Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Aug 1.
Published in final edited form as: Drug Alcohol Depend. 2020 Jun 18;213:108125. doi: 10.1016/j.drugalcdep.2020.108125

Presence of a social peer enhances acquisition of remifentanil self-administration in male rats

Rebecca S Hofford 1,2,*, Paige N Bond 1, Jonathan J Chow 1, Michael T Bardo 1
PMCID: PMC7371539  NIHMSID: NIHMS1605155  PMID: 32590212

Abstract

Background:

Social peers influence human drug use at every stage of addiction. Using a dual-compartment apparatus that allows for limited social contact, recent work has shown that cocaine self-administration is enhanced when two rats are trained to self-administer at the same time compared to rats trained alone or trained in the presence of a saline self-administration control peer. The current study measured social influence on self-administration of the short-acting opioid remifentanil using a dual-compartment operant conditioning chamber.

Methods:

Adult male rats were placed in one of five groups: (1) REMI-REMI group, in which both rats self-administered remifentanil; (2) REMI-SAL group, in which rats self-administered remifentanil in the presence of a peer that self-administered saline; (3) SAL-REMI group, in which rats self-administered saline in the presence of a peer that self-administered remifentanil; and (4) REMI ALONE and (5) SAL ALONE groups, in which rats administered their respective drugs alone (no peer). Self-administration was measured using a 2-lever procedure during acquisition, maintenance, increasing fixed-ratio, and dose-response phases.

Results:

The presence of a social peer enhanced drug intake during acquisition, regardless of the drug exposure of their peer. Additionally, active lever position significantly affected remifentanil intake during acquisition and maintenance, with the greatest influence occurring when the active lever was close to the peer.

Conclusion:

The presence of a social peer in the drug-taking context potentiates remifentanil self-administration, regardless of the peer’s drug access. Future studies utilizing the dual-compartment apparatus will help elucidate the neural mechanisms underlying social influence on opioid abuse.

Keywords: social, remifentanil, self-administration, peer, opioid

1. Introduction

Peer social interactions influence drug addiction at every stage of the disorder. Adolescents are especially prone to peer influence and one of the strongest predictors of substance use during the teenage years is close relationships with drug-using peers (Branstetter et al., 2011). Conversely, individuals that have healthy relationships with drug-free friends and family are significantly less likely to develop substance use problems (Forster et al., 2015). While capturing human social dynamics in rodents is challenging, a few models have been developed that allow interrogation of social influences on drug self-administration.

Generally, the presence of social peers in the home cage during development decreases self-administration of drugs from multiple classes, including psychostimulants (Bardo et al., 2001; Green et al., 2010), alcohol (McCool and Chappell, 2009), and opioids (Hofford et al., 2017), demonstrating that conspecifics in the home cage protect against self-administration that is conducted in a different context. However, presence of a peer in the drug-taking context also influences drug self-administration. In general, rats choose social interaction over either methamphetamine or heroin in an operant choice procedure (Venniro et al., 2018; Venniro and Shaham, 2020). Even when another rat is present in the drug-taking context, but they cannot physically interact, self-administration can be affected. In rats trained to stable rates of amphetamine self-administration, the introduction of another rat into an adjacent chamber separated from the self-administering rat by clear Plexiglas was found to cause a temporary rate-dependent change in responding (Gipson et al. 2011). When the response rate was low using a high dose of amphetamine (0.1 mg/kg/infusion), introduction of a peer increased self-administration, a phenomenon known as social facilitation. In contrast, when the response rate was high using a low dose of amphetamine (0.01 mg/kg/infusion), introduction of a peer decreased self-administration (Gipson et al., 2011). While these models recapitulate some of the influence of social peers on drug-taking in humans, they do not address the finding that interaction with drug-using peers increases the risk of drug use because only one animal is allowed to self-administer at a time.

More recently, Smith and colleagues have developed a paradigm that allows two rats to self-administer drugs at the same time in adjacent operant conditioning chambers (Lacy et al., 2014; Smith, 2012). In this paradigm, two separate self-administration chambers are joined together at one side wall and are divided by a wire screen partition that allows for visual, olfactory, auditory, and limited tactile contact with a peer, while preventing rats from engaging with each other’s operant manipulanda. Using this apparatus, results show that cocaine self-administration is enhanced when both rats are trained at the same time compared to when rats are trained alone. In addition, cocaine self-administration is suppressed when rats are paired with a peer that self-administers saline (Smith, 2012). This finding generalizes to models of problematic intake, as rats paired with a cocaine self-administering peer exhibit greater escalation than rats trained with a saline self-administering peer (Robinson et al., 2016).

To date, most of this work using a dual-compartment apparatus to assess social influences on drug taking in male pairs has been conducted using stimulant drugs (cocaine or amphetamine). It is currently unclear if the effects of social influence on self-administration generalize to other drugs such as opioids. In one study, Smith and colleagues examined heroin self-administration in male-female and female-female dyads across the estrous cycle (Lacy et al., 2016). Heroin intake in females was decreased during proestrus, but heroin intake was also reduced in males when their female peers were in either estrous or proestrus. However, the self-administration behavior of male-female dyads is likely modulated by factors beyond social interaction per se, including sexual receptiveness. While both cocaine and opioids both possess high abuse liability, they have opposite effects on social play behavior when given at low doses, with cocaine reducing and morphine enhancing play (Achterberg et al., 2014; Vanderschuren et al., 1995). Hence, while the effects of social rearing in the home cage seem to transcend drug class (Bardo et al., 2001; Green et al., 2010; Hofford et al., 2017; McCool and Chappell, 2009), the current study determined if the presence of a same-sex (male) social peer in the drug-taking context alters acquisition of remifentanil self-administration. Remifentanil was chosen because it has an ultrashort plasma half-life (Haidar et al., 1997), and thus engenders high rates of responding that allow for fine-grained dose-response analyses.

