THE EFFECTS OF AN NMDA RECEPTOR ANTAGONIST
IN THE CEREBRAL VENTRICLES ON THE
SELF-ADMINISTRATION OF COCAINE IN RATS
Kristina Andrea Uban
B.S., Arizona State University, 2002
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
This thesis for the Master of Arts
Kristina Andrea Uban
has been approved
Richard M. Allen, Ph. D.
Peter Kaplan, Ph. D.
David Albeck, Ph. D.
Uban, Kristina Andrea (M.A., Clinical Psychology)
The Effects of an NMDA Receptor Antagonist in the Cerebral Ventricles on the Self-
Administration of Cocaine in Rats
Thesis directed by Assistant Professor Richard M. Allen
Past research provides evidence for the role of glutamate in the transition from casual
cocaine use to chronic consumption. The underlying neurobiological mechanism of this
transition, along with the specific site in the brain where it occurs, is not fully understood. The
present study examined the role of a competitive NMDA receptor antagonist (LY235959), when
infused into the cerebral ventricles, on the reinforcing effects of cocaine (0.33mg/infusion) in rats.
The effects of acute and chronic administration of LY235959 into the cerebral ventricles were
investigated under both fixed ratio (FR1) and progressive ratio (PR1) schedules of reinforcement.
A methodology was established for: 1) implanting site specific cannulae into the right ventricle; 2)
performing acute infusions; 3) determining an appropriate dose for chronic administration of the
NMDA antagonist; and 4) histological verification of cannula placement. A significant increase in
self-administration for cocaine under a 6 hour fixed-ratio (FR1) schedule of reinforcement over
ten sessions was found with the continuous administration of 0.2 pg and 0.3 pg LY235959 into the
cerebral ventricles. The findings from the present study with intracranial administration of
LY235959 were similar to findings from past studies with systemic administration. The findings
support a hypoglutamatergic basis in the transition from occasional cocaine use to the chronic
consumption of cocaine over time.
This abstract accurately represents the content of the candidates thesis. I recommend its
Richard M. Allen
To my amazing mother, father, sister, and grandparents. Without your unconditional love and
never ending support, none of this would have been possible.
Many thanks to my advisor, Richard M. Allen, for his wonderful supervision, encouragement of
independence, and having confidence in me. I also wish to thank David Albeck for his
contribution to my training in the lab, and serving as a mentor in numerous areas of my academic
experience. Special thanks to Dorothy Yamamoto, Beth Atwood, and Kevin Blauth for not only
their contributions, but also friendships. Thank you to Peter Kaplan for serving as a member of
my committee and contributing feedback.
Table of Contents
1.1 Animal Models of Drug Addiction......................................................9
1.2 Long and Short Access to Cocaine.....................................................9
1.3 Schedules of Reinforcement..........................................................11
1.4 Evidence for the Role of Glutamate in the Escalation of Cocaine Consumption..........12
1.5 Present Study.......................................................................15
2. Materials and Methods................................................................16
2.1 Animals and housing.................................................................16
2.3 Self-administration procedure.......................................................16
2.4 Catheter surgery....................................................................17
2.5 Experiment 1: Acute infusions of LY235959...........................................17
2.6 Dose determination study............................................................18
2.7 Acute ICV infusions.................................................................19
2.8 Histological examination............................................................19
2.9 Statistical analyses................................................................19
2.10 Experiment 2: Chronic administration of LY235959...................................20
2.11 LY235959 pump......................................................................21
2.12 Saline control.....................................................................21
2.13 Histological examination...........................................................21
2.14 Statistical analyses...............................................................21
2.15 Problems with catheter maintenance.................................................22
2.16 Validity and reliability of measures...............................................24
3.1 Effect of acute infusions of LY235959 on cocaine self-administration................26
3.2 Effect of chronic administration of LY235959 on cocaine self-administration.........26
3.3 Establishment of mini-pump implantation procedure...................................27
3.4 Establishment of histological examination...........................................27
4.1 Practical issues of methodology......................................................29
4.2 Future implications..........................................................................30
Appendix: Diagnostic Statistical Manual IV R Criteria for Substance Dependence..................32
Animal Care and Use Statement................................................................... 33
List of Figures
Figure 1.1 Catheter with plastic tubing cover over guide cannula.
Figure 1.2 Plastic cap, tubing cover, guide cannula.
Figure 1.3 Catheter with plastic cap over tubing cover.
Figure 1.4 New open plastic cap, old closed plastic cap, catheter with plastic tubing cover.
Cocaine self-administration data
Figure 2.1 Acute Infusions ICV
Figure 2.2 ICV Pump Escalation Data
Figure 2.3 ICV Pump Progressive-ratio Data
Drug addiction is a complex problem, which leaks into every aspect of an individuals
life. Defining aspects of drug addiction include an uncontrollable craving, compulsiveness, and
continued use despite adverse consequences. People addicted to drugs often perform behaviors
specific to obtaining drug, where their behaviors are limited to obtaining, using, and recovering
from drugs. This behavioral pattern is inflexible, and results in losing control over their drug use.
Characteristics of drug addiction are comparable to those of a chronic disease (Deroche-Gamonet,
2004), and therefore require long-term treatment.
Cocaine is a powerful stimulant that has direct affects on the brain (NIDA, 1999).
According to the National Survey on Drug Use and Health, an estimated 5.9 million Americans
have used cocaine during the past year in 2002 and 2003 (NSDUH, 2005). Even more, 1.5 million
of those people were considered dependent on or abusing cocaine during the year of 2002 (NIDA,
1999). According to NIDA, conservative estimates found that, every $1 invested in addiction
treatment programs yields a return of between $4 and $7 in reduced drug-related crime, criminal
justice costs, and theft alone (NIDA, 2004). This estimate does not take into account health care
or other related costs. Even more, interpersonal conflicts and quality of life can be drastically
improved, therefore benefiting the larger community.
People who are recovering from addiction have a high propensity to relapse, even after
years of sobriety (NIDA, 2004). Of all the people in recovery from withdrawal, around 90% use
again (Deroche-Gamonet et al., 2004). The chronic and compulsive use of drugs considerably
alters the chemistry of the brain, thus affecting behavior even after long periods of abstinence. In
order to provide better treatment, the role of the neurobiological mechanisms underlying cocaine
addiction must be better understood. This understanding can lead to the development of
pharmacotherapies to successfully treat addiction and prevent relapse.
Animal Models of Drug Addiction
Given that the actual drug-seeking behavior is the common feature of addiction for all
drugs, the neural mechanisms that should be investigated are those that are related to the
reinforcing properties responsible for prolonged chug-seeking behavior (Stolerman, 1992). Rats
display comparable characteristics of addiction, and are valuable for studies that aim at better
understanding the underlying neurobiological mechanisms of drug addiction, or more generally,
the brains reward system. Drugs that are abused in humans also act as reinforcers in rats
(Stolerman, 1992). There are a few exceptions; drugs that are typically non-addictive in humans,
Cocaine Use and NMDA Receptor 10
such as hallucinogens, are not self-administered by rats. The Diagnostic and Statistical Manual
provides criteria for both substance abuse and dependence (appendix 1). According to Deroche-
Gamonet et al., there are three diagnostic criteria for substance dependence, as stated in the DSM
IV, that are comparable to behavior observed in rats that have learned how to self-administer
cocaine (2004). For one, rats with a history of exposure to cocaine continue to work for cocaine
even while a stop signal is on, indicating that no drug will be available. Secondly, rats with a
history of cocaine SA work harder than naive rats under a progressive ratio (PR) schedule, where
an increasingly large degree of work is required from the rats in order to receive cocaine. This
increase in break point, or number of presses during a PR schedule, is inferred as increased
motivation for the drug (Hodos, 1961; Richardson & Roberts, 1996). The third criterion is defined
by continued use of drug despite extremely aversive consequences. For example, Mien an electric
shock is paired with the administration of cocaine, rats with cocaine SA experience will continue
to self-administer (Vanderschuren and Everitt, 2004). Using the extinction-reinstatement
procedure, rats that displayed all three of the previously mentioned behaviors were found to be
more likely to press for cocaine after going through the extinction process (Deroche-Gamonet et
al., 2004). When given the choice between food and cocaine, rats with extended history of cocaine
self-administration selected cocaine.
