Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jun 5.
Published in final edited form as: Psychopharmacology (Berl). 2022 Nov 4;239(12):3963–3973. doi: 10.1007/s00213-022-06270-x

Individual differences in cocaine seeking during voluntary abstinence predicts cocaine relapse and the circuitry mediating relapse

Yasmin Padovan-Hernandez 1,2,3, Giselle Rojas 1, Lizhen Wu 1, Lori A Knackstedt 1,2
PMCID: PMC10240883  NIHMSID: NIHMS1899438  PMID: 36329194

Abstract

Rationale

There are no FDA-approved treatments to facilitate recovery from cocaine use disorder. Contingency management offers non-drug reinforcers to encourage abstinence and is effective at reducing drug seeking during treatment, but once discontinued, relapse rates increase.

Objectives

We sought to establish a choice-based rodent model of voluntary abstinence (VA) from cocaine to test the ability of ceftriaxone, an antibiotic consistently shown to prevent relapse to cocaine seeking in rodents, to attenuate relapse after discontinuation of VA, and to investigate relapse-induced neuronal activation via c-Fos expression.

Methods

Male Sprague–Dawley rats self-administered sucrose pellets for 5 days and intravenous cocaine for 12 days. Rats then underwent 14 days of voluntary or forced abstinence. VA sessions entailed the opportunity to choose between sucrose and cocaine delivery in discrete trials (20 trials/day). Ceftriaxone (or vehicle) was administered during the last 7 days of abstinence. During a relapse test, only the cocaine-paired lever was available and presses on the lever delivered cocaine-paired cues.

Results

There were more presses on the sucrose lever during VA, but cocaine intake did not decline to zero. Ceftriaxone had no effect on cocaine intake during VA. Neither ceftriaxone nor VA reduced cocaine seeking during the relapse test, and cocaine intake during VA positively correlated with cocaine seeking during the test in vehicle-treated animals. Relapse-induced c-Fos expression was found to be greater in the ventral orbitofrontal cortex following VA.

Conclusions

Sucrose availability leads to a decrease in, but not cessation of, cocaine seeking and a differential engagement of the circuitry underlying relapse.

Keywords: OFC, Abstinence, Reinstatement, VTA, Contingency management

Introduction

Cocaine use disorder (CUD) is characterized by high rates of relapse even after long periods of abstinence and poses significant threats to personal and societal well-being. There are currently no FDA-approved pharmacological treatments to reduce cocaine use despite decades of research into its underlying neurobiology. In the absence of an effective pharmacotherapy, psychosocial interventions are the current standard treatment for most substance use disorders, including CUD. Among the most successful and most used interventions is contingency management (CM) (Dutra et al. 2008). CM involves a systematic offering of non-drug reinforcers (e.g., vouchers redeemable for goods and services in the community or access to employment) to reinforce drug abstinence (Kampman 2019). Once CM is discontinued, however, it is not uncommon for patients to relapse (Davis et al. 2016; Higgins et al. 2004). Clinical researchers have attempted to use typical pharmacological treatments such as methadone to prolong abstinence once CM is discontinued, without success (Petty et al. 2015). Thus, there is a need for an effective pharmacotherapy that would maintain abstinence once CM is discontinued.

The neurobiology underlying relapse following CM is generally unknown, and until recently, an animal model for this human condition had not been established (Venniro et al. 2016). In such models, voluntary abstinence (VA) is achieved through mutually exclusive choices between a desirable reward (e.g., food or access to a conspecific) and a drug reinforcer for 2 weeks, followed by removal of the ability to respond for the alternative reward during a drug-seeking “relapse” test. Prior to the present paper, such models employing palatable food as the competing reinforcer have used methamphetamine, heroin, and fentanyl as the drug reinforcers (Caprioli et al. 2015a, b; Caprioli et al. 2017; Reiner et al. 2020; Venniro et al. 2017a, b; Venniro et al. 2017a, b). Using this model, a number of brain regions have been implicated in relapse following VA that are congruent with those involved in relapse following forced abstinence (FA), including the anterior insular cortex and central amygdala (Reiner et al. 2020; Venniro et al. 2017a, b), nucleus accumbens (Rossi et al. 2020), and the dorsal striatum (Caprioli et al. 2017). Interestingly, unlike FA, relapse to methamphetamine seeking following VA has been shown to selectively activate the dorsomedial but not dorsolateral striatum, indicating that VA alters the neural circuitry engaged during relapse (Caprioli et al. 2017).

Ceftriaxone is a β-lactam antibiotic that has consistently been shown to attenuate relapse to cocaine-seeking behaviors after extinction training and forced abstinence (Bechard et al. 2018, 2021; Knackstedt et al. 2010; Sari et al. 2009). Ceftriaxone also attenuates the reinstatement of methamphetamine (Abulseoud et al. 2012), ethanol (Alhaddad et al. 2014; Qrunfleh et al. 2013), nicotine (Alajaji et al. 2013), and hydrocodone intake and conditioned place preference (Alshehri et al. 2018). Ceftriaxone has been shown to reduce cocaine relapse-induced glutamate efflux in the nucleus accumbens core (Trantham-Davidson et al. 2012) as well as cocaine-evoked dopamine levels in pretreated rats challenged with cocaine (Barr et al. 2015). Additionally, previous studies have shown that ceftriaxone does not alter sucrose consumption (Sari et al. 2011). This selectivity, versatility, and consistency in attenuating relapse to drug seeking makes ceftriaxone a good candidate as a possible pharmacotherapy to reduce drug seeking after CM is discontinued.

