)
Veh – Veh
Veh – Yoh
Pra 0.5 – Yoh
Pra 1 – Yoh
Pra 2 – Yoh
8.18 ± 0.69
5.78 ± 0.74
5.90 ± 0.54
6.57 ± 0.56
7.15 ± 0.73
14.12 ± 1.27
11.50 ± 1.40
10.31 ± 1.25
11.23 ± 1.09
10.11 ± 1.35
25.48 ± 2.02
17.59 ± 1.91
16.72 ± 2.30
22.78 ± 2.59
21.01 ± 2.25
Guanfacine
(α2 adrenergic ago)
Veh – Veh
Veh – Yoh
Gua 0.125 – Yoh
Gua 0.25 – Yoh
Gua 0.5 – Yoh
8.20 ± 1.29
6.01 ± 0.66
7.41 ± 0.89
7.86 ± 1.40
8.94 ± 1.67
14.69 ± 1.86
10.58 ± 1.48
12.74 ± 1.85
13.44 ± 1.50
13.64 ± 1.24
21.96 ± 4.10
17.59 ± 3.29
19.04 ± 3.20
21.71 ± 3.09
24.17 ± 3.41
SCH 39166
(D1/5 ant)
Veh – Veh
Veh – Yoh
SCH 0.015 – Yoh
SCH 0.0625 – Yoh
6.43 ± 0.79
5.55 ± 0.54
7.45 ± 1.28
7.08 ± 0.70
11.08 ± 1.45
12.32 ± 1.45
12.77 ± 1.73
10.55 ± 1.39
20.32 ± 2.38
20.53 ± 2.98
18.67 ± 2.78
20.33 ± 3.05
Naloxone
(MOR ant)
Veh – Veh
Veh – Yoh
Nal 0.5 – Yoh
Nal 1.0 – Yoh
Nal 2.0 – Yoh
7.86 ± 0.65
5.70 ± 0.50
5.74 ± 0.35
4.59 ± 0.34
5.68 ± 0.51
10.35 ± 1.12
3.19 ± 2.51
19.40 ± 2.74
17.25 ± 1.91
16.93 ± 2.57
20.04 ± 2.49
Drug
Antalarmin
(CRF1 ant)
9.20 ± 0.96
10.68 ± 1.17
9.28 ± 0.85
9.11 ± 0.83
Values represent mean (SEM) premature response latencies for the 15, 30, and 60 s premature
delays for test sessions of combined-drug experiments. The 1 and 4 s delays were not analyzed
124
due to minimal responding at these delays. The yohimbine (Yoh) dose was 0 or 2.5 mg/kg; other
doses are in mg/kg. * p < 0.05 compared to Veh-Veh (0-0 mg/kg).
Ago, agonist; ant, antagonist; Ant, antalarmin; CRF1, corticotropin releasing factor receptor 1;
D1/5, dopamine 1/5 receptor family; GC, glucocorticoid receptor; Gua, guanfacine, Mif,
mifepristone; MOR, mu opioid receptor; Nal, naloxone; Pra, prazosin; SCH, SCH 39166
5.5 Discussion
We show that yohimbine dose-dependently increases premature responding on a response
inhibition task, while having no effect on any of the other variables measured. Our findings are
consistent with other studies, using both rodents and humans (Swann et al. 2005; Sun et al.
2010), and add to preclinical evidence that this effect is not task specific. Our results also fit with
data showing that impulsivity predicts both hazardous drinking and smoking in individuals who
report high, but not low, stress levels (Fields et al. 2009; Fox et al. 2010), and with broader
findings that stress impairs several aspects of cognitive functioning including attentional set
shifting (Liston et al. 2006; Liston et al. 2009; Butts et al. 2013), risky decision making (Buckert
et al. 2014), and perseverative responding (Schwager et al. 2014).
Whether yohimbine induces a stress-like state in our manipulation is unclear. Yohimbine
is used as a pharmacological stressor because of its anxiogenic effects in humans and animals
(Johnston and File 1989; Stine et al. 2002), and its physiological profile, which is similar to
footshock stress in animals (Funk et al. 2006). Thus, it is widely assumed that yohimbine’s
behavioural effects are stress-related; however they may be mediated directly via HPA activity,
indirectly on extra-hypothalamic noradrenergic activity, or by an action on other
neurotransmitter systems. The fact that CRF1 antagonism does not block yohimbine-induced
corticosterone secretion suggests extra-hypothalamic effects of yohimbine, at least on this
measure (Marinelli et al. 2007; Gozzi et al. 2013). Additionally, yohimbine-induced premature
125
responding in the 5-choice serial reaction time task (5-CSRTT) is mediated, in part, through
stress-related CREB-phosphorylation in the orbitofrontal cortex (Sun et al. 2010). Sun and
colleagues’ data suggest stress-related effects of yohimbine, rather than non-specific increases in
noradrenergic activity; however it is still unclear whether the noradrenergic system is involved at
all. Indeed, yohimbine’s actions on some drug-taking behaviours associated with impulsivity are
independent of noradrenergic activity (Davis et al. 2008; Lê et al. 2009), which may explain why
we did not observe a role for either noradrenergic agents in our study. This possibility is further
supported by the fact that we did not observe a yohimbine-induced block in impulsive action by
opioid antagonism, a system implicated in both noradrenergic stress responding and impulsive
action (Drolet et al. 2001; Pattij et al. 2009; Befort et al. 2011; Mahoney et al. 2013).
In fact, yohimbine-induced impulsive action was not blocked by any of the drugs we
tested. One likely explanation is that yohimbine’s actions on different systems are synergistic.
This possibility fits with the fact that impulsivity is mediated by several neurotransmitter
systems, which act to synergistically alter inhibitory control (Koskinen et al. 2003; Agnoli and
Carli 2011; Agnoli and Carli 2012; Jupp and Dalley 2014). Indeed, it may be that, in addition to
α2 noradrenergic receptor agonism, dopaminergic and/or serotonergic receptor activity drives
yohimbine action on our task (Millan et al. 2000). Yohimbine acts as an agonist at 5-HT1a
receptors (Millan et al. 2000); however systemic 5-HT1a receptor agonism has no effect on
premature responding, and intra-prefrontal agonism decreases impulsivity in the 5-CSRTT
(Winstanley et al. 2003). Therefore, it is more likely that synergistic effects would be due to
yohimbine’s activation of D1/5 receptors, which are involved in increased impulsivity (Millan et
al. 2000; Pattij et al. 2007b; Loos et al. 2010). We tested dose ranges that were previously shown
126
to block yohimbine-induced behaviour, so it is unlikely that we did not reach the effective dose
range.
It is possible that yohimbine-induced effects are driven by systems that we did not
examine, such as orexin and neuropeptide Y (NPY). Orexin neurons are involved in the acute
stress response (Johnson et al. 2010), and orexin1 receptor activity is involved in stress-induced
drug intake and reinstatement (Boutrel et al. 2005; Lawrence et al. 2006; Gilpin et al. 2008;
Goltz et al. 2011). In contrast, NPY, which opposes the CRF-mediated stress response and
reduces anxiety-like behaviour, attenuates stress-induced reinstatement (Cippitelli et al. 2010;
Gilpin 2012). Therefore these systems may be involved in mediating yohimbine-induced
impulsive action in opposing manors. This is highly speculative, and whether orexin or NPY
systems are involved in impulse control at all has yet to be determined, although NPY gene
polymorphisms have been implicated in ADHD (Lesch et al. 2011).
Additionally, the cognitive mechanisms involved in yohimbine-induced impulsivity still
require elucidation. It is unlikely that stress is simply altering the ability to time the length of the
premature delay since this phase is unpredictable in our task. Interestingly, it was shown that
yohimbine reinstates (increases) lever pressing both in animals that were previously trained with
lever pressing resulting in cue presentation + food pellet (reward) delivery, and surprisingly, in
animals that were previously trained with lever pressing resulting in cue presentation but in the
absence of reward (Chen et al. 2014). Thus, these authors argued that yohimbine increases
reactivity to all visual/auditory cues. Although speculative, it is possible that increased leverpressing behaviour reflects increased impulsive action, thus fitting with the possibility that, in
stress-induced reinstatement models, stress increases reinstatement by increasing impulsivity. In
both cases yohimbine may be increasing salience to cues, thus driving lever-pressing behaviour.
127
Acute social stress increases reward salience (Chaijale et al. 2014) so it is also possible that
yohimbine specifically increases impulsivity by increasing salience for reward-associated cues,
including the discriminative stimulus (lever light).
In conclusion, yohimbine’s mechanism of action is complex and behaviour-specific. Our
findings support the fact that stress can induce impulsive action, but suggests that none of the
neurotransmitter systems measured here act in isolation to mediate the effect of yohimbineinduced impulsive action.
128
General Discussion
6.1 Summary of findings
Opioid receptor involvement in impulsive action
Taken together, our results show a role for opioid and stress systems in impulsive action.
More specifically, pharmacological activation of delta opioid receptors (DORs) increased
impulsivity in rats in a Variable delay response inhibition (RI) task (Chapter 2) (Befort et al.
2011; Mahoney et al. 2013). Additionally, while morphine (MOR agonist) did not alter
impulsivity in a Variable delay RI task (Chapter 2) (Befort et al. 2011), it increased impulsivity
in Fixed delay versions of the same task (Chapter 3) (Mahoney et al. 2013). Finally, neither
constitutive M/DOR KO mice, nor mice lacking M/DORs in GABAergic forebrain neurons,
expressed an impulsivity phenotype (Chapter 4).
The fact that pharmacological activation of DORs increased premature responding was
surprising, given that DOR KO mice are also more impulsive (Olmstead et al. 2009; Befort et al.
2011). We argued that these seemingly contradictory findings could be attributed to differences
in task parameters across experiments. Specifically, optimal responding in the rat and mouse RI
tasks may require unique circuitry that are differentially altered by DORs. In support of this,
MOR activation uniquely altered premature responding across tasks (Fixed versus Variable
delay). Moreover, these animals exhibited distinct probabilities of responding prematurely that
was dependent on the premature delay length (Fixed 4- versus 60-s tasks). The probability of
premature responding increased across the premature delay interval in the Fixed 4-s task, and
remained stable across the delay in the Fixed 60-s task. Morphine did not alter these overall
129
patterns of responding, but exacerbated the degree of change over the delay intervals. Thus,
MORs may activate distinct systems when the premature delay is short versus long. Rats and
mice were tested on different delay lengths (1-60 s in rats; and 1-8 s in mice), which could
explain the different pattern of responding in these two experiments. Either way, these data
highlight the importance of task parameters on behavioural responding.
As discussed in Chapter 2, pharmacological and genetic manipulations often yield
different behavioural outcomes. For instance, MOR KO mice are less anxious and show more
depressive-like behaviour (Filliol et al. 2000), yet pharmacological activation of MORs reduces
depressive-like behaviour in rodents (Tejedor-Real et al. 1995; Besson et al. 1996; Filliol et al.
2000; Yang et al. 2011; Berrocoso et al. 2013; Lutz and Kieffer 2013). Discordant findings
across pharmacological and genetic manipulations can likely be attributed to lack of receptor
binding specificity and changes to other neurotransmitter systems (e.g., sensitization),
respectively. In support of this, Oprd1 gene (i.e., coding for DOR) deletion modifies genes
coding for GABA, glutamate, and dopaminergic transporters, receptors and receptor subunits,
and Oprd1 KO mice have potentiated dopaminergic and glutamatergic activity in striatal
circuitry (Le Merrer et al. 2013). In other words, increased impulsive action observed in DOR
KO mice (Olmstead et al. 2009) may be a direct result of altered dopaminergic or glutamatergic
gene expression, as opposed to indirect effects of these systems due to lack of DORs.
We reported that neither constitutive M/DOR KO mice (CMV mice) exhibited any
impulsivity phenotype, unlike previously reported data showing that MOR and DOR KO mice
show decreased and increased impulsivity, respectively (Olmstead et al. 2009). In Chapter 4, we
discuss how mouse strain, genetic background, their interaction, and genetic construction may
unexpectedly alter behavioural phenotypes. Taken together, these findings emphasize the
130
consideration of pharmacological and genetic manipulations when interpreting complex
behaviours, such as impulsivity.
Stress involvement in impulsive action
We observed that acute pharmacological stress increases impulsive action (Chapter 5),
similar to previous findings (Sun et al. 2010). Interestingly, basal levels of impulsive action
(premature responding) and anxiety-like behaviour (time on the open arms) do not correlate with
each other (Molander et al. 2011), nor do basal impulsivity and novelty-induced corticosterone
secretion (Molander et al. 2011). It may be that heightened levels of anxiety or stress, rather than
differences in basal levels of either, mediate impulsivity. This possibility is supported by
findings that impulsivity predicts both hazardous drinking and smoking in individuals who report
high, but not low, stress levels (Fields and Collins 2009; Fox et al. 2010). Additionally, both high
anxiety and high impulsive action phenotypes predict escalated drug taking (Belin et al. 2008;
Dilleen et al. 2012), suggesting that perhaps extreme levels, but not basal or low levels, are risk
factors for drug-related behaviour.
In Chapter 5 we discussed the possibility that reinstated lever pressing for drug following
stress may reflect increased impulsivity. Recent findings show that yohimbine-induced lever
pressing following extinction may not reflect reinstatement for the previously reinforced drug or
food reward (Chen et al. 2014). These findings oppose the widely accepted assumption that
increased lever pressing following extinction reflects drug (or food) seeking, and raises the
possibility that impulsivity is mediating stress-induced lever pressing in reinstatement
paradigms. Indirect support for this comes from the fact that drugs used to treat impulse control
disorders (e.g., guanfacine) reduce stress-induced drug craving and relapse in people with
substance abuse disorder (Fox et al 2012). In contrast, animals categorized as high and low on
131
impulsive choice do not differ in yohimbine-induced reinstated cocaine seeking (Regier et al.
2014); whether impulsive action mediates yohimbine-induced reinstatement remains to be
determined.
6.2 Neuropharmacological mechanisms
Opioid-induced impulsive action
The fact that DOR, but not MOR, activation alters impulsive responding when the delay
is variable suggests that these opioid receptor subtypes differentially alter impulsive responding.