Compulsive drug-seeking is characteristic of people with addictions, who use drugs
despite the adverse hurdles to acquire the drug and often detrimental consequences of their use.
This hallmark behavior of addiction has been seen in rats with a history of cocaine self-
administration. Drug seeking was suppressed in rats with little experience of cocaine SA when an
unpleasant foot-shock was paired with the cocaine (Louk et al., 2004; Vanderschuren and Everitt,
2004). However, rats with extended cocaine SA experience did not show suppression in drug-
seeking when cocaine SA was paired with the foot shock (Louk et al., 2004; Vanderschuren and
Everitt, 2004). Interestingly, compulsive seeking in rats is not seen when sucrose, a highly
desirable natural reinforcer, is used in place of cocaine. These findings imply that compulsive-
seeking, despite aversive consequences, is specific to drugs of abuse and does not apply to natural
reinforcers. External adversities may be powerless over drug-seeking behavior in rats that have an
extended history of cocaine self-administration (Vanderschuren & Everitt, 2004). These findings
suggest that the incentive value of cocaine is significantly increased with chronic exposure (Louk
et al., 2004; Vanderschuren & Everitt, 2004).
Addiction in humans is extremely complex. Studies using animal models focus on a
single neurobiological or behavioral aspect of addiction, as wells as other related parts. The
Cocaine Use and NMDA Receptor 11
present study focuses on the escalation of cocaine consumption. The changes that accompany the
transition from casual use to chronic consumption were investigated. The amount of cocaine
consumed, along with changes in motivation for drug, were also examined.
Long and Short Access to Cocaine
Ahmed and Koob found that rats allowed uncontrolled access to cocaine self-
administration resulted in a greater escalation of drug intake (1998). Rats that self-administer
cocaine tend to take a steady amount of cocaine when placed in an operant chamber for two hours.
Cocaine self-administration is controlled with a short-access (2 hour) model, therefore the degree
of exposure to cocaine is limited. Ahmed and Koob found that unrestricted access to cocaine in
the self-administration procedure better represents the escalation of intake over time (1998). The
long-access model allows for more daily access to the drug making it a valid model for human
drug use. Further, the increase in escalation seen in rats resulting from the long-access model,
allows for measurement of increases in cocaine consumption. The long-access model in rats is
representative of the escalation of drug intake seen in humans, when a transition from drug use to
drug addiction can be observed (Ahmed & Koob, 1998). Past studies allowing rats unlimited
access to self-administer cocaine resulted in a rapid escalation of intake, an irregular rate of
responding, and death. The long-access model has been shown to result in healthy weight
maintenance, as well as a stable rate of escalation. The present study implicates both the short and
long access models. Short access to cocaine self-administration is useful for the initial phase of
self-administration training, when the subjects first learn the association between pressing the
lever and the cocaine infusion. The long-access was used in the testing phase, where the
acceleration in the amount of cocaine consumed was measured.
Schedules of Reinforcement
Two schedules of reinforcement were used in the present study: continuous reinforcement and
progressive-ratio-schedule of reinforcement. Under a continuous reinforcement schedule, a
behavior (lever press) is followed by the direct consequence of the behavior (infusion of cocaine).
This schedule of reinforcement is thought to be a measure of the reinforcing effects of drugs. Rats
tend to regulate the intake of drug consumption with changes in the dose of cocaine. As the dose
of cocaine increases, the number of responses for cocaine by rats has been found to decrease
proportionately, suggesting that, animals regulate their intoxication around some endogenous
reference or hedonic set point (Ahmed & Koob, 1998, p. 299). This phenomenon is observed
Cocaine Use and NMDA Receptor 12
with both the short access and long access models. However, the rats allowed long access to
cocaine administer more cocaine during initial 10 minutes of the session, and then stabilize at a
higher rate of administering cocaine (1998). According to Ahmed and Koob, this increase in
escalation of cocaine self-administration cannot be explained by tolerance or sensitization alone,
but rather reflects, an acquired need state for a higher level of cocaine intoxication (1998, p.
299). The transition from drug use to drug addiction may be facilitated by this acquisition and
maintenance of the need for a higher level of cocaine intoxication, or an increase in hedonic set
Rats placed in an operant box during a progressive-ratio schedule of reinforcement are
required to press an increasing number of times in order to receive a single reinforcement, and are
used as a measure of incentive, or motivation for drug. Break-point is the number of
reinforcements achieved under a PR schedule. Higher doses of cocaine result in increased
pressing over a session. Paterson and Markou (2003) propose that an increase in hedonic set
point, may be accompanied by increased motivation to self-administer cocaine (p. 2229).
Evidence supporting this includes the greater resistance to extinction for the long-access rats
during a progressive -ratio (PR) schedule of reinforcement (2003). Increased break-points during
a PR schedule with saline were found for long-access rats when compared to short-access rats,
suggesting that there is also an increase in motivation for stimuli associated with cocaine (2003).
These findings correspond with reports from humans with addiction who experience an increase in
motivation for acquiring and using drug (2003). Recent evidence suggests that chronic cocaine
use disrupts neurotransmission within regions of the brain that are associated with reinforcement
(Kalivas et al., 2003; Tang et al., 2004).
Evidence for the Role of Glutamate in the Escalation of Cocaine Consumption
The underlying neurobiological mechanism responsible for the increased reinforcing
power of cocaine is still not understood. Past research has established the vital role of the
mesocorticolimbic dopamine pathway in producing the pleasurable effects produced by cocaine.
The main mechanism of action is the blockade of dopamine re-uptake in the pre-synaptic neuron,
resulting in elevated levels of available dopamine (Heikkila et al., 1975; Hummel & Unterwald,
2002). The natural effects of released dopamine are intensified and prolonged. However,
dopamine antagonists alone have not proven affective in reducing cocaine consumption or
preventing relapse in humans. This suggests that more than just dopamine is important in cocaine
consumption. The other neural transmitter systems affected by cocaine use, such as the release of
Cocaine Use and NMDA Receptor 13
glutamate, are not fully understood (Hummel & Unterwals, 2002). The pathway through which
dopaminergic and glutamaneric neurons project include the ventral tegmental area (VTA), nucleus
accumbens (NA), prefrontal cortex (PFC), amygdala, and hippocampus. Recent studies provide
support for the considerable role of glutamate transmission within the reward pathway (Kalivas,
2003). Glutamate may play a part in the neurobiological adaptations associated with chronic
cocaine use (Kalivas, 2003).
Recent evidence supporting the role of glutamate in the escalation of cocaine consumption has
emerged. Differences in both the efficacy of cocaine to elevate dopamine levels in the NA and
rate of cocaine metabolism were compared between rats allowed only short-access and those given
long-access to cocaine (Ahmed et at., 2003). Neither of these factors explained differences
between the short-access and long-access rats, suggesting that another factor independent of
dopamine is responsible for the increased rate of escalation of cocaine intake over time (Ahmed et
at., 2003). Looking at reinstatement of cocaine SA, or relapse, alterations in glutamate
transmission seem to be a necessary component, but not dopamine (Cornish & Kalivas, 2000).