While punishment- and social interaction-based models of voluntary abstinence from cocaine have been established (Farrell et al. 2019; Venniro et al. 2021), a palatable food-based model has not been used for the study of cocaine relapse to date. In the present study, we sought to develop such a model, test the efficacy of choice-based abstinence with and without ceftriaxone to attenuate cocaine relapse in this model, and further elucidate the circuitry underlying relapse following VA training. To this end, we trained male Sprague–Dawley rats to self-administer sucrose pellets and intravenous cocaine. Rats were then assigned to either forced abstinence or voluntary abstinence, where they were able to choose between both reinforcers. Animals received either ceftriaxone (200 mg/kg) or vehicle injections (i.p.) during the last 7 days of abstinence, followed by a cue-primed cocaine relapse test and assessment of relapse-induced neuronal activation via c-Fos expression in a number of brain regions previously established to be involved in cocaine relapse.

Methods

Animals

Adult male Sprague–Dawley rats (n = 29; 8 weeks of age at arrival) were obtained from Charles River (Raleigh, NC, USA). Rats were individually housed in a temperature- and humidity-controlled vivarium and maintained on a 12-h reverse-light cycle. All procedures were carried out during the dark phase of the cycle. Animals were provided with 20 g of standard rat chow daily, and water was available ad libitum. All work was approved by the Institutional Animal Care and Use Committees of the University of Florida.

Drugs

Cocaine was provided by the NIDA Controlled Substances Program (Research Triangle Institute, NC, USA). Cocaine was dissolved in 0.9% physiological saline at a concentration of 4 mg/mL. Ceftriaxone (“Cef”; Sigma-Aldrich, St. Louis, MO) was prepared in 0.9% physiological saline and administered intraperitoneally (i.p.) at a dose of 200 mg/kg in a volume of 1 mL/kg. Vehicle was also administered at a volume of 1 mL/kg.

Surgery

Animals were anesthetized with ketamine (87.5 mg/kg, i.p.), and xylazine (5 mg/kg, i.p.) administered in a volume of 1 mL/kg. Catheters (SILASTIC tubing, ID 0.51 mm, OD 0.94 mm, Dow Corning, Midland, MI) were implanted and secured into the jugular vein with suture thread (Surgical Specialties Corp., Wyomissing, PA). The catheter tubing was passed subcutaneously and exited from the back where it was connected to a cannula (Plastics One, Roanoke, VA, USA) embedded in a rubber harness (Instech, Plymouth Meeting, PA, USA). This harness was worn by the rat for the duration of the experiment. Carprofen (5 mg/kg, subcutaneous) was administered on the day of surgery and for the three following days for analgesia. Heparin (100 units/mL in 0.1 mL) was administered for the duration of the experiment. Catheter patency was tested periodically with methohexital sodium (10 mg/mL; Eli Lilly, Indianapolis, IN, USA).

Sucrose self-administration

Rats first underwent sucrose self-administration (see timeline in Fig. 1a). Sessions began with the illumination of a yellow house light followed 10 s later by insertion of the sucrose lever on the opposite side of the chamber (Fig. 1b). The house light remained on for the duration of the session and served as the sucrose DS + . Rats self-administered sucrose pellets (45 mg; Bio-serv) during daily 2-h sessions under a fixed ratio 1 (FR-1) reinforcement schedule, where one lever press delivered 5 sucrose pellets and illuminated a red stimulus light (CS +) over the sucrose lever. The sound of the sucrose pellets hitting the metal tray inside the operant chamber was also sucrose-associated CS + . The number of pellets delivered was chosen so that the number of presses would be approximately the same during both sucrose and cocaine self-administration. Sucrose pellet delivery was followed by a 20-s timeout during which time the lever was retracted. Animals were required to meet the criteria of 5 days with: (1) at least ten reinforcer deliveries and (2) consumption of ≥ 80% pellets delivered. Once the criteria were met, rats began cocaine self-administration.

Fig. 1.

Fig. 1

Open in a new tab

Timeline and experimental conditions. a Timeline of experiment. b During sucrose self-administration training, the white house light was illuminated, and presses on the left lever delivered sucrose and the illumination of a red light over the lever. d During cocaine self-administration training, a blue house light was illuminated, and presses on the right lever yielded cocaine and cocaine-associated cues (white stimulus light over the lever and 2900 Hz tone). d During voluntary abstinence, both the blue and white house light were illuminated and presses on the sucrose and cocaine lever yielded their respective reinforcers and reinforcer-associated cues

Cocaine self-administration

Rats were trained to self-administer cocaine during daily 2-h sessions in the same operant chambers in which sucrose self-administration took place. Sessions began with the illumination of a blue house light (cocaine DS +) positioned on the wall adjacent to the cocaine lever (Fig. 1c), followed 10 s later by the insertion of the “cocaine lever” and an inactive lever. The DS + remained on for the duration of the session. Presses on the cocaine lever resulted in intravenous cocaine delivery (0.35 mg/infusion in 0.1 mL) and presentation of discrete cocaine-associated cues (CS +): a tone (2900 Hz tone, 5 s) and the illumination of a white stimulus light above the cocaine lever. Following the infusion, the cocaine lever was retracted for the duration of the 20-s timeout. Inactive lever presses had no programmed consequences but were recorded. Upon meeting cocaine self-administration criteria (10 or more infusions for at least 12 days), rats were assigned to undergo either voluntary or forced abstinence such that cocaine and sucrose intake did not differ between conditions.

