D

Cv

Pa

b

a

ARRAA

KACCIRSY

1

tmteis22(

3

h0

ARTICLE IN PRESSG ModelAD-5222; No. of Pages 7

Drug and Alcohol Dependence xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Drug and Alcohol Dependence

j ourna l h o me pa ge: www.elsev ier .com/ locate /drugalcdep

ocaine-, caffeine-, and stress-evoked cocaine reinstatement in highs. low impulsive rats: Treatment with allopregnanolone

aul S. Regiera,∗, Alexander B. Claxtonb, Natalie E. Zlebnika,b, Marilyn E. Carrollb

Graduate Program in Neuroscience, University of Minnesota, Minneapolis, MN 55455, United StatesDepartment of Psychiatry, University of Minnesota, Minneapolis, MN 55455, United States

r t i c l e i n f o

rticle history:eceived 11 March 2014eceived in revised form 1 July 2014ccepted 2 July 2014vailable online xxx

eywords:llopregnanoloneaffeineocaine

mpulsivityeinstatementelf-administrationohimbine

a b s t r a c t

Background: Previous research indicates that individual differences in traits such as impulsivity, avidityfor sweets, and novelty reactivity are predictors of several aspects of drug addiction. Specifically, ratsthat rank high on these behavioral measures are more likely than their low drug-seeking counterpartsto exhibit several characteristics of drug-seeking behavior. In contrast, initial work suggests that the lowdrug-seeking animals are more reactive to negative events (e.g., punishment and anxiogenic stimuli).The goal of this study was to compare high and low impulsive rats on reinstatement of cocaine-seekingbehavior elicited by cocaine (COC) and by negative stimuli such as the stress-inducing agent yohimbine(YOH) or a high dose of caffeine (CAFF). An additional goal was to determine whether treatment withallopregnanolone (ALLO) would reduce reinstatement (or relapse) of cocaine-seeking behavior underthese priming conditions.Methods: Female rats were selected as high (HiI) or low (LoI) impulsive using a delay-discounting task.After selection, they were allowed to self-administer cocaine for 12 days. Cocaine was then replacedwith saline, and rats extinguished lever responding over 16 days. Subsequently, rats were pretreatedwith either vehicle control or ALLO, and cocaine seeking was reinstated by injections of COC, CAFF, orYOH.

Results: While there were no phenotype differences in maintenance and extinction of cocaine self-administration or reinstatement under control treatment conditions, ALLO attenuated COC- andCAFF-primed reinstatement in LoI but not HiI rats.Conclusions: Overall, the present findings suggest that individual differences in impulsive behavior mayinfluence efficacy of interventions aimed to reduce drug-seeking behavior.

© 2014 Elsevier Ireland Ltd. All rights reserved.

. Introduction

Vulnerability to drug addiction and relapse to drug seeking afterermination of use is determined by both genetic and environ-

ental factors. Studies from several laboratories have indicatedhat factors such as sex (Anker and Carroll, 2010b, 2011; Beckert al., 2012; Carroll and Anker, 2010), age (Spear, 2000), impuls-vity (Carroll et al., 2008, 2009, 2012; Perry and Carroll, 2008),weet preference (Dess et al., 1998, 2000, 2005; Carroll et al.,

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

008, 2012), novelty reactivity (Flagel et al., 2009; Kabbaj et al.,000), prenatal stress (Frye et al., 2011), and avidity for exerciseLarson and Carroll, 2005) predict vulnerability to drug-seeking

∗ Corresponding author at: Graduate Program in Neuroscience, 6-145 Jackson Hall,21 Church Street SE, Minneapolis, MN 55455, United States. Tel.: +1 612 626 6474.

E-mail address: regi0011@umn.edu (P.S. Regier).

ttp://dx.doi.org/10.1016/j.drugalcdep.2014.07.001376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.

behavior, particularly with stimulant drugs. These vulnerabil-ity factors (female > male, adolescent > adult, high impulsive > lowimpulsive, high sweet intake > low sweet intake, high novelty reac-tivity > low novelty reactivity, higher avidity for exercise > loweravidity for exercise) predict elevated drug seeking throughout sev-eral phases of drug self-administration, such as initiation (Daviset al., 2008; Perry et al., 2005; Perry and Carroll, 2008; Piazza et al.,1989), escalation (Anker et al., 2009b, 2010; Perry et al., 2008),resistance to extinction after termination of drug access in rats(Belin et al., 2011; Perry et al., 2008), reinstatement of responding(relapse) after termination of drug access (Larson and Carroll, 2005;Perry et al., 2006, 2008), and they affect motivation levels under aprogressive-ratio schedule (Belin et al., 2011; Carroll et al., 2002).

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

Some of these predictors of drug self-administration, like impulsi-vity (Dallery and Raiff, 2007; Krishnan-Sarin et al., 2007; Yoon et al.,2007), also predict susceptibility to drug abuse in humans (Elmanet al., 2001). For example higher vs. lower measures of impulsive

ING ModelD

2 hol D

b2

malweeoreascaiosom(ctbp

toeaIptbctifd2iaedoTlcp2ia

r

TR

ARTICLEAD-5222; No. of Pages 7

P.S. Regier et al. / Drug and Alco

ehavior predict self-report of more rewarding drug effects (de Wit,009; Perry and Carroll, 2008).

Recent research on individual differences in drug-seeking ani-als indicates corresponding differences in reactivity to negative

spects related to drugs of abuse. For example, animals selected forow impulsivity (LoI vs. HiI), low saccharin intake (LoS vs. HiS), as

ell as adults (vs. adolescents), and males (vs. females) not onlyxhibit less drug seeking than their counterparts, but they alsoxhibit more sensitivity to punishment by histamine, greater signsf withdrawal from drugs of abuse, greater taste aversion, greateresponse to anxiogenic stimuli, and greater acoustic startle (Carrollt al., 2009; Carroll and Holtz, 2014; Holtz et al., 2013, 2014; Holtznd Carroll, 2014; McLaughlin et al., 2011; Radke et al., 2014, butee Radke et al., 2013), while their counterparts (HiI, HiS, adoles-ents, and females, respectively) have greater response to positivespects of drugs (i.e., greater drug-seeking and drug-taking behav-ors, see review by Carroll et al., 2009). This raises the questionf to what extent are individual differences in drug seeking under-tood based on differential reactions to positive or negative aspectsf drugs. Furthermore, these differing behavioral characteristicsay be accompanied by underlying neurobiological characteristics

Flagel et al., 2010; Kabbaj and Akil, 2001; Regier et al., 2012) thatould interact with intervention techniques, such as treatmentshat aim to reduce drug seeking. Therefore, differential vulnera-ility to drug seeking may be an important factor for customizingrevention and treatment strategies for drug abuse in humans.