Lever pressing was stimulated by the administration of an NMDA receptor agonist into the NA of
rats that had gone through extinction (Cornish et at., 1999). However, the administration of an
NMDA receptor antagonist into the NA also reinstated drug-seeking in rats that had gone through
extinction (Park et at., 2002).
Some evidence from past studies suggests a role for enhanced glutamate in escalation of
cocaine consumption. Kalivas and Duffy (1995) stimulated D1 receptors in rats by perfusing the
D1 agonist, SKF-82958, into the VTA. Increased amounts of glutamate released resulted from the
administration of SKF-82958 in the VTA. Administration of cocaine resulting in increased D1
receptor stimulation was found to augment the release of glutamate (Kalivas and Duffy, 1995).
The administration of an NMDA receptor agonist into the NA decreased responding in rats for
cocaine suggesting an enhancement of the rewarding effects of cocaine, (Cornish et at., 1999) or a
decrease in motivation to SA cocaine. Extracellular glutamate concentrations in both the nucleus
accumbens core (NAc) and ventral tegmental area (VTA) increase with repeated exposure to
cocaine (Pierce et at., 1996; Reid et at., 1997; Kalivas & Duffy, 1998). These elevated levels of
glutamate are observed only in rats displaying behavioral sensitization to cocaine (Pierce et at.,
1996). Behavioral sensitization is a phenomenon that occurs after repeated exposure to cocaine,
where the positive reinforcing value and psychomotor stimulating effects of cocaine are
increasingly augmented. Specifically, glutamate receptors in these regions have been found to be
hyper responsive after exposure to cocaine, possibly underlying behavioral sensitization to cocaine
Cocaine Use and NMDA Receptor 14
within the mesocorticolimbic pathway (White et al., 1995; Zhang et al., 1997). Alterations in
glutamate may have an important role in the transition from casual cocaine use to addiction.
Wang et al., found changes in protein levels of iGluR subunits in the ventral tegmental area
(VTA), core of the nucleus accumbens (NAc), and the prefrontal cortex (PFC) accompanying
withdrawal and binge use of cocaine in rats (2004). After one day of withdrawal from cocaine
self-administration, rats displaying behavioral sensitization had elevated GluRl and NMDAR1
levels in the VTA compared to those displaying no behavioral sensitization, who had equivalent
receptor levels to the control rats receiving saline only (Churchill et cd., 1999). After three weeks
of withdrawal, cocaine sensitized rats had significantly elevated levels of GluRl in the nucleus
accumbens compared to their non-sensitized and saline counterparts (Churchill et al., 1999).
These findings suggest that long-term augmentations in glutamate receptor levels that result from
cocaine consumption occur only in those displaying behavioral sensitization. Behavioral changes,
such as sensitization, appear to parallel alterations in dopamine and glutamate in the nucleus
accumbens (Churchill et al., 1999).
There is also recent evidence for the role of decreased glutamate. In a study investigating the
rewarding actions of blocking dopamine re-uptake and blocking NMDA receptors, rats were
trained to press a lever resulting in brain-stimulation of the nucleus accumbens (Carlezon & Wise,
1996). Increases in the frequency of lever-pressing occurred after intra-cranial administration of
dizocilpine (MK-801) and CPP, both NMDA receptor antagonists. These finding suggest that the
blockade of NMDA receptors itself may be rewarding independent from dopamine effects. In a
study using mice, two subclasses of glutaminergic antagonist were self-administered into the
VTA, suggesting that when centrally administered, they can be highly rewarding (David et al.,
1998). The mice were then treated with a dopaminergic D2 antagonist prior to the session,
resulting in a fast extinction of lever-pressing. These findings suggest that the interaction between
dopamine and glutamate in the VTA significantly affect rewarding properties (David et al., 1998).
Evidence for the role of both increased and decreased glutamate in cocaine addiction with
humans corresponds to past findings from animal research. Tang et al., examined brain tissue of
cocaine overdose victims and found up-regulation in NMDAR1, GluR2, and GluR5 receptors in
the VTA (2003). Human subjects reported enhanced subjective effects of cocaine after taking
memantine, a non-competitive NMDA antagonist (Collins et al., 1998). Humans reported
attenuation in the euphoria produced by cocaine with modafinal treatment, a glutamate enhancing
agent (Dackis & OBrien, 2003). Human patients seeking treatment for cocaine abuse were able
to remain abstinent from cocaine for a greater period of time when taking modafinal than those
Cocaine Use and NMDA Receptor 15
who did not receive the drug (Dackis et al., 2005). This evidence suggests that an increase in
glutamate decreases the subjective rewarding effects of cocaine. However, the glutamate
enhancing mechanism of modafmal is still not understood.
The complex role of glutamate in cocaine consumption is not fully understood. One
conceptualization of how both enhanced and decreased glutamate affect consumption of cocaine
has been offered by past research: basal glutamate tone in the nucleus accumbens is
conceptualized as background noise against which the effects of cocaine and cocaine-paired
signals are played (McFarland et al. 2003; Kalivas, 2004). With repeated cocaine exposure, the
background noise becomes quieter (i.e., basal glutamate levels fall) as the salience of cocaine
signals are increasingly amplified.
Previous research has studied the effects of the systemic administration of a non-competitive
NMDA receptor antagonist, LY235959, on cocaine consumption (Allen et al., 2005). Rats
receiving an NMDA antagonist continuously for two weeks through a subcutaneously implanted
osmotic pump, during long-access (6 hours) self-administration sessions, escalate intake of
cocaine at a greater rate than the control subjects (Allen et al., 2005). For the proposed study, it
was hypothesized that the reduction in glutamate tone may also underlie escalation of cocaine
consumption. Changes in motivation to consume cocaine was also measured. The first step in
understanding the glutamate altering mechanism of an NMDA receptor antagonist, LY235959, is
to determine the site of action of this effect The present study investigated this effect to establish
whether or not it is centrally mediated. An NMDA receptor antagonist was delivered directly into
the central nervous system in order to observe the behavioral effects relating to cocaine SA. This
was accomplished by placing a guide cannula, connected to an osmotic pump containing the
NMDA antagonist, into the rats cerebral ventricles. The effects of glutamate on cocaine SA were
found to occur centrally, so the specific sites of action within the brain were investigated. Further,
the methodology used in this study was tested for use in future studies. Determining the role of
glutamate will enhance our overall understanding of cocaine addiction and may lead to improved
pharmacological treatments for humans.
Material and Methods
Animals and housing
Male Sprague Dawley rats weighing between 250-375g were purchased from Harlan
(Indianapolis, IN). The rats were housed at the University of Colorado at Denver and Health
Sciences Center, Downtown Denver campus. A 12 hour light/dark cycle was used with the lights
on at 7 a.m. Animals had unrestricted access to water and food prior to the start of an experiment.
During an experiment, water was restricted to 30ml each weekday (M-Th) and 110ml over the
weekends (F-Sun). All procedures were performed during the light cycle. Procedures were
approved by the CU-Denver Institutional Animal Care Committee and in accordance with the
National Research Councils Guidelines for the Care and Use of Mammals in Neuroscience and
Behavioral Research (2005).
Self-administration sessions were conducted in individual Plexiglas operant chambers (29
X 24 X 21cm) contained within sound-attenuating cubicles (MED Associates, St. Albans, VT).