Voluntary abstinence

During the voluntary abstinence (VA) period, rats were placed into the same operant chambers with both the sucrose and cocaine DS + lights illuminated. VA lasted for 14 days, with one session per day; each session was comprised of 20 discrete trials. In each trial, all three levers were available: the sucrose lever, cocaine lever, and inactive lever (Fig. 1d). Upon the execution of a choice (pressing the sucrose or cocaine lever), the corresponding reinforcer and CS + were delivered, and the levers immediately retracted. The discrete-choice trials were separated by an intertrial interval of 6 min, during which time the cocaine and sucrose levers remained retracted; due to technical reasons, the inactive lever did not retract. For each discrete-choice trial, if no choice was made within 5 min, the cocaine and sucrose levers retracted, and a new trial began after the standard intertrial interval. During the last 7 days of voluntary abstinence, rats were administered either ceftriaxone (200 mg/kg; n = 8) or vehicle (n = 9) immediately following the session. This regimen was selected based on published findings that this dose of ceftriaxone reduces relapse to cocaine seeking after 5 days of treatment, with a range of administration from 5 to 10 days (e.g., Knackstedt et al. 2010; Sari et al. 2009; Stennett et al. 2020).

Forced abstinence

Rats (n = 8) assigned to 14 days of forced abstinence (FA) were handled daily but did not go back to the operant chamber until test day. During the last 7 days of forced abstinence, rats were treated with vehicle akin to the vehicle-treated VA rats. A ceftriaxone-treated group was not included here, as ceftriaxone has previously been demonstrated to attenuate such relapse and alter relapse-induced c-fos expression after forced abstinence (Bechard & Knackstedt 2019; Bechard et al. 2021).

Cued cocaine-seeking test

Following 14 days of voluntary or forced abstinence, rats were placed into the same operant chambers used for self-administration and voluntary abstinence for a 2-h cue-induced relapse test. During the test, the cocaine DS + (blue house light) was illuminated, and only the cocaine and inactive levers were available. Presses on the cocaine lever resulted in the delivery of drug-associated cues but no drug. Inactive lever presses had no programmed consequences but were recorded.

Immunohistochemistry

Immediately following relapse testing, animals were euthanized with Euthasol (1 mL/kg, i.p.) and transcardially perfused with 4% paraformaldehyde (PFA) following phosphate-buffered saline (PBS). Brains were extracted and kept in 4% PFA solution for 24 h, followed by 48 h in 20% sucrose in PBS. Brains were then frozen and kept in a ‒ 80 °C freezer until sliced on a cryostat at 30 μm. Free-floating Sects. (2 sections/region/rat) were blocked in 2% normal donkey serum followed by incubation in rabbit anti-Fos (1:10 000; EMD Millipore) overnight. During day two of the procedure, biotinylated donkey anti-rabbit secondary antibody (1:500; Jackson Immuno) was applied, followed by avidin–biotin complex (1:500; Vector Laboratories) and 3,3-diaminobenzidine (Vector Laboratories). The tissue was then mounted onto slides, air-dried, and cover slipped. Images were captured with a Tucsen monochrome CCD camera attached to an Olympus BX51 microscope. Bilateral images were acquired from 2 sections/region. Image J software (NIH) was used to quantify the number of c-Fos+ cells/mm2 within each region.

Statistical analysis

Analyses were conducted using GraphPad Prism (9.3.1). The number of infusions, active lever presses, and inactive lever presses were compared between groups later receiving Cef/Veh and FA/VA to ensure there were no pretreatment differences in cocaine or sucrose self-administration behavior prior to treatment. To do so, sucrose and cocaine self-administration training data were analyzed using a repeated measures (RM) ANOVA, with time as the within-subjects factor and group as the between-subjects factor. A 3-way (treatment × reinforcer × time) ANOVA was used to examine treatment (Cef vs. Veh) differences in reinforcer seeking during voluntary abstinence. A median split was conducted on the mean percent responses (of 20 possible trials) on the cocaine lever during the last 5 days of VA; this time period was chosen because behavior was observed to have stabilized at this time. Rats that fell below the median were deemed “sucrose-preferring,” and those above the median were deemed “cocaine-preferring”. The mean percent responses on the cocaine lever were compared between sucrose- and cocaine-preferring rats that received ceftriaxone or vehicle with a 2-way ANOVA. The relationship between the amount of cocaine intake during VA and lever presses during relapse was examined using Pearson’s correlation. c-Fos expression per mm2 was calculated and averaged between the two hemispheres and 2 slices and was compared between groups with one-way ANOVAs.

Results

Sucrose and cocaine self-administration

Four rats were eliminated from the experiment due to catheter failure. The remaining 25 rats met both sucrose and cocaine self-administration criteria. No difference in sucrose seeking was observed in animals later assigned to forced/voluntary abstinence type or ceftriaxone/vehicle treatment as two-way ANOVAs showed no significant time × group interaction [F (8, 88) = 1.023, p = 0.4247; Fig. 2a]. The same was observed for presses on the cocaine lever [F (22, 242) = 1.038, p = 0.4187; Fig. 2b] and inactive lever [F (22, 242) = 0.3948, p = 0.9939; Fig. 2c]. Main effects of time were observed for all three levers [sucrose: F (4, 88) = 4.908, p = 0.0013; cocaine: F (11, 242) = 3.654, p < 0.0001; inactive lever: F (l1, 242) = 2.243, p = 0.0130] as animals familiarized themselves with the consequences of lever presses.

Fig. 2.