To date, there have been very few preclinical animal studieshat have addressed negative aspects of drugs of abuse in termsf the development and treatment of drug abuse, and there areven fewer studies in humans (Anker and Carroll, 2011; Carrollnd Holtz, 2014; Holtz et al., 2013; Holtz and Carroll, 2011, 2014).n terms of prevention, it is particularly important to know howositive vs. negative aspects of drugs will affect relapse in groupshat differentially administer drugs. Therefore, interventions coulde matched to differences in self-administration and the specificonditions that elicit relapse. For example, studies have shownhat female animals tend to show greater drug-seeking behav-or than male animals (see review by Carroll et al., 2009), andemale animals are more responsive to interventions that reducerug-seeking behavior including pharmacological (Campbell et al.,002; Carroll et al., 2001; Cosgrove and Carroll, 2004) and behav-

oral methods, such as access to an alternative reinforcer (Cosgrovend Carroll, 2003) or opportunity for aerobic exercise (Cosgrovet al., 2002; Zlebnik et al., 2014), available concurrently with therug. However, in recent studies of rats selectively bred for highr low saccharin intake (HiS vs. LoS), an opposite effect occurred.he lower self-administering LoS rats reduced their rate of esca-ation of cocaine intake, while their higher self-administering HiSounterparts increased their cocaine escalation when treated withrogesterone (Anker et al., 2012) or baclofen (Holtz and Carroll,011). Thus, initial evidence suggests that phenotypic differences

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

n self-administration might determine the success of interventionsimed to reduce drug seeking.

The purpose of the present study was to further examine theeduction of reinstatement to cocaine seeking evoked by cocaine

able 1einstatement groups and order of priming events.

Priming condition N

Cocaine (COC) 24Pretreatment

Prime

Caffeine (CAFF) 24Pretreatment

Prime

Yohimbine (YOH) 19Pretreatment

Prime

PRESSependence xxx (2014) xxx–xxx

(COC), caffeine (CAFF), and yohimbine (YOH) in high (HiI) vs. low(LoI) impulsive rats treated with allopregnanolone (ALLO) or avehicle control (VEH). Female rats were selected for HiI vs. LoIimpulsivity based on their performance on a delay-discounting taskthat was determined by their preference for a small-immediate(HiI) vs. large-delayed food reward (LoI) as described previously(Perry et al., 2005; Perry and Carroll, 2008). Allopregnanolone hasbeen used in previous rat studies to reduce cocaine- (Anker et al.,2009a; Anker and Carroll, 2010a) and methamphetamine- (Holtzet al., 2012) induced reinstatement of drug seeking in rats. SinceALLO, progesterone, and their precursor, pregnenolone have beenshow to have anxiolytic effects (Anker and Carroll, 2010a, 2011;Anker et al., 2007; Concas et al., 2000; Llaneza and Frye, 2009;Schneider and Popik, 2007), it was hypothesized that it may pro-duce a greater reduction of stress-induced (e.g., YOH and a highdose of CAFF priming injections) responding in LoI vs. HiI animals,since LoI rats may be more sensitive to anxiogenic stimuli or othernegative aspects of drugs (Holtz et al., 2013; Holtz and Carroll, 2011,2014). Female rats were used in this study, as they show a greaterdifference between HiI and LoI measures of cocaine seeking thanmale rats (Perry et al., 2008); thus, they would provide a higherbaseline of reinstatement for reduction by ALLO.

All rats were trained to self-administer iv cocaine, and theirdrug-seeking behavior was extinguished by substituting a salinesolution. Subsequently, reinstatement responding was comparedwhen different groups of rats were treated with ALLO or VEH beforeCOC, CAFF, or YOH. The priming conditions were selected to sim-ulate conditions that might elicit relapse in abstinent human drugabusers, such as COC itself, physiological stress (YOH), or a highdose of CAFF. Caffeine was included as a priming agent to serveas a model of a stimulant that is widely used in the human pop-ulation and for its anxiogenic characteristics when consumed athigh doses. The selected dose of CAFF (40 mg/kg) was hypothesizedto have an anxiogenic effect based on evidence from previous lit-erature. For example, an anxiogenic effect of 20 mg/kg CAFF wasfound in rats tested on an elevated plus maze (Gulick and Gouild,2009; Silva and Frussa-Filho, 2000), conditioned place and tasteaversions were reported in rats treated with 20 mg/kg (Steigerwaldet al., 1988; Myers and Izbicki, 2006) and 32 mg/kg (Vishwanathet al., 2011) CAFF, and other measures of stress such as CAFF with-drawal have been reported (Bhorkar et al., 2014). Based on Perryet al. (2008) it was hypothesized that HiI rats would reinstate res-ponding on the drug lever more than LoI rats to a positive priminginjection (COC), and LoI would exceed HiI rats on reinstatement toanxiogenic stimuli (YOH and a high dose of CAFF).

2. Methods

2.1. Subjects

Adult female Wistar rats (total n = 67) were used for this experiment (seeTable 1). Estrous cycle was not monitored to prevent disruption of cocaine-

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

maintained behavior by repeated vaginal lavage (Walker et al., 2002); therefore,results can be generalized across all phases of the estrous cycle. After a minimum of 3days of acclimation following arrival to the laboratory, rats were housed individuallyin plastic holding cages and moved daily into experimental chambers for delay-discounting testing. During the self-administration period following completion of

Reinstatement priming sequence

VEH VEH VEH ALLOSAL COC SAL COC

VEH VEH VEH ALLOSAL CAFF SAL CAFF

VEH VEH VEH ALLOSAL YOH SAL YOH

ING ModelD

hol D

drtttAp

2

twh(aseptTsdohlalotmbf5at9wacM

2

TrcBwlLh(

(egtt(thoipwic

p2lctmcwiw

significant main effect of phenotype or day and no significantphenotype × day interaction for responses (Table 2) or infusions(Table 3).

Table 2Mean (±SEM) responses during maintenance and extinction.

Phase Blocks of 4 days HiI responses LoI responses

MaintenanceDays 1–4 50.60 ± 1.60 50.07 ± 1.32Days 5–8 49.63 ± 1.91 53.55 ± 1.87Days 9–12 46.52 ± 1.35 53.30 ± 1.76

ARTICLEAD-5222; No. of Pages 7

P.S. Regier et al. / Drug and Alco

elay discounting, rats were housed in experimental chambers. For all conditions,ats were housed in temperature (24 ◦C) – and humidity-controlled rooms wherehere was a 12-h light/dark cycle (lights on at 6:00 am). Use of these animals forhis protocol (1008A87754) was approved by the University of Minnesota Institu-ional Animal Care and Use Committee and was accredited by the Association forssessment and Accreditation of Laboratory Animal Care (AAALAC). Recommendedrinciples of animal care were followed (National Research Council, 2003).