This is supported by evidence that DOR and MOR KO mice show opposing phenotypes
(Olmstead et al. 2009). In addition, DOR activation increases premature responding in the
Variable delay version of the RI task, whereas MOR activation has no effect (Chapter 2) (Befort
et al. 2011). What exactly these different opioid receptor subtypes are doing is not clear, but the
differential distribution of MORs and DORs across different populations of GABAergic neurons
(Mansour et al. 1995; Atwood et al. 2014) may underlie their influence on impulsive responding
(see Figure 6.1 for a schematic of opioid distribution on GABAergic neurons). Although we did
not asses this possibility, striatal dopaminergic and GABAergic mechanisms may mediate MORinduced impulsive action. Indeed, MOR antagonism in the NAccC blocks amphetamine (DA
agonist)-induced impulsive action in the 5-CSRTT (Wiskerke et al. 2011a). Moreover, morphine
and amphetamine produce similar patterns of responding in the RI task (Hayton et al. 2012;
Mahoney et al. 2013). As described in detail in the General Introduction, MOR activity in the
striatum increases DA in the NAcb (Di Chiara and Imperato 1988) through both direct activity of
MORs on presynaptic GABAergic terminals that synapse on dopaminergic neurons, and indirect
activity via GABAergic interneurons (Mansour et al. 1995). Additionally, MOR activity in the
132
VTA may regulate impulsive action by increasing striatal DA activity (Devine et al 1993). With
respect to GABA, high impulsive action rats have reduced GAD65/67 (GABA precursor) protein
expression in the NAcbC (Caprioli et al. 2014). Furthermore, viral-mediated GAD65/67
knockdown in the NAcbC increases impulsive action, as does GABAA and GABAB agonism
(Oliver et al. 2009; Caprioli et al. 2014). To date, the roles of D/MORs have not been examined
in relation to GABA-mediated impulsive action. It is unclear which of these mechanisms
regulates impulsive action.
The mechanisms of DOR-mediated impulsive action is not understood. It is possible that,
like serotonergic activity, both overstimulation (e.g., pharmacological manipulation) and
hypoactivity (e.g., M/DOR KO) of DORs impairs inhibitory control (Dalley et al. 2002b; Dalley
et al. 2008). In addition, DORs may also interact with dopaminergic (described above) and/or
cholinergic systems. Indeed, DORs inhibit cholinergic release within the striatum (Mansour et al.
1995), and DORs on cholinergic interneurons in the NAcbS are involved in other cue-associated,
goal-directed behaviours (Bertran-Gonzalez et al. 2013; Laurent et al. 2014). The role of DORcholinergic interactions in impulsive action has yet to be determined.
133
Figure 6.1 Schematic of opioid receptor distribution in regions associated with impulsive action.
Both MORs and DORs are located on GABAergic medium spiny projecting neurons in the
nucleus accumbens (NAcb) core and shell, and on GABAergic interneurons in the NAcb,
hippocampus (HP), and ventral tegmental area (VTA). Additionally, these receptors are on
glutamatergic prefrontal cortex (PFC) and hypothalamic projecting neurons. Ach, acetylcholine;
DA, dopamine; EA, extended amygdala; Glu, Glutamate; VP, ventral pallidum. Red lines,
GABAergic neuron; Blue lines, Glutamatergic neuron; Green lines, Dopaminergic neuron.
Stress-induced impulsive action
Surprisingly, none of the agents we tested blocked yohimbine-induced impulsive action
(Chapter 5). It is unlikely that yohimbine-induced impulsive action is mediated by DA and 5HT
receptor subtypes that we did not examine. This is because D2/3 antagonism does not affect
impulsive responding (van Gaalen et al. 2006), and these receptors not involved in yohimbineinduced reinstatement (Brown et al. 2012). Similarly, yohimbine has agonist effects at 5HT1a
receptors (Millan et al. 2000), yet systemic 5HT1a receptor agonism has no effect on premature
responding, and intra-prefrontal 5HT1a agonism decreases (rather than increases) impulsivity in
the 5-CSRTT (Winstanley et al. 2003).
134
It is also possible that we saw no effect on impulsivity with different antagonist drugs
because the dose ranges we employed were too small. This seems unlikely since the same dose
ranges of these agents block other yohimbine-induced behavioural changes (Marinelli et al.
2007; Lê et al. 2009; Lê et al. 2011; Simms et al. 2012; Brown et al. 2012). That said, the role of
systemic injections of glucocorticoid and CRF1 receptor antagonists on impulsivity has not been
examined, and was a secondary aim of this study. It is possible that these receptors are not
involved in impulsivity, and thus, do not mediate yohimbine-induced impulsivity.
In Chapter 5, we argue that it is most likely that yohimbine-induced impulsive action
involves several neurotransmitter systems that act in a synergistic fashion, which is a possibility
that merits further examination. Moreover, the circuitry of stress-induced impulsivity is still
relatively unknown, but likely involves prefrontal regions (Sinha 2008). OFC activity mediates
yohimbine-induced premature responding in the 5-CSRTT (Sun et al. 2010) and yohimbine
increases DA levels in mPFC (Chen et al. 2014). In contrast, yohimbine does not increase DA in
the NAcb (Chen et al. 2014), indirectly suggesting that this region is not involved in yohimbineinduced impulsive action, at least via dopaminergic activity. Together, these findings point to
prefrontal mechanisms in mediating yohimbine-induced impulsivity in our task, and broadly
suggest that physiological responses to stress may alter prefrontal pathways resulting in
increased impulsive responding.
Cognitive mechanims of opioid- and stress-induced impulsive action
There are several possible cognitive mechanisms of opioid- and stress-induced
impulsivity. In Chapter 3, we speculate that morphine impairs the rats’ ability to time the
premature phase in the Fixed 4-s task. This would explain why MOR activation increases
impulsive action in the Fixed, but not Variable, delay task. It is possible that this process also
135
drives MOR-induced impulsive action in the 5-CSRTT in which animals are predominantly
tested with a set inter trial interval (i.e., ‘premature phase’) of 5 s (Pattij et al. 2009; Wiskerke et
al. 2011a). This interpretation also fits with the fact that, along with mediating impulsive action,
cortico-striatal-hippocampal circuitry regulates interval timing (Bannerman et al. 1998; Matell
and Meck 2004; Meck et al. 2008; Singh et al. 2011; Yin and Meck 2014).
In contrast to the Fixed delay RI task, animals are unable to time the premature phase
interval in the Variable delay task. Therefore, inhibitory control in the Variable delay task likely
relies on divergent cognitive mechanisms. During baseline or drug-free conditions, inhibitory
circuits may favour impulse control processing. However, when under stress or following opioid
activation, “pro-impulsivity” mechanisms may drive behaviour by increasing 1.) non-specific
reactivity to various cues, 2.) salience of cues associated with reward, and/or 3.) reward salience.
Recent evidence suggests that yohimbine may non-specifically increase reactivity to cues
(i.e., magazine light), regardless of their association with contingent reinforcement (e.g., food
reward following a lever pressing response) (Shin et al. 2010; Chen et al. 2014). It may be that
yohimbine increases general, non-specific, reactivity to various cues presented in the RI task,
ultimately resulting in impaired inhibitory control. Conversely, yohimbine and/or M/DOR
activation may increase impulsive action more specifically by increasing the salience of cues
associated with the premature phase or reward (Cardinal et al. 2000; Tomie et al. 2008). This
latter possibility fits with findings that both a high impulsive action phenotype (Lovic et al.
2011) and M/DOR activation (Difeliceantonio and Berridge 2012; Laurent et al. 2014) increase
salience to cues associated with food and drug reward. Alternatively, yohimbine and/or opioid
activation may increase impulsivity by increasing reward salience (i.e., the reward becomes more
reinforcing or motivating). In support of this, acute social stress increases reward salience in
136
rodents (Chaijale et al. 2014), an effect that is also observed in humans assessed by risky
decision making tasks (Mather and Lighthall 2012; Starcke and Brand 2012). Similarly, NAcb
MOR activation increases cue-induced sucrose seeking (Peciña and Berridge 2013; Castro and
Berridge 2014), which is interpreted as increased motivation to seek sucrose elicited by cues that
were previously associated with its presentation (Peciña and Berridge 2013; Castro and Berridge
2014). These data suggest that reward and cue salience processing may not be fully distinct, and
that MOR signalling may mediate both of these processes.
It is unclear which, if any, of these cognitive processes mediate opioid- or stress-induced
impulsive action. Although it is impossible to dissociate these processes with our measure of
impulsive action, the RI task requires several associative processes that could alter cue and/or
motivational processes, which in turn may drive impulsive responding. As discussed in Section
6.3 (e.g., 6.3.5) below, individual differences relating to cue and reward processing predict
different aspects of drug taking behaviour. Thus, dissociating these processes in rodent models of
impulsive action may ultimately have clinical relevance.
6.3 Theoretical considerations: parsing impulsive action from other behavioural measures
High impulsivity is one of several endophenotypes that is a predisposing vulnerability
factor for addiction-like behaviour. Other endophenotypes include risky-decision making,
increased locomotor reactivity, impaired reversal learning, anxiety-like behaviour, and Pavlovian
reward cue responding. In the following sections, I will briefly describe the aforementioned
behaviours, and describe how they relate to impulsive action. Finally, in section 6.3.6, I will
describe how each of these endophenotypes uniquely contributes to the development and
expression of addiction-like behaviour. More specifically, I will discuss how propensity to
137
acquire drug-taking behaviour, escalated intake, and relapse are all mediated in part by distinct
endophenotypes.
Risky decision making
Risky decision making, and its associated risky behaviour, involve decisions about
different risks and rewards. In many ways, risky decision making is more similar to impulsive
choice than impulsive action, in that animal models of risky decision making force a choice
between differentially reinforced options. However, unlike impulsive choice tasks, risky decision
making tasks assess sensitivity to risk by forcing a choice between uncertain/risky (high-value
reward) and certain/safe (low-value reward) choices. Several commonly used risky decision
making tasks include a rodent version of the Iowa gambling task (Zeeb et al. 2009), a
probabilistic discounting task (St Onge and Floresco 2009), and a risky decision making task
(Mitchell et al. 2011). In these tasks, the ‘risk’ is lost reward opportunity (Zeeb et al. 2009; St
Onge and Floresco 2009) or shock (Simon et al. 2009).
In the rodent gambling task, animals learn to associate reward probabilities with each of 4
decks of cards (Zeeb et al. 2009). Two decks produce large rewards (greater number of sucrose
pellets), but result in larger punishments (longer time-out periods), whereas the remaining two
decks are smaller reward options (fewer sucrose pellets) with less punishment (shorter time-out
period) (Zeeb et al. 2009). Since sessions are a fixed duration, high risk options result in fewer
rewards overall. The probabilistic discounting task is a simplified version of the rodent gambling
task that assesses reward uncertainty (St Onge and Floresco 2009). In the probabilistic risktaking task, animals must choose between either a high-value (more pellets) or low-value (fewer
pellets) reward; however the high-value reward is associated with increasing probabilities of not
receiving reward (St Onge and Floresco 2009). Finally, in the risky decision making task (Simon
138
et al. 2009), rodents choose between a small reward with no probability of receiving a shock and
a larger reward with varying shock probabilities (Simon et al. 2009; Mitchell et al. 2011).
Perhaps not surprisingly, reward choices in the risky decision making (shock) and
probabilistic risk-taking (no shock) tasks correlate (Simon et al. 2009), suggesting that
probabilistic decision making involving either reward or punishment assessment may be
mediated by similar mechanisms. Conversely, responding in delay discounting tasks measuring
impulsive choice does not correlate with responding in either the risky decision making task or
the probabilistic risk-taking task (Simon et al. 2009; Mitchell et al. 2011; Shimp et al. 2014),
suggesting that the process of selecting immediate versus delayed rewards is different from the
process of selecting safe versus risky choices. The relationship between risky decision making
and impulsive action has not been examined systematically, but it would follow that impulsive
action is distinct from risky decision making, given that the latter involves choice assessments.
Finally, unlike other behaviours described in Section 6.3, the role of risky decision making as an
endphophenotype to drug addiction has not been established, and merits further research.
Locomotor reactivity
In preclinical literature, sensation seeking/risk taking was one of the first phenotypes
associated with drug-related behaviour (Piazza et al. 1989; Piazza et al. 1990; Dellu et al. 1996).
In this model, animals are categorized as high or low responders based on degree of locomotor
response habituation (reactivity) to a novel environment (Piazza et al. 1990; Dellu et al. 1996;
Belin et al. 2011). Specifically, high responders are more reactive (i.e., exhibit less habituation)
compared to low responders; based on these findings, high responders are thought of as a rodent
equivalent of human sensation seekers (Dellu et al. 1996). Many lines of evidence suggest that
impulsive action and sensation seeking are distinct, namely that high responder rats are not more
139
impulsive in the 5-CSRTT than low responder rats (Dalley et al. 2007; Belin et al. 2008).
Furthermore, while high responder animals show increased exploratory locomotion and a
prolonged corticosterone secretion following exposure a novel environment (Piazza et al. 1991),
high impulsive action animals do not (Molander et al. 2011).
Reversal learning
Reversal learning is thought to be a measure of cognitive flexibility, which broadly refers
to the ability to adapt to changing conditions, for instance, by altering behaviour (Stalnaker et al.
2009). In rodent models, reversal learning is characterized as the ability to inhibit a response that
was previously rewarded. In reversal learning tasks, animals first learn to discriminate between
two stimuli/locations, and to associate one of the stimuli/locations with food reinforcement
(Izquierdo and Jentsch 2012). Once animals learn to discriminate these options, the
reinforcement contingencies are reversed, and the previously correct response is no longer
reinforcing. In these tasks, poor reversal learners perseverate on the previously learned
contingency, and require more trials to learn the reversal. Reversal learning requires executive
functioning, including assessment of expected outcomes, rule shifting (i.e., contingency
reversal), and the ability to inhibit responses that follow contingency shifts. In this way,
perseverative responding during the reversal phase is thought to assess action inhibition in a
similar way to premature responding in impulsive action tasks (Crews and Boettiger 2009;
Izquierdo and Jentsch 2012). Indeed, high impulsive people show deficits in reversal learning
(Franken et al. 2008; Romer et al. 2009).
140
Anxiety-like behaviour
Anxiety-like behaviour is usually assessed by open arm activity on an elevated plus
maze. Studies examining the direct relationship between anxiety and impulsivity are limited;
however one study found no relationship between impulsive action and anxiety-like behaviour,
or between impulsivity and novelty-induced corticosterone responses (Molander et al. 2011). A
more comprehensive discussion on the relationship between anxiety/stress and impulsive action
can be found in previous Sections 6.1.2, 6.2.2, and 6.2.3.
Pavlovian reward cue respond
In classical conditioning, conditioned stimuli (CS) that predict, or are associated with,
rewards can also acquire properties that in themselves motivate behaviour; these are referred to
as incentive stimuli (Berridge and Robinson 1998; Saunders and Robinson 2013). When a CS
(e.g., lever + tone) is repeatedly paired with food reward delivery, a small portion of rats will
spend relatively more time interacting with the CS (“sign-trackers”), suggesting that the cue
itself has become rewarding. Other rats tend to interact more with the magazine, which has
previous been more directly associated with food delivery (“goal-trackers”) (Tomie et al. 1989;
Tomie et al. 2008; Saunders and Robinson 2013). Sign trackers show increased impulsive action
(Lovic et al. 2011; Olshavsky et al. 2014), pointing to a relationship that warrants further
investigation.