2. Materials and Methods

2.1. Animals

Adult male Sprague-Dawley rats (PND 70–80) were purchased from Envigo-Harlan (Indianapolis, IN) and were single-housed upon arrival. Colony space was AAALAC accredited and was humidity- and temperature-controlled on a 24 h light:dark cycle (lights on 0700). Food and water were available ad libitum in the home cage throughout the entire experiment. Experimental procedures occurred during the light phase. All animal procedures were approved by the IACUC at University of Kentucky and all practices conformed to the “Guide for the Care and Use of Laboratory Animals” (National Research Council 2010).

2.2. Surgical Procedures

After one week of acclimation, rats underwent surgery for placement of a jugular catheter. Briefly, rats were anesthetized with a ketamine (Butler Schein, Dublin OH) /xylazine (Akorn, Inc., Decatur IL) /acepromazine (Boehringer Ingelheim, St. Joseph MO) cocktail (75/7.5/0.75 mg/kg; 0.15 ml/100g body weight; i.p.). A catheter was implanted into the right jugular vein, threaded under the skin, and exited the body through an incision on the scalp. The catheter port was attached to the skull using four jeweler’s screws and dental acrylic. Rats were allowed to recover for one week before experimental procedures began. During this time, rats received a 5.5 mg/kg carprofen injection (s.c.) on the day following surgery and daily catheter infusions of 0.2 ml gentamicin in sterile saline (10.15 mg/ml, Abraxis BioScience, Los Angeles CA) followed by 0.2 ml of post-flush solution [containing 1% gentamicin (10.15 mg/ml) and 3% heparin (1000 USP units/ml, Abraxis BioScience, Los Angeles CA) in sterile saline] starting on recovery day 2 and occurring daily after self-administration sessions.

2.3. Remifentanil Self-administration

2.3.1. Apparatus

Self-administration sessions were conducted in custom-built operant conditioning chambers (custom built by MED Associates, St. Albans VT). Two standard chambers were combined into one larger apparatus to form a dual-compartment operant conditioning chamber (Fig. 1A). The dual-compartment chamber allowed two rats to individually respond to operant stimuli while maintaining visual, auditory, olfactory, and limited somatosensory contact through a wire screen (1.27 cm, 19 gauge) partition (Weiss et al., 2018). Each side of the dual-compartment chamber contained two retractable levers with a white cue light above each lever, a house light mounted on the back wall, and a syringe pump for drug delivery (PHM-100; MED Associates).

Figure 1: Experimental design.

Figure 1:

Open in a new tab

(A) Illustration of the dual-compartment operant conditioning apparatus. (B) Experimental timeline. (C) Timeline of autoshaped acquisition sessions. (D) Timeline of regular sessions.

2.3.2. Experimental Groups

Rats were randomly assigned to one of 5 treatment groups as follows: (1) REMI-REMI rats self-administered remifentanil and their peer also self-administered remifentanil; (2) REMI-SAL rats self-administered remifentanil and their peer had access to saline only; (3) SAL-REMI rats had access to saline only and their peer self-administered remifentanil; (4) REMI ALONE rats self-administered remifentanil without a peer; and (5) SAL ALONE rats had access to saline without a peer. Paired rats always had the same peer throughout the study. Note that REMI-SAL and SAL-REMI rats were partners, with the REMI-SAL rat self-administering remifentanil and the SAL-REMI rat self-administering saline. This contrasts with the REMI-REMI dyads, where both rats self-administered remifentanil concomitantly.

2.3.3. Assignment of Active Lever Position

Active lever configuration was counterbalanced across rats within each treatment group. For each individual rat within each group, the resulting configurations were classified as “close” or “far” based on their distance from the wire partition (see Fig. 1A).

2.4. Experimental Timeline

2.4.1. Acquisition

The entire experimental timeline is provided in Fig. 1B. During the first 7 days of training, each daily session was broken into 4 phases: (1) habituation; (2) autoshaping; (3) rest; and (4) operant self-administration (Fig. 1C); peers were present throughout all phases for REMI-REMI, REMI-SAL, and SAL-REMI groups. Upon placement into the operant chamber, rats experienced 5 min of dark to acclimate them to their environment and to allow social exploration of their peer (‘habituation’), depending on group assignment. The start of the autoshaping phase was signaled with illumination of the houselight and extension of the inactive lever into the chamber. On a variable interval 90-sec schedule [range of intervals 4” to 213” (Fleshler and Hoffman, 1962)], the active lever was extended into the chamber for 15 sec. Upon lever press or after 15 sec had elapsed, the active lever retracted, an infusion of 3 μg/kg/infusion remifentanil or saline was delivered (3.4 sec duration, 0.1 ml volume) and both cue lights turned on for the duration of the infusion. Rats received 10 infusions over 15 min during this phase (‘autoshaping’). Immediately after autoshaping, rats entered into a 15-min period where only the inactive lever remained extended and the houselight was turned off (‘rest’). The start of the 60-min self-administration phase was signaled with re-illumination of the houselight and insertion of the active lever; the inactive lever remained extended throughout this phase. The self-administration phase allowed rats to earn 3 μg/kg/infusion remifentanil or saline on an FR1 schedule. Active lever presses also resulted in illumination of the cue lights and retraction of the active lever for the duration of the infusion with no additional time out. Autoshaped acquisition training sessions occurred once daily for 7 consecutive days. Each chamber operated independently with the exception that the autoshaping and self-administration phases of each session terminated simultaneously for rats in dyads (REMI-REMI, REMI-SAL, and SALREMI).