.2. Delay discounting

Adult female rats were tested on a delay-discounting task for food in experimen-al chambers described previously (Perry et al., 2008). These chambers were housedithin a wooden sound-attenuating enclosure equipped with a ventilation fan. Eachad a port for a water bottle for ad libitum water access and a 45-mg pellet feederCoulbourn Instrument, Lehigh Valley, PA) that was mounted on a stainless steel wallnd attached to a pellet delivery trough. On either side of the pellet feeder were twotandard response levers mounted about 4 cm above the cage floor. One lever deliv-red one pellet immediately after a lever press, and the other lever delivered threeellets after an adjusting delay. The lever that delivered one pellet immediately andhe lever that delivered three pellets after an adjusting delay alternated each day.he delay was adjusted based on the rat’s choices. A lever press on the immediateide decreased the delay by 1 s, and a lever press on the delayed side increased theelay by 1 s. The rats were tested at the same time 7 days a week for 2-h sessionsr 60 trials (four-trial blocks), whichever came first. Sessions began with a whiteouse light (4.6 W) and one of the sets of tricolored stimulus (situated above both

evers) lights being lit. When lit, the stimulus lights indicated that the lever wasctive. The first two trials consisted of a forced choice on each lever with stimulusights lit above the correct lever. The third and fourth trials consisted of a free choicen either lever with stimulus lights lit above both levers. At the end of the session,he final delay was recorded and used for the starting delay on the following day. A

ean adjusted delay (MAD) score was calculated by taking the total delay dividedy the number of free choice trials. Once a rat completed at least 50 trials per sessionor 5 days, and the difference in MAD scores across those 5 days was no greater than

s, an average MAD score across the 5 days was calculated. The rat was selecteds HiI when the average MAD score was <9 s, and the rat was selected as LoI whenhe average MAD score was >13 s. Even though it was rare, rats that fell in between

and 13 s were excluded from the study. The rationale for these selection criteriaas determined by a previous study in which Perry et al. (2005) used several rats

nd found a bimodal distribution that matched well to this range of numbers. Dataollection and experimental programming were controlled by PC computers andED-PC software (Med Associates, St. Albans, VT).

.3. Cocaine self-administration

After rats were selected as HiI or LoI, they underwent catheterization surgery.hey were anesthetized with ketamine (60 mg/kg) and xylazine (10 mg/kg) andeceived atropine (0.15 mL) and doxapram (5 mg/kg) to facilitate respiration. Ahronic indwelling polyurethane catheter (MRE-040-S-20, Braintree Scientific, Inc.,raintree, MA) was implanted in the right jugular vein. The other end of the catheteras led subcutaneously to an incision made medial and 1 cm caudal from the scapu-

ae and was connected to the cannula embedded in an infusion harness (Instechaboratories, Plymouth Meeting, PA). For 3 days after surgery, rats were giveneparin (0.2 mL, 50 units/mL, iv) to prevent clotting in the catheter and baytril2.0 mg/kg, iv) to prevent infection.

After the 3-day recovery period, rats were trained to self-administer cocaine0.4 mg/kg) in experimental chambers identical to the delay-discounting procedure,xcept that instead of a pellet dispenser, there was a holder for a jar containinground food. In addition, there was a syringe pump that contained a 30-mL syringehat delivered COC or SAL into the operant chamber via a tether (C31CS; Plas-ics One, Roanoke, VA), connected on one end to the rat harness and to a swivel050-0022, Alice King Chatham, Hawthorne, CA) on the other end. The rats wererained under an FR-1 reinforcement schedule during daily 2-h training sessions. Aouse light automatically turned on at 9 am every morning signaling the beginningf the session. During this time, a lever press on one lever (active lever) resultedn activation of tricolored stimulus lights above the lever and activation of theump, which delivered intravenous cocaine at a volume of 0.025 mL/100 g bodyeight (duration = 1 s/100 g). Responses on the other lever (inactive lever) resulted

n illumination of stimulus lights above the lever but had no other programmedonsequences.

During acquisition, responding by rats was facilitated by baiting the lever witheanut butter and providing non-contingent infusions. Rats were required to earn5 or more infusions a session and to maintain a 2:1 ratio of active vs. inactive

ever responses for three consecutive sessions. Once rats could reliably meet theseriteria without the lever being baited or non-contingent infusions being provided,hey entered the maintenance phase, where their responses and infusions were

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

onitored and recorded for 12 days. During this time, catheters and tethers werehecked daily for leaks with a heparin/saline solution, and every 5–7 days cathetersere checked for patency by flushing with an iv solution (Caine et al., 1999) contain-

ng 30 mg/mL ketamine, 1.5 mg/mL midazolam, and saline (0.1–0.2 mL, iv). Patencyas inferred by loss of righting reflex. After 12 days of rats maintaining at least 25

PRESSependence xxx (2014) xxx–xxx 3

infusions per session with a 2:1 active to inactive lever responses, the extinctionperiod began.

The extinction phase lasted for 16 days, and the consequences of lever respon-ding remained identical to maintenance, except that COC was replaced with SAL.After extinction, and for 3 days prior to reinstatement, the stimulus lights, syringepump, and house light were disconnected. Responses on both levers during thispre-reinstatement period were recorded but had no programmed consequences.

Subsequently, the rats entered into the reinstatement period, where they weredivided into three separate groups. Lights and pump remained off during this time,and rats received alternating ip injections of saline (SAL) and drug solution (COC,CAFF, or YOH) with pretreatment 30 min before each daily session with either peanutoil (VEH) or ALLO (15 mg/kg). Each drug and VEH or ALLO combination was admin-istered only once, with SAL and VEH combination being administered on days inbetween the administration of drug and VEH or ALLO combination (Table 1).

2.4. Drugs

Cocaine HCL was provided by the National Institute of Drug Abuse (ResearchTriangle Institute, Research Triangle Park, NC). It was dissolved in sterile 0.9% saline,and the anticoagulant heparin (1 mL heparin/200 mL of saline) was added to preventthrombin accumulation. Caffeine and ALLO were obtained from Sigma Aldrich (St.Louis, MO) and dissolved in saline (40 mg/mL) and peanut oil (15 mg/kg), respec-tively. Yohimbine (2.5 mg/mL) was obtained from Lloyd Laboratories (Shenandoah,IA) and came in an injectable form.

2.5. Data analysis

Primary dependent measures included MADs during the delay discounting task,responses and infusions during maintenance and extinction of self-administration,and responses during reinstatement of cocaine-seeking behavior. Repeated meas-ures were days during maintenance and extinction and injection type duringreinstatement. Outlying values within each group that were two standard devi-ations outside of the mean were excluded from analysis. MADs were comparedbetween HiI and LoI rats using an unpaired two-tailed Student’s t-test. For mainte-nance and extinction, responses and infusions were averaged across 4-day blocksto reduce variability and the number of post hoc contrasts and were analyzedusing a two-factor mixed ANOVA (phenotype × days). For reinstatement, groupsthat received different priming injections (e.g., COC, CAFF, YOH) were analyzedseparately using two-factor mixed ANOVA (phenotype × priming injection, e.g.,VEH/COC, ALLO/COC). Dunn’s (Bonferroni) procedure was used for post hoc analy-ses, and results were considered significant if p < 0.05. All analyses were completedusing GBStat (Dynamic Microsystems, Silver Spring, MD).