Fig. 2

Open in a new tab

Prior to the initiation of abstinence, there were no differences in presses on the sucrose, cocaine, or inactive lever. Presses on the sucrose lever (a), cocaine lever (b), or inactive lever (c) during reinforcement training did not differ between animals later assigned to forced abstinence (n = 8), voluntary abstinence with ceftriaxone treatment (n = 8), or vehicle-treated voluntary abstinence (n = 9) groups

Voluntary abstinence

During VA, there was no significant reinforcer × treatment × time interaction [Fig. 3a]. However, there was a reinforcer × time interaction [F (13, 195) = 4.902, p < 0.0001]. Post-hoc tests found that overall, more choices were made on the sucrose lever than on the cocaine lever during the last 5 days of VA (days 10–14). A median split conducted on the percent presses on the cocaine lever during the last 5 days of VA revealed two populations of rats: sucrose- and cocaine-preferring. A two-way ANOVA found a main effect of time on reinforcer choice [F (1, 13) = 126.8, p < 0.001; Fig. 0.3b], but no reinforcer choice × treatment interaction, further indicating that ceftriaxone did not affect reinforcement choice during voluntary abstinence. A mixed effects analysis showed that the mean latency to choose the cocaine lever was higher for the sucrose-preferring rats on several days [F (1,15) = 5.278, p = 0.0364; Fig. 3c], supporting two behaviorally distinct phenotypes and suggesting more impulsive decision-making in cocaine-preferring animals. There were no differences observed for the latency to choose the sucrose lever (Fig. 3d). We further assessed the timing of presses on the cocaine lever over all 20 trials; finding reinforcer choice did not shift from early to late trials; and thus, satiation was not the motivation to switch to the other reinforcer; this was not affected by preferred lever or ceftriaxone treatment (Fig. 3e). No sucrose pellets were left in the hopper at the conclusion of the VA sessions.

Fig. 3.

Fig. 3

Open in a new tab

A subset of rats continued to choose cocaine during voluntary abstinence. a Ceftriaxone had no effect on reinforcer choice during voluntary abstinence. Ceftriaxone- and vehicle-treated rats made more responses on the sucrose lever than the cocaine lever in the last 5 days of VA. b A median split conducted on the number of cocaine reinforcers earned during voluntary abstinence revealed two distinct populations of rats: sucrose-preferring (n = 10) and cocaine-preferring (n = 7). Ceftriaxone treatment did not influence this distribution. Mean latency prior to (c) cocaine and (d) sucrose choices across VA days supports behaviorally distinct phenotypes based on preference. e Heatmap of the percent of choices (out of the total 20 trials) that were on the cocaine lever in vehicle-treated rats (left), sucrose-preferring (middle), and cocaine-preferring (right). There was no evidence of trial-dependent differences in choice between sucrose and cocaine

Cued cocaine seeking

A one-way ANOVA found that neither ceftriaxone nor abstinence type had a statistically significant effect on presses on the cocaine lever during the cued cocaine-seeking test [F (2, 21) = 3.239, p = 0.059; Fig. 4a]. One VA-Veh rat displayed active lever presses that were greater than two standard deviations from the mean; data from this rat was excluded from all analyses. Inactive lever presses were low and did not differ across groups (Fig. 4b). Interestingly, a positive correlation between cocaine intake during VA and presses on the cocaine lever during the test were observed in rats treated with vehicle {r = 0.72, p = 0.0247; Fig. 4c). However, presses on the cocaine lever during the test did not correlate with total cocaine intake during VA in ceftriaxone-treated animals (r = 0.11, p > 0.05; Fig. 4d].

Fig. 4.

Fig. 4

Open in a new tab

Cocaine relapse was not altered by voluntary abstinence or ceftriaxone treatment. a The number of presses on the cocaine lever and on the inactive lever (b) did not differ between groups during the relapse test. For vehicle-treated rats (c), but not ceftriaxone-treated rats (d), the number of cocaine infusions during voluntary abstinence was positively correlated with lever pressing during the relapse test

c-Fos expression

Tissue from three rats could not be processed for immunohistochemistry due to improperly perfused tissue (VA-Cef: 2 rats; VA-Veh: 1 rat). Additionally, individual sections from some brain regions were damaged during immunohistochemistry and could not be used for c-Fos quantification. The number of rats/group that remained in the analyses are depicted in Fig. 5. There were more c-Fos+ cells in the vOFC of vehicle-treated voluntary abstinence animals compared to those assigned to forced abstinence [F (2, 16) = 3.74, p = 0.0465; Fig. 5a], with post-hoc tests finding a significant increase in c-Fos expression in the VA condition relative to the FA condition, implicating higher engagement of this region as a result of VA training. There were no group differences in c-Fos expression in the lOFC (Fig. 5b), PL cortex (Fig. 5d), and nucleus accumbens or DMS (not shown). There was an effect of treatment on VTA c-Fos expression [F (2, 17) = 7.274, p = 0.0052; Fig. 5d], with the VA-Cef group displaying greater c-Fos expression than both VA-Veh and FA groups (ps < 0.05). When using the median split to divide rats into cocaine- or sucrose-preferring and comparing c-Fos expression between such groups with a one-way ANOVA, group differences were present in the VTA [F (4, 15) = 20.01, p < 0.0001; Fig. 5e). Consistent with the group level data (Fig. 5d), independent of whether ceftriaxone-treated rats preferred cocaine or sucrose, ceftriaxone-treated rats displayed greater c-Fos expression than the FA group. However, this analysis revealed a dichotomy between sucrose- and cocaine-preferring rats treated with vehicle, with the sucrose-preferring rats displaying greater c-Fos expression relative to cocaine-preferring rats in the VA condition and rats in the FA condition. Such effects of the cocaine-preferring phenotype on c-Fos expression were not detected in other brain regions, likely due to low power in the VA condition (n’s ≤ 4). When considering only vehicle-treated rats, presses on the cocaine-lever during the test were positively correlated with vOFC c-Fos expression (r = 0.65, p = 0.0160; Fig. 5f).

Fig. 5.