It is worth noting that although sign-tracking, high reactivity, and anxiety-like behaviour
do not correlate with each other (Flagel et al. 2009; Molander et al. 2011; Vanhille et al. 2014),
yet animals bred as high- and low-responders to novelty tend to be sign trackers and goal
trackers, respectively (Flagel et al. 2010). In addition, these high responder rats also show
141
increased impulsive action (Flagel et al. 2010). These findings in selectively bred lines of rats
suggest that shared genetic influences contribute to these behaviours. Nonetheless, it is likely
that each of these processes have dissociable neurobiological underpinnings that uniquely
contribute to the observed endophenotypes.
Vulnerability to addiction: unique contribution of behavioural endophenotypes
Broadly, the endophenotypes discussed in Section 6.3 are predisposing risk factors for
various aspects of drug-taking behaviour (Table 6.1). More specifically, clinical and preclinical
research suggests that each of these endophenotypes may confer vulnerability to specific aspects
of drug-taking behaviour, from initial intake to relapse following abstinence.
A high responder phenotype predicts the propensity to acquire cocaine selfadministration during initial phases of drug access (indexed by rate of acquisition) (Piazza and
Le Moal 1998). In contrast, high responders do not exhibit escalated drug intake that is thought
to measure compulsive, habitual responding (indexed by continued drug seeking in spite of
negative consequences) (Belin et al. 2009). Similar to high responders, poor reversal learners
have greater propensity to initially self-administer cocaine and respond for the drug at higher
rates than good reversal learners (Cervantes et al. 2013). Despite this finding, it is not clear
whether a poor reversal learning phenotype predicts drug-taking behaviour following chronic
drug use (i.e., escalation or reinstatement).
In contrast to high locomotor reactivity and poor reversal learning, high impulsivity
predicts the transition to escalated, compulsive drug use (Dalley et al. 2007; Belin et al. 2008;
Diergaarde et al. 2008; Besson et al. 2013), but not the initial propensity to acquire cocaine selfadministration or maintain drug intake (Dalley et al. 2007; Belin et al. 2008; Perry et al. 2008;
Besson et al. 2013). These findings highlight that high impulsivity is a crucial endophenotype
142
that makes some individuals more vulnerable to drugs of abuse by directly interacting with drug
action. More specifically, unlike a high responding phenotype, these findings suggest that
impulsivity does not necessarily affect initial drug intake, but rather, once drug is on board, there
is an interaction between impulsivity and drug-action in high, but not low, impulsive, animals. In
clinical populations, increased impulsivity, but not sensation-seeking, is increased in both
stimulant abusers and their non-drug using biological siblings, supporting the fact that
impulsivity, but not sensation-seeking, may be an endophenotype for stimulant addiction in
clinical populations (Ersche et al. 2010). Similar to high impulsivity rats, high anxiety rats also
have a greater propensity to transition to escalated cocaine self-administration (Dilleen et al.
2012).
Finally, a sign-tracking phenotype is associated with increased sensitivity and motivation
for cues associated with drug-related stimuli, which are particularly important for reinstatement
(Flagel et al. 2009; Flagel et al. 2010). For instance, sign trackers show greater cue-induced
reinstatement and increased resistance to extinction (when the discriminative cue is present)
compared to goal trackers (Tomie et al. 1989; Tomie et al. 2008; Lovic et al. 2011; Saunders and
Robinson 2013). Like sign tracking, high impulsivity also predicts reinstatement of cocaine
seeking (Koob and Le Moal 2008; Economidou et al. 2009; Pattij and De Vries 2013). Since sign
trackers show increased impulsive action (Lovic et al. 2011; Olshavsky et al. 2014), it is possible
that these two phenotypes act together to promote cue-induced reinstatement; however this has
not been examined.
Overall, high responding and impaired reversal learning predict intake during initial
access, high impulsivity and anxiety drive escalated, compulsive intake, and sign-tracking and
high impulsivity predict reinstatement. Thus, it is likely that these endophenotypes and their
143
associated neuropharmacological mechanisms differentially interact with drug to drive distinct
aspects of the addiction process (Izquierdo and Jentsch 2012; Jenstch et al. 2014; Jupp and
Dalley 2014).
Table 6.1 Role of behavioural phenotypes to drug take during different phases of drug
addiction
Endophenotype
Acquisition of Drug
Self-Administration
High
Responding
High Impulsive Poor Reversal
Action
Learning
High Anxiety
Sign Tracking
↑
=
↑
=
(Piazza et al.
(Belin et al. (Cervantes et al. (Dilleen et al.
1989; Piazza et
2008;
2013)
2012)
al. 1990; Belin Economidou et
et al. 2008) al. 2009; Besson
et al. 2013; Bird
and Schenk
2013)
=
(Saunders and
Robinson
2010)
?
?
↑
(Beckmann et
al. 2011)
Maintenance of
Drug SelfAdministration
?
=
↑
(Dalley et al. (Cervantes et al.
2007; Belin et
2013)
al. 2008; Anker
et al. 2009;
Besson et al.
2013)
Escalated Intake
?
=
(Dalley et al.
2007; Anker et
al. 2009)
?
↑
(Dilleen et al.
2012)
?
Motivation for SelfAdministration
=
(Belin et al.
2008)
=
(Belin et al.
2008)
?
?
?
Compulsive Drug
Use
=
(Belin et al.
2008)
↑
(Belin et al.
2008)
?
?
?
Reinstatement
=
↑
(Sutton et al. (Economidou et
2000)
al. 2009; Bird
and Schenk
2013)
?
↑
(Lovic et al.
2011)
↑
(Saunders and
Robinson
2010; Barker
et al. 2012)
↑ denotes increased behaviour; = denotes no change; ? denotes a relationship that has not been
examined
144
6.4 Conclusions and clinical relevance
Taken together, these data highlight a novel role of DORs in impulsive action in rats and
contribute to growing evidence of opioid involvement in impulsivity. We did not observe an
impulsivity phenotype either in mice lacking M/DORs of GABAergic forebrain neurons or in
constitutive M/DOR KO mice. Therefore it is difficult to conclude that opioid receptors on
GABAergic forebrain neurons mediate impulsive action. These data highlight the importance of
taking behavioural task parameters and genetic versus pharmacological manipulations into
account when studying complex cognitive mechanisms such as impulsivity. Finally, we show
that yohimbine, a pharmacological stressor, increases impulsive action, which is likely mediated
by synergistic action of several neurotransmitter systems.
Elucidating the neuropharmacological mechanims of impulsivity may allow for the
development of effective therapeutic agents to treat maladaptive impulsivity. Our results suggest
that opioid system antagonism may be one of these targets. Naloxone and naltrexone (MOR
antagonists) are both agents found in clinically available pharmacotherapies for opiate
dependence (Orman and Keating 2009). To date, opioid antagonists (e.g., naltrexone) have been
used somewhat successfully to decrease pathological gambling (Dannon et al. 2005; Grant et al.
2009; Bosco et al. 2012; Rosenberg et al. 2013; Bartley and Bloch 2013), suggesting that opioid
antagonists are an efficacious treatment option for some impulse control disorders. Whether
opioid antagonism can attenuate symptoms associated with other impulse control disorders is not
well established in clinical literature, but in rodent models, co-administration of naltrexone
reduces the rewarding effects of methylphenidate, a pharmacotherapy for ADHD (Zhu et al.
2011). These data suggest that a combination drug therapy may be effective in reducing the
145
misuse and abuse potential of methylphenidate (Wilens et al. 2008), while maintaining efficacy
in treating ADHD, and possibly other impulse control disorders.
Additionally, in the last decade, evidence has emerged showing a role for functional
M/DOR heterodimers in producing a pain state (Gomes et al. 2000; Gomes et al. 2004; Kabli et
al. 2010; Milan-Lobo and Whistler 2011). These findings have led to the hypothesis that
pharmacological antagonism of these heterodimers may reduce chronic pain (Haack and
McCarty 2011). Although these M/DOR heterodimers have not been examined with respect to
impulsive action, M/DOR heterodimers may have similar therapeutic potentials for impulse
control disorders.
Understanding how individual differences alter the propensity for stress to increase
impulsivity may also prove to be relevant to clinical treatments of impulse control disorders,
including addiction. Furthermore, since high impulsivity increases drug-related behaviour in
people with perceived high stress, pharmacotherapies that target the stress system may act to
reduce drug-related behaviour through impulsivity-related mechanisms. Overall, the data
presented in this thesis are particularly important for the understanding and intervention of
psychiatric disorders that are characterized by impaired impulse control, and open the possibility
of treatment options that target opioid and stress systems.
146
References
Abela AR, Chudasama Y (2014) Noradrenergic α2A-receptor stimulation in the ventral
hippocampus reduces impulsive decision-making. Psychopharmacology (Berl) 231:521–31.
Agnoli L, Carli M (2011) Synergistic interaction of dopamine D1 and glutamate N-methyl-Daspartate receptors in the rat dorsal striatum controls attention. Neuroscience 185:39–49.
Agnoli L, Carli M (2012) Dorsal-striatal 5-HT1A and 5-HT1C receptors control impulsivity and
perseverative responding in the 5-choice serial reaction time task. Psychopharmacology
(Berl) 219:633–45.
Almeida M, Lee R, Coccaro EF (2010) Cortisol responses to ipsapirone challenge correlate with
aggression, while basal cortisol levels correlate with impulsivity, in personality disorder and
healthy volunteer subjects. J Psychiatr Res 44:874–80.
Ambroggi F, Turiault M, Milet A, et al. (2009) Stress and addiction: glucocorticoid receptor in
dopaminoceptive neurons facilitates cocaine seeking. Nat Neurosci 12:247–249.
American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders,
5th ed. Arlington, VA.
Anderson S, Eisenstat D, Shi L, Rubenstein J (1997) Interneuron migration from basal forebrain
to neocortex: dependence on Dlx genes. Science 278:474–476.
Anker JJ, Perry JL, Gliddon LA, Carroll ME (2009) Impulsivity predicts the escalation of
cocaine self-administration in rats. Pharmacol Biochem Behav 93:343–8.
Atwood BK, Kupferschmidt D, Lovinger DM (2014) Opioids induce dissociable forms of longterm depression of excitatory inputs to the dorsal striatum. Nat Neurosci 17:540–549.
Austin M, Kalivas P (1990) Enkephalinergic and GABAergic modulation of motor activity in the
ventral pallidum. J Pharmacol Exp Ther 252:1370–1377.
Baarendse PJJ, Vanderschuren LJMJ (2012) Dissociable effects of monoamine reuptake
inhibitors on distinct forms of impulsive behavior in rats. Psychopharmacology (Berl)
219:313–326.
Baldacchino A, Balfour DJK, Matthews K (2014) Impulsivity and opioid drugs: differential
effects of heroin, methadone and prescribed analgesic medication. Psychol Med 1–13. doi:
10.1017/S0033291714002189
Bannerman DM, Yee BK, Good MA, et al. (1998) Double dissociation of function within the
hippocampus: A comparison of dorsal, ventral, and complete hippocampal cytotoxic
lesions. Behav neurocience 113:1170–88.
147
Bari A, Eagle DM, Mar AC, et al. (2009) Dissociable effects of noradrenaline, dopamine, and
serotonin uptake blockade on stop task performance in rats. Psychopharmacology (Berl)
205:273–83.
Bari A, Mar AC, Theobald DE, et al. (2011) Prefrontal and monoaminergic contributions to stopsignal task performance in rats. J Neurosci 31:9254–9263.
Bari A, Robbins TW (2013) Inhibition and impulsivity: behavioral and neural basis of response
control. Prog Neurobiol 108:44–79.
Barker JM, Torregrossa MM, Taylor JR (2012) Low prefrontal PSA-NCAM confers risk for
alcoholism-related behavior. Nat Neurosci 15:1356–8.
Bartley CA, Bloch MH (2013) Meta-analysis: pharmacological treatment of pathological
gambling. Expert Rev Neurother 13:887–94.
Basar K, Sesia T, Groenewegen H, et al. (2010) Nucleus accumbens and impulsivity. Prog
Neurobiol 92:533–557.
Beckmann JS, Marusich JA, Gipson CD, Bardo MT (2011) Novelty seeking, incentive salience
and acquisition of cocaine self-administration in the rat. Behav Brain Res 216:159–65.
Befort K, Mahoney MK, Chow C, et al. (2011) Effects of delta opioid receptors activation on a
response inhibition task in rats. Psychopharmacology (Berl) 214:967–976.
Befort K, Mattéi M, Roeckel N, Kieffer B (1994) Chromosomal localization of the delta opioid
receptor gene to human 1p34.3-p36.1 and mouse 4D bands by in situ hybridization.
Genomics 20:143–145.
Belin D, Balado E, Piazza PV, Deroche-Gamonet V (2009) Pattern of intake and drug craving
predict the development of cocaine addiction-like behavior in rats. Biol Psychiatry 65:863–
868.
Belin D, Berson N, Balado E, et al. (2011) High-novelty-preference rats are predisposed to
compulsive cocaine self-administration. Neuropsychopharmacology 36:569–79.
Belin D, Mar AC, Dalley JW, et al. (2008) High impulsivity predicts the switch to compulsive
cocaine-taking. Science 320:1352–1355.
Berridge KC, Robinson TE (1998) What is the role of dopamine in reward: hedonic impact,
reward learning, or incentive salience? Brain Res Brain Res Rev 28:309–69.
Berrocoso E, Ikeda K, Sora I, et al. (2013) Active behaviours produced by antidepressants and
opioids in the mouse tail suspension test. Int J Neuropsychopharmacol 16:151–62.
148
Bertran-Gonzalez J, Laurent V, Chieng BC, et al. (2013) Learning-related translocation of δopioid receptors on ventral striatal cholinergic interneurons mediates choice between goaldirected actions. J Neurosci 33:16060–71.
Besson A, Privat A, Eschalier A, Fialip J (1996) Effects of morphine, naloxone and their
interaction in the learned-helplessness paradigm in rats. Psychopharmacology (Berl)
123:71–8.
Besson M, Belin D, McNamara R, et al. (2010) Dissociable control of impulsivity in rats by
dopamine D2/3 receptors in the core and shell subregions of the nucleus accumbens.
Neuropsychopharmacology 35:560–569.
Besson M, Pelloux Y, Dilleen R, et al. (2013) Cocaine modulation of frontostriatal expression of
Zif268, D2, and 5-HT2c receptors in high and low impulsive rats.
Neuropsychopharmacology 38:1963–1973.
Bird J, Schenk S (2013) Contribution of impulsivity and novelty-seeking to the acquisition and
maintenance of MDMA self-administration. Addict Biol 18:654–64.
Bolivar VJ, Cook MN, Flaherty L (2001) Mapping of quantitative trait loci with
knockout/congenic strains. Genome Res 11:1549–52.
Bornovalova MA, Lejuez CW, Daughters SB, et al. (2005) Impulsivity as a common process
across borderline personality and substance use disorders. Clin Psychol Rev 25:790–812.
Bosco D, Plastino M, Colica C, et al. (2012) Opioid antagonist naltrexone for the treatment of
pathological gambling in Parkinson disease. Clin Neuropharmacol 35:118–20.