3. Results

3.1. Mean adjusted delay

Confirming behavioral phenotype, mean HiI MAD scores(4.71 ± 0.33) were significantly lower than mean LoI MAD scores(20.67 ± 1.44) (t(65) = 11.17, p < 0.0001).

3.2. Maintenance

Mean responses and infusions (4-day blocks) for thecocaine self-administration maintenance period are shown inTables 2 and 3. Results of the two-factor ANOVA revealed no

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

Extinction

Days 1–4 29.52 ± 2.27 32.42 ± 2.62Days 5–8 11.28 ± 0.99 8.68 ± 0.80Days 9–12 6.63 ± 0.63 5.4 ± 0.69Days 13–16 3.90 ± 0.43 3.76 ± 0.50

ARTICLE IN PRESSG ModelDAD-5222; No. of Pages 7

4 P.S. Regier et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx

Table 3Mean (±SEM) infusions during maintenance and extinction.

Phase Blocks of 4 days HiI infusions LoI infusions

MaintenanceDays 1–4 40.41 ± 1.03 38.40 ± 0.91Days 5–8 39.79 ± 1.06 41.24 ± 1.14Days 9–12 38.44 ± 1.07 42.79 ± 1.37

Days 1–4 22.38 ± 1.66 26.22 ± 2.06Days 5–8 8.54 ± 1.66 6.38 ± 0.55

3

omwia

appppb

3

3pst(nH

3cni(nn

Fss

Fig. 2. Mean (±SEM) responses for caffeine- (CAFF) primed reinstatement ofcocaine-seeking behavior. Pretreatment with allopregnanolone (ALLO) attenuatedcocaine-seeking responses in low (LoI) but not high (HiI) impulsive rats. *p < 0.05.

Extinction Days 9–12 4.69 ± 0.70 3.88 ± 0.44Days 13–16 2.75 ± 0.32 2.81 ± 0.36

.3. Extinction

Table 2 displays the mean number of responses (4-day blocks)ver the 16-day extinction period. While there was no significantain effect of phenotype or a phenotype × day interaction, thereas a significant main effect of day (F(3,267) = 171.80, p < 0.0001),

ndicating a decrease in responding over the extinction period forll rats.

Saline infusions (Table 3) over the extinction period werenalyzed similarly. Results indicated no significant main effect ofhenotype but a significant main effect of day (F(3,267) = 128.37,

< 0.0001) and phenotype × day interaction (F(3,267) = 2.74, < 0.05). Post hoc analyses revealed a notable decline in res-onding in all rats over the extinction period but no differencesetween the HiI and LoI rats.

.4. Reinstatement

.4.1. Cocaine-primed reinstatement. Cocaine-seeking responsesrimed by cocaine are displayed in Fig. 1. Results indicated noignificant main effect of phenotype or phenotype × priming injec-ion interaction, but there was a main effect of priming injectionF(1,45) = 10.41, p < 0.01). Post hoc analyses revealed that ALLO sig-ificantly reduced COC-primed responding in LoI (p < 0.01) but notiI rats.

.4.2. Caffeine-primed reinstatement. Results for CAFF-inducedocaine seeking are displayed in Fig. 2. While there was no sig-ificant main effect of phenotype and no phenotype × priming

njection interaction, there was a main effect of priming injection

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

F(1,43) = 12.49, p < 0.01). Post hoc analyses revealed that ALLO sig-ificantly reduced CAFF-induced responding in LoI (p < 0.05) butot HiI rats.

ig. 1. Mean (±SEM) responses for cocaine- (COC) primed reinstatement of cocaine-eeking behavior. Pretreatment with allopregnanolone (ALLO) attenuated cocaine-eeking responses in low (LoI) but not high (HiI) impulsive rats. *p < 0.05.

Fig. 3. Mean (±SEM) responses for yohimbine- (YOH) primed reinstatement ofcocaine-seeking behavior in low (LoI) and high (HiI) impulsive rats. No significantdifferences were found.

3.4.3. Yohimbine-primed reinstatement. Yohimbine-inducedcocaine-seeking responses are displayed in Fig. 3. There were nosignificant main effects of phenotype or priming injection and nophenotype × priming injection interaction.

4. Discussion

One goal of this study was to compare HiI and LoI ratson a cocaine self-administration, extinction, and reinstatement(relapse) procedure to determine whether there were phenotypedifferences (HiI vs. LoI) in response to treatment with ALLO (vs.VEH). Previous work with HiI and LoI rats indicated that HiI ratsacquired cocaine self-administration faster than LoI rats (Perryet al., 2008), they escalated their intake more than LoI rats (Ankeret al., 2009b), and they showed higher levels of reinstatement res-ponding than LoI rats when given a priming injection of 15 mg/kgbut not 10 mg/kg cocaine (Perry et al., 2008). The reinstatementresults from the present study differed from previous results thatshowed greater reinstatement in HiI vs. LoI, which may have beendue to slightly different reinstatement procedures. Little is known,however, regarding how HiI vs. LoI and other animals that differen-

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

tially administer drugs respond to interventions aimed to reducedrug seeking. The present results indicated that LoI rats reducedtheir drug seeking more than HiI rats when treated with ALLO(Figs. 1 and 2).

ING ModelD

hol D

rprbstcldmatto

LptasraboecrsLicp2ei

idpcabamoTmtarAaa22tai

epiCps

ARTICLEAD-5222; No. of Pages 7

P.S. Regier et al. / Drug and Alco

The result that ALLO attenuated reinstatement more in LoIats than HiI rats was consistent with recent studies involvinghenotypes, such as high (HiS) vs. low (LoS) saccharin-preferringats that were treated with progesterone (Anker et al., 2012) oraclofen (Holtz and Carroll, 2011) during long-access to cocaineelf-administration. There was a more marked effect of proges-erone and baclofen in LoS than HiS rats compared to their vehicleontrols. In these previous studies, as well as the present study, theower drug-seeking phenotype had better success than the higherrug-seeking phenotype with interventions aiming to reduce drug-otivated responding. The baseline rates of behavior before ALLO

dministration were equal in all of these studies, with no pheno-ypic differences in maintenance or extinction; thus, the differentreatment effects could not be explained by different baseline ratesf drug seeking.