Fig. 5

Open in a new tab

Relapse induced c-Fos expression. a c-Fos expression in the vOFC is increased in the voluntary abstinence (VA) – vehicle (VEH) condition relative to the forced abstinence (FA) – Veh condition. However, c-Fos expression in the lOFC (b) and PL cortex (c) did not differ by condition. d A history of ceftriaxone (Cef) treatment resulted in increased relapse-induced c-Fos expression in the VTA. e Vehicle-treated cocaine-preferring rats displayed less c-Fos expression in the VTA than the other VA conditions. f vOFC c-fos expression positively correlated with the number of presses on the cocaine lever in the Veh-treated rats. *p < 0.05 vs. FA-VEH. #p < 0.05 vs. VEH-CP

Discussion

Here, we adapted a model of palatable food-based voluntary abstinence used previously to induce abstinence from methamphetamine and opioids (Caprioli et al. 2015a, b; Caprioli et al. 2015a, b; Reiner et al. 2020; Venniro et al. 2017a, b; Venniro et al. 2020; Venniro et al. 2017a, b) for use with cocaine. Interestingly, by the end of the 14-day voluntary abstinence period, while there were overall more choices on the sucrose lever than on the cocaine lever, complete cocaine abstinence was not attained at the group level. Rats that underwent voluntary abstinence did not display altered seeking during a cocaine relapse test relative to rats that had undergone forced abstinence. Ceftriaxone treatment had no effect on the number of presses on the cocaine lever during voluntary abstinence or during the cocaine relapse test.

The persistent choice for the cocaine lever observed here is contrary to results previously shown for palatable food-based voluntary abstinence from methamphetamine and fentanyl (Caprioli et al. 2015a, b; Caprioli et al. 2015a, b; Reiner et al. 2020; Venniro et al. 2017a, b; Venniro et al. 2020; Venniro et al. 2017a, b). This may be due to the use of sucrose pellets as the alternative reinforcer in the present work; complete abstinence from methamphetamine and fentanyl has been observed when the alternative reinforcer was palatable food (test diet: 12.7% fat, 66.7% carbohydrate, and 20.6% protein). Additionally, the present study makes use of a high dose of cocaine that is at bottom of the descending limb of the dose response curve, while previous studies (Caprioli et al. 2015a, b; Reiner et al. 2020; Venniro et al. 2017a, b; Venniro et al. 2020; Venniro et al. 2017a, b) used doses of methamphetamine or fentanyl at the peak of or on the ascending limb, respectively, of the dose response curves for self-administration (Clemens et al. 2006; Stevenson et al. 2020).

While the persistent cocaine seeking observed here is not in agreement with data from voluntary abstinence models for methamphetamine and opioids, it is similar to behavior observed in choice models involving cocaine. For example, the Ahmed group has found that presses on the cocaine lever do not decline to zero when rats are given the opportunity to respond for intravenous cocaine or a sweet solution following 15–19 days of cocaine self-administration (Canchy et al. 2020; Cantin et al. 2010). This group reports a much lower percentage (~ 15%) of rats that continue to display a cocaine preference (defined by exhibiting > 60% presses on cocaine lever), while another study found a similar prevalence (~ 16%) of cocaine-preferring male rats when using a criteria of > 50% presses on the cocaine lever (Perry et al. 2013). Here, we found that approximately 40% of animals continue to show a greater than 50% preference for cocaine over sucrose, and 23.5% show greater than 60% preference. The greater prevalence of rats exhibiting a persistent choice for cocaine here is potentially due to a higher dose of cocaine used (0.35 mg/infusion) compared to the dose employed by the prior studies (0.25 mg/infusion and 0.4 mg/kg/infusion). Akin to the present results, over the first 5 choice trials, a 25–30% preference for cocaine is exhibited by rats choosing between regular chow pellets and the same dose of cocaine used in the present study (Tunstall & Kearns 2014). However, in a slightly different model that did not entail a cocaine self-administration training period prior to the choice phase, a preference for sucrose was not shifted by increasing doses of cocaine (0.25–1.5 mg/kg/infusion; Lenoir et al. 2007). Interestingly, even a lower cocaine dose (0.2 mg/infusion) produced continued responding for cocaine in a subset of rats in a punishment-based model of voluntary abstinence (Farrell et al. 2019). Unlike prior studies (Canchy et al. 2020; Cantin et al. 2010; Perry et al. 2013), the present study involved food restriction, which has the potential to increase the motivational value of sucrose pellets; yet, a subset of rats chose cocaine over sucrose pellets. We did not find evidence that rats displayed within-session shifts in sucrose-cocaine choices, indicating that neither satiation nor the anorectic effects of cocaine (Vandaele et al. 2015) led to rats switching from one reinforcer to another from trial 1 to 20. Taken together, this indicates that the dose of cocaine, satiation, and anorectic effects of cocaine do not influence the preference for continued cocaine choices when an alternative reinforcer or a punishment is delivered, but rather, a subset of rats possess unique neurobiology which mediates these choices.

Akin to human behavior, when the alternate reinforcer was not available, rats resumed presses on the cocaine lever. While there was no lasting effect of voluntary abstinence on cocaine seeking during the cued seeking test at the group level, the amount of cocaine intake during voluntary abstinence positively correlated with cued cocaine seeking during the test. Thus, the success of an alternative reinforcer at reducing cocaine seeking predicts later seeking when the alternative reinforcer is discontinued. This supports the idea that combining pharmacotherapy with contingency management approaches could yield greater cocaine abstinence, if the pharmacotherapy is able to reduce cocaine seeking during CM. Unfortunately, while ceftriaxone has consistently been successful at attenuating reinstatement to cocaine seeking when rats are tested after several days of drug-abstinence, whether instrumental extinction training or forced abstinence occurred (Smaga et al. 2020), this effect was not observed here following voluntary abstinence.