2.4.2. Maintenance and Increasing Fixed Ratio Requirements

Starting on the day following acquisition, rats continued to self-administer on an FR1 for an additional 7 days, followed by 3 days at FR2, 3 days at FR3, and 3 days at FR5. All sessions started with a 5-min dark period with peers present (depending on group assignment), followed by illumination of the houselight and insertion of the active and inactive levers to begin the session. Sessions during this phase and all subsequent phases lasted 65 min total (5 min dark and 60 min operant self-administration); see Fig. 1D.

2.4.3. Dose Response

Starting the day after the increasing FR phase, rats underwent assessment of their dose-response curves with peers present as before (depending on group assignment). All remifentanil self-administering rats (REMI-REMI, REMI-SAL, and REMI ALONE) received additional remifentanil doses in the following order: 0.1, 1, 0, 0.3, and 10 μg/kg/infusion. Rats had access to each dose for 2 consecutive days. Saline control rats (SAL-REMI and SAL ALONE) lever pressed for saline throughout the entire experiment.

2.4.4. Exclusion Criteria

Rats were removed from the study if they lost headmount integrity or catheter patency. Upon removal, their data was excluded from the entire experimental phase they were currently in and all subsequent phases. Additionally, rats in dyads were excluded from analysis when their peer was removed from the study. Subject sizes for each phase are included in Tables 1 and 2.

Table 1.

Sample size for analysis of social effects.

Phase Group n
Acquisition REMI-REMI 10
REMI-SAL 8
SAL-REMI 8
REMI ALONE 8
SAL ALONE 8
Maintenance REMI-REMI 10
REMI-SAL 8
SAL-REMI 8
REMI ALONE 7
SAL ALONE 8
Inc. FR REMI-REMI 10
REMI-SAL 6
SAL-REMI 6
REMI ALONE 7
SAL ALONE 7
Dose-Response REMI-REMI 8
REMI-SAL 5
SAL-REMI 5
REMI ALONE 6
SAL ALONE 7
Open in a new tab
Table 2.

Sample sizes for lever configuration analysis.

Phase Lever Conf (total n) Group n / gp
Acquisition C (11) REMI-REMI 6
REMI-SAL 5
F (7) REMI-REMI 4
REMI-SAL 3
Maintenance C (11) REMI-REMI 6
REMI-SAL 5
F (7) REMI-REMI 4
REMI-SAL 3
Inc. FR C (9) REMI-REMI 6
REMI-SAL 3
F (7) REMI-REMI 4
REMI-SAL 3
Dose-Response C (8) REMI-REMI 6
REMI-SAL 2
F (5) REMI-REMI 2
REMI-SAL 3
Open in a new tab

2.5. Statistical Analyses

Active and inactive lever presses during acquisition, maintenance, increasing FR, and dose-response phases were modeled separately using linear mixed-effects (LME), with subject as a random factor, group as a between-subjects fixed factor, and session as a within-subjects fixed factor (Gelman and Hill, 2006). Remifentanil administering groups (REMI-REMI, REMI-SAL, and REMI ALONE) were analyzed separately from saline self-administering groups (SAL-REMI and SAL ALONE) since the current experiment sought to determine whether remifentanil self-administration could be influenced by the presence and drug experience of a peer in the drug-taking context and an a priori expectation that remifentanil groups would have much greater responding than saline groups. For an additional analysis of drug intake during the dose-response phase, demand curves were fit to remifentanil consumption using the formula: log Q = log(Q0)+k*(e(-αQ0C) −1), where Q equaled consumption, Q0 equaled consumption at zero cost, C equaled unit price (lever presses needed to earn 1 μg/kg), k was a constant that was shared among groups, and α was the slope of the function (Hursh and Silberberg, 2008). Only rats in the remifentanil self-administering groups were used in this analysis; the demand model did not fit saline consumption data. Nonlinear mixed-effect (NLME) models were used to calculate the best fit parameters k, Q0, and α, as well as for comparison of Q0 and α between groups using subject as a random factor and group as a between-subjects fixed factor (Pinheiro et al., 2009).

For analyzing the effect of active lever position, active and inactive lever presses during acquisition, maintenance, increasing FR, and dose response from remifentanil paired rats in dyads (REMI-REMI and REMI-SAL) were analyzed using LME with subject as a random factor, lever configuration as a between-subjects fixed factor, and session as a within-subjects fixed factor. Since these analyses were meant to identify an effect of lever position on remifentanil intake independent of treatment group and given that individual subject size was small and there were no differences in lever pressing between REMI-REMI and REMI-SAL at any phase, lever position data from REMI-REMI and REMI-SAL groups were combined. Demand curves were calculated for the dose-response phase and differences between groups were assessed using NLME. P values less than 0.05 were deemed statistically significant. In the presence of significant interactions, differences between each group were assessed using Bonferroni-corrected pairwise comparison of slope using linear regression (Searle et al., 1980) for acquisition, maintenance, and increasing FR. Portions of Fig. 1 were constructed with Biorender.com.