While these initial studies with the HiI vs. LoI and HiS vs.oS rats suggest greater reductions in drug-seeking behavior byharmacological interventions in the low drug-seeking pheno-ype, studies comparing other vulnerability characteristics, suchs sex, show different results. For example, female rats generallyhow greater drug self-administration behavior than male rats (seeeviews; Anker and Carroll, 2011; Carroll et al., 2008, 2009; Carrollnd Anker, 2010), but when treated with medications such asaclofen (Campbell et al., 2002), ketoconazole (Carroll et al., 2001),r behavioral interventions such as a nondrug rewards (Cosgrovet al., 2002), female rats consistently showed greater reductions inocaine or opioid self-administration than male rats. This oppositeesponse (females > males) to interventions aimed to reduce drugelf-administration compared to the other phenotypes (LoS > HiS,oI > HiI) may be related to hormonal status (e.g., progesteronen females). Progesterone and its metabolites (e.g., ALLO) reduceocaine-seeking behavior through several phases of the addictionrocess in female rats more than male rats (see Anker and Carroll,010b; Carelli, 2002). Thus, endogenous progesterone may havenhanced the other pharmacological and behavioral interventionsn female rats but not male rats.

The present results with ALLO are consistent with previous find-ngs that progesterone and its metabolite ALLO (Anker et al., 2009a)ecreased cocaine-primed reinstatement in female rats. Elevatedrogesterone levels in freely cycling female rats were also asso-iated with decreased cocaine-induced reinstatement (Feltensteinnd See, 2007). Finasteride, a 5-alpha reductase inhibitor thatlocks the metabolism of progesterone to ALLO also blocked thettenuating effects of progesterone on cocaine-primed reinstate-ent (Anker et al., 2009a). Thus, progesterone’s inhibitory effect

n cocaine seeking in females may be through its metabolite, ALLO.he behavioral effects of ALLO may be explained by its indirectodulation of dopamine in the nucleus accumbens and ventral

egmental areas of the brain (Carelli, 2002; Rutsuko et al., 2004)nd by acting as a potent positive allosteric modulator of GABAA-eceptors (for reviews see Lambert et al., 1995; Gasoir et al., 1999).nother mechanism of ALLO, that is not mutually exclusive, is itsbility to alter stress- and cocaine-related hypothalamic pituitarydrenal (HPA) activation, which is increased by stress- (Stewart,000; Weiss et al., 2001), and cue-induced (Goeders and Clampitt,002) reinstatement. ALLO has been shown to reduce HPA activa-ion by decreasing corticotropin-releasing hormone levels (Lynchnd Sofuoglu, 2010; Patchev et al., 1995) and preventing increasesn plasma corticosterone following stress (Owens et al., 1992).

Another goal of the present study was to determine whether theffectiveness of ALLO to reduce reinstatement was related to theriming condition: specifically, whether it was considered reward-

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

ng or anxiogenic. We used three different priming conditions, COC,AFF, and YOH to represent positive (COC) or negative (CAFF, YOH)harmacological stimuli that have been reported to elicit drug-eeking behavior formerly rewarded by cocaine. In previous studies

PRESSependence xxx (2014) xxx–xxx 5

of individual differences in drug seeking, results demonstrated dif-ferential responding to the positive and negative aspects of drugtaking in HiS and LoS rats. While HiS rats exceeded LoS rats indrug intake (Perry et al., 2006; Anker et al., 2012) and reinstate-ment of drug-seeking behavior (Perry et al., 2006), LoS rats (vs.HiS) revealed greater reactivity to negative events, such as foodrestriction (Carroll et al., 2012) and withdrawal from ethanol (Desset al., 2005), glucose (Yakovenko et al., 2011) morphine (Holtz et al.,2014), and cocaine (Radke et al., 2014). Given the previous findingthat lower drug-seeking animals respond more to negative aspectsof drugs compared to higher drug-seeking animals, one potentialexplanation for the results that ALLO reduced reinstatement morein LoI than HiI rats is that these pharmacological manipulationsproduced more of a stress response in LoI rats compared to HiIrats. Cocaine has been shown to induce stress hormones, which isattenuated by ALLO (Frye, 2007), and it is possible that COC and ahigh dose of CAFF differentially affect stress pathways in LoI andHiI rats. Previous work has shown an induction of ALLO by COC(Kohtz et al., 2010). Therefore, another explanation is that HiI andLoI rats could produce different levels of ALLO in response to COCand CAFF, so that treatment with ALLO might affect HiI and LoI dif-ferently. Further work is needed to investigate potential differencesof stress hormones and levels of ALLO in response to pharmacolog-ical manipulations in HiI and LoI rats.

While current results yielded phenotype differences in responseto treatment with ALLO (Figs. 1 and 2), results did not demonstratedifferential reinstatement of cocaine-seeking behavior primed byCOC, CAFF or YOH injections under control conditions in the HiI andLoI rats. Further work is needed to fully examine the type of primingstimuli and the role of their positive (COC) and negative (high doseof CAFF, YOH) pharmacological effects. The finding of similar resultsin HiI vs. LoI and HiS vs. LoS rats (i.e., that behavior was alleviatedby treatment with ALLO in females) is important, because com-mon substances such as CAFF may provoke substantial relapse incocaine-abstinent individuals, and common neurobiological mech-anisms may underlie these effects. Improved strategies to reducedrug-seeking behavior are needed and translational significancemay result from a better understanding of these individual differ-ences and a range of environmental events that prompt relapse.

In summary, the present results indicate that LoI rats exhibitedgreater reductions than HiI rats in COC and CAFF-primed reinstate-ment due to treatment of ALLO. Recent evidence suggests a role forALLO and its precursors, progesterone (Anker et al., 2010) and pre-gnenolone (Valee et al., 2014), in the development and treatmentof various forms of drug addiction. The present findings emphasizethe potential success of matching effective interventions to reducedrug seeking to individual differences in drug self-administration.Results from this study show greater reduction of reinstatement forlower drug-seeking animals using a novel therapeutic approach,ALLO, that attenuated COC- and CAFF-induced reinstatement ofcocaine-seeking behavior. Further studies will be needed in orderto discover interventions that will be successful in reducing drug-motivated behavior in higher drug-seeking animals.

Role of funding source

This study was supported by NIDA grants R01 DA003240(M.E.C.) and K05 DA015267 (M.E.C.).