On the wall opposite to the levers, a speaker for white noise (90 dB) and a tone presentation
speaker (Sonalert Tone Generator, 2900 Hz) were mounted 12 cm above the floor. Six cm above
the tone speaker, a light (100 mA) was mounted. Two retractable levers were mounted on a wall
of the chamber. A food or fluid delivery dish sits between levers. Two stimulus lights were
positioned 6 cm above the levers. A computer-controlled syringe pump was operated by an
electronic circuit activated by a lever press, which controlled all cocaine infusions. All behavioral
data were collected and stored by a computer using MED-PC for Windows software (MED
The beginning of the session was signaled by a light in the chamber. White noise was
activated at the start of the session to eliminate outside disturbances. When the active lever had
been pressed by the subject, an additional light appeared while a predetermined dose of drug was
delivered. The second light remained on for 15 seconds, during which the active lever became
inactive. This process was maintained by a computer, which also collected data. A second lever
in the chamber was inactive throughout the sessions. The subjects were connected to a swivel
device that allowed free roaming of the chamber during the session. A coil leash attached the rats
catheter to the drug delivering tube at the swivel device. A syringe with cocaine located outside of
Cocaine Use and NMDA Receptor 17
the chamber was attached to the tubing, and pumped out the correct dose of drug for each active
lever press. The outside doors to the chamber contained a small glass opening allowing the
experimenters to observe the subject without interfering.
All subjects underwent surgery for catheter implantation. An indwelling, intravenous
catheter was implanted into the left jugular vein. The catheter allowed the rats to self-administer
drug during each session in the operant chamber. Subjects were first anesthetized with ketamine
(100 mg/kg) and xylazine (20 mg/kg). An injection of atropine was given after the subject had
been anesthetized. The subjects hair was then shaved in two places to accommodate incisions; 1)
a square on the back just below the shoulder blades and 2) a small spot to the left of the sternum
and up to the bottom of the jaw bone. Both incision sites were then sterilized with sterile
povidone-iodine (10%) and isopropyl alcohol (70%). A horizontal incision was first made on the
back with an approximate length of 3.8 cm. The second incision was made at a slight vertical
angle above the jugular vein on the right side of the rats neck with an approximate length of 10
mm. A guide cannula was slipped under the skin over the right shoulder blade connecting the two
incisions. The end of the catheter containing the tubing was then slid under the skin using the
guide cannula. Next, the jugular vein was located a small incision was made on its surface,
through which the tubing of the catheter was inserted. The incision above die vein was then
covered with a small piece of Bard surgical PTFE mesh and stitched up with silk suture. The
back-plate of the catheter was then placed into the incision on the back so that the guide cannula
portion protruded from the skin allowing for the self-administration of cocaine. Nylon sutures
were used to secure the back-plate into place. The wounds were dressed with povidone-iodine.
During recovery, subjects were given unlimited access to water and food. An additional
water bottle containing 5ml Childrens Tylenol in 80ml of water was also provided. All subjects
were carefully monitored each day to ensure a healthy recovery. Rats were returned to their
individual housing cage and allowed 1 week to recover before the start of self-administration
Experiment 1: Acute infusions of LY235959
The effect of acute infusions of LY235959ICV on the self-administration for cocaine
under a progressive-ratio (PR1) schedule of reinforcement was first investigated. A progressive
ratio schedule of reinforcement requires the subjects to press an increasing number of times for a
Cocaine Use and NMDA Receptor 18
single infusion, and is used to measure motivation to consume drug. A second surgery occurred 2-
3 weeks after the start of cocaine self-administration training. Only subjects with working
catheters and stable responding were selected for the dose determination phase of the experiment.
Subjects were anesthetized with ketamine (100 mg/kg) and xylazine (20 mg/kg). An injection of
atropine was given after the subject had been anesthetized. To implant intracranial cannulae, first
the dorsal side of the head was shaved, just posterior to the eyes, and ending anterior to the neck.
A stereotaxic apparatus was used to locate the ventricle in the right hemisphere. To do this, the
subject was secured into the ear bars of the stereotaxic apparatus. The shaved area was sterilized
using alternating povodine-iodine and alcohol swabs. A unilateral incision approximately 1.75
centimeters was made on top of the head, beginning anterior to the neck to the midpoint between
the eyes. The skull was exposed at the site of the incision. The junction of the coronal and
sagittal sutures at the top of the skull (Bregma) was located with a dummy infuser, and coordinates
were recorded. The coordinates for the right ventricle were calculated from Bregma -0.92 mm
(AP -.10mm, ML -.15mm, DV -,30mm). A small hole was carefully drilled dorsal of the right
ventricle using a hand-held MultiPro Cordless DREMEL drill (7.2V Model 7700 Type 1),
allowing the guide cannula to extend into the specific site of the brain. The DREMEL drill head
was approximately 1 mm in diameter. The dummy cannula was re-aligned with the hole to verify
accuracy of the drilling process. Three surrounding holes were also placed into the skull to secure
the guide cannula with screws (approximately 1 mm in diameter, 3.2mm length): 1) anterior to the
cannula in right hemisphere, 2) posterior to the cannula in right hemisphere, and 3) lateral to the
cannula in left hemisphere. The guide cannula was placed into the arm of the stereotaxic
apparatus, and bregma was again located. The coordinates were calculated a second time, serving
as a check for accuracy. Once the guide cannula was placed over the hole above the ventricles the
guide cannula was lowered into the ventricles. Instant adhesive (LOCT1TE 382) was used to
secure the screws to the guide cannula. Accelerator (LOCTITE 712) was used hardening the
adhesive instantly. The arm was detached from the guide cannula and removed. Dental cement
was made using acrylic powder mixed with acrylic liquid (Jet Acrylic, LANG). The cement was
carefully dripped around the secured guide cannula into the incision covering the exposed skull.
The rats recovered for 48 hours.
Dose determination study
We first needed to determine an effective dose of LY235959 for intraventricular
administration. Potential doses were first chosen after an extensive review of the literature.
Cocaine Use and NMDA Receptor 19
Eleven rats were used to determine the appropriateness of the dose of the NMDA antagonist. The
dose we chose for the escalation studies was based on well tolerated doses that produced an effect
on PR responding when administered acutely ICV.
Acute ICV infusions
The methodology for successful infusions was established. A lOpl syringe attached to
flexible plastic tubing (.020 inch inner diameter, TYGON) was connected to an infuser. The
tubing was filled with sterile water prior to infusions. The infuser was placed into the guide
cannula, and 5 pi vehicle administered over 30 seconds. The infuser was left in the guide cannula
for 1 minute to ensure that the vehicle was dispersed. The subject was then placed into the operant
chamber, where their session was activated 5 minutes after the infusion. Behavioral changes were
recorded, along with the quality of the infusion.
After the completion of the experiment, the rats were sacrificed for histological
examination using a dye, cresyl violet, to stain the track of the guide cannula. All verification was
Behavior was analyzed looking at number of lever presses during a self-administration
session under a progressive-ratio of reinforcement. The data collected was analyzed looking at
both the cumulative number of lever presses and breakpoint (the number of responses before 60
minutes of no responding). All data was analyzed using SPSS, version 14.0. The mean
differences between the varying doses of LY235959 and saline infusions were analyzed using
repeated-measures ANOVA. The within-subjects factor was LY235959 dose. The dose variable
included the saline infusion and the 0.03 gg, 0.1 pg, and 0.3 gg doses of the NMDA receptor
antagonist (LY235959). The dependent variable of interest, motivation for the drug, was
measured by the amount of work a rat was willing to exert during separate sessions under a PR
schedule of reinforcement.
An alpha of less than .05 was committed to in order to decrease the probability of
concluding that the treatment had an effect when in fact there is no true effect. A large enough
sample size to detect a real effect using a two-tailed test is 6 10 subjects. Past studies involving
Cocaine Use and NMDA Receptor 20
systemic self-administration have been powerful enough to find significant differences using a
similar number of subjects.