3. Results

3.1. Presence of a social peer during remifentanil self-administration increases drug intake

In both remifentanil and saline self-administering rats during acquisition, LME analysis of active lever presses identified a significant effect of session F(1, 22.98) = 44.58 and F(1, 14) = 8.27, respectively (both p < 0.05); see Figs. 2A and 3A. While remifentanil self-administering groups increased their active lever presses across sessions, saline self-administering groups decreased their active lever presses across sessions. A significant interaction between group and session was only found in remifentanil self-administering groups, F(2, 22.98) = 3.56, p < 0.05. Post hoc analysis revealed that both REMI-REMI and REMI-SAL groups showed a greater increase in active lever presses across sessions compared to REMI ALONE (each p < 0.017), but REMI-REMI and REMI-SAL groups did not differ from each other. Ninety-five percent confidence intervals (CIs) of active lever presses during acquisition were: REMI-REMI (5.50 – 37.64, mean = 27.57), REMI-SAL (8.60 – 45.69, mean = 27.14), REMI ALONE (7.84 – 20.09, mean = 13.96), SAL-REMI (0.84 – 2.45, mean = 1.64), and SAL ALONE (0.37 – 1.63, mean = 1.00). Analysis of inactive lever presses also yielded a significant effect of session during acquisition in remifentanil self-administering groups F(1, 23.06) = 11.99 (Fig. 2A) and in saline self-administering groups F(1, 14) = 4.63, both p < 0.05 (Fig. 3A). In both cases, inactive lever presses decreased across sessions.

Figure 2: Effect of peers on remifentanil self-administration.

Figure 2:

Open in a new tab

(A) Active and inactive lever presses for remifentanil during autoshaped acquisition in different treatment groups. Among remifentanil self-administering groups, active lever pressing was higher in REMI-REMI and REMI-SAL groups compared to the REMI ALONE group across sessions, each p < 0.017. (B) Active and inactive lever presses for remifentanil during maintenance on FR1 in different treatment groups. (C) Active and inactive lever presses for remifentanil across increasing FR requirements in different treatment groups. (D) Active and inactive lever presses for remifentanil during the dose response phase in different treatment groups. (E) Remifentanil consumption at various unit prices graphed as an economic demand function in different remifentanil self-administration groups. Numbers in parentheses indicate n size.

Figure 3: Effect of peers on saline self-administration.

Figure 3:

Open in a new tab

(A) Active and inactive lever presses for saline during autoshaped acquisition in different treatment groups. (B) Active and inactive lever presses for saline during maintenance on FR1 in different treatment groups. (C) Active and inactive lever presses for saline with increasing FR requirement in different treatment groups. (D) Active and inactive lever presses for saline during the dose response phase in different treatment groups. Numbers in parentheses indicate n size. (Note difference in scale on Y axis compared to remifentanil self-administering groups in Fig 2.)

Following autoshaped training, analysis of social effects during maintenance indicated that remifentanil self-administering groups increased active lever presses across sessions F(1, 22.09) = 11.18, p < 0.05, as did saline self-administering groups F(1, 14) = 4.69, p < 0.05; see Figs. 2B and 3B. No group main effect was found in either remifentanil- or saline self-administering groups. Ninety-five percent CIs of active lever presses during maintenance: REMI-REMI (56.64 – 73.72, mean = 65.18), REMI-SAL (63.74 – 84.07, mean = 73.91), REMI ALONE (45.74 – 57.47, mean = 51.60), SAL-REMI (1.48 – 2.49, mean = 1.98), and SAL ALONE (0.98 – 1.60, mean = 1.29). Given that the CIs obtained between REMI-SAL and REMI ALONE groups did not overlap, it is possible that a larger number of subjects might have revealed a significant effect. Analysis of inactive lever presses during maintenance did not yield any significant effects in remifentanil or saline-administering groups (Figs. 2B and 3B).

During the increasing FR phase, a significant increase in active lever pressing was found in remifentanil self-administering groups, F(1, 19.92) = 44.62, p < 0.05 (Fig. 2C), but not in saline self-administering groups (Fig. 3C). There was no significant difference among remifentanil self-administering groups during the increasing FR phase. Ninety-five percent CIs of active lever presses during increasing FR were: REMI-REMI (155.50 – 253.80, mean 204.7), REMI-SAL (141.20 – 249.90, mean = 195.50), REMI ALONE (118.90 – 185.10, mean = 152.00), SAL-REMI (2.30 – 3.33, mean = 2.82), and SAL ALONE (2.38 – 3.87, mean = 3.13). Analysis of inactive lever presses during the increasing FR phase did not yield any significant effects in remifentanil or saline-administering groups (Figs. 2C and 3C). Infusions earned during acquisition, maintenance, and increasing FR from individual rat dyads (REMI-REMI and REMI-SAL / SAL-REMI) are presented in Fig. 4. Visual examination of infusions earned within REMI-REMI dyads indicate that some dyads had nearly identical daily remifentanil intake early in training (e.g. Fig. 4A and 4D), but some dyads differed in their intake over days (e.g. Fig. 4B).

Figure 4: Individual self-administration profiles for paired rats during acquisition, maintenance, and increasing FR.

Figure 4:

Open in a new tab

Infusions earned during autoshaped acquisition, FR1 maintenance, and increasing FR for paired rats in (A-E) REMI-REMI group and (F-M) REMI-SAL/SAL-REMI groups. Dotted vertical lines separate different phases of the experiment. Absence of data from the increasing FR phase in (F, G) due to rats meeting exclusion criteria. Otherwise, individual missing data points are due to rats coming unhooked during the session.