Bothe GWM, Bolivar VJ, Vedder MJ, Geistfeld JG (2004) Genetic and behavioral differences
among five inbred mouse strains commonly used in the production of transgenic and
knockout mice. Genes Brain Behav 3:149–57.
Boutrel B, Kenny PJ, Specio SE, et al. (2005) Role for hypocretin in mediating stress-induced
reinstatement of cocaine-seeking behavior. Proc Natl Acad Sci U S A 102:19168–73.
Bowers BJ, Wehner JM (2001) Ethanol consumption and behavioral impulsivity are increased in
protein kinase Cgamma null mutant mice. J Neurosci 21:RC180.
Brevers D, Cleeremans A, Verbruggen F, et al. (2012) Impulsive action but not impulsive choice
determines problem gambling severity. PLoS One 7:e50647. doi:
10.1371/journal.pone.0050647
Briand LA, Blendy JA (2011) Molecular and genetic substrates linking tress and addiction. Brain
Res 1314:219–34.
149
Broom D, Jutkiewicz E, Folk J, et al. (2002a) Nonpeptidic delta-opioid receptor agonists reduce
immobility in the forced swim assay in rats. Neuropsychopharmacology 26:744–755.
Broom D, Jutkiewicz E, Rice K, et al. (2002b) Behavioral effects of delta-opioid receptor
agonists: potential antidepressants? Jpn J Pharmacol 90:1–6.
Broos N, Diergaarde L, Schoffelmeer ANM, et al. (2012a) Trait impulsive choice predicts
resistance to extinction and propensity to relapse to cocaine seeking: a bidirectional
investigation. Neuropsychopharmacology 37:1377–1386.
Broos N, Schmaal L, Wiskerke J, et al. (2012b) The relationship between impulsive choice and
impulsive action: a cross-species translational study. PLoS One 7:e36781. doi:
10.1371/journal.pone.0036781
Brown ZJ, Kupferschmidt D, Erb S (2012) Reinstatement of cocaine seeking in rats by the
pharmacological stressors, corticotropin-releasing factor and yohimbine: role for D1/5
dopamine receptors. Psychopharmacology (Berl) 224:431–40.
Buckholtz JW, Treadway MT, Cowan RL, et al. (2010) Dopaminergic network differences in
human impulsivity. Science 329:532.
Burton CL, Fletcher PJ (2012) Age and sex differences in impulsive action in rats : the role of
dopamine and glutamate. Behav Brain Res 230:21–33.
Butts K, Floresco S, Phillips A (2013) Acute stress impairs set-shifting but not reversal learning.
Behav Brain Res 252:222–229.
Caprioli D, Sawiak S, Merlo E, et al. (2014) Gamma aminobutyric acidergic and neuronal
structural markers in the nucleus accumbens core underlie trait-like impulsive behavior.
Biol Psychiatry 75:115–123.
Cardinal RN, Pennicott DR, Sugathapala CL, et al. (2001) Impulsive choice induced in rats by
lesions of the nucleus accumbens core. Science 292:2499–2501.
Cardinal RN, Robbins TW, Everitt BJ (2000) The effects of d-amphetamine, chlordiazepoxide,
a-flupenthixol and behavioural manipulations on choice of signalled and unsignalled
delayed reinforcement in rats. Psychopharmacology (Berl) 152:362–375.
Cassin SE, von Ranson KM (2005) Personality and eating disorders: a decade in review. Clin
Psychol Rev 25:895–916.
Castellanos FX, Tannock R (2002) Neuroscience of attention-deficit/hyperactivity disorder: the
search for endophenotypes. Nat Rev Neurosci 3:617–28.
150
Castro DC, Berridge KC (2014) Opioid hedonic hotspot in nucleus accumbens shell: mu, delta,
and kappa maps for enhancement of sweetness “liking” and “wanting.” J Neurosci
34:4239–4250.
Cervantes MC, Laughlin RE, Jentsch JD (2013) Cocaine self-administration behavior in inbred
mouse lines segregating different capacities for inhibitory control. Psychopharmacology
(Berl) 229:515–25.
Chaijale NN, Snyder K, Arner J, et al. (2014) Repeated social stress increases reward salience
and impairs encoding of prediction by rat locus coeruleus neurons.
Neuropsychopharmacology doi: 10.1038/npp.2014.200
Chamberlain SR, Muller U, Blackwell AD, et al. (2006) Neurochemical modulation of response
inhibition and probabilistic learning in humans. Science 311:861–863.
Chamberlain SR, Robbins TW (2013) Noradrenergic modulation of cognition: therapeutic
implications. J Psychopharmacol 27:694–718.
Charbogne P, Kieffer BL, Befort K (2014) 15 years of genetic approaches in vivo for addiction
research: opioid receptor and peptide gene knockout in mouse models of drug abuse.
Neuropharmacology 76:204–17.
Charney DA, Zikos E, Gill KJ (2010) Early recovery from alcohol dependence: factors that
promote or impede abstinence. J Subst Abuse Treat 38:42–50.
Chen Y, Fiscella KK, Bacharach SZ, et al. (2014) Effect of yohimbine on reinstatement of
operant responding in rats is dependent on cue contingency but not food reward history.
Addict Biol. doi: 10.1111/adb.12164
Chieng B, Williams J (1998) Increased opioid inhibition of GABA release in nucleus accumbens
during morphine withdrawal. J Neurosci 18:7033–7039.
Chu Sin Chung P, Keyworth HL, Martin-Garcia E, et al. (2014) A novel anxiogenic role for the
delta opioid receptor expressed in GABAergic forebrain neurons. Biol Psychiatry. doi:
10.1016/j.biopsych.2014.07.033
Chu Sin Chung P, Kieffer BL (2013) Delta opioid receptors in brain function and diseases.
Pharmacol Ther 140:112–20.
Chudasama Y, Doobay M, Liu Y (2012) Hippocampal-prefrontal cortical circuit mediates
inhibitory response control in the rat. J Neurosci 32:10915–10924.
Chudasama Y, Passetti F, Rhodes S, et al. (2003) Dissociable aspects of performance on the 5choice serial reaction time task following lesions of the dorsal anterior cingulate,
infralimbic and orbitofrontal cortex in the rat: differential effects on selectivity, impulsivity
and compulsivity. Behav Brain Res 146:105–119.
151
Cippitelli A, Damadzic R, Hansson AC, et al. (2010) Neuropeptide Y (NPY) suppresses
yohimbine-induced reinstatement of alcohol seeking. Psychopharmacology (Berl) 208:417–
426.
Claes L, Vandereycken W, Vertommen H (2005) Impulsivity-related traits in eating disorder
patients. Pers Individ Dif 39:739–749.
Cobos I, Borello U, Rubenstein J (2007) Dlx transcription factors promote migration through
repression of axon and dendrite growth. Neuron 54:873–888.
Cobos I, Calcagnotto M, Vilaythong A, et al. (2005) Mice lacking Dlx1 show subtype-specific
loss of interneurons, reduced inhibition and epilepsy. Nat Neurosci 8:1059–1068.
Cole BJ, Robbins TW (1987) Amphetamine impairs the discriminative performance of rats with
dorsal noradrenergic bundle lesions on a 5-choice serial reaction time task: new evidence
for central dopaminergic-noradrenergic interactions. Psychopharmacology (Berl) 91:458–
466.
Cook M, Bolivar V, McFadyen M, Flaherty L (2002) Behavioral differences among 129
substrains: Implications for knockout and transgenic mice. Behav Neurosci 116:600–611.
Crews FT, Boettiger CA (2009) Impulsivity, frontal lobes and risk for addiction. Pharmacol
Biochem Behav 93:237–247.
Crusio WE (2004) Flanking gene and genetic background problems in genetically manipulated
mice. Biol Psychiatry 56:381–385.
Dalley J, Everitt BJ, Robbins TW (2011) Impulsivity, compulsivity, and top-down cognitive
control. Neuron 69:680–694.
Dalley J, Theobald D, Pereira E, et al. (2002a) Specific abnormalities in serotonin release in the
prefrontal cortex of isolation-reared rats measured during behavioural performance of a task
assessing visuospatial attention and impulsivity. Psychopharmacology (Berl) 164:329–340.
Dalley JW, Fryer TD, Brichard L, et al. (2007) Nucleus accumbens D2/3 receptors predict trait
impulsivity and cocaine reinforcement. Science 315:1267–1270.
Dalley JW, Mar AC, Economidou D, Robbins TW (2008) Neurobehavioral mechanisms of
impulsivity: fronto-striatal systems and functional neurochemistry. Pharmacol Biochem
Behav 90:250–260.
Dalley JW, Theobald DE, Eagle DM, et al. (2002b) Deficits in impulse control associated with
tonically-elevated serotonergic function in rat prefrontal cortex. Neuropsychopharmacology
26:716–28.
152
Dannon PN, Lowengrub K, Musin E, et al. (2005) Sustained-release bupropion versus naltrexone
in the treatment of pathological gambling. J Clin Psychopharmacol 25:593–596.
Davis AR, Shields AD, Brigman JL, et al. (2008) Yohimbine impairs extinction of cocaineconditioned place preference in an alpha2-adrenergic receptor independent process. Learn
Mem 15:667–76.
De Wit H (2009) Impulsivity as a determinant and consequence of drug use: a review of
underlying processes. Addict Biol 14:22–31.
Dellu F, Piazza P V, Mayo W, et al. (1996) Novelty-seeking in rats - biobehavioral
characteristics and possible relationship with the sensation-seeking trait in man.
Neuropsychobiology 34:136–45.
Di Chiara G, Imperato A (1988) Opposite effects of mu and kappa opiate agonists on dopamine
release in the nucleus accumbens and in the dorsal caudate of freely moving rats. J
Pharmacol Exp Ther 244:1067–1080.
Diergaarde L, Pattij T, Poortvliet I, et al. (2008) Impulsive choice and impulsive action predict
vulnerability to distinct stages of nicotine seeking in rats. Biol Psychiatry 63:301–8.
Difeliceantonio AG, Berridge KC (2012) Which cue to “want”? Opioid stimulation of central
amygdala makes goal-trackers show stronger goal-tracking, just as sign-trackers show
stronger sign-tracking. Behav Brain Res 230:399–408.
Dilleen R, Pelloux Y, Mar AC, et al. (2012) High anxiety is a predisposing endophenotype for
loss of control over cocaine, but not heroin, self-administration in rats.
Psychopharmacology (Berl) 222:89–97.
Doetschman T (2008) Influence of Genetic Background on Genetically Engineered Mouse
Phenotypes. In: Kuhn R, Wurst W (eds) Gene Knockout Protoc. Methods Mol. Biol., 2nd
ed. Humana Press, pp 423–433.
Dom G, D’haene P, Hulstijn W, Sabbe B (2006) Impulsivity in abstinent early-and late-onset
alcoholics: differences in self-report measures and a discounting task. Addiction 101:50–59.
Drolet G, Dumont ÉC, Gosselin I, et al. (2001) Role of endogenous opioid system in the
regulation of the stress response. Prog Neuropsychopharmacol Biol Psychiatry 25:729–741.
Eagle DM, Baunez C (2010) Is there an inhibitory-response-control system in the rat? Evidence
from anatomical and pharmacological studies of behavioral inhibition. Neurosci Biobehav
Rev 34:50–72.
Economidou D, Pelloux Y, Robbins TW, et al. (2009) High impulsivity predicts relapse to
cocaine-seeking after punishment-induced abstinence. Biol Psychiatry 65:851–856.
153
Economidou D, Theobald DEH, Robbins TW, et al. (2012) Norepinephrine and dopamine
modulate impulsivity on the five-choice serial reaction time task through opponent actions
in the shell and core sub-regions of the nucleus accumbens. Neuropsychopharmacology
37:2057–2066.
Erbs E, Faget L, Scherrer G, et al. (2012) Distribution of delta opioid receptor-expressing
neurons in the mouse hippocampus. Neuroscience 221:203–13.
Ersche KD, Turton AJ, Pradhan S, et al. (2010) Drug addiction endophenotypes: impulsive
versus sensation-seeking personality traits. Biol Psychiatry 68:770–3.
Evenden JL (1999) Varieties of impulsivity. Psychopharmacology (Berl) 146:348–361.
Fernando ABP, Economidou D, Theobald DE, et al. (2012) Modulation of high impulsivity and
attentional performance in rats by selective direct and indirect dopaminergic and
noradrenergic receptor agonists. Psychopharmacology (Berl) 219:341–352.
Fields S, Collins C, Lerras K, Reynolds B (2009) Dimensions of impulsive behavior in
adolescent smokers and nonsmokers. Exp Clin Psychopharmacol 17:302–311.
Filliol D, Ghozland S, Chluba J, et al. (2000) Mice deficient for delta- and mu-opioid receptors
exhibit opposing alterations of emotional responses. Nat Genet 25:195–200.
Flagel SB, Akil H, Robinson TE (2009) Individual differences in the attribution of incentive
salience to reward-related cues: implications for addiction. Neuropharmacology 56 Suppl
1:139–148.
Flagel SB, Robinson TE, Clark JJ, et al. (2010) An animal model of genetic vulnerability to
behavioral disinhibition and responsiveness to reward-related cues: implications for
addiction. Neuropsychopharmacology 35:388–400.
Fletcher PJ, Rizos Z, Noble K, Higgins GA (2011) Impulsive action induced by amphetamine,
cocaine and MK801 is reduced by 5-HT(2C) receptor stimulation and 5-HT(2A) receptor
blockade. Neuropharmacology 61:468–477.
Fowler SC, Pinkston J, Vorontsova E (2009) Timing and space usage are disrupted by
amphetamine in rats maintained on DRL 24-s and DRL 72-s schedules of reinforcement.
Psychopharmacology (Berl) 204:213–225.
Fox HC, Bergquist KL, Hong KI, Sinha R (2007) Stress-induced and alcohol cue-induced
craving in recently abstinent alcohol-dependent individuals. Alcohol Clin Exp Res 31:395–
403.
Fox HC, Bergquist KL, Peihua G, Rajita S (2010) Interactive effects of cumulative stress and
impulsivity on alcohol consumption. Alcohol Clin Exp Res 34:1376–85.
154
Fox HC, Seo D, Tuit K, et al. (2012) Guanfacine effects on stress, drug craving and prefrontal
activation in cocaine dependent individuals: preliminary findings. J Psychopharmacol
26:958–72.
Franken I, Strien J, Nijs I, Muris P (2008) Impulsivity is associated with behavioral decisionmaking deficits. Psychiatry Res 158:155–63.
Franowicz JS, Kessler LE, Borja CMD, et al. (2002) Mutation of the alpha 2A-adrenoceptor
impairs working memory performance and annuls cognitive enhancement by guanfacine. J
Neurosci 22:8771–8777.
Funk D, Li Z, Lê A (2006) Effects of environmental and pharmacological stressors on c-fos and
corticotropin-releasing factor mRNA in rat brain: relationship to the reinstatement of
alcohol seeking. Neuroscience 138:235–43.