Ceftriaxone had no effect on cocaine choice during voluntary abstinence, consistent with its inability to reduce cocaine self-administration in rats (Sondheimer & Knackstedt 2011). One of the known mechanisms through which ceftriaxone reduces cocaine seeking after a drug-free period is the restoration of glutamate homeostasis and prevention of accumbal glutamate efflux that is characteristic of cocaine relapse (Trantham-Davidson et al. 2012). Notably, GLT-1 and xCT protein levels are decreased in the nucleus accumbens core following 14–21 days of withdrawal from cocaine self-administration and can be rescued by ceftriaxone treatment (Knackstedt et al. 2010). In our study, animals did not fully abstain from cocaine, and thus likely did not possess the withdrawal-induced changes necessary for ceftriaxone to successfully reduce drug seeking in the relapse test. The use of a positive control condition (FA-Cef) would have strengthened this conclusion; however, to date, there are more than a dozen reports that ceftriaxone attenuates cocaine seeking after a cocaine-free period (Smaga et al. 2020). While there was a positive correlation between cocaine intake during VA and cued cocaine seeking in vehicle-treated rats, the same relationship was not observed in ceftriaxone-treated rats. Thus, while not attenuating either relapse or the choice between sucrose and cocaine, ceftriaxone treatment disrupted the relationship between such cocaine choices and lever presses during the relapse test. This is potentially due to ceftriaxone’s influence on the neurochemistry and/or neurocircuitry mediating cocaine seeking, such as the finding that ceftriaxone treatment increased relapse-induced c-Fos expression in the VTA. We previously found that rats with a history of ceftriaxone treatment display increased VTA c-fos expression when assessed following a cocaine + cue-primed reinstatement test (Stennett et al. 2020). Thus, the present results are consistent with the literature that a history of ceftriaxone treatment increases relapse-induced VTA c-fos expression. However, the significance of these findings is unclear; as the ceftriaxone-induced increase in VTA c-fos was accompanied by reduced cocaine + cue-primed reinstatement following extinction training, the same reduction was not observed here when the relapse test did not follow a drug-abstinence period. Thus, the behavioral and neurochemical consequences of ceftriaxone-induced increases in VTA c-fos are currently unknown.

There was an increase in c-Fos expression in the vOFC of VA-Veh rats following the cued seeking test that is congruent with previous results emphasizing the role of the vOFC → piriform cortex pathway in relapse to fentanyl seeking following VA (Reiner et al. 2020). The use of c-fos analysis here was to investigate whether voluntary abstinence and/or ceftriaxone alter relapse-induced neuronal activity. Other work has focused on the roles of the OFC in mediating cocaine choice, finding, for example, that non-overlapping populations of OFC neurons mediate cocaine vs. saccharin choice with cocaine-preferring rats displaying a greater population of action-encoding OFC neurons (Guillem & Ahmed 2018). While there was no effect of voluntary abstinence on cued relapse, there was a positive correlation between vOFC c-fos expression and relapse when examining vehicle-treated rats, potentially indicating that this brain region mediates relapse. Ceftriaxone disrupted the positive correlation between vOFC c-fos expression and cocaine seeking during the cue test, just as it disrupted the relationship between cocaine intake and cued cocaine seeking. We propose that ceftriaxone’s neurochemical effects in the VTA, or in other unexamined brain regions mediating relapse, underlie these effects, such that decreased vOFC activity does not then yield decreased relapse in the ceftriaxone condition. Caprioli et al. (2017) found that DMS neuronal ensembles are necessary for cued methamphetamine seeking after voluntary abstinence. While we did not find group-level differences in DMS c-Fos expression here, this does not discount the possibility that the DMS is involved in cocaine seeking after voluntary abstinence. Future work should explore the circuitry of seeking in a voluntary abstinence model in the absence of ceftriaxone treatment.

In conclusion, our findings agree with the literature in finding that rats continue to choose cocaine over sucrose at a population level. Individual differences in sucrose-cocaine choice were detected, and while the minority of rats continued to choose cocaine during voluntary abstinence, this cocaine intake led to greater cued-induced cocaine seeking. Rats readily pressed the lever for cocaine-associated cues after VA was discontinued, in agreement with prior work with fentanyl and methamphetamine and akin to the inability of contingency management to produce lasting decreases in drug seeking in humans (Davis et al. 2016; Higgins et al. 2004). Future work should test the circuitry of cued cocaine seeking after voluntary abstinence in a choice-based model with the goal of developing treatments to extend the ability of CM to reduce drug seeking after its cessation.

Acknowledgements

Figure 1 was adapted from “Skinner Box”, by BioRender.com (2022). Retrieved from https://app.biorender.com/biorender-templates.

Funding

YPH and GR were supported by NIH T34GM118272. LK is supported by NIH DA045140. These funding agencies had no role in experimental design, interpretation of data, or in the selection of this journal for publication.

Footnotes

Consent for publication All authors consent that the manuscript be submitted for publication.

Competing interests The authors declare no competing interests.

Data availability

Data is available upon request.