LME analysis of the dose response phase indicated a significant main effect of dose on active lever presses F(2.74, 41.02) = 13.33, p < 0.05 and inactive lever presses F(2.24, 33.08) = 6.56, p < 0.05 in remifentanil self-administering group (Fig. 2D), but not the saline self-administering controls (Fig. 3D). A NLME analysis of demand, with a shared k of 2.98, revealed no significant differences among remifentanil self-administering groups in either α or Q0 (Fig. 2E). Additionally, a significant group x dose interaction effect was found on inactive lever presses in the remifentanil self-administering groups F(10, 74) = 2.45, p < 0.05 (Fig. 2D); however, post hoc analysis did not identify any significant pairwise comparisons. There was no significant difference in inactive lever presses between saline self-administering groups (Fig. 3D). Ninety-five percent CIs of active lever presses during the dose-response phase were: REMI-REMI (155.8 – 513.4, mean = 334.6), REMI-SAL (77.64 – 426.50, mean = 252.10), REMI ALONE (126.60 – 372.90, mean = 249.80), SAL-REMI (1.58 – 4.26, mean = 2.92), and SAL ALONE (1.86 – 4.14, mean = 3.00). Visual examination of infusions earned during the dose-response phase indicated that all dyads had different patterns of remifentanil intake across doses, regardless of dyad type (Fig. 5).

Figure 5: Individual dose-response profiles for paired rats.

Figure 5:

Open in a new tab

Infusions earned during the dose-response phase for paired rats in (A-E) REMI-REMI group and (F-J) REMI-SAL/SAL-REMI groups. Missing data points are from rats coming unhooked during the session.

3.2. Position of the active lever influences remifentanil intake in social dyads

To determine whether active lever position affected remifentanil self-administration in the social dyads (REMI-REMI and REMI-SAL), active and inactive lever presses were analyzed based on each rats’ active lever position. Overall, there were more dyads in the close configuration (n =11 at start of experiment) compared to far (n=7 at start), but each configuration contained similar numbers of REMI-REMI and REMI-SAL (close: REMI-REMI n = 6, REMI-SAL n = 5; far: REMI-REMI n = 4, REMI-SAL n = 3 at start of the experiment); see Table 2. For responses during acquisition, LME analysis identified a main effect of session, F(1, 9.23, 30.82) = 27.28, p < 0.05, and a significant lever position x session interaction F(6, 96) = 2.47, p < 0.05; see Fig. 6A. Analysis of group differences revealed that lever pressing was significantly greater across sessions when the levers were close compared to far from the wire partition (p < 0.05). There was no significant effect of lever configuration on inactive lever presses (Fig. 6A). Ninety-five percent CIs of active lever presses during acquisition were: close configuration (7.79 – 49.20, mean = 28.49) and far configuration (5.52 – 29.42, mean = 17.47).

Figure 6: Effect of lever position on remifentanil self-administration.

Figure 6:

Open in a new tab

(A) Active and inactive lever presses during autoshaped acquisition in REMI-REMI and REMI-SAL groups combined with the levers being close or far from the wire screen partition; active lever pressing was significantly higher in the close compared to the far lever position across sessions, p < 0.05. (B) Active and inactive lever presses during maintenance on FR1 in REMI-REMI and REMI-SAL groups combined with the levers being close or far from the wire screen partition. (C) Active and inactive lever presses across increasing FR requirements in REMI-REMI and REMI-SAL groups combined with the levers being close or far from the wire screen partition. (D) Active and inactive lever presses at various doses in REMI-REMI and REMI-SAL groups combined with the levers being close or far from the wire screen partition. (E) Remifentanil consumption at various unit prices graphed as an economic demand function in REMI-REMI and REMI-SAL groups combined with the levers being close or far from the wire screen partition. Numbers in parentheses indicate n size.

For maintenance, lever configuration did not affect remifentanil intake, with a main effect of session only, F(3.904, 61.81) = 4.08, p < 0.05 (Fig. 6B). Lever position also did not affect inactive lever presses (Fig. 6B). Ninety-five percent CIs of active lever presses during maintenance were: close configuration (70.32 – 78.64, mean = 74.48) and far configuration (52.60 – 74.30, mean = 63.45). Lever position also did not affect remifentanil intake during the increasing FR phase (main effect of session only, F(2.05, 28.74) = 29.66, p < 0.05, Fig. 6C), and there were no significant effects on inactive lever presses during this phase (Fig. 6C). Ninety-five percent CIs of active lever presses during the increasing FR phase were: close configuration (159.80 – 260.70, mean = 210.20) and far configuration (139.70 – 240.70, mean = 190.20).

LME analysis of the dose response phase did not reveal a significant main effect of lever position or lever position x dose interaction, although there was a significant main effect of dose on active lever presses F(2.38, 25.60) = 6.17, p < 0.05 (Fig. 6D); there were no effects of lever position on inactive lever presses. When converted to a demand curve using a shared k of 2.00, NLME did not identify any significant difference in α or Q0 (Fig. 6E). Ninety-five percent CIs of active lever presses during the dose-response phase were: close configuration (146.00 – 524.80, mean = 335.40) and far configuration (81.83 – 471.00, mean = 276.40).

4. Discussion

This study demonstrates that the acquisition of opioid self-administration is subject to social influence. Specifically, the presence of a social peer accelerated acquisition of remifentanil self-administration during initial training, although the difference in responding dissipated after extended training. Similar to the results of Smith (2012) that showed a role of social influence on cocaine self-administration, when both peers had access to remifentanil (REMI-REMI group), they self-administered more than rats that self-administered alone (REMI ALONE group). However, the current study found that rats that self-administered remifentanil in the presence of a saline self-administering peer (REMI-SAL group) also self-administered more remifentanil than REMI ALONE rats. This latter finding contrasts with the results of Smith (2012) which showed that cocaine self-administration was inhibited when rats were paired with a saline self-administering peer (Smith, 2012).