Gaveriaux-Ruff C, Nozaki C, Nadal X, et al. (2011) Genetic ablation of delta opioid receptors in
nociceptive sensory neurons increases chronic pain and abolishes opioid analgesia. Pain
152:1238–48.
Gilpin NW (2012) Corticotropin-releasing factor (CRF) and neuropeptide Y (NPY): effects on
inhibitory transmission in central amygdala, and anxiety- & alcohol-related behaviors.
Alcohol 46:329–337. doi: 10.1016/j.alcohol.2011.11.009
Gilpin NW, Stewart RB, Badia-Elder NE (2008) Neuropeptide Y suppresses ethanol drinking in
ethanol-abstinent, but not non-ethanol-abstinent, Wistar rats. Alcohol 42:541–551.
Goeders NE (1997) A neuroendocrine role in cocaine reinforcement. Psychoneuroendocrinology
22:237–259.
Goeders NE (2002) The HPA axis and cocaine reinforcement. Psychoneuroendocrinology
27:13–33.
Goltz C Von Der, Koopmann A, Dinter C, et al. (2011) Hormones and behavior involvement of
orexin in the regulation of stress, depression and reward in alcohol dependence. Horm
Behav 60:644–650.
Gomes I, Gupta A, Filipovska J, et al. (2004) A role for heterodimerization of mu and delta
opiate receptors in enhancing morphine analgesia. Proc Natl Acad Sci U S A 101:5135–9.
Gomes I, Jordan BA, Gupta A, et al. (2000) Heterodimerization of mu and delta opioid
receptors : A role in opiate synergy. J. Neurosci. 20:RC110.
Goudriaan AE, Yücel M, van Holst RJ (2014) Getting a grip on problem gambling: what can
neuroscience tell us? Front Behav Neurosci. doi: 10.3389/fnbeh.2014.00141
155
Gozzi A, Lepore S, Vicentini E, et al. (2013) Differential effect of orexin-1 and CRF-1
antagonism on stress circuits: a fMRI study in the rat with the pharmacological stressor
Yohimbine. Neuropsychopharmacology 38:2120–30.
Grant JE, Kim SW, Odlaug BL (2009) A double-blind, placebo-controlled study of the opiate
antagonist, naltrexone, in the treatment of kleptomania. Biol Psychiatry 65:600–606.
Greenhouse S, Geisser S (1959) On methods in the analysis of profile data. Psychometrika
24:95–112.
Greenwald MK, Lundahl LH, Steinmiller CL (2013) Yohimbine increases opioid-seeking
behavior in heroin-dependent, buprenorphine-maintained individuals. Psychopharmacology
(Berl) 225:811–24.
Haack KKV, McCarty N a. (2011) Functional consequences of GPCR heterodimerization:
GPCRs as allosteric modulators. Pharmaceuticals 4:509–523.
Haass-Koffler CL, Bartlett SE (2012) Stress and addiction: contribution of the corticotropin
releasing factor (CRF) system in neuroplasticity. Front. Mol. Neurosci. 5:91
doi/10.3389/fnmol.2012.00091
Hamilton KR, Ansell EB, Reynolds B, et al. (2013) Self-reported impulsivity, but not behavioral
choice or response impulsivity, partially mediates the effect of stress on drinking behavior.
Stress 16:3–15.
Harrison AA, Everitt BJ, Robbins TW (1997) Central 5-HT depletion enhances impulsive
responding without affecting the accuracy of attentional performance: interactions with
dopaminergic mechanisms. Psychopharmacology (Berl) 133:329–342.
Hayaki J, Stein MD, Lassor JA, et al. (2005) Adversity among drug users: relationship to
impulsivity. Drug Alcohol Depend 78:65–71.
Hayes DJ, Jupp B, Sawiak SJ, et al. (2014) Brain γ-aminobutyric acid: a neglected role in
impulsivity. Eur J Neurosci. doi: 10.1111/ejn.12485
Hayton SJ, Lovett-Barron M, Dumont EC, Olmstead MC (2010) Target-specific encoding of
response inhibition: increased contribution of AMPA to NMDA receptors at excitatory
synapses in the prefrontal cortex. J Neurosci 30:11493–11500.
Hayton SJ, Maracle AC, Olmstead MC (2012) Opposite effects of amphetamine on impulsive
action with fixed and variable delays to respond. Neuropsychopharmacology 37:651–659.
Hayton SJ, Olmstead MC (2009) Fractionating animal models of motor impulsivity: reconciling
the neurochemistry of disinhibition. In: Granon S (ed) Endophenotypes of psychiatric and
neurodegenerative disorders in animal models. Transworld Research Network, Kerala India.
156
Herman JP, James P (2012) Neural pathways of stress integration: relevance to alcohol abuse.
Alcohol Res 34:441–447.
Heyman GM, Gibb SP (2006) Delay discounting in college cigarette chippers. Behav Pharmacol
17:669–679.
Izquierdo A, Jentsch JD (2012) Reversal learning as a measure of impulsive and compulsive
behavior in addictions. Psychopharmacology (Berl) 219:607–620.
Jenstch JD, Ashenhurst JR, Cervantes MC, et al. (2014) Dissecting impulsivity and its
relationships to drug addictions. Ann N Y Acad Sci. doi: 10.1111/nyas.12388
Jentsch JD, Pennington ZT (2014) Reward, interrupted: inhibitory control and its relevance to
addictions. Neuropharmacology 76:479–86.
Jentsch JD, Taylor JR (1999) Impulsivity resulting from frontostriatal dysfunction in drug abuse:
implications for the control of behavior by reward-related stimuli. Psychopharmacology
(Berl) 146:373–390.
Jiang Z, North R (1992) Pre- and postsynaptic inhibition by opioids in rat striatum. J Neurosci
12:356–361.
Johnson PL, Truitt W, Fitz SD, et al. (2010) A key role for orexin in panic anxiety. Nat Med
16:111–5.
Johnson S, North R (1992) Two types of neurone in the rat ventral tegmental area and their
synaptic inputs. J Physiol 450:455–468.
Johnston AL, File SE (1989) Yohimbine’s anxiogenic action: evidence for noradrenergic and
dopaminergic sites. Pharmacol Biochem Behav 32:151–6.
Joos L, Goudriaan AE, Schmaal L, et al. (2013) The relationship between impulsivity and
craving in alcohol dependent patients. Psychopharmacology (Berl) 226:273–83.
Jupp B, Caprioli D, Dalley J (2013a) Highly impulsive rats: modelling an endophenotype to
determine the neurobiological, genetic and environmental mechanisms of addiction. Dis
Model Mech 6:302–311.
Jupp B, Caprioli D, Saigal N, et al. (2013b) Dopaminergic and GABA-ergic markers of
impulsivity in rats: evidence for anatomical localisation in ventral striatum and prefrontal
cortex. Eur J Neurosci 37:1519–28.
Jupp B, Dalley JW (2014) Convergent pharmacological mechanisms in impulsivity and
addiction: insights from rodent models. Br J Pharmacol. doi: 10.1111/bph.12787
157
Jutkiewicz E (2006) The antidepressant -like effects of delta-opioid receptor agonists. Mol Interv
6:162–169.
Kabli N, Martin N, Fan T, et al. (2010) Agonists at the δ-opioid receptor modify the binding of
µ-receptor agonists to the µ-δ receptor hetero-oligomer. Br J Pharmacol 161:1122–36. d
Kalivas PW, Volkow N, Seamans J (2005) Unmanageable motivation in addiction: a pathology
in prefrontal-accumbens glutamate transmission. Neuron 45:647–50.
Karlsgodt KH, Lukas SE, Elman I (2003) Psychosocial stress and the duration of cocaine use in
non-treatment seeking individuals with cocaine dependence. Am J Drug Alcohol Abuse
29:539–551.
Kim S (1998) Opioid antagonists in the treatment of impulse-control disorders. J Clin Psychiatry
59:159–164.
Kirby KN, Petry NM (2004) Heroin and cocaine abusers have higher discount rates for delayed
rewards than alcoholics or non-drug-using controls. Addiction 99:461–471.
Klenowski P, Morgan M, Bartlett SE (2014) The role of delta opioid receptors in learning and
memory underlying the development of addiction. Br J Pharmacol. doi: 10.1111/bph.12618
Kollins SH (2003) Delay discounting is associated with substance use in college students. Addict
Behav 28:1167–1173.
Koob GF (2008) A role for brain stress systems in addiction. Neuron 59:11–34.
Koob GF (2009) Neurobiological substrates for the dark side of compulsivity in addiction.
Neuropharmacology 56 Suppl 1:18–31.
Koob GF, Kreek MJ (2007) Stress, dysregulation of drug reward pathways, and the transition to
drug dependence. Am J Psychiatry 164:1149.
Koob GF, Le Moal M (2008) Addiction and the brain antireward system. Annu Rev Psychol
59:29–53.
Koob GF, Volkow ND (2010) Neurocircuitry of addiction. Neuropsychopharmacology 35:217–
238.
Koopmans G, Blokland A, Vannieuwenhuijzen P, Prickaerts J (2003) Assessment of spatial
learning abilities of mice in a new circular maze. Physiol Behav 79:683–693.
Koskinen T, Haapalinna A, Sirviö J (2003) Alpha-adrenoceptor-mediated modulation of 5-HT2
receptor agonist induced impulsive responding in a 5-choice serial reaction time task.
Pharmacol Toxicol 92:214–225.
158
Kramer TJ, Rilling M (1970) Differential reinforcement of low rates: a selective critique.
Psychol Bull 74:225.
Kreek MJ, Koob GF (1998) Drug dependence: stress and dysregulation of brain reward
pathways. Drug Alcohol Depend 51:23–47.
Kreek MJ, Nielsen DA, Butelman ER, LaForge KS (2005) Genetic influences on impulsivity,
risk taking, stress responsivity and vulnerability to drug abuse and addiction. Nat Neurosci
8:1450–1457.
Kupchik YM, Scofield MD, Rice KC, et al. (2014) Cocaine dysregulates opioid gating of GABA
neurotransmission in the ventral pallidum. J Neurosci 34:1057–66.
Lakso M, Sauer B, Mosinger B, et al. (1992) Targeted oncogene activation by site-specific
recombination in transgenic mice. Proc Natl Acad Sci U S A 89:6232–6236.
Laurent V, Bertran-Gonzalez J, Chieng BC, Balleine BW (2014) Δ-Opioid and dopaminergic
processes in accumbens shell modulate the cholinergic control of predictive learning and
choice. J Neurosci 34:1358–69.
Lawrence AJ, Cowen MS, Yang H-J, et al. (2006) The orexin system regulates alcohol-seeking
in rats. Br J Pharmacol 148:752–759.
Lê AD, Funk D, Harding S, et al. (2009) The role of noradrenaline and 5-hydroxytryptamine in
yohimbine-induced increases in alcohol-seeking in rats. Psychopharmacology (Berl)
204:477–488.
Lê AD, Funk D, Juzytsch W, et al. (2011) Effect of prazosin and guanfacine on stress-induced
reinstatement of alcohol and food seeking in rats. Psychopharmacology (Berl) 218:89–99.
Le Merrer J, Becker JAJ, Befort K, et al. (2009) Reward processing by the opioid system in the
brain. Physiol Rev 89:1379–1412.
Le Merrer J, Faget L, Matifas A (2012) Cues predicting drug or food reward restore morphineinduced place conditioning in mice lacking delta opioid receptors. Psychopharmacology
(Berl) 223:99–106.
Le Merrer J, Rezai X, Scherrer G, et al. (2013) Impaired hippocampus-dependent and facilitated
striatum-dependent behaviors in mice lacking the δ opioid receptor.
Neuropsychopharmacology 38:1050–9.
Le Moine C, Kieffer B, Gaveriaux-Ruff C, et al. (1994) Delta-opioid receptor gene expression in
the mouse forebrain: Localization in cholinergic neurons of the striatum. Neuroscience
62:635–40.
159
Le Roy I, Pothion S, Mortaud S, et al. (2000) Loss of aggression, after transfer onto a C57BL/6J
background, in mice carrying a targeted disruption of the neuronal nitric oxide synthase
gene. Behav Genet 30:367–73.
Lesch K-P, Selch S, Renner TJ, et al. (2011) Genome-wide copy number variation analysis in
attention-deficit/hyperactivity disorder: association with neuropeptide Y gene dosage in an
extended pedigree. Mol Psychiatry 16:491–503.
Li Z, Zharikova A, Vaughan CH, et al. (2010) Intermittent high-dose ethanol exposures increase
motivation for operant ethanol self-administration: Possible neurochemical mechanism.
Brain Res 1310:142–153.
Liston C, McEwen BS, Casey BJ (2009) Psychosocial stress reversibly disrupts prefrontal
processing and attentional control. Proc Natl Acad Sci U S A 106:912–7.
Liston C, Miller MM, Goldwater DS, et al. (2006) Stress-induced alterations in prefrontal
cortical dendritic morphology predict selective impairments in perceptual attentional setshifting. J Neurosci 26:7870–7874.
Liu J, Ghattas I, Liu S, et al. (1997) Dlx genes encode DNA-binding proteins that are expressed
in an overlapping and sequential pattern during basal ganglia differentiation. Dev Dyn
210:498–512.
Logan GD (1994) On the ability to inhibit thought and action: A users’ guide to the stop signal
paradigm. In: Dagenbach D, Carr T (eds) Inhib. Process. attention, Mem. Lang. Academic
Press, San Diego, pp 189–239
Logan GD, Cowan WB (1984) On the ability to inhibit thought and action: a theory of an act of
control. Psychol Rev 91:295–327.
Loos M, Pattij T, Janssen MCW, et al. (2010) Dopamine receptor D1/D5 gene expression in the
medial prefrontal cortex predicts impulsive choice in rats. Cereb Cortex 20:1064–1070.
Lord JA, Waterfield A, Hughes J, Kosterlitz H (1977) Endogenous opioid peptides: multiple
agonists and receptors. Nature 267:495–9.
Love TM, Stohler CS, Zubieta J (2009) Positron emission tomography measures of endogenous
opioid neurotransmission and impulsiveness traits in humans. Arch Gen Psychiatry
66:1124–34.
Lovic V, Saunders BT, Yager LM, Robinson TE (2011) Rats prone to attribute incentive salience
to reward cues are also prone to impulsive action. Behav Brain Res 223:255–261.
Lu L, Shepard JD, Scott Hall F, Shaham Y (2003) Effect of environmental stressors on opiate
and psychostimulant reinforcement, reinstatement and discrimination in rats: a review.
Neurosci Biobehav Rev 27:457–491.
160
Lutz P-E, Kieffer BL (2013) Opioid receptors: distinct roles in mood disorders. Trends Neurosci
36:195–206.
Ma CL, Qi XL, Peng JY, Li BM (2003) Selective deficit in no-go performance induced by
blockade of prefrontal cortical a2-adrenoceptors in monkeys. Neuroreport 14:1013.