Experiment 2: Chronic administration of LY235959
The second aim of the study was to determine the role of continuous administration of
LY235959ICV on the escalation of cocaine consumption under a 6 hour fixed-ratio (FR) schedule
of reinforcement. Subjects self-administered cocaine intravenously during an FR1 schedule (2
hours) for five days each week (M-F) for two to three weeks during training. After stabilization of
responding, continuous ICV administration of the NMDA antagonist (LY235959) began. All
intracranial cannula minipump assembly surgeries were performed on a Friday, allowing subjects
72 h recovery before the escalation phase of the experiment. Subjects with working catheters and
stable responding were selected for the escalation phase of the experiment. Subjects were
anesthetized with ketamine (100 mg/kg) and xylazine (20 mg/kg). An injection of atropine was
given after the subject had been anesthetized. To implant intracranial cannulae with an osmotic
minipump attached, first the dorsal side of the head was shaved, just posterior to the eyes, and
ending anterior to the neck. A stereotaxic apparatus was used to locate the ventricles in the right
hemisphere. The subject was secured into the ear bars of the stereotaxic apparatus. The shaved
area was sterilized using alternating povodine-iodine and alcohol swabs. A unilateral incision
approximately 1 V* centimeters was made, beginning anterior to the neck to the midpoint between
the eyes. The skull was exposed at the site of the incision. Bregma was located with a dummy
infuser, and coordinates were recorded. A small hole was cautiously drilled dorsal of the right
ventricle using a drill, allowing die guide cannula to extend into the brain. The dummy cannula
was re-aligned with the hole to verify accuracy of die drilling process. Three surrounding holes
were also placed into the skull to secure the guide cannula with screws. Subjects were implanted
with an Alzec osmotic minipump designed for intracranial infusions into the cerebral ventricles,
which continuously administered either an NMDA competitive receptor antagonist (LY235959) or
saline. The minipumps were implanted underneath the skin on the neck, where the tubing
extended into an incision in the skull. The guide cannula attached to the minipump was placed
into the arm of the stereotaxic apparatus, and bregma was again located. The coordinates were
calculated a second time, serving as a check for accuracy. Once the guide cannula was placed
over the hole above the ventricles, the tubing connected to the minipump was carefully tucked
under the skin, and die guide cannula lowered into the ventricles. Adhesive was used to secure the
Cocaine Use and NMD A Receptor 21
guide cannula to the screws. Dental cement was carefully dripped around the secured guide
cannula into the incision covering the exposed skull. The rats recovered for 48 hours.
During recovery from both catheter and stereotaxic surgeries, subjects were given
unlimited access to water and food. An additional water bottle containing 5ml Childrens Tylenol
in 80ml of water was also provided. All subjects were carefully monitored each day to ensure a
The pump administered either 0.2 pg or 0.3 pg LY235959 per day for 2 weeks into the
right ventricle. During this process, the subjects were allowed daily long access (6 hours) to SA
cocaine during weekdays for two weeks. All subjects were tested under a progressive-ratio (PR)
schedule of reinforcement on Saturday following each week of long access sessions with the pump
(total of 2 PR sessions). One purpose of this study was to determine whether or not behavioral
adaptations of increased cocaine consumption can be replicated using IC V administration of
LY235959 rather than systemic administration. Such a finding would demonstrate that NMDA
receptor antagonist facilitation of escalation of cocaine consumption is a centrally mediated effect,
and demonstrate the feasibility of this procedure for subsequent studies designed to locate a
specific neurobiological site of action for this phenomenon.
A control group received saline instead of the NMDA antagonist administered
continuously through an osmotic pump. All methods were the same as the treatment group. This
group served as a saline control group for statistical comparison.
After the completion of the experiment, the rats were sacrificed for histological
examination using cresyl violet to stain the track of the guide cannula. All verification was
Behavior was analyzed looking at a subjects cumulative number of lever presses during
a 6 hour self-administration session under a fixed ratio of reinforcement. The data collected under
a progressive-ratio of reinforcement was analyzed looking at both the cumulative number of lever
Cocaine Use and NMDA Receptor 22
presses, and breakpoint (the number of responses before 60 minutes of no responding). All data
was analyzed using SPSS, version 14.0. The mean differences between the drug and saline groups
were analyzed using a two-way repeated-measure ANOVA. The independent variables were dose
and day. The dose variable included the saline group, the 0.2 jig per day dose, and the 0.3 pg per
day dose of the NMDA receptor antagonist (LY235959). The between-group differences were
analyzed with data from all three conditions. The day variable included 5-10 individual long
access sessions. The first dependent variable, escalation of cocaine consumption, was measured
by the number of presses made by the subject during a 6 hour fixed ratio schedule. Key time
points, such as the 5th and 10th sessions, were examined. The percent increase in lever presses
during the first hour of responding for each day during escalation was compared to baseline
pressing prior to pump implantation. The second dependent variable of interest, motivation for the
drug, was measured by the amount of work a rat was willing to exert during 2 separate sessions
using a progressive ratio schedule (PR). The number of reinforcements received during a PR
session prior to a 60 minute period of no responding is called breakpoint. Both PR1 sessions after
the 5th and 10th days of escalation were also analyzed for statistical significance. Tukeys least
significant difference method of analysis was used for post-hoc comparisons.
An alpha of less than .05 was committed to in order to decrease the probability of
concluding that the treatment had an effect when in fact there is no true effect. Six to ten rats in
each condition using a two-tailed test is a large enough sample size to detect a real effect. Past
studies involving systemic self-administration have been powerful enough to find significant
differences using a similar number of subjects.
Problems with catheter maintenance
All catheters used during this experiment were constructed by students working in the lab. In
2004, approximately 15 percent of subjects had catheter failure due to blockage or leaking in our
laboratory. During the first half of the present experiment, the percentage of rats with blocked or
leaky catheters increased significantly to approximately 27 percent. The failure of the catheter
occurred early, preventing many subjects from completing the experiment. In 2004,
approximately 60 percent of subjects successfully completed at least 30 sessions prior to catheter
failure, compared to only 16 percent in 2005. Many catheters were completely clear when flushed
after the experiment in the evening, but were found to be blocked the next morning. The process
of investigating the cause of the blocked catheters was thorough, because the cause was not clear.
Cocaine Use and NMDA Receptor 23
New tubing supplies used in the construction of catheters were ordered to ensure that the
failure of the catheters was not a result of faulty or brittle tubing. The next round of subjects with
new tubing still had a high attrition rate. Our second step was to re-train all lab assistants in the
construction of catheters, to rule out human error. All lab assistants attended a training session
with the lab supervisor to review the steps in the construction process. The old catheters that had
become blocked were extracted and examined to determine the exact location of the blockage.
The blockage appeared to be occurring in the bend of the cannula, and not in the tubing. The lab
supervisor bent all cannulae before the catheters were constructed to ensure that the bend was
smooth and not pinched. Again, new catheters were constructed, but the low proportion of rats
able to complete the experiment before their catheter failed was unchanged.
At this phase, many precautionary steps were taken. Typically, during the first week of
recovery after catheter implantation, the catheters are not flushed until the first session. We tried
flushing the catheters with saline twice during the week of recovery to reduce the possibility of
blood build-up during the week of recoveiy. The syringes and needle tips used to flush the
catheters before and after sessions were replaced after each use in order to reduce the chance of
small debris entering the catheter and clogging the passage. The tubing that connects the syringe
containing drug through the operant chamber to the catheter was also replaced more frequently to
reduce possible debris from coming loose and entering the catheter. Additionally, a former lab
technician was asked to construct ten catheters independently, to compare the attrition rate to those
subjects implanted with our catheters. Still, there was no change in the rate of rats successfully
completing the experiment prior to catheter failure.