There are several noteworthy differences between Smith (2012) and the current study, including the experimental design and drug tested. First, given the relatively small sample sizes used in the current study, it is possible that adding more subjects could reveal group effects at later stages of self-administration. For example, comparison of the 95% CIs for the REMI-SAL and REMI ALONE groups during maintenance showed that they did not overlap. Second, the current study employed an autoshaping protocol to train the rats to lever press for drug (Carroll and Lac, 1993), while most previous studies used food training (Lacy et al., 2016; Peitz et al., 2013; Robinson et al., 2016; Smith, 2012). The drug autoshaping procedure used here eliminates the need for food training, which can complicate interpretation of the acquisition results because the active lever is associated with two different reinforcers (food and drug). For this experiment, the autoshaping training procedure ensured that all rats received similar experience with remifentanil and associated cues early in training, despite any potential distractions present in the drug-taking environment (e.g. presence of a peer).

Additionally, rats in the current study were housed individually and lived outside the drug-taking context, whereas the cocaine self-administering dyads in Smith (2012) lived in their operant chambers. Previous studies have shown that isolated rats demonstrate a greater social place preference than group-housed rats (Douglas et al., 2004). Thus, the incentive value of social peers may have been greater in the current study. Interestingly, previous studies have shown that psychostimulant and heroin intake differs depending on whether self-administration occurs in the home cage or in a separate context (Badiani et al., 2011), with heroin intake being higher when administered in the home cage and stimulant intake (cocaine and amphetamine) being higher when it occurs in a separate environment (Caprioli et al., 2008, 2007). Since the current study with remifentanil occurred in a separate operant chamber, it would be interesting to examine peer influence on remifentanil self-administration in the home environment.

Another explanation for the discrepancy in results between the current study and the study by Smith (2012) may relate to a difference between remifentanil and cocaine. Opioids and cocaine produce contrasting sympathetic effects (Pitts and Marwah, 1988; Rollins et al., 2014) and, subsequently, have either depressive or stimulant qualities, respectively. However, remifentanil is a short-acting opioid that has short-lived suppression effects and engenders high rates of self-administration (Panlilio and Schindler, 2000), thus making it unlikely that the discrepant results reflect differential stimulant properties of cocaine and remifentanil.

Another relevant difference between remifentanil and cocaine is their potential effects on social behaviors. Systemic or intra-accumbens injections of the classic μ opioid morphine increases social play (Trezza et al., 2011; Vanderschuren et al., 1995), whereas systemic injections of cocaine and other psychostimulants attenuate social play (Achterberg et al., 2014). In addition to their play-enhancing effects, opioids have been implicated in social attention (Chelnokova et al., 2016), approach to social stimuli (Wöhr and Schwarting, 2009), and enhancement of social reward (Achterberg et al., 2018). Opioid administration may enhance attention to the peer, which may then cause more opioid release and thus, increase the incentive value of remifentanil and its paired cues. The ability of opioids to enhance attention to peers could explain the faster acquisition of remifentanil self-administration present in both REMI-REMI and REMI-SAL groups compared to REMI ALONE.

There were several instances of significant session effects on active lever pressing in rats that self-administered saline throughout the experiment (SAL-REMI and SAL ALONE groups). However, the magnitude of change across sessions was not as large as that observed in rats self-administering remifentanil (REMI-REMI, REMI-SAL, and REMI ALONE groups) and there were no significant group or group x session effects involving the SAL-REMI and SAL ALONE groups. Since rats will lever press at low rates for a visual stimulus (Cain et al., 2006; Marx et al., 1955), it is likely that the response-contingent cue light illumination served as a weak reinforcer in saline self-administering rats. In contrast to the remifentanil self-administering groups, the presence of a peer did not enhance responding for visual novelty, since there were no differences in active lever responding between SAL-REMI and SAL ALONE groups.

An important novel finding from the current study is that the position of the active lever relative to the wire screen partition affected acquisition of remifentanil self-administration. Prior studies have employed a one-lever model, where only one lever was present in each adjacent compartment of the dual operant apparatus, with both active levers close to the wire screen partition (Lacy et al., 2014; Smith, 2012). It was stated in those previous reports, but not experimentally demonstrated, that this single active lever configuration was necessary to achieve social influences on cocaine self-administration. The current study suggests that this is partially true for remifentanil, although all remifentanil groups eventually acquired self-administration regardless of the active lever position. Importantly, however, remifentanil self-administering rats in dyads (REMI-REMI and REMI-SAL groups) that had active levers close to the partition self-administered significantly more remifentanil during acquisition. Thus, remifentanil elicits greater self-administration when social dyads are trained with both levers positioned close to the partition, thus allowing both self-administering rats to be in close proximity to each other while accessing drug.

Finally, it is worth noting that both the social effect and the lever position effect occurred early in training but dissipated over the course of the study, consistent with our previous work showing that social influences on amphetamine self-administration decrease over time with repeated exposure to the same peer (Gipson et al., 2011). Group differences present only during training reflect a difference in acquisition rate, with partnered rats learning faster than non-partnered rats and partnered rats close to the partition learning faster than rats far from the partition. The lack of effect during later phases of self-administration suggests that neither peers nor lever proximity to a peer alters remifentanil sensitivity (no shift in the dose response curve) or motivation for remifentanil (no change in responding as FR value increases) under the conditions tested.