Madden GJ, Petry NM, Badger GJ, Bickel WK (1997) Impulsive and self-control choices in
opioid-dependent patients and non-drug-using control participants: drug and monetary
rewards. Exp Clin Psychopharmacol 5:256–62.
Mahoney MK, Silveira MM, Olmstead MC (2013) Increased impulsive action in rats: effects of
morphine in a short and long fixed-delay response inhibition task. Psychopharmacology
(Berl) 230:569–77.
Mansour A, Fox C, Akil H, Watson S (1995) Opioid-receptor mRNA expression in the rat CNS:
anatomical and functional implications. Trends Neurosci 18:22–29.
Marinelli PW, Funk D, Juzytsch W, et al. (2007) The CRF1 receptor antagonist antalarmin
attenuates yohimbine-induced increases in operant alcohol self-administration and
reinstatement of alcohol seeking in rats. Psychopharmacology (Berl) 195:345–355.
Martin G, Nie Z, Siggins G (1997) mu-Opioid receptors modulate NMDA receptor-mediated
responses in nucleus accumbens neurons. J Neurosci 17:11–22.
Martin W, Eades C, Thompson J, et al. (1976) The effects of morphine- and nalorphine- like
drugs in the nondependent and morphine-dependent chronic spinal dog. J Pharmacol Exp
Ther 197:517–532.
Matell MS, Meck WH (2004) Cortico-striatal circuits and interval timing: coincidence detection
of oscillatory processes. Brain Res Cogn Brain Res 21:139–70.
Mather M, Lighthall NR (2012) Both risk and reward are processed differently in decisions made
under stress. Curr Dir Psychol Sci 21:36–41.
Matthes H, Maldonado R, Simonin F, et al. (1996) Loss of morphine-induced analgesia, reward
effect and withdrawal symptoms in mice lacking the mu-opioid-receptor gene. Nature
383:819–823.
McClure GY, Wenger GR, McMillan DE (1997) Effects of drugs on response duration
differentiation. V: differential effects under temporal response differentiation schedules. J
Pharmacol Exp Ther 281:1357–1367.
McEwen BS (2008) Central effects of stress hormones in health and disease: understanding the
protective and damaging effects of stress and stress mediators. Eur J Pharmacol 583:174–
185.
161
Meck WH (1983) Selective adjustment of the speed of internal clock and memory processes. J
Exp Psychol Anim Behav Process 9:171–201.
Meck WH, Penney TB, Pouthas V (2008) Cortico-striatal representation of time in animals and
humans. Curr Opin Neurobiol 18:145–52.
Milan-Lobo L, Whistler JL (2011) Heteromerization of the μ- and δ-opioid receptors produces
ligand-biased antagonism and alters μ-receptor trafficking. J Pharmacol Exp Ther 337:868–
75.
Millan MJ, Newman-Tancredi A, Audinot V, et al. (2000) Agonist and antagonist actions of
yohimbine as compared to fluparoxan at alpha(2)-adrenergic receptors (AR)s, serotonin (5HT)(1A), 5-HT(1B), 5-HT(1D) and dopamine D(2) and D(3) receptors. Significance for the
modulation of frontocortical monoaminergic transmission and depressive states. Synapse
35:79–95.
Milstein JA, Dalley JW, Robbins TW (2010) Methylphenidate-induced impulsivity:
Pharmacological antagonism by beta-adrenoreceptor blockade. J Psychopharmacol 24:309–
21.
Mitchell JM, Tavares VC, Fields HL, et al. (2007) Endogenous opioid blockade and impulsive
responding in alcoholics and healthy controls. Neuropsychopharmacology 32:439–449.
Mitchell MR, Vokes CM, Blankenship AL, et al. (2011) Effects of acute administration of
nicotine, amphetamine, diazepam, morphine, and ethanol on risky decision-making in rats.
Psychopharmacology (Berl) 218:703–712.
Molander AC, Mar AC, Norbury A, et al. (2011) High impulsivity predicting vulnerability to
cocaine addiction in rats: some relationship with novelty preference but not novelty
reactivity, anxiety or stress. Psychopharmacology (Berl) 215:721–31.
Monory K, Massa F, Egertová M, et al. (2006) The endocannabinoid system controls key
epileptogenic circuits in the hippocampus. Neuron 51:455–466.
Moran-Santa Maria MM, McRae-Clark A, Baker NL, et al. (2014) Yohimbine administration
and cue-reactivity in cocaine-dependent individuals. Psychopharmacology (Berl). doi:
10.1007/s00213-014-3555-9
Moreno M, Economidou D, Mar AC, et al. (2013) Divergent effects of D(2/3) receptor activation
in the nucleus accumbens core and shell on impulsivity and locomotor activity in high and
low impulsive rats. Psychopharmacology (Berl) 228:19–30.
Murphy ER, Fernando ABP, Urcelay GP, et al. (2012) Impulsive behaviour induced by both
NMDA receptor antagonism and GABAA receptor activation in rat ventromedial prefrontal
cortex. Psychopharmacology (Berl) 219:401–410.
162
Nagy A (2000) Cre recombinase: the universal reagent for genome tailoring. Genesis 26:99–109.
Newman JP, Widom CS, Nathan S (1985) Passive avoidance in syndromes of disinhibition:
psychopathy and extraversion. J Pers Soc Psychol 48:1316–1327.
Nielsen CKC, Simms JJA, Bito-Onon JJ, et al. (2012) The delta opioid receptor antagonist,
SoRI‐9409, decreases yohimbine stress-induced reinstatement of ethanol-seeking. Addict
Biol 17:224–234.
Nieto M, Guen S, Kieffer B, et al. (2005) Physiological control of emotion-related behaviors by
endogenous enkephalins involves essentially the delta opioid receptors. Neuroscience
135:305–313.
Nigg JT, Wong MM, Martel MM, et al. (2006) Poor response inhibition as a predictor of
problem drinking and illicit drug use in adolescents at risk for alcoholism and other
substance use disorders. J Am Acad Child Adolesc Psychiatry 45:468–75.
Oliver YP, Ripley TL, Stephens DN (2009) Ethanol effects on impulsivity in two mouse strains:
similarities to diazepam and ketamine. Psychopharmacology (Berl) 204:679–92.
Olmstead MC (2006) Animal models of drug addiction: Where do we go from here? Q J Exp
Psychol 59:625–653.
Olmstead MC, Ouagazzal AM, Kieffer BL (2009) Mu and delta opioid receptors oppositely
regulate motor impulsivity in the signaled nose poke task. PLoS One 4:e4410.
Olshavsky ME, Shumake J, Rosenthal AA, et al. (2014) Impulsivity, risk-taking, and
distractibility in ratsexhibiting robust conditioned orienting behaviors. J Exp Anal Behav
102:162–78.
Orban PC, Chui D, Marth JD (1992) Tissue- and site-specific DNA recombination in transgenic
mice. Proc Natl Acad Sci U S A 89:6861–6865.
Orman JS, Keating GM (2009) Buprenorphine/naloxone: a review of its use in the treatment of
opioid dependence. Drugs 69:577–607.
Paine TA, Slipp LE, Carlezon WA (2011) Schizophrenia-like attentional deficits following
blockade of prefrontal cortex GABAA receptors. Neuropsychopharmacology 36:1703–13.
Paine TA, Tomasiewicz HC, Zhang K, Carlezon WA (2007) Sensitivity of the five-choice serial
reaction time task to the effects of various psychotropic drugs in Sprague-Dawley rats. Biol
Psychiatry 62:687–93.
Pathan H, Williams J (2012) Basic opioid pharmacology: an update. Br J Pain 6:11–16.
163
Pattij T, De Vries TJ (2013) The role of impulsivity in relapse vulnerability. Curr Opin
Neurobiol 23:700–5.
Pattij T, Janssen MCW, Schepers I, et al. (2007a) Effects of the cannabinoid CB1 receptor
antagonist rimonabant on distinct measures of impulsive behavior in rats.
Psychopharmacology (Berl) 193:85–96.
Pattij T, Janssen MCW, Vanderschuren LJMJ, et al. (2007b) Involvement of dopamine D1 and
D2 receptors in the nucleus accumbens core and shell in inhibitory response control.
Psychopharmacology (Berl) 191:587–98. =
Pattij T, Schetters D, Janssen MCW, et al. (2009) Acute effects of morphine on distinct forms of
impulsive behavior in rats. Psychopharmacology (Berl) 205:489–502.
Pattij T, Schetters D, Schoffelmeer ANM, et al. (2012) On the improvement of inhibitory
response control and visuospatial attention by indirect and direct adrenoceptor agonists.
Psychopharmacology (Berl) 219:327–340.
Pattij T, Vanderschuren LJMJ (2008) The neuropharmacology of impulsive behaviour. Trends
Pharmacol Sci 29:192–199.
Peciña S, Berridge KC (2013) Dopamine or opioid stimulation of nucleus accumbens similarly
amplify cue-triggered “wanting” for reward: entire core and medial shell mapped as
substrates for PIT enhancement. Eur J Neurosci 37:1529–40.
Perrine S, Hoshaw B, Unterwald E (2006) Delta opioid receptor ligands modulate anxiety-like
behaviors in the rat. Br J Pharmacol 147:864–872.
Perrine S, Sheikh I, Nwaneshiudu C, et al. (2008) Withdrawal from chronic administration of
cocaine decreases delta opioid receptor signaling and increases anxiety- and depression-like
behaviors in the rat. Neuropharmacology 54:355–364.
Perry J, Carroll M (2008) The role of impulsive behavior in drug abuse. Psychopharmacology
(Berl) 200:1–26.
Perry JL, Larson EB, German JP, et al. (2005) Impulsivity (delay discounting) as a predictor of
acquisition of IV cocaine self-administration in female rats. Psychopharmacology (Berl)
178:193–201.
Perry JL, Nelson SE, Carroll ME (2008) Impulsive choice as a predictor of acquisition of IV
cocaine self- administration and reinstatement of cocaine-seeking behavior in male and
female rats. Exp Clin Psychopharmacol 16:165–177.
Pezze M, Dalley JW, Robbins TW (2007) Differential roles of dopamine D1 and D2 receptors in
the nucleus accumbens in attentional performance on the five-choice serial reaction time
task. Neuropsychopharmacology 32:273–83.
164
Pezze M, McGarrity S, Mason R, et al. (2014) Too little and too much: hypoactivation and
disinhibition of medial prefrontal cortex cause attentional deficits. J Neurosci 34:7931–
7946.
Pezze MA, Dalley JW, Robbins TW (2009) Remediation of attentional dysfunction in rats with
lesions of the medial prefrontal cortex by intra-accumbens administration of the dopamine
D(2/3) receptor antagonist sulpiride. Psychopharmacology (Berl) 202:307–13.
Piazza PV, Deminière J, Maccari S, et al. (1990) Individual reactivity to novelty predicts
probability of amphetamine self-administration. Behav Pharmacol 1:339–345.
Piazza PV, Deminiere J-M, Le Moal ML, Simon H (1989) Factors that predict individual
vulnerability to amphetamine self-administration. Science 245:1511–1513.
Piazza PV, Le Moal ML (1998) The role of stress in drug self-administration. Trends Pharmacol
Sci 19:67–74.
Piskorowski R, Chevaleyre V (2013) Delta-opioid receptors mediate unique plasticity onto
parvalbumin-expressing interneurons in area CA2 of the hippocampus. J Neurosci
33:14567–78.
Pothuizen H, Jongen-Rêlo A, Feldon J, Yee B (2005) Double dissociation of the effects of
selective nucleus accumbens core and shell lesions on impulsive-choice behaviour and
salience learning in rats. Eur J Neurosci 22:2605–2616.
Pozzi L, Baviera M, Sacchetti G, et al. (2011) Attention deficit induced by blockade of N-methyl
d-aspartate receptors in the prefrontal cortex is associated with enhanced glutamate release
and cAMP response element binding protein phosphorylation: role of metabotropic
glutamate receptors 2/3. Neuroscience 176:336–348.
Pradhan AA, Befort K, Nozaki C, et al. (2011) The delta opioid receptor: an evolving target for
the treatment of brain disorders. Trends Pharmacol Sci 32:581–590.
Puig MV, Gulledge AT (2011) Serotonin and prefrontal cortex function: neurons, networks, and
circuits. Mol Neurobiol 44:449–64.
Quednow BB, Csomor PA, Chmiel J, et al. (2008) Sensorimotor gating and attentional setshifting are improved by the mu-opioid receptor agonist morphine in healthy human
volunteers. Int J Neuropsychopharmacol 11:655–669.
Regier PS, Claxton AB, Zlebnik NE, Carroll ME (2014) Cocaine-, caffeine-, and stress-evoked
cocaine reinstatement in high vs. low impulsive rats: treatment with allopregnanolone. Drug
Alcohol Depend. doi: 10.1016/j.drugalcdep.2014.07.001
Rezaï X, Faget L, Bednarek E, et al. (2012) Mouse δ opioid receptors are located on presynaptic
afferents to hippocampal pyramidal cells. Cell Mol Neurobiol 32:509–16.
165
Richards JB, Lloyd DR, Kuehlewind B, et al. (2013) Strong genetic influences on measures of
behavioral-regulation among inbred rat strains. Genes Brain Behav 12:490–502.
Robbins TW (2002) The 5-choice serial reaction time task: Behavioural pharmacology and
functional neurochemistry. Psychopharmacology (Berl) 163:362–380.
Robinson ESJ (2012) Blockade of noradrenaline re-uptake sites improves accuracy and impulse
control in rats performing a five-choice serial reaction time tasks. Psychopharmacology
(Berl) 219:303–312.
Robinson ESJ, Dalley JW, Theobald DEH, et al. (2008) Opposing roles for 5-HT2A and 5-HT2C
receptors in the nucleus accumbens on inhibitory response control in the 5-choice serial
reaction time task. Neuropsychopharmacology 33:2398–406.
Robinson ESJ, Eagle DM, Economidou D, et al. (2009) Behavioural characterisation of high
impulsivity on the 5-choice serial reaction time task: specific deficits in “waiting” versus
“stopping.” Behav Brain Res 196:310–316.
Rogers R, Baunez C, Everitt BJ, Robbins TW (2001) Lesions of the medial and lateral striatum
in the rat produce differential deficits in attentional performance. Behav neurocience
115:799–811.
Romer D, Betancourt L, Giannetta JM, et al. (2009) Executive cognitive functions and
impulsivity as correlates of risk taking and problem behavior in preadolescents.
Neuropsychologia 47:2916–2926.
Rosenberg O, Dinur LK, Dannon PN (2013) Four-year follow-up study of pharmacological
treatment in pathological gamblers. Clin Neuropharmacol 36:42–5.
Ross EL, Yoon JH, Mahoney JJ, et al. (2013) The impact of self-reported life stress on current
impulsivity in cocaine dependent adults. Prog Neuropsychopharmacol Biol Psychiatry
46:113–9.