References

  1. Abulseoud OA, Miller JD, Wu J, Choi D-S, Holschneider DP (2012) Ceftriaxone upregulates the glutamate transporter in medial prefrontal cortex and blocks reinstatement of methamphetamine seeking in a condition place preference paradigm. Brain Res 1456:14–21. 10.1016/j.brainres.2012.03.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alajaji M, Bowers MS, Knackstedt L, & Damaj MI (2013). Effects of the beta-lactam antibiotic ceftriaxone on nicotine withdrawal and nicotine-induced reinstatement of preference in mice. Psychopharmacology. 225(3). 10.1007/s00213-013-3047-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Alhaddad H, Das SC, Sari Y (2014) Effects of ceftriaxone on ethanol intake: a possible role for xCT and GLT-1 isoforms modulation of glutamate levels in P rats. Psychopharmacol 231(20):4049–4057. 10.1007/s00213-014-3545-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bechard AR, Knackstedt LA (2019) The effects of Pavlovian cue extinction and ceftriaxone on cocaine relapseafter abstinence. Drug Alcohol Depend 197:83–86. 10.1016/j.drugalcdep.2019.01.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Alshehri FS, Hakami AY, Althobaiti YS, Sari Y (2018) Effects of ceftriaxone on hydrocodone seeking behavior and glial glutamate transporters in P rats. Behav Brain Res 347:368–376. 10.1016/j.bbr.2018.03.043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Barr JL, Rasmussen BA, Tallarida CS, Scholl JL, Forster GL, Unterwald EM, Rawls SM (2015) Ceftriaxone attenuates acute cocaine-evoked dopaminergic neurotransmission in the nucleus accumbens of the rat. Br J Pharmacol 172(22):5414–5424. 10.1111/bph.13330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bechard AR, Hamor PU, Schwendt M, Knackstedt LA (2018) The effects of ceftriaxone on cue-primed reinstatement of cocaine-seeking in male and female rats: estrous cycle effects on behavior and protein expression in the nucleus accumbens. Psychopharmacol 235(3):837–848. 10.1007/s00213-017-4802-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bechard AR, Logan CN, Mesa J, Padovan-Hernandez Y, Blount H, Hodges VL, Knackstedt LA (2021) Role of prefrontal cortex projections to the nucleus accumbens core in mediating the effects of ceftriaxone on cue-induced cocaine seeking. Addict Biol 26(2):e12928. 10.1111/adb.1292812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Canchy L, Girardeau P, Durand A, Vouillac-Mendoza C, & Ahmed SH (2020) Pharmacokinetics trumps pharmacodynamics during cocaine choice: a reconciliation with the dopamine hypothesis of addiction. Neuropsychopharmacol 10.1038/s41386-020-0786-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cantin L, Lenoir M, Augier E, Vanhille N, Dubreucq S, Serre F, Vouillac C, Ahmed SH (2010) Cocaine is low on the value ladder of rats: possible evidence for resilience to addiction. PLoS ONE 5(7):e11592. 10.1371/journal.pone.0011592 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Caprioli D, Venniro M, Zeric T, Adhikary S, Madangopal R, Marchant N, Lucantonio F, Schoenbaum G, Bossert J, Shaham Y (2015a) Effect of the novel positive allosteric modulator of mGluR2 AZD8529 on incubation of methamphetamine craving after prolonged voluntary abstinence in a rat model. Biol Psychiat 78(7):463–473 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Caprioli D, Venniro M, Zhang M, Bossert J, Warren B, Hope B, Shaham Y (2017) Role of dorsomedial striatum neuronal ensembles in incubation of methamphetamine craving after voluntary abstinence. J Neurosci 37(4):1014–1027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Caprioli D, Zeric T, Thorndike E, Venniro M (2015b) Persistent palatable food preference in rats with a history of limited and extended access to methamphetamine self-administration. Addict Biol 20(5):913–926 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Clemens KJ, Cornish JL, Hunt GE, McGregor IS (2006) Intravenous methamphetamine self-administration in rats: effects of intravenous or intraperitoneal MDMA co-administration. Pharmacol Biochem Behav 85(2):454–463. 10.1016/j.pbb.2006.09.015 [DOI] [PubMed] [Google Scholar]
  15. Davis DR, Kurti AN, Skelly JM, Redner R, White TJ, Higgins ST (2016) A review of the literature on contingency management in the treatment of substance use disorders, 2009–2014. Prev Med 92:36–46. 10.1016/j.ypmed.2016.08.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dutra L, Stathopoulou G, Basden SL, Leyro TM, Powers MB, Otto MW (2008) A meta-analytic review of psychosocial interventions for substance use disorders. Am J Psychiatry 165(2):179–187. 10.1176/appi.ajp.2007.06111851 [DOI] [PubMed] [Google Scholar]
  17. Farrell MR, Ruiz CM, Castillo E, Faget L, Khanbijian C, Liu S, Schoch H, Rojas G, Huerta MY, Hnasko TS, Mahler SV (2019) Ventral pallidum is essential for cocaine relapse after voluntary abstinence in rats. Neuropsychopharmacol 44(13):2174–2185. 10.1038/s41386-019-0507-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Guillem K, Ahmed SH (2018) Preference for cocaine is represented in the orbitofrontal cortex by an increased proportion of cocaine use-coding neurons. Cereb Cortex 28(3):819–832. 10.1093/cercor/bhw398 [DOI] [PubMed] [Google Scholar]
  19. Trantham-Davidson H, LaLumiere RT, Reissner KJ, Kalivas PW, Knackstedt LA (2012) Ceftriaxone normalizes nucleus accumbens synaptic transmission, glutamate transport, and export following cocaine self-administration and extinction training. J Neurosci 32(36):12406–12410. 10.1523/JNEUROSCI.1976-12.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Higgins ST, Heil SH, Lussier JP (2004) Clinical implications of reinforcement as a determinant of substance use disorders. Annu Rev Psychol 55(1):431–461. 10.1146/annurev.psych.55.090902.142033 [DOI] [PubMed] [Google Scholar]