Contributors

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

All authors have approved of and contributed to the finalmanuscript. M.E.C. and P.S.R. were involved with the design of theexperiment and graphic presentation. M.E.C., P.S.R., and N.E.Z. wereinvolved with the data analysis. P.S.R. and A.B.C. were involved with

ING ModelD

6 hol D

dw

C

A

KRW

R

A

A

A

A

A

A

A

A

B

B

B

C

C

C

C

C

C

C

C

C

C

ARTICLEAD-5222; No. of Pages 7

P.S. Regier et al. / Drug and Alco

irecting the daily experimental procedures. P.S.R., A.B.C., and N.E.Z.ere involved with the data collection.

onflict of interest

No conflict declared.

cknowledgments

The authors would like to thank Justin Anker, Thomas Baron,atie Bressler, Nathan Holtz, Seth Johnson, Sean Navin, Amy Saykao,achael Turner, Aneal Rege, Tyler Rehbein, Troy Velie, and Jeremyilliams for technical assistance.

eferences

nker, J.J., Carroll, M.E., 2011. Females are more vulnerable to drug abuse than males:evidence from preclinical studies and role of ovarian hormones. In: Neill, J.C.,Kulkarni, J. (Eds.), Biological Basis of Sex Differences in Psychopharmacology,vol. 8. Springer, London, pp. 73–96.

nker, J.J., Carroll, M.E., 2010a. Sex differences in the effects of allopregnanoloneon yohimbine-induced reinstatement of cocaine seeking in rats. Drug AlcoholDepend. 107, 264–267.

nker, J.J., Carroll, M.E., 2010b. The role of progestins in the behavioral effects ofcocaine and other drugs of abuse: human and animal research. Neurosci. Biobe-hav. Rev. 35, 315–333.

nker, J.J., Holtz, N.A., Carroll, M.E., 2012. Effects of progesterone on escalation ofintravenous cocaine self-administration in rats selectively bred for high or lowsaccharin intake. Behav. Pharmacol. 23, 205–2010.

nker, J.J., Holtz, N.A., Zlebnik, N., Carroll, M.E., 2009a. Effects of allopregnanoloneon the reinstatement of cocaine-seeking behavior in male and female rats. Psy-chopharmacology (Berl.) 203, 63–72.

nker, J.J., Larson, E.B., Gliddon, L.A., Carroll, M.E., 2007. Effects of progesterone onthe reinstatement of cocaine-seeking behavior in female rats. Exp. Clin. Psy-chopharmacol. 15, 472–480.

nker, J.J., Perry, J.L., Gliddon, L.A., Carroll, M.E., 2009b. Impulsivity predicts the esca-lation of cocaine self-administration in rats. Pharmacol. Biochem. Behav. 93,343–348.

nker, J.J., Zlebnik, N.E., Carroll, M.E., 2010. Differential effects of allopregnanoloneon the escalation of cocaine self-administration and sucrose intake in femalerats. Psychopharmacology (Berl.) 212, 419–429.

ecker, J.B., Perry, A.N., Westenbroek, C., 2012. Sex differences in the neural mech-anisms mediating addiction: a new synthesis and hypothesis. Biol. Sex Differ. 3,14–35.

elin, D., Berson, N., Balado, E., Piazza, P.V., Deroche-Gamonet, V., 2011. High-noveltypreference rats are predisposed to compulsive cocaine self-administration. Neu-ropsychopharmacology 36, 569–579.

horkar, A.A., Dandekar, M.P., Nekhate, K.T., Subhedar, N.K., Kokare, D.M., 2014.Involvement of the central melanocortin system in the effects of caffeine onanxiety-like behavior in mice. Life Sci. 95, 72–80.

aine, S.B., Negus, S.S., Mello, N.K., 1999. Method for training operant responding andevaluating cocaine self-administration behavior in mutant mice. Psychophar-macology (Berl.) 147, 22–24.

ampbell, U.C., Morgan, A.D., Carroll, M.E., 2002. Sex differences in the effect ofbaclofen on the acquisition of cocaine self-administration in rats. Drug AlcoholDepend. 66, 61–69.

arelli, R.M., 2002. Nucleus accumbens cell firing during goal-directed behaviors forcocaine vs. ‘natural’ reinforcement. Physiol. Behav. 76, 379–387.

arroll, M.E., Anker, J.J., 2010. Sex differences and ovarian steroid hormones in ani-mal models of drug dependence. Horm. Behav. 58, 44–56.

arroll, M.E., Anker, J.J., Perry, J.L., 2009. Modeling risk factors for nicotine and otherdrug abuse in the preclinical laboratory. Drug Alcohol Depend. 104, S70–S78.

arroll, M.E., Campbell, U.C., Heideman, P., 2001. Ketoconazole suppresses foodrestriction-induced increases in heroin self-administration in rats: sex differ-ences. Exp. Clin. Psychopharmacol. 9, 307–316.

arroll, M.E., Holtz, N.A., 2014. The relationship between feeding and drug-seekingbehaviors. In: Brewerton, T.D., Dennis, A.B. (Eds.), Eating Disorders, Addic-tions and Substance Use Disorders: Research, Clinical and Treatment Aspects.Springer-Verlag, New York.

arroll, M.E., Holtz, N.A., Zlebnik, N.E., 2012. Saccharin preference in rats: relationto impulsivity and comorbid drug intake. In: Avena, N.M. (Ed.), Animal Modelsof Eating Disorders. Springer-Verlag, New York, pp. 201–234, 380.

arroll, M.E., Morgan, A.D., Anker, J.J., Perry, J.L., 2008. Selective breeding for differ-ential saccharin intake as an animal model of drug abuse. Behav. Pharmacol. 19,

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

435–460.arroll, M.E., Morgan, A.D., Campbell, U.C., Lynch, W.J., Dess, N.K., 2002. Intravenous

cocaine and heroin self-administration in rats selectively bred for differentialsaccharin intake: phenotype and sex differences. Psychopharmacology (Berl.)161, 304–313.

PRESSependence xxx (2014) xxx–xxx

Concas, A., Porcu, P., Sogliano, C., Serra, M., Purdy, R.H., Biggio, G., 2000. Caffeine-induced increases in the brain and plasma concentrations of neuroactive steroidsin the rat. Pharmacol. Biochem. Behav. 66, 39–45.

Cosgrove, K.P., Carroll, M.E., 2003. Differential effects of a nondrug reinforcer, sac-charin, on oral self-administration of phencyclidine (PCP) in male and femalerhesus monkeys. Psychopharmacology (Berl.) 170, 9–16.

Cosgrove, K.P., Carroll, M.E., 2004. Effects of bremazocine on oral phencyclidine (PCP)self-administration in male and female rhesus monkeys. Exp. Clin. Psychophar-macol. 12, 111–117.

Cosgrove, K.P., Hunter, R., Carroll, M.E., 2002. Wheel-running attenuates intravenouscocaine self-administration in rats: sex differences. Pharmacol. Biochem. Behav.73, 663–671.

Dallery, J., Raiff, B.R., 2007. Delay discounting predicts cigarette smoking in a lab-oratory model of abstinence reinforcement. Psychopharmacology (Berl.) 190,485–496.

Davis, B.A., Clinton, S.M., Akil, H., Becker, J.B., 2008. The effects of novelty-seekingphenotypes and sex differences on acquisition of cocaine self-administration inselectively bred high-responder and low-responder rats. Pharmacol. Biochem.Behav. 90, 331–338.