Roychowdhury S, Peña-Contreras Z, Tam J, et al. (2012) Α₂- and Β-adrenoceptors involvement
in nortriptyline modulation of auditory sustained attention and impulsivity.
Psychopharmacology (Berl) 222:237–45.
Saal D, Dong Y, Bonci A, Malenka RC (2003) Drugs of abuse and stress trigger a common
synaptic adaptation in dopamine neurons. Neuron 37:577–582.
Sarnyai Z, Shaham Y, Heinrichs SC (2001) The role of corticotropin-releasing factor in drug
addiction. Pharmacol Rev 53:209–243.
Saunders BT, Robinson TE (2013) Individual variation in resisting temptation: implications for
addiction. Neurosci Biobehav Rev 37:1955–75.
166
Saunders BT, Robinson TE (2010) A cocaine cue acts as an incentive stimulus in some but not
others: implications for addiction. Biol Psychiatry 67:730–6.
Schank JR, Ryabinin AE, Giardino WJ, et al. (2012) Stress-related neuropeptides and addictive
behaviors: Beyond the usual suspects. Neuron 76:192–208.
Schwager A, Haack A, Taha S (2014) Impaired flexibility in decision making in rats after
administration of the pharmacological stressor yohimbine. Psychopharmacology (Berl).
Semenova S, Markou A (2007) The effects of the mGluR5 antagonist MPEP and the mGluR2/3
antagonist LY341495 on rats’ performance in the 5-choice serial reaction time task.
Neuropharmacology 52:863–72.
Sesack SR, Grace AA (2010) Cortico-basal ganglia reward network: microcircuitry.
Neuropsychopharmacology 35:27–47.
Shaham Y, Rajabi H, Stewart J (1996) Relapse to heroin-seeking in rats under opioid
maintenance: the effects of stress, heroin priming, and withdrawal. J Neurosci 16:1957–
1963.
Shalev U, Erb S, Shaham Y (2010) Role of CRF and other neuropeptides in stress-induced
reinstatement of drug seeking. Brain Res 1314:15–28.
Shimp KG, Mitchell MR, Beas S, et al. (2014) Affective and cognitive mechanisms of risky
decision making. Neurobiol Learn Mem. doi: 10.1016/j.nlm.2014.03.002
Shin R, Cao J, Webb SM, Ikemoto S (2010) Amphetamine administration into the ventral
striatum facilitates behavioral interaction with unconditioned visual signals in rats. PLoS
One 5:e8741. doi: 10.1371/journal.pone.0008741
Simms JA, Haass-Koffler CL, Bito-Onon J, et al. (2012) Mifepristone in the central nucleus of
the amygdala reduces yohimbine stress-induced reinstatement of ethanol-seeking.
Neuropsychopharmacology 37:906–918. doi: 10.1038/npp.2011.268
Simon NW, Gilbert RJ, Mayse JD, et al. (2009) Balancing risk and reward: a rat model of risky
decision making. Neuropsychopharmacology 34:2208–2217.
Simonin F, Gavériaux-Ruff C, Befort K, et al. (1995) kappa-Opioid receptor in humans: cDNA
and genomic cloning, chromosomal assignment, functional expression, pharmacology, and
expression pattern in the central nervous system. Proc Natl Acad Sci U S A 92:7006–7010.
Singh T, Mcdannald MA, Takahashi YK, et al. (2011) The role of the nucleus accumbens in
knowing when to respond. Learn Mem (Cold Spring Harb NY) 18:85–87.
Sinha R (2008) Chronic stress, drug use, and vulnerability to addiction. Ann N Y Acad Sci
1141:105–130.
167
Sinha R, Catapano D, O’Malley S (1999) Stress-induced craving and stress response in cocaine
dependent individuals. Psychopharmacology (Berl) 142:343–51.
Sinha R, Garcia M, Paliwal P, et al. (2006) Stress-induced cocaine craving and hypothalamicpituitary-adrenal responses are predictive of cocaine relapse outcomes. Arch Gen Psychiatry
63:324–331.
Sirviö J, Mazurkiewicz M, Haapalinna A, et al. (1994) The effects of selective alpha-2
adrenergic agents on the performance of rats in a 5-choice serial reaction time task. Brain
Res Bull 35:451–455.
Smith DR, Gallagher M, Stanton ME (2007) Genetic background differences and nonassociative
effects in mouse trace fear conditioning. Learn Mem 14:597–605.
Solanto M V, Abikoff H, Sonuga-Barke E, et al. (2001) The ecological validity of delay aversion
and response inhibition as measures of impulsivity in AD/HD: a supplement to the NIMH
multimodal treatment study of AD/HD. J Abnorm Child Psychol 29:215–28.
Spanagel R, Herz A, Shippenberg TS (1990) The effects of opioid peptides on dopamine release
in the nucleus accumbens: An in vivo microdialysis study. J Neurochem 55:1734–1740.
Spina L, Longoni R, Mulas A, et al. (1998) Dopamine-dependent behavioural stimulation by
non-peptide delta opioids BW373U86 and SNC 80: 1. Locomotion, rearing and stereotypies
in intact rats. Behav Pharmacol 9:1–8.
St Onge JR, Floresco SB (2009) Dopaminergic modulation of risk-based decision making.
Neuropsychopharmacology 34:681–97.
Stalnaker TA, Takahashi Y, Roesch MR, Schoenbaum G (2009) Neural substrates of cognitive
inflexibility after chronic cocaine exposure. Neuropharmacology 56 Suppl 1:63–72.
Starcke K, Brand M (2012) Decision making under stress: a selective review. Neurosci Biobehav
Rev 36:1228–48.
Stine SM, Southwick SM, Petrakis IL, et al. (2002) Yohimbine-induced withdrawal and anxiety
symptoms in opioid-dependent patients. Biol Psychiatry 51:642–651.
Stock D, Ellies D, Zhao Z, et al. (1996) The evolution of the vertebrate Dlx gene family. Proc
Natl Acad Sci U S A 93:10858–10863.
Stühmer T, Puelles L, Ekker M, Rubenstein J (2002) Expression from a Dlx gene enhancer
marks adult mouse cortical GABAergic neurons. Cereb Cortex 12:75–85.
Sun H, Green TA, Theobald DEH, et al. (2010) Yohimbine increases impulsivity through
activation of cAMP response element binding in the orbitofrontal cortex. Biol Psychiatry
67:649–656.
168
Sutton M, Karanian D, Self D (2000) Factors that determine a propensity for cocaine-seeking
behavior during abstinence in rats. Neuropsychopharmacology 22:626–641.
Svoboda KR, Adams CE, Lupica CR (1999) Opioid receptor subtype expression defines
morphologically distinct classes of hippocampal interneurons. J Neurosci 19:85–95.
Swann AC, Birnbaum D, Jagar AA, et al. (2005) Acute yohimbine increases laboratorymeasured impulsivity in normal subjects. Biol Psychiatry 57:1209–1211.
Swerdlow NR, Caine SB, Geyer MA (1991) Opiate-dopamine interactions in the neural
substrates of acoustic startle gating in the rat. Prog Neuropsychopharmacol Biol psychiatry
15:415–426.
Tejedor-Real P, Mico J a, Maldonado R, et al. (1995) Implication of endogenous opioid system
in the learned helplessness model of depression. Pharmacol Biochem Behav 52:145–52.
Tomie A, Brooks W, Zito B (1989) Sign-tracking: the search for reward. In: Kein S, MowrerR
(eds) Contemporary Learning Theories - Pavlovian Conditioning and the Status of
Traditional Learning Theory. Lawrence Erlbaum Associates, Hillsdale, NJ, pp 191–226.
Tomie A, Grimes K, Pohorecky LA (2008) Behavioral characteristics and neurobiological
substrates shared by pavlovian sign-tracking and drug abuse. Brain Res Rev 58:121–135.
Tziortzis D, Mahoney JJ, Kalechstein AD, et al. (2011) The relationship between impulsivity and
craving in cocaine- and methamphetamine-dependent volunteers. Pharmacol Biochem
Behav 98:196–202.
Ukai M, Okuda A (2003) Endomorphin-1, an endogenous μ-opioid receptor agonist, improves
apomorphine-induced impairment of prepulse inhibition in mice. Peptides 24:741–744.
Umhau JC, Schwandt ML, Usala J, et al. (2011) Pharmacologically induced alcohol craving in
treatment seeking alcoholics correlates with alcoholism severity, but is insensitive to
acamprosate. Neuropsychopharmacology 36:1178–86.
Ungless MA, Argilli E, Bonci A (2010) Effects of stress and aversion on dopamine neurons:
implications for addiction. Neurosci Biobehav Rev 35:151–6.
Urcelay GP, Dalley JW (2012) Linking ADHD, impulsivity, and drug abuse: a
neuropsychological perspective. Curr Top Behav Neurosci 9:173–197.
Van Gaalen MM, Brueggeman RJ, Bronius PFC, et al. (2006) Behavioral disinhibition requires
dopamine receptor activation. Psychopharmacology (Berl) 187:73–85.
Van Holst RJ, van den Brink W, Veltman DJ, Goudriaan AE (2010) Why gamblers fail to win: a
review of cognitive and neuroimaging findings in pathological gambling. Neurosci
Biobehav Rev 34:87–107.
169
Vanhille N, Belin-Rauscent A, Mar AC, et al. (2014) High locomotor reactivity to novelty is
associated with an increased propensity to choose saccharin over cocaine: new insights into
the vulnerability to addiction. Neuropsychopharmacology. doi: 10.1038/npp.2014.204
Vendruscolo LF, Barbier E, Schlosburg JE, et al. (2012) Corticosteroid-dependent plasticity
mediates compulsive alcohol drinking in rats. J Neurosci 32:7563–7571.
Verdejo A, Toribio I, Orozco C, et al. (2005) Neuropsychological functioning in methadone
maintenance patients versus abstinent heroin abusers. Drug Alcohol Depend 78:283–288.
Verdejo-García A, Lawrence AJ, Clark L (2008) Impulsivity as a vulnerability marker for
substance-use disorders: Review of findings from high-risk research, problem gamblers and
genetic association studies. Neurosci Biobehav Rev 32:777–810.
Von Diemen L, Bassani DG, Fuchs SC, et al. (2008) Impulsivity, age of first alcohol use and
substance use disorders among male adolescents: a population based case-control study.
Addiction 103:1198–1205.
Wang H, Cuzon VC, Pickel VM (2003) Ultrastructural localization of delta-opioid receptors in
the rat caudate-putamen nucleus during postnatal development: relation to synaptogenesis. J
Comp Neurol 467:343–53.
Weibel R, Reiss D, Karchewski L, et al. (2013) Mu opioid receptors on primary afferent nav1.8
neurons contribute to opiate-induced analgesia: Insight from conditional knockout mice.
PLoS One 8:e74706. doi: 10.1371/journal.pone.0074706
Winstanley CA, Chudasama Y, Dalley JW, et al. (2003) Intra-prefrontal 8-OH-DPAT and
M100907 improve visuospatial attention and decrease impulsivity on the five-choice serial
reaction time task in rats. Psychopharmacology (Berl) 167:304–314.
Winstanley CA, Dalley JW, Theobald DE, Robbins TW (2004) Fractionating impulsivity:
Contrasting effects of central 5-HT depletion on different measures of impulsive behavior.
Neuropsychopharmacology 29:1331–1343.
Winstanley CA, Eagle DM, Robbins TW (2006a) Behavioral models of impulsivity in relation to
ADHD: Translation between clinical and preclinical studies. Clin Psychol Rev 26:379–395.
Winstanley CA, Theobald DEH, Dalley JW, et al. (2006b) Double dissociation between
serotonergic and dopaminergic modulation of medial prefrontal and orbitofrontal cortex
during a test of impulsive choice. Cereb Cortex 16:106–114.
Winstanley CA, Zeeb FD, Bedard A, et al. (2010) Dopaminergic modulation of the orbitofrontal
cortex affects attention, motivation and impulsive responding in rats performing the fivechoice serial reaction time task. Behav Brain Res 210:263–272.
170
Wiskerke J, Schetters D, van Es IE, et al. (2011a) {micro}-opioid receptors in the nucleus
accumbens shell region mediate the effects of amphetamine on inhibitory control but not
impulsive choice. J Neurosci 31:262–272.
Wiskerke J, Stoop N, Schetters D, et al. (2011b) Cannabinoid CB1 receptor activation mediates
the opposing effects of amphetamine on impulsive action and impulsive choice. PLoS One
6:e25856. doi: 10.1371/journal.pone.0025856
Yang Q-Z, Lu S-S, Tian X-Z, et al. (2011) The antidepressant-like effect of human opiorphin via
opioid-dependent pathways in mice. Neurosci Lett 489:131–5. doi:
10.1016/j.neulet.2010.12.002
Yin B, Meck WH (2014) Comparison of interval timing behaviour in mice following dorsal or
ventral hippocampal lesions with mice having δ-opioid receptor gene deletion. Philos Trans
R Soc Lond B Biol Sci. doi: 10.1098/rstb.2012.0466
Zeeb FD, Robbins TW, Winstanley CA (2009) Serotonergic and dopaminergic modulation of
gambling behavior as assessed using a novel rat gambling task. Neuropsychopharmacology
34:2329–2343.
Zerucha T, Stühmer T, Hatch G, et al. (2000) A highly conserved enhancer in the Dlx5/Dlx6
intergenic region is the site of cross-regulatory interactions between Dlx genes in the
embryonic forebrain. J Neurosci 20:709–721.
Zhu J, Spencer T, Kachroo A, et al. (2011) Methylphenidate and μ opioid receptor interactions: a
pharmacological target for prevention of stimulant abuse. Neuropharmacology 61:283–292.
171
Appendix I
Responding in drug-free yohimbine test sessions in the response inhibition task
Behavioural Measure
Dose
(mg/kg)
Premature
Responses (%)
Omissions
Non-reinforced
Lever Presses
Correct Response
Latency (s)
Vehicle
0.625
1.25
2.5
5.0
27.80 ± 5.56
24.81 ± 5.47
22.11 ± 4.87
27.86 ± 4.87
21.79 ± 4.64
0.92 ± 0.29
1.25 ± 0.99
0.33 ± 0.19
1.33 ± 0.51
1.50 ± 0.67
137.00 ± 45.12
148.58 ± 67.74
161.08 ± 69.69
143.58 ± 53.99
203.67 ± 61.64
1.96 ± 0.22
1.76 ± 0.16
1.68 ± 0.13
2.01 ± 0.18
1.79 ± 0.15
Premature Response Latency (s)
Delay (s)
Dose
(mg/kg)
15
30
60
Vehicle
0.625
1.25
2.5
5.0
6.34 ± 0.56
5.28 ± 0.72
6.48 ± 0.35
6.72 ± 0.93
7.82 ± 0.61
12.04 ± 1.20
10.87 ± 1.05
15.15 ± 2.04
10.51 ± 1.38
13.24 ± 0.86
23.11 ± 2.18
21.19 ± 2.11
25.75 ± 2.28
24.45 ± 2.06
25.87 ± 2.26
Values represent mean (SEM) total omissions, total non-reinforced lever presses, and correct response latency
collapsed across delay, and premature response latency across delay. Premature delay latencies for the 1 and 4 s
delay were not analyzed due to minimal responding at these delays. None of the yohimbine doses significantly
differed from Vehicle (0 mg/kg).