  21. Kampman KM (2019) The treatment of cocaine use disorder. Sci Adv. 5(10). 10.1126/sciadv.aax1532 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lenoir M, Serre F, Cantin L, Ahmed SH (2007) Intense sweetness surpasses cocaine reward. PLoS ONE 2(8):e698. 10.1371/journal.pone.0000698 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Knackstedt LA, Melendez RI, Kalivas PW (2010) Ceftriaxone restores glutamate homeostasis and prevents relapse to cocaine-seeking. Biol Psychiat 67(1):81–84. 10.1016/j.biopsych.2009.07.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Perry AN, Westenbroek C, Becker JB (2013) The development of a preference for cocaine over food identifies individual rats with addiction-like behaviors. PLoS One 8(11):e79465. 10.1371/journal.pone.0079465 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Petry NM, Alessi SM, Barry D, Carroll KM (2015) Standard magnitude prize reinforcers can be as efficacious as larger magnitude reinforcers in cocaine-dependent methadone patients. J Consult Clin Psychol 83(3):464–472. 10.1037/a0037888 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Qrunfleh AM, Alazizi A, Sari Y (2013) Ceftriaxone, a beta-lactam antibiotic, attenuates relapse-like ethanol-drinking behavior in alcohol-preferring rats. J Psychopharmacol (oxford, England) 27(6):541–549. 10.1177/0269881113482529 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Reiner D, Lofaro O, Applebey S, Korah H, Venniro M, Cifani C, Bossert J, Shaham Y (2020) Role of projections between piriform cortex and orbitofrontal cortex in relapse to fentanyl seeking after palatable food choice-induced voluntary abstinence. J Neurosci 40(12):2485–2497 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Rossi LM, Reverte I, Ragozzino D, Badiani A, Venniro M, Caprioli D (2020) Role of nucleus accumbens core but not shell in incubation of methamphetamine craving after voluntary abstinence. Neuropsychopharmacol 45(2):256–265. 10.1038/s41386-019-0479-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Sari Y, Sakai M, Weedman JM, Rebec GV, Bell RL (2011) Ceftriaxone, a beta-lactam antibiotic, reduces ethanol consumption in alcohol-preferring rats. Alcohol Alcohol (oxford, Oxfordshire) 46(3):239–246. 10.1093/alcalc/agr023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Sari Y, Smith KD, Ali PK, Rebec GV (2009) Upregulation of Glt1 attenuates cue-induced reinstatement of cocaine-seeking behavior in rats. J Neurosci 29(29):9239–9243. 10.1523/JNEUROSCI.1746-09.2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Smaga I, Fierro D, Mesa J, Filip M, Knackstedt LA (2020) Molecular changes evoked by the beta-lactam antibiotic ceftriaxone across rodent models of substance use disorder and neurological disease. Neurosci Biobehav Rev 115:116–130. 10.1016/j.neubiorev.2020.05.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sondheimer I, Knackstedt LA (2011) Ceftriaxone prevents the induction of cocaine sensitization and produces enduring attenuation of cue- and cocaine-primed reinstatement of cocaine-seeking. Behav Brain Res 225(1):252–258. 10.1016/j.bbr.2011.07.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Stennett BA, Padovan-Hernandez Y, Knackstedt LA (2020) Sequential cocaine-alcohol self-administration produces adaptations in rat nucleus accumbens core glutamate homeostasis that are distinct from those produced by cocaine self-administration alone. Neuropsychopharmacol 45(3):441–450. 10.1038/s41386-019-0452-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Stevenson GW, Giuvelis D, Cormier J, Cone K, Atherton P, Krivitsky R, Warner E, StFaurent B, Dutra J, Bidlack JM, Szabò F, Polt R, Bilsky EJ (2020) Behavioral pharmacology of the mixed-action delta-selective opioid receptor agonist BBI-11008: studies on acute, inflammatory and neuropathic pain, respiration, and drug self-administration. Psychopharmacol 237(4):1195–1208. 10.1007/s00213-019-05449-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tunstall BJ, Kearns DN (2014) Reinstatement in a cocaine vs food choice situation: reversal of preference between drug and non-drug rewards. Addiction Biolo. 19(5):838–848. 10.1111/adb.12054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Vandaele Y, Cantin F, Serre F, Vouillac-Mendoza C, Ahmed S (2015) Choosing under the influence: a drug-specific mechanism by which the setting controls drug choices in rats. Neuropsychopharmacol 41:646–657 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Venniro M, Caprioli D, & Shaham Y (2016) Chapter 2 - Animal models of drug relapse and craving: From drug priming-induced reinstatement to incubation of craving after voluntary abstinence. In Ekhtiari H & Paulus MP (Eds.), Progress in Brain Research (Vol. 224, pp. 25–52). Elsevier, 10.1016/bs.pbr.2015.08.004 [DOI] [PubMed] [Google Scholar]
  38. Venniro M, Caprioli D, Zhang M, Whitaker F, Zhang S, Warren B, Cifani C, Marchant N, Yizhar O, Bossert J, Chiamulera C, Morales M, Shaham Y (2017a) The anterior insular cortex→entral amygdala glutamatergic pathway is critical to relapse after contingency management. Neuron 96(2)414–427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Venniro M, Panlilio LV, Epstein DH, & Shaham Y (2021) The protective effect of operant social reward on cocaine self-administration, choice, and relapse is dependent on delay and effort for the social reward. Neuropsychophannacol, 46(13), Article 13. 10.1038/s41386-021-01148-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Venniro M, Russell TI, Ramsey LA, Richie CT, Fesscher HMB, Giovanetti SM, Messing RO, Shaham Y (2020) Abstinence-dependent dissociable central amygdala microcircuits control drug craving. Proc Natl Acad Sci USA 117(14):8126–8134. 10.1073/pnas.2001615117 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Venniro M, Zhang M, Shaham Y, Caprioli D (2017b) Incubation of methamphetamine but not heroin craving after voluntary abstinence in male and female rats. Neuropsychopharmacol 42:1126–1135. 10.1038/npp.2016.287 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data is available upon request.

RESOURCES