5. Conclusion

The current study demonstrated that acquisition of remifentanil self-administration is enhanced in the presence of a peer, regardless of whether the peer self-administered remifentanil or saline. The influence of the peer was most pronounced when the active levers were placed in close proximity to the wire screen partition so that self-administration occurred close to the adjacent peer. Identifying the factors involved in social influences on drug self-administration in rodents provides support for using dual-compartment operant chambers to investigate opioid taking in peer groups. Future studies employing these methods will provide a step toward understanding the neural systems that underlie social influences on opioid use disorder.

Highlights.

  • The study found that rats took more drug in the presence of another rat

  • Drug intake did not differ in rats paired with a drug-taking or a saline-taking rat

  • Rats took more drug when their active lever was closer to their peer

Acknowledgements

The authors would like to thank Dr. Lindsey Hammerslag for her thoughtful insights on statistical design.

Funding provided by grants: National Institute of Health grant number DA041755 to MTB and BBRF Young Investigator Award to RSH.

Role of Funding Source

Nothing declared.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest

No conflict declared.

References

  1. Achterberg EJM, Trezza V, Siviy SM, Schrama L, Schoffelmeer ANM, Vanderschuren LJMJ, 2014. Amphetamine and cocaine suppress social play behavior in rats through distinct mechanisms. Psychopharmacology (Berl). 231, 1503–1515. 10.1007/s00213-013-3272-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Achterberg EJM, van Swieten MMH, Houwing DJ, Trezza V, Vanderschuren LJMJ, 2018. Opioid modulation of social play reward in juvenile rats. Neuropharmacology. 10.1016/j.neuropharm.2018.09.007 [DOI] [PubMed]
  3. Badiani A, Belin D, Epstein D, Calu D, Shaham Y, 2011. Opiate versus psychostimulant addiction: the differences do matter. Nat. Rev. Neurosci 12, 685–700. 10.1038/nrn3104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bardo MT, Klebaur JE, Valone JM, Deaton C, 2001. Environmental enrichment decreases intravenous self-administration of amphetamine in female and male rats. Psychopharmacology (Berl). 155, 278–284. 10.1007/s002130100720 [DOI] [PubMed] [Google Scholar]
  5. Branstetter SA, Low S, Furman W, 2011. The Influence of Parents and Friends on Adolescent Substance Use: A Multidimensional Approach. J. Subst. Use 16, 150–160. 10.3109/14659891.2010.519421 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cain ME, Green TA, Bardo MT, 2006. Environmental enrichment decreases responding for visual novelty. Behav. Processes 73, 360–366. 10.1016/j.beproc.2006.08.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Caprioli D, Celentano M, Paolone G, Lucantonio F, Bari A, Nencini P, Badiani A, 2008. Opposite environmental regulation of heroin and amphetamine self-administration in the rat. Psychopharmacology (Berl). 198, 395–404. 10.1007/s00213-008-1154-3 [DOI] [PubMed] [Google Scholar]
  8. Caprioli D, Paolone G, Celentano M, Testa A, Nencini P, Badiani A, 2007. Environmental modulation of cocaine self-administration in the rat. Psychopharmacology (Berl). 192, 397–406. 10.1007/s00213-007-0717-z [DOI] [PubMed] [Google Scholar]
  9. Carroll ME, Lac ST, 1993. Autoshaping i.v. cocaine self-administration in rats: effects of nondrug alternative reinforcers on acquisition. Psychopharmacology (Berl). 110, 5–12. 10.1007/BF02246944 [DOI] [PubMed] [Google Scholar]
  10. Chelnokova O, Laeng B, Løseth G, Eikemo M, Willoch F, Leknes S, 2016. The μ-opioid system promotes visual attention to faces and eyes. Soc. Cogn. Affect. Neurosci 11, 1902–1909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Douglas LA, Varlinskaya EI, Spear LP, 2004. Rewarding properties of social interactions in adolescent and adult male and female rats: Impact of social versus isolate housing of subjects and partners. Dev. Psychobiol 45, 153–162. 10.1002/dev.20025 [DOI] [PubMed] [Google Scholar]
  12. Fleshler M, Hoffman HS, 1962. A progression for generating variable-interval schedules. J. Exp. Anal. Behav 5, 529–530. 10.1901/jeab.1962.5-529 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Forster M, Grigsby TJ, Bunyan A, Unger JB, Valente TW, 2015. The Protective Role of School Friendship Ties for Substance Use and Aggressive Behaviors Among Middle School Students. J. Sch. Health 85, 82–89. 10.1111/josh.12230 [DOI] [PubMed] [Google Scholar]
  14. Gelman A, Hill J, 2006. Data analysis using regression and multilevel/hierarchical models. Cambridge University Press. [Google Scholar]