Dess, N.K., Arnal, J., Chapman, C.D., Siebel, S., VenderWeele, D.A., Green, K., 2000.Exploring adaptations to famine: rats selectively bred for differential saccha-rin intake differ on deprivation-induced hyperactivity and emotionality. Int. J.Comp. Psychol. 13, 34–512.

Dess, N.K., Badia-Elder, N.E., Thiele, T.E., Kiefer, S.W., Blizard, D.A., 1998. Ethanolconsumption in rats selectively bred for differential saccharin intake. Alcohol16, 275–278.

Dess, N.K., O’Neil, P., Chapman, C.D., 2005. Ethanol withdrawal and proclivity areinversely related in rats selectively bred for differential saccharin intake. Alcohol37, 9–22.

de Wit, H., 2009. Impulsivity as a determinant and consequence of drug use: a reviewof underlying processes. Addict. Biol. 14, 22–31.

Elman, I., Krause, S., Karlsgodt, K., Schoenfiel, D.A., Gollub, R.L., Beriter, H.C.,Gastfriend, D.R., 2001. Clinical outcomes following cocaine infusion innontreatment-seeking individuals with cocaine dependence. Biol. Psychiatry 49,553–555.

Feltenstein, M.W., See, R.E., 2007. Plasma progesterone levels and cocaine-seekingin freely cycling female rats across the estrous cycle. Drug Alcohol Depend. 89,183–189.

Flagel, S.B., Akil, H., Robinson, T.E., 2009. Individual differences in the attributionof incentive salience to reward-related cues; implications for addiction. Neu-ropharmacology 56, 139–148.

Flagel, S.B., Robinson, T.E., Clark, J.J., Clinton, S.M., Watson, S.J., Seeman, P., Phillips,P.E., Akil, H., 2010. An animal model of genetic vulnerability to behavioral disin-hibition and responsiveness to reward-related cues: implications for addiction.Neuropsychopharmacology 35, 388–400.

Frye, C.A., Paris, J.J., Osborne, D.M., Campbell, J.C., Kippin, T.E., 2011. Prenatal stressalters progestogens to mediate susceptibility to sex-typical, stress-sensitive dis-orders, such as drug abuse: a review. Front. Psychiatry 2, 1–15.

Frye, C.A., 2007. Progestins influence motivation, reward, conditioning, stress,and/or response to drugs of abuse. Pharmacol. Biochem. Behav. 2, 209–219.

Gasoir, M., Carter, R.B., Witkin, J.M., 1999. Neuroactive steroids: potential thera-peutic use in neurological andpsychiatric disorders. Trends Pharmacol. Sci. 20,107–112.

Goeders, N.E., Clampitt, D.M., 2002. Pothetial role for the hypothalamo-pituitary-adrenal axis in the conditioned reinforcer-induced reinstatment of extinguishedcocaine seeking in rats. Psychopharmacology 16, 222–232.

Gulick, D., Gouild, T.J., 2009. Effects of ethanol and caffeine on behavior in C57BL/6mice in the plus-maze discriminative avoidance task. Behav. Neurosci. 123,1271–1278.

Holtz, N.S., Anker, J.J., Regier, P.S., Claxton, A., Carroll, M.E., 2013. Cocaine self-administration punished by i.v. histamine in rat models of high and low drugabuse vulnerability: effect of saccharin preference, impulsivity, and sex. Physiol.Behav. 122, 32–38.

Holtz, N.A., Carroll, M.E., 2014. Animal models of addiction: genetic influences. In:Kim, Y.-K., Gewirtz, J. (Eds.), Animal Models for Behavior Genetics Research.Handbook of Behavior Genetics, vol. 7. Springer, London, UK (in press).

Holtz, N.A., Carroll, M.E., 2011. Baclofen has opposite effects on escalation of i.v.cocaine self-administration in rats selectively bred for high (HiS) and low (LoS)saccharin intake. Pharmacol. Biochem. Behav. 100, 275–283.

Holtz, N.A., Lozama, A., Prisinzano, T.E., Carroll, M.E., 2012. Reinstatement ofmethamphetamine seeking in male and female rats treated with modafinil andallopregnanolone. Drug Alcohol Depend. 120, 233–237.

Holtz, N.A., Radke, A.K., Zlebnik, N.E., Harris, A.C., Carroll, M.E., 2014. Intracranial self-stimulation reward thresholds during morphine withdrawal in rats selectivelybred for high (HiS) and low (LoS) saccharin intake. Brain Res. (submitted forpublication).

Kabbaj, M., Akil, H., 2001. Individual differences in novelty-seeking behavior in rats:a c-fos study. Neuroscience 106, 535–545.

Kabbaj, M., Devine, D.P., Savage, V.R., Akil, H., 2000. Neurobiological correlates ofindividual differences in novelty-seeking behavior in the rat: differential expres-sion of stress-related molecules. J. Neurosci. 20, 6983–6988.

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

Kohtz, A.S., Paris, J.J., Frye, C.A., 2010. Low doses of cocaine decrease, and high dosesincrease, anxiety-like behavior and brain progestogen levels among intact rats.Horm. Behav. 57, 474–480.

Krishnan-Sarin, S., Reynolds, B., Duhig, A.M., Smith, A., Liss, T., McFetridge, A., Cav-allo, D.A., Carroll, K.M., Potenza, M.N., 2007. Behavioral impulsivity predicts

ING ModelD

hol D

L

L

L

L

M

M

N

O

P

P

P

P

P

P

R

ARTICLEAD-5222; No. of Pages 7

P.S. Regier et al. / Drug and Alco

treatment outcome in a smoking cessation program for adolescent smokers.Drug Alcohol Depend. 88, 79–82.

ambert, J.J., Belelli, D., Hill-Venning, C., Peters, J.A., 1995. Neurosteroids and GABAAreceptor function. Trends Pharmacol. Sci. 16, 295–303.

arson, E.B., Carroll, M.E., 2005. Wheel-running as a predictor of cocaine self-administration and reinstatement in female rats. Pharmacol. Biochem. Behav.82, 590–600.

laneza, D.C., Frye, C.A., 2009. Progestogens and estrogen influence impulsive bury-ing and avoidant freezing behavior of naturally cycling and ovariectomized rats.Pharmacol. Biochem. Behav. 93, 337–342.

ynch, W.J., Sofuoglu, M., 2010. Role of progesterone in nicotine addiction: evidencefrom initiation to relapse. Exp. Clin. Psychopharmacol. 18, 451–461.

cLaughlin, I.B., Dess, N.K., Chapman, C.D., 2011. Modulation of methylphenidateeffects on wheel running and acoustic startle by acute food deprivation in com-mercially and selectively bred rats. Pharmacol. Biochem. Behav. 97, 500–508.