172
Appendix II
ANOVA summary table for drug-free yohimbine test sessions
Behavioural Measure
df
F
p
ηp2
Omissions
Dose
4, 44
1.00
0.42
-
Non-reinforced Lever Presses
Dose
4, 44
1.93
0.16
-
Correct Latency
Dose
Delay
Dose x Delay
4, 44
4, 44
16, 176
1.63
9.46
1.20
0.18
< 0.001*
0.32
0.46
-
Premature Latency
Dose
Delay
Dose x Delay
4, 44
2, 22
8, 88
4.13
108.17
0.55
0.06
< 0.001*
0.68
0.91
-
* denotes a significant effect
173
Appendix III
Effects of corticotropin releasing factor, glucocorticoid, adrenergic, and dopaminergic
agents on yohimbine-induced responding on measures of the response inhibition task in
drug-free test sessions
Dose
(mg/kg)
Premature
Responses
(%)
Omissions
Non-reinforced
Lever Presses
Correct
Response
Latency (s)
Veh – Veh
Veh – Yoh
Ant 10 – Yoh
Ant 20 – Yoh
22.28 ± 7.91
25.30 ± 8.52
23.84 ± 8.94
23.06 ± 8.80
2.25 ± 1.07
3.50 ± 1.78
3.13 ± 2.35
2.88 ± 1.70
151.13 ± 62.62
158.88 ± 43.11
115.38 ± 36.68
125.63 ± 36.48
1.89 ± 0.26
1.86 ± 0.27
1.85 ± 0.22
1.90 ± 0.24
Veh – Veh
Veh – Yoh
Ant 5 – Yoh
Ant 15 – Yoh
13.89 ± 4.37
15.34 ± 5.55
21.21 ± 5.07
18.29 ± 4.88
2.43 ± 2.27
1.86 ± 1.06
0.43 ± 0.30
0.86 ± 0.60
62.63 ± 28.34
51.13 ± 19.10
74.75 ± 34.12
70.88 ± 34.83
1.64 ± 0.25
1.69 ± 0.20
1.50 ± 0.25
1.96 ± 0.65
Mifepristone
(GC ant)
Veh – Veh
Veh – Yoh
Mif 7.5 –Yoh
Mif 15 – Yoh
Mif 30 – Yoh
12.13 ± 2.62
11.25 ± 2.07
15.32 ± 3.00
14.04 ± 2.51
14.99 ± 3.51
1.00 ± 0.52
0.50 ± 0.31
0.50 ± 0.31
0.50 ± 0.31
0.80 ± 0.42
19.44 ± 8.35
21.90 ± 13.10
28.78 ± 15.37
17.44 ± 4.62
25.11 ± 9.20
1.78 ± 0.18
1.74 ± 0.17
1.71 ± 0.18
1.89 ± 0.18
1.90 ± 0.21
Prazosin
(α1 adrenergic ant)
Veh – Veh
Veh – Yoh
Pra 0.5 – Yoh
Pra 1 – Yoh
Pra 2 – Yoh
28.51 ± 5.04
26.10 ± 5.25
28.40 ± 5.24
28.02 ± 4.82
25.57 ± 4.90
4.31 ± 2.29
3.06 ± 1.11
2.81 ± 1.26
2.94 ± 0.99
4.31 ± 1.56
109.33 ± 26.95
109.93 ± 26.36
105.53 ± 22.28
106.00 ± 23.29
108.20 ± 24.08
2.00 ± 0.20
2.08 ± 0.16
2.06 ± 0.13
2.27 ± 0.39
2.10 ± 0.11
Guanfacine
(α2 adrenergic ago)
Veh – Veh
Veh – Yoh
Gua 0.125 – Yoh
Gua 0.25 – Yoh
Gua 0.5 – Yoh
40.19 ± 5.16
32.99 ± 5.84
31.87 ± 6.46
31.66 ± 8.40
30.00 ± 5.61
2.75 ± 1.05
2.50 ± 0.73
6.63 ± 4.25
5.00 ± 4.44
2.00 ± 0.98
114.00 ± 32.21
117.25 ± 36.13
107.50 ± 31.35
162.00 ± 63.94
117.63 ± 40.76
2.08 ± 0.36
2.01 ± 0.21
2.11 ± 0.24
1.85 ± 0.27
1.90 ± 0.32
SCH 39166
(D1/5 ant)
Veh – Veh
Veh – Yoh
SCH 0.015 – Yoh
SCH 0.0625 – Yoh
27.80 ± 5.88
27.59 ± 4.97
25.21 ± 4.95
26.13 ± 5.25
1.56 ± 0.76
2.56 ± 1.02
2.25 ± 1.00
2.56 ± 1.13
181.25 ± 53.40
170.00 ± 47.35
169.13 ± 51.21
131.00 ± 34.47
1.78 ± 0.12
1.99 ± 0.16
1.88 ± 0.12
1.87 ± 0.12
Drug
Antalarmin
(CRF1 ant)
Values represent mean (SEM) total omissions, total non-reinforced lever presses, and correct response latency
collapsed across delay for drug-free test sessions of combined-drug experiments. The yohimbine dose was 0 or 2.5
mg/kg; other drug doses are in mg/kg. None of the drug combinations significantly differed from their respective
Veh-Veh (0 – 0 mg/kg) or Veh-Yoh (0 – 2.5 mg/kg) doses.
Ago, agonist; ant, antagonist; Ant, antalarmin; CRF1, corticotropin releasing factor receptor 1; D 1/5, dopamine 1/5
receptor family; GC, glucocorticoid receptor; Gua, guanfacine, Mif, mifepristone; Pra, prazosin; SCH, SCH 39166
174
Appendix IV
Effects of corticotropin releasing factor, glucocorticoid, adrenergic, and dopaminergic
agents on yohimbine-induced premature response latencies in the response inhibition task
in drug-free test sessions
Premature Delay (s)
Dose
(mg/kg)
15
30
60
Veh – Veh
Veh – Yoh
Ant 10 – Yoh
Ant 20 – Yoh
4.77 ± 0.78
5.93 ± 0.76
6.42 ± 0.68
5.30 ± 0.58
12.36 ± 2.08
12.85 ± 2.49
11.93 ± 1.33
13.90 ± 2.72
28.36 ± 4.10
20.53 ± 3.34
20.22 ± 4.28
22.97 ± 4.20
Veh – Veh
Veh – Yoh
Ant 5 – Yoh
Ant 15 – Yoh
5.28 ± 0.96
7.70 ± 0.39
7.12 ± 0.61
8.40 ± 0.48
12.63 ± 2.52
10.88 ± 1.72
10.93 ± 2.29
13.79 ± 2.21
36.47 ± 4.57
25.53 ± 5.06
28.54 ± 2.78
24.81 ± 3.06
Mifepristone
(GC ant)
Veh – Veh
Veh – Yoh
Mif 7.5 –Yoh
Mif 15 – Yoh
Mif 30 – Yoh
6.35 ± 1.44
7.12 ± 1.15
6.03 ± 0.47
5.00 ± 0.58
4.42 ± 1.28
12.20 ± 1.60
10.38 ±1.69
10.24 ± 2.33
13.48 ± 1.87
10.42 ± 1.56
30.32 ± 2.37
23.25 ± 2.46
24.74 ± 1.82
25.04 ± 2.31
29.27 ± 3.18
Prazosin
(α1 adrenergic ant)
Veh – Veh
Veh – Yoh
Pra 0.5 – Yoh
Pra 1 – Yoh
Pra 2 – Yoh
7.32 ± 0.53
7.06 ± 0.80
7.03 ± 0.70
5.43 ± 0.57
7.30 ± 0.79
14.78 ± 1.19
15.58 ± 1.51
13.32 ± 1.32
13.02 ± 1.62
14.22 ± 1.04
22.98 ± 1.82
23.57 ± 3.00
23.52 ± 3.24
23.67 ± 2.48
24.32 ± 1.89
Guanfacine
(α2 adrenergic ago)
Veh – Veh
Veh – Yoh
Gua 0.125 – Yoh
Gua 0.25 – Yoh
Gua 0.5 – Yoh
6.93 ± 0.56
6.65 ± 0.58
7.88 ± 0.80
8.22 ± 1.16
8.43 ± 1.20
12.35 ± 0.86
14.58 ± 1.21
15.33 ± 1.85
13.46 ± 1.92
14.80 ± 1.91
21.65 ± 3.05
21.18 ± 2.22
23.03 ± 3.49
20.58 ± 2.71
22.45 ± 3.65
SCH 39166
(D1/5 ant)
Veh – Veh
Veh – Yoh
SCH 0.015 – Yoh
SCH 0.0625 – Yoh
6.53 ± 0.62
6.60 ± 0.79
5.65 ± 0.68
6.58 ± 0.68
10.19 ± 1.27
11.04 ± 1.40
11.68 ± 1.42
11.46 ± 1.23
23.65 ± 3.87
20.83 ± 2.09
25.03 ± 2.34
22.91 ± 2.65
Drug
Antalarmin
(CRF1 ant)
Values represent mean (SEM) premature response latency for the 15, 30, and 60 s premature delays for drug-free
test sessions of combined-drug experiments. The 1 and 4 s delays were not analyzed due to minimal responding at
these delays. The yohimbine dose was 0 or 2.5 mg/kg; other drugs are in mg/kg. None of the agents significantly
differed from veh-veh (0 – 0 mg/kg) or veh-yoh (0 – 2.5 mg/kg) drug combinations.
Ago, agonist; ant, antagonist; Ant, antalarmin; CRF1, corticotropin releasing factor receptor 1; D 1/5, dopamine 1/5
receptor family; GC, glucocorticoid receptor; Gua, guanfacine, Mif, mifepristone; Pra, prazosin; SCH, SCH 39166
175
Appendix V
ANOVA summary table for drug-free test sessions for corticotropin releasing factor,
glucocorticoid, adrenergic, and dopaminergic agents on yohimbine-induced responding on
measures of the response inhibition task
Drug
Antalarmin
(CRF1 ant)
0 – 20 mg/kg
df
F
p
ηp2
Dose
Delay
Dose x Delay
3, 21
4, 28
12, 84
0.58
8.72
0.95
0.60
< 0.05*
0.45
0.56
-
Dose
3, 21
0.33
0.71
-
Non-reinforced Lever Presses
Dose
3, 21
0.83
0.44
-
Dose
Delay
Dose x Delay
3, 21
4, 28
12, 84
0.11
2.68
0.68
0.87
0.10
0.61
-
Dose
Delay
Dose x Delay
3, 21
2, 14
6, 42
0.74
24.85
2.06
0.54
< 0.01*
0.15
0.78
-
Premature Responses (%)
Dose
Delay
Dose x Delay
3, 21
4, 28
12, 84
2.83
12.19
1.84
0.063
< 0.01
0.15
0.64
-
Omissions
Dose
3, 18
0.51
0.53
-
Non-reinforced Lever Presses
Dose
3, 21
0.60
0.50
-
Correct Latency
Dose
Delay
Dose x Delay
3, 21
4, 28
12, 84
0.49
6.86
1.73
0.53
< 0.01*
0.17
0.49
-
Behavioural Measure
Premature Responses (%)
Omissions
Correct Latency
Premature Latency
Antalarmin
(CRF1 ant)
0 – 15 mg/kg
176
Premature Latency
Dose
Delay
Dose x Delay
Mifepristone
(GC ant)
3, 21
2, 14
6, 42
1.18
81.43
1.66
0.34
<0.001*
0.22
0.92
0.19
4, 36
4, 36
16, 224
1.19
35.90
1.00
0.33
< 0.001*
0.43
0.46
-
4, 36
0.65
0.63
-
4, 32
0.33
0.71
-
4, 56
4, 56
16, 224
1.16
10.54
1.47
0.36
< 0.001*
0.11
0.43
4, 56
2,28
8, 112
1.27
208.24
1.92
0.30
< 0.001*
0.063
0.94
-
4, 60
4, 60
16, 240
0.75
29.63
0.94
0.56
< 0.001*
0.46
0.66
-
4, 60
0.76
0.46
-
4, 60
0.56
0.47
-
4, 60
4, 60
16, 240
0.25
4.17
0.92
0.91
0.036*
0.55
0.22
-
4, 60
2, 30
8, 120
0.54
63.37
0.38
0.61
< 0.001*
0.93
0.81
-
4, 28
4, 28
1.31
35.52
0.29
< 0.001*
0.84
Premature Responses (%)
Dose
Delay
Dose x Delay
Omissions
Dose
Non-reinforced Lever Presses
Dose
Correct Latency
Dose
Delay
Dose x Delay
Premature Latency
Dose
Delay
Dose x Delay
Prazosin
(α1 adrenergic
Premature Responses (%)
Dose
Delay
Dose x Delay
ant)
Omissions
Dose
Non-reinforced Lever Presses
Dose
Correct Latency
Dose
Delay
Dose x Delay
Premature Latency
Dose
Delay
Dose x Delay
Guanfacine
(α2 adrenergic
ago)
Premature Responses (%)
Dose
Delay
177
Dose x Delay
16, 112
1.39
0.27
-
4, 28
0.59
0.67
-
4, 28
1.39
0.28
-
4, 28
4, 28
16, 112
0.29
7.48
1.79
0.88
< 0.001*
0.16
0.52
4, 28
2, 14
8, 56
0.65
34.95
0.40
0.63
< 0.001*
0.78
0.83
-
3, 45
4, 60
12, 180
0.74
22.79
0.71
0.54
< 0.001*
0.63
0.60
-
3, 45
0.87
0.46
-
3, 45
3.54
0.061
0.19
3, 45
4, 60
12, 180
1.15
12.75
1.06
0.34
< 0.001*
0.40
0.46
-
3, 45
2, 30
6, 90
0.65
43.64
0.84
0.59
< 0.001*
0.46
0.74
-
Omissions
Dose
Non-reinforced Lever Presses
Dose
Correct Latency
Dose
Delay
Dose x Delay
Premature Latency
Dose
Delay
Dose x Delay
SCH 39166
(D1/5 ant)
Premature Responses (%)
Dose
Delay
Dose x Delay
Omissions
Dose
Non-reinforced Lever Presses
Dose
Correct Latency
Dose
Delay
Dose x Delay
Premature Latency
Dose
Delay
Dose x Delay
* denotes a significant effect
Ago, agonist; ant, antagonist; Ant, antalarmin; CRF1, corticotropin releasing factor receptor 1; D 1/5, dopamine 1/5
receptor family; GC, glucocorticoid receptor; Gua, guanfacine, Mif, mifepristone; Pra, prazosin; SCH, SCH 39166