The blocked catheters were again extracted and closely examined. The metal tube comprising
the cannula was straightened with pliers and a small wire pushed through. Consistently, a dark
particle was pushed through the extended cannula. This particle was identified as a piece of the
plastic tubing used to cover the cannula opening and prevent blockage (figure 1). Over this tubing
cover, another small plastic cap with a closed top is screwed on to keep the tubing cover attached
(figure 1.2). Our conclusion was that the plastic cap was pushing against the tubing cover, causing
small pieces to break off and stick in the bend of the cannula, blocking any solution from
continuing through the passage (figure 1.3).
New plastic caps with an open top were ordered (figure 1.4). Like the former caps, the new
caps ensured that the tubing cover would not be pulled off over night. The open caps did not place
unnecessary pressure against the tubing, therefore should not cause it to break off into the cannula.
All rats received the open cap, and the successful completion of the experiment with a working
Cocaine Use and NMDA Receptor 24
catheter significantly increased to approximately 90%. Catheter failure has been nearly
Validity and Reliability of Measures
The formal equivalence of the SA model can be assessed by its face validity. The escalation
of cocaine SA in rats looks like the phenomenon of addiction in humans. Psychomotor activity is
enhanced in rats as it is with humans. Rats with a history of cocaine SA will eventually begin to
groom themselves less; similarly, humans slowly begin to disregard appearance as their addiction
progresses. Using the long access model, the binge-like pattern of cocaine SA resembles that of a
person with chronic cocaine use (Ahmed & Koob, 1998). In the presence of an environmental
stimuli associated with cocaine use, rats will exhibit behaviors similar to those after cocaine use,
as humans report experiencing the perceptual effects of cocaine without actual use. Despite
evidence for the formal equivalence of the SA model, we can never really know what the rat is
experiencing; other forms of validity, such as correlation or functional equivalence, must also
coincide with the model (Katz & Higgins, 2002).
The predictive validity of the self-administration model can be supported by the observation
that the drugs self-administered by rats in the laboratory correlate with those abused by people
(Stolerman, 1992). Rats with a history of cocaine SA are more likely to press the lever after
extinction; people tend to be at higher risk for relapse if they have greater experience with cocaine
use. The functional equivalence of the S A model in rats refers to the similarity in the
neurobiological mechanisms underlying cocaine escalation in rats and humans. Post-mortem
analyses of humans who died from an overdose of cocaine have investigated alterations of proteins
in specific brain regions based on past animal research. An up-regulation of certain ionotropic
glutamate receptor subtypes in cocaine overdose victims were found in the brain, demonstrating,
that some of the adaptations that occur in animal models of drug abuse are applicable to die
human brain (Tang et al. 2003, p. 921). Another way of assessing functional equivalence is to
see if manipulation of controlling variables has comparable effects in the laboratory with those
occurring in real life (Katz & Higgins, 2002). One controlling variable is history of cocaine use.
Rats with a greater history of cocaine use will exert more work for cocaine under a PR schedule of
reinforcement. With chronic cocaine use, people tend to be willing to work harder for cocaine.
Unlike studies with human participants, animal models tend to be more reliable. All rats are
housed in the same vicinity, and receive equal amounts of food and water. A computer program is
used to collect all data, unlike the often variable self-report measures or assessment tools with
Cocaine Use and NMDA Receptor 25
human subjects. The same behavioral outcome can be measured across studies. The dose of drug
used during each session can be controlled. Using the SA model, the same escalation of cocaine
consumption can be consistently observed in the rats.
Effect of acute infusions of LY235959 on cocaine self-administration
The effect of varying doses of the NMDA antagonist on responding for cocaine during a
5-h session under a progressive ratio schedule of reinforcement (PR) is presented in figure 2.1.
The breakpoints and total infusions did not differ significantly from each other. Rats (n = 8)
infused with 0.03 pg/5pl LY235959 self-administered a mean of 17.3 1.1 (mean S.E.M.,
throughout) 0.33 mg cocaine infusions during a session. The mean number of infusions for
subjects (n = 8) receiving 0.1 pg/5pl LY235959 was 16.8 2.7. Subjects receiving 0.3 pg/5pl
LY235959 pressed the lever a mean of 10.6 2.4. A repeated-measures ANOVA revealed
significant differences in lever presses as a result of the varying dosage, F (3) = 3.89, p < .05.
Paired-samples t-tests resulted in significant differences between the 0.03 pg/5pl and 0.3 pg/5pl
doses of LY235959 only, t (7) = 3.06, g < .05. The doses 0.45 pg/5pl and 0.6 pg/5pl resulted in
extreme motor impairment when delivered into the lateral ventricles. As a result of the
impairment caused by the infusions, the sessions were ended early and no data pertaining to lever
pressing was collected. The impairment caused by the acute infusion of the higher LY235959
doses (0.45 pg/5pl and 0.6 pg/5pl) was not recovered from and produced irreversible damage,
unlike the temporary impairment observed when administered systemically (Allen etai, 2005).
Effect of chronic administration of LY235959 on cocaine self-administration
Figure 2.2 shows the effect of chronic administration of both vehicle and LY235959 into
the right ventricle on 0.33 mg per infusion cocaine self-administration. The average number of
infusions on the first day of escalation for the saline pump were fairly uniform (68.5 3.3, n =
11). The first day of the escalation phase was variable among both the 0.2 pg LY235959 pump
group (54.9 13.8, n = 6) and 0.3 pg (60.3 5.1, n = 4). A repeated-measures ANOVA revealed
a significant main effect of treatment condition over the first five days of escalation, F_(2,18) =
3.54, g_= .05. The number or infusions received varied significantly over the first five days of
escalation, F.(4,18) = 9.97, g_< .001. The interaction between the day and treatment condition
during the initial five days of escalation was also significant, F_(8, 18) = 2.39, g_< .05. Post-hoc
analysis using the least significant difference method (LSD) revealed significant differences in
escalation during the first five days of escalation between the vehicle control and 0.3 pg
LY235959 pump groups (mean difference = 17.12, g_< .05).
Escalation was also analyzed over ten days (figure 2.3). The average number of presses
on day ten of escalation were as follow: vehicle pump group (72.3 3.8, n = 9); 0.2 pg LY235959
Cocaine Use and NMD A Receptor 27
(87.0 7.3, n = 6); and 0.3 fig LY235959 (109.9 30.5, n = 4). No significant main effect of
treatment condition was found, F (2,16) = 2.92, g = .08. The number of infusions received
differed significantly between days, F (9,18) = 6.99, g < .001. There was a significant interaction
between treatment condition and day, F (18,16) = 2.12, g < .01. A significant difference over ten
days of escalation between the vehicle and 0.3 pg LY235959 pumps were revealed by post-hoc
comparisons (LSD), mean difference = 22.52, g < .05.
Total infusions received under a progressive-ratio schedule of reinforcement were
compared for all three conditions prior to pump implantation and after both day 5 and 10 of
escalation (figure X). No significant effects of treatment were found between the three conditions,
F (2,10) = .96, g = .41. Differences in break points between the treatment conditions were also
not significant after analysis with independent /-tests.
Independent /-tests were used to analyze key days of interest between the saline pump,
0.2 gg LY235959 pump, and 0.3 gg LY235959 treatment conditions. No significant differences
existed between both FR (/ = -.28, g > .05) and PR (/ = .06, g > .05) sessions prior to treatment.