yers, K.P., Izbicki, E.V., 2006. Reinforcing and aversive effects of caffeine measuredby flavor preference conditioning in caffeine-naive and caffeine-acclimated rats.Physiol. Behav. 88, 585–596.

ational Research Council, 2003. Guidelines for the Care and Use of Mammals inNeuroscience and Behavioral Research. The National Academies, Washington,DC, pp. 209.

wens, M.J., Ritchie, J.C., Nemeroff, C.B., 1992. 5 alpha-pregnane-3 alpha, 21-diol-20-one(THDOC) attenuates mild stress-induced increases in plasma corticosteronevia a non-glucocorticoid mechanism: comparison with alprazolam. Brain Res.573, 353–355.

atchev, V.K., Hayashi, S., Orikasa, C., Almeida, O.B., 1995. Implications of estrogen-dependent brain organization for gender differences in hypothalamo-pituitary-adrenal regulation. FASEB J. 9, 419–423.

erry, J.L., Carroll, M.E., 2008. The role of impulsive behavior in drug abuse. Psy-chopharmacology (Berl.) 200, 1–26.

erry, J.L., Dess, N.K., Morgan, A.D., Anker, J.J., Carroll, M.E., 2006. Escalation of IVcocaine self-administration and reinstatement of cocaine-seeking behavior inrats selectively bred for high and low saccharin intake. Psychopharmacology(Berl.) 186, 235–245.

erry, J.L., Larson, E.B., German, J.P., Madden, G.J., Carroll, M.E., 2005. Impuls-ivity (delay discounting) as a predictor of acquisition of i.v. cocaineself-administration in female rats. Psychopharmacology (Berl.) 178, 193–201.

erry, J.L., Nelson, S.E., Carroll, M.E., 2008. Impulsive choice as a predictor of acqui-sition of i.v. cocaine self-administration and reinstatement of cocaine-seekingbehavior in male and female rats. Exp. Clin. Psychopharmacol. 16, 165–177.

iazza, P.V., Deminiere, J.M., Le Moal, M., Simon, H., 1989. Factors that pre-

Please cite this article in press as: Regier, P.S., et al., Cocaine-, caffeine-, arats: Treatment with allopregnanolone. Drug Alcohol Depend. (2014)

dict individual vulnerability to amphetamine self-administration. Science 245,1511–1513.

adke, A.K., Holtz, N.A., Gewirtz, J.C., Carroll, M.E., 2013. Reduced emotional signsof opiate withdrawal in animals selectively bred for low (LoS) versus high (HiS)saccharin intake. Psychopharmacology (Berl.) 227, 117–126.

PRESSependence xxx (2014) xxx–xxx 7

Radke, A.K., Zlebnik, N.E., Carroll, M.E., 2014. Cocaine reward and withdrawal in ratsselectively bred for low (LoS) versus high (HiS) saccharin intake. Psychophar-macology (submitted for publication).

Regier, P.S., Carroll, M.E., Meisel, R.L., 2012. Cocaine-induced c-Fos expression in ratsselectively bred for high or low saccharin intake and in rats selected for high orlow impulsivity. Behav. Brain Res. 233, 271–279.

Rutsuko, I., Robbins, T.W., Everitt, B., 2004. Differential control over cocaine-seeking behavior by nucleus accumbens core and shell. Nat. Neurosci. 7,389–397.

Schneider, T., Popik, P., 2007. Attenuation of estrous cycle-dependent marble bury-ing in female rats by acute treatment with progesterone and antidepressants.Psychoneuroendocrinology 21, 651–659.

Silva, R.H., Frussa-Filho, R., 2000. The plus-maze discriminative avoidance task: anew model to study memory–anxiety interactions. Effects of chlordiazepoxideand caffeine. J. Neurosci. Methods 102, 117–125.

Spear, L.P., 2000. The adolescent brain and age-related behavioral manifestations.Neurosci. Biobehav. Rev. 24, 417–463.

Steigerwald, E.S., Rusiniak, K.W., Eckel, D.L., O’Reagan, M.H., 1988. Aversive condi-tioning properties of caffeine in rats. Pharmacol. Biochem. Behav. 31, 579–584.

Stewart, J., 2000. Pathways to relapse: the neurobiology of drug- and stress-inducedrelapse to drug-taking. J. Psychiatry Neurosci. 25, 125–136.

Valee, M., Vitiello, S., Bellocchio, L., Hebert-Chatelain, E., Monlezun, S., Martin-Garcia,E., Kasaetz, F., Baillie, G.L., Panin, F., Cathala, A., Roullot-Lacarriere, V., Fabre,S., Hurst, D.P., Lynch, D.L., Shore, D.M., Deroche-Gamonet, V., Spampinato, U.,Revest, J.-M., Maldonado, R., Reggio, P.H., Ross, R.A., Marsiacano, G., Piazza, P.V.,2014. Pregnenolone can protect the brain from cannabis intoxication. Science343, 2014.

Walker, Q.D., Nelson, C.J., Smith, D.C., Kuhn, M., 2002. Vaginal lavage attenuatescocaine-stimulated activity and establishes place preference in rats. Pharmacol.Biochem. Behav. 73, 743–752.

Weiss, F., Ciccocioppo, R., Parsons, L.H., Katner, S., Liu, X., Zorilla, E.P., Valdez, G.R., Bn-Shahar, O., Angeletti, S., Richter, R.R., 2001. Compulsive drug-seeking behaviorand relapse. Neuroadaptation, stress, and conditioning factors. Ann. N. Y. Acad.Sci. 937, 1–26.

Vishwanath, J.M., Desko, A.G., Riley, A.L., 2011. Caffeine-induced taste aversions inLewis and Fischer rat strains: differential sensitivity to the aversive effects ofdrugs. Pharmacol. Biochem. Behav. 100, 66–72.

Yoon, J.H., Higgins, S.T., Heil, S.H., Sugarbaker, R.J., Thomas, C.S., Badger, G.J., 2007.Delay discounting predicts postpartum relapse to cigarette smoking amongpregnant women. Exp. Clin. Psychopharmacol. 15, 176–186.

Yakovenko, V., Speidel, E.R., Chapman, C.D., Dess, N.K., 2011. Food dependence in

nd stress-evoked cocaine reinstatement in high vs. low impulsive, http://dx.doi.org/10.1016/j.drugalcdep.2014.07.001

rats selectively bred for low versus high saccharin intake: implications for “foodaddiction”. Appetite 57, 397–400.

Zlebnik, N.E., Saykao, A.T., Carroll, M.E., 2014. Effects of combined exercise andprogesterone treatments on cocaine seeking in male and female rats. Psy-chopharmacology (Berl.), http://dx.doi.org/10.1007/s00213-014-3513-6.