Independent /-tests revealed significant differences in mean infusions between the vehicle pump
(71.3 4.7) and 0.2 gg LY235959 pump (71.3 4.7) conditions (/ = -2.60, g < .05) on day 6 of
escalation The 0.3 gg LY235959 pump condition had significantly greater number of infusions
received than the vehicle pump condition on day 2 (/ = -2.88, g < .05), day 3 (/ = -3.19, g < .01),
and day 4 of escalation (/ = -3.26, g < .01). Differences were found between the 0.2 gg LY235959
pump and 0.3 gg LY235959 pump conditions on day 2 (/ = -2.28, g < .05), day 3 (/ = -2.38, g <
.05), and day 4 of escalation (/ = -2.79, g < .05), where the higher LY235959 dose performed
significantly more lever presses.
Establishment of mini-pump implantation procedure
The placement of the catheters in die mid-back area limited the distance away from the
guide cannula from which the pump sat. Therefore, the frill length of the tubing was not needed,
and the excess was curled around to avoid kinks. Initial histological examinations revealed that
the dental cement was leaking onto the excess tubing connecting the pump to the guide cannula,
causing it to break or become blocked. To prevent this, later surgeries incorporated 2-3 stitches at
the posterior end of the incision, frilly covering the tubing, and the area requiring dental cement
was reduced. Additionally, the stitches were pressed down prior to application, preventing the
cement from coming into contact with the tubing. The tubing was shortened from the original
Cocaine Use and NMDA Receptor 28
length of 15cm to approximately 5 cm to minimize excess tubing pressing against the soft cement
Establishment of histological examination
For the chronic administration of LY235959 subjects, the pump was exposed in the
euthanized subjects, and examined for any visible kinks, leaks, or blockage. A syringe containing
a dye, cresyl violet, connected to a short piece of tubing was attached to the hollow needle of the
minipump. The dye was injected into the exposed tubing, staining the specific site of the brain
located at the end point of the cannula. The volume of cresyl violet necessary to stain the track of
the guide cannula was determined. A volume of 0.15ml of cresyl violet was the amount found to
successfully stain the site without excess dye leaking into surrounding areas of the brain. The
brain was removed and sliced to locate the dyed section and determine the exact end point of the
The present study investigated the role of a competitive NMDA receptor antagonist
(LY235959) administered continuously into the cerebral ventricles (ICV) on cocaine self-
administration in rats given long access (6 hours). The present data suggest that the escalation of
cocaine consumption was facilitated by the intracranial continuous administration of LY235959.
A trend in increased responding under a progressive-ratio schedule of reinforcement was found for
both LY235959 treatment conditions. A significant difference between the vehicle and 0.2 pg
LY235959 pump conditions in cocaine self-administration was found on day 6 of escalation only.
Lever-pressing was significantly greater in the 0.3 pg LY235959 pump group than the vehicle
condition group on days 2, 3, and 4 of escalation. The two varying doses of LY235959 were
significantly different on days 2, 3, and 4, where the higher dose pressed more over a 6 hour
session. A significant effect of treatment condition was found for days 1 through 5 of escalation
only. An effect of day was also found to be significant over the ten days of escalation, where
more infusions where being taken by all groups with increased cocaine self-administration
experience. An interaction between the day of escalation and treatment condition was found,
suggesting that the effects of LY235959 treatment on lever-pressing differ across days. Subjects
receiving continuous administration of LY235959 had fewer lever-presses on the first day, but
continued to surpass the mean infusions of the saline group after day three of escalation.
The data collected from the 0.3 pg LY235959 pump treatment suggest that this dose of
LY235959 is needed to observe the centrally mediated behavioral effects of the NMDA antagonist
on cocaine self-administration. Further, the steep dose-response curve of LY235959 when
administered into a specific site of the CNS suggests that higher doses of LY235959 will not be
tolerated well. Although there were significant results obtained, a larger sample size of the
LY235959 treated groups would better illuminate the effects of the NMDA receptor antagonists
on cocaine self-administration. More data are needed for meaningful statistical analyses.
Practical issues of methodology
The cerebral ventricles were the specific site of interest in this study in order to determine
whether or not the facilitation of escalation of cocaine consumption resulting from treatment with
LY235959 was a centrally mediated effect. The ventricles were chosen because of their size and
proximity to the exterior of the skull compared to the ultimate target, the nucleus accumbens. The
nucleus accumbens is thought to be a primary mediator of cocaine addiction, and is the eventual
Cocaine Use and NMDA Receptor 30
aim for the continuous administration of an NMDA receptor antagonist. The process of
establishing methodology prior to executing this aim was a major goal of the present study.
Many practical issues were established during this study for implementation in future studies.
An appropriate dose of LY235959 was first found. Acute infusions of varying doses of
LY235959 significantly affected the number of infusions earned in subjects with a progressive-
ratio schedule of reinforcement. 0.1 pg LY235959 delivered acutely increased lever-pressing
slightly compared to saline infusions, while 0.3 pg LY235959 decreased lever-pressing. The
dose-response curve for the NMDA receptor antagonist, LY235959, was found to have a very
steep curve. The extreme motor impairment of high LY235959 doses (0.45 pg and 0.6 pg) was
not recovered from. The high doses were found to produce permanent damage. The difference in
volume between damaging and tolerable doses delivered ICV was only 0.15 pg. The dose of 0.3
pg LY235959 was enough to produce an effect when administered acutely ICV.
Once a reasonable dose was determined, the methodology for mini-pump implantation was
developed. Histological verifications revealed problems with this process initially. The plastic
tubing connecting the pump to the guide cannula was found to be either detached or blocked for
some subjects. Instant adhesive was used to secure the tubing to the guide cannula, and prevented
it from becoming loose and separating. Excess tubing had become cemented in with the hardened
acrylic powder, causing a block in the flow of the vehicle from the pump to the cannula. To solve
this problem, the tubing was shortened, stitches were used to close up the incision closest to the
tubing, and the passage from the opening to the tubing was pressed down, preventing the fluid
cement from covering the plastic tubing. Another practical issue regarding the methodology
included establishing an appropriate volume of dye, cresyl violet, to inject into the track of the
cannula during histological verification for visible staining of the ventricles.
Within the reward pathway of the brain, the nucleus accumbens has been identified as a center
of rewarding processes. The involvement of dopamine in the nucleus accumbens on the rewarding
effects of cocaine has also been established (Wise & Hoffman, 1992; Koob, 1992). However,
recent pharmacotherapies targeted at altering dopamine transmission have not been of therapeutic
value for people with cocaine addictions. There is a need for better understanding of other
important neurotransmitters involved in cocaine addiction and relapse.
Evidence supports the role of glutamate transmission within the nucleus accumbens in relapse
(Cornish, 1999). Although the stimulation of dopamine receptors within the nucleus accumbens
Cocaine Use and NMDA Receptor 31
has been found to reinstate drug-seeking for cocaine in rats, Cornish & Kalivas found that the
transmission of glutamate within the nucleus accumbens, and not dopamine, is important in
relapse following the systemic administration of cocaine (2000). The complex interaction
between dopamine and glutamate in the nucleus accumbens is not fully understood. Further, the
affect of this interaction on cocaine self-administration is unknown.
Systemic administration of LY235959 has been found to modulate cocaine self-administration
in rats when allowed short-access to self-administer (Allen et al., 2005; Pulvirenti, 1992; Pierce &
Kalivas, 1997). Present findings from this study support the role of glutamate transmission within
the central nervous system in facilitating the escalation of cocaine self-administration when given
long-access. Future studies involving the administration of glutamate altering treatments into
specific sites of the brain, such as the nucleus accumbens, are valuable for understanding the exact
neurobiological mechanism of the neurotransmitter on chronic cocaine consumption.
Cornish et al. found when subjects received intra-accumbens administration of NMDA and
AMPA receptor antagonists, they pressed fewer times under a 2 hour fixed-ratio session (1999).
The left-ward shift in lever-pressing was interpreted as a result of the enhanced rewarding
properties of cocaine. Permitting rats long access to cocaine typically results in increased
consumption (Ahmed & Koob, 1998). Future studies investigating glutamate and the escalation of
cocaine self-administration should implement the long-access model of drug intake, while
administering glutamate enhancing agents and antagonists directly into the site of interest within
the central nervous system. If glutamate transmission is a key factor in both the transition from
casual cocaine use to chronic consumption, and relapse to drug-taking, then the pharmacological
alteration of glutamate transmission may be a beneficial treatment of cocaine addiction.
The findings of the present study support the hypothesis that the chronic blockade of NMDA
receptors increases the escalation of cocaine self-administration. The motivation to consume
cocaine may also increase along with the facilitation of the escalation of cocaine consumption. A
hypoglutamatergic basis for some changes related to cocaine addiction is concordant with the
present findings. The mechanism of action for this phenomenon appears to be centrally mediated.
Future studies should target other key sites of action related to the rewarding properties of cocaine,
such as the nucleus accumbens, ventral tegmental area, and pre-frontal cortex.
Diagnostic Statistical Manual IV R Criteria for Substance Dependence
Maladaptive pattern of substance use, leading to clinical impairment or distress,
occurring over 12 month period.
Three or more of the following:
Need for increased amounts of substance to achieve intoxication
Markedly diminished effect with continued use of same amount
Characteristic withdrawal (w/d) syndrome (substance dependent)
Substance or substitute taken to relieve or avoid w/d symptoms
Substance taken in larger amounts or for longer period than was intended.
Persistent desire or unsuccessful efforts to cut down or control substance use; difficulty
stopping or limiting drug intake.
Great deal of time spent in activities related to acquiring, using, or recovering from the
substance; extremely high motivation to use.
Important social, occupational, or recreational activities are given up or reduced because
of substance use.
Continued substance use despite knowledge of having persistent or recurrent physical or
psychological problem that is likely to have been caused or exacerbated by substance use;
use despite harmful consequences.
University of Colorado at Denver and Health Sciences Center
Department of Biology
Location: North Classroom 3014
Campus Box 171, P.O. Box 173364
Denver, Colorado 80217-3364
Phone: 303-556-8440, Fax: 303-556-4352
UVI UAL CARE AND USE STATEMENT
ing the IACUC application from Dr. Richard Allen, Psychology, on the protocol using
S|prague-Dawley rats to examine whether increased cocaine consumption is related to
s in glutamatergic neurotransmission in corticomesolimbic pathways of the central
iervo|u|s system (for a NIDA grant application).
itocol for animal care and use has been approved by the University of Colorado at Denver
jional Animal Care and Use Committee. Through its general assurance to the Public
Service (Assurance #A3658-01, which expires 9-30-2007) the University assures the
flnlg agency that it will comply with the Public Health Service Policy on Humane Care and
f Laboratory Animals by Awardee Institutions.
r J. Stith
Â£ f Department, University of Colorado-Denver
Campuses: Downtown Denver Fitzsimons at Aurora Ninth and Colorado
Catheter with plastic tubing cover over guide cannula.
Plastic cap, tubing cover, guide cannula.
Catheter with plastic cap over tubing cover.
Cocaine Use and NMDA Receptor 35
Left, new open plastic cap Middle, old dosed plastic cap. Right,
catheter with plastic tubing cover.
Acute Infusions LY235959 ICV
Cocaine Use and NMDA Receptor 37
Cocaine Use and NMD A Receptor 38
0.2 LY235959 Pump
0.3 LY235959 Pump
Ahmed, S. H., & Koob, G. F. (1998). Transition from moderate to excessive drug intake: change
in hedonic set point. Science, Vol.282, 298-300.
Ahmed, S. H., Lin, D., Koob, G. F., & Parsons, L. H., (2003). Escalation of cocaine self
administration does not depend on altered cocaine-induced nucleus accumbens dopamine
levels. J. of Neurochemistry, Vol. 86, 102-113.
Carlezon, W. A. Jr., & Wise, R. A., (1996). Microinjections of phencyclidine (PCP) and related
drugs into nucleus accumbens shell potentiate medial forebrain bundle brain stimulation
reward. Psychopharmacology, Vol. 128, 413-420.
Carlezon, W. A. Jr., & Wise, R. A., (1996). Rewarding actions of phencyclidine and related drugs
in nucleus accumbens shell and frontal cortex. J. of Neuroscience, Vol. 16(9), 3112
Collins, E. D Ward, A. S., McDowell, D. M., Foltin, R. W & Fischman, M. W., (1998). The
effects of memantine on the subjective, reinforcing and cardiovascular effects of cocaine
in humans. Behavioural Pharmacology, Vol. 9, 587-598.
Cornish, J. L., Duffy, P., & Kalivas, P. W., (1999). A role of nucleus accumbens glutamate
transmission in the relapse to cocaine-seeking behavior. Neuroscience, Vol. 93(4), 1359
Cornish, J. L., & Kalivas, P. W (2000). Glutamate transmission in the nucleus accumbens
mediates relapse in cocaine addiction. J. of Neuroscience, Vol. 20, RC89 (1-5).
Dackis, C., & OBrien, C (2003). Glutamatergic agents for cocaine dependence. Annuals N.Y.
Academy of Science, Vol. 1300, 328-345.
Dackis, C. A., Kampman, K. M., Lynch, K. G., Pettinati, H. M., & OBrien, C. P., (2005). A
double-blind, placebo-controlled trial of modafinil for cocaine dependence.
Neuropsychopharmacology, Vol. 30, 205-211.
David,V., Durkin, T. P., & Cazala, P., (1998). Rewarding effects elicited by the microinjection of
either AMPA or NMDA glutamatergic antagonists into the ventral tegmental area
revealed by an intracranial self-administration paradigm in mice. European Journal of
Neuroscience, Vol 10, 1394-1402.
Deroche-Gamonet, V., Belin, D Pier, V. P., (2004). Evidence for addiction-like
behavior in the rat. Science, Vol. 305, 1014-1016.
Kalivas, P. W & Duffy, P., (1995). Di Receptors modulate glutamate transmission in the ventral
tegmental area. J. of Neuroscience, Vol. 15(7), 5379-5388.
Paterson, N. E., & Markou, A., (2003). Increased motivation for self-administered cocaine after
escalated cocaine intake. NeuroReport, Vol. 14(17), 2229-2239.
Cocaine Use and NMDA Receptor 40
Pierce, R. C., Bell, K., Dufly, P., & Kalivas, P. W., (1996). Repeated cocaine augments excitatory
amino acid transmission in the nucleus accummbens only in rats having developed
behavioral sensitization. J. of Neuroscience, Vol. 16(4), 1550-1560.
Stolerman, Ian. (1992). Drugs of abuse: behavioural priniciples, methods and terms. TIPS
Reviews, Vol. 13, 170-176.
Tang, W., Fasulo, W. H., Mash, D. C., & Hemby, S. E., (2003). Molecular profiling of midbrain
dopamine regions in cocaine overdose victims. J. Neurochemistry, Vol. 85, 911-924.
Tang, W., Wesley, M., Freeman, W. M., Liang, B., & Hemby, S. E., (2004). Alterations in
ionotropic glutamate receptor subunits during binge cocaine self-administration and
withdrawal in rats. J. of Neurochemistry, Vol. 89, 1021-1033.
Vanderschuren, L. J. M. J., & Everitt, B. J., (2004). Drug seeking becomes compulsive after
prolonged cocaine self-administration. Science, Vol. 305, 1017-1019.