  15. Gipson CD, Yates JR, Beckmann JS, Marusich JA, Zentall TR, Bardo MT, 2011. Social facilitation of d-amphetamine self-administration in rats. Exp. Clin. Psychopharmacol 19, 409–419. 10.1037/a0024682 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Green TA, Alibhai IN, Roybal CN, Winstanley CA, Theobald DEH, Birnbaum SG, Graham AR, Unterberg S, Graham DL, Vialou V, Bass CE, Terwilliger EF, Bardo MT, Nestler EJ, 2010. Environmental Enrichment Produces a Behavioral Phenotype Mediated by Low Cyclic Adenosine Monophosphate Response Element Binding (CREB) Activity in the Nucleus Accumbens. Biol. Psychiatry 67, 28–35. 10.1016/j.biopsych.2009.06.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Haidar SH, Moreton JE, Liang Z, Hoke JF, Muir KT, Eddington ND, 1997. Evaluating a Possible Pharmacokinetic Interaction between Remifentanil and Esmolol in the Rat. J. Pharm. Sci 86, 1278–1282. 10.1021/js970079e [DOI] [PubMed] [Google Scholar]
  18. Hofford RS, Chow JJ, Beckmann JS, Bardo MT, 2017. Effects of environmental enrichment on self-administration of the short-acting opioid remifentanil in male rats. Psychopharmacology (Berl). 10.1007/s00213-017-4734-2 [DOI] [PMC free article] [PubMed]
  19. Hursh SR, Silberberg A, 2008. Economic Demand and Essential Value. Psychol. Rev 115, 186–198. 10.1037/0033-295X.115.1.186 [DOI] [PubMed] [Google Scholar]
  20. Lacy RT, Strickland JC, Feinstein MA, Robinson AM, Smith MA, 2016. The effects of sex, estrous cycle, and social contact on cocaine and heroin self-administration in rats. Psychopharmacology (Berl). 233, 3201–3210. 10.1007/s00213-016-4368-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lacy RT, Strickland JC, Smith MA, 2014. Cocaine self-administration in social dyads using custom-built operant conditioning chambers. J. Neurosci. Methods 236, 11–18. 10.1016/j.jneumeth.2014.07.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Marx MH, Henderson RL, Roberts CL, 1955. Positive reinforcement of the bar-pressing response by a light stimulus following dark operant pretests with no after effect. J Comp Physiol Psychol. 48, 73–76. [DOI] [PubMed] [Google Scholar]
  23. McCool BA, Chappell AM, 2009. Early Social Isolation in Male Long-Evans Rats Alters Both Appetitive and Consummatory Behaviors Expressed during Operant Ethanol Self-administration. Alcohol. Clin. Exp. Res 33, 273–282. 10.1111/j.1530-0277.2008.00830.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Panlilio LV, Schindler CW, 2000. Self-administration of remifentanil, an ultra-short acting opioid, under continuous and progressive-ratio schedules of reinforcement in rats. Psychopharmacology (Berl). 150, 61–66. 10.1007/s002130000415 [DOI] [PubMed] [Google Scholar]
  25. Peitz GW, Strickland JC, Pitts EG, Foley M, Tonidandel S, Smith MA, 2013. Peer Influences on Drug Self-Administration: An Econometric Analysis in Socially Housed Rats. Behav. Pharmacol 24, 114–123. 10.1097/FBP.0b013e32835f1719 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Pinheiro J, Bates D, DebRoy S, Sarkar D, 2009. nlme: Linear and Nonlinear Mixed Effects Models.
  27. Pitts DK, Marwah J, 1988. Cocaine-elicited mydriasis in the rat: pharmacological comparison to clonidine, D-amphetamine and desipramine. J. Pharmacol. Exp. Ther 247, 815 LP – 823. [PubMed] [Google Scholar]
  28. Robinson AM, Lacy RT, Strickland JC, Magee CP, Smith MA, 2016. The Effects of Social Contact on Cocaine Intake Under Extended-Access Conditions In Male Rats. Exp. Clin. Psychopharmacol 24, 285–296. 10.1037/pha0000078 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rollins MD, Feiner JR, Lee JM, Shah S, Larson M, 2014. Pupillary Effects of High-dose Opioid Quantified with Infrared Pupillometry. Anesthesiol. J. Am. Soc. Anesthesiol 121, 1037–1044. 10.1097/ALN.0000000000000384 [DOI] [PubMed] [Google Scholar]
  30. Searle SR, Speed FM, Milliken GA, 1980. Population Marginal Means in the Linear Model : An Alternative to Least Squares Means. Am. Stat 34, 216–221. 10.1016/j.bjps.2017.12.020 [DOI] [Google Scholar]
  31. Smith MA, 2012. Peer influences on drug self-administration: Social facilitation and social inhibition of cocaine intake in male rats. Psychopharmacology (Berl). 224, 81–90. 10.1007/s00213-012-2737-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Trezza V, Damsteegt R, Achterberg EJM, Vanderschuren LJMJ, 2011. Nucleus Accumbens μ-Opioid Receptors Mediate Social Reward. J. Neurosci 31, 6362 LP – 6370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Vanderschuren LJMJ, Spruijt BM, Van Ree JM, Niesink RJM, 1995. Effects of morphine on different aspects of social play in juvenile rats. Psychopharmacology (Berl). 117, 225–231. 10.1007/BF02245191 [DOI] [PubMed] [Google Scholar]
  34. Venniro M, Shaham Y, 2020. An operant social self-administration and choice model in rats. Nat. Protoc 15, 1542–1559. 10.1038/s41596-020-0296-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Venniro M, Zhang M, Caprioli D, Hoots JK, Golden SA, Heins C, Morales M, Epstein DH, Shaham Y, 2018. Volitional social interaction prevents drug addiction in rat models. Nat. Neurosci 21, 1520–1529. 10.1038/s41593-018-0246-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Weiss VG, Yates JR, Beckmann JS, Hammerslag LR, Bardo MT, 2018. Social reinstatement: a rat model of peer-induced relapse. Psychopharmacology (Berl). 235, 3391–3400. 10.1007/s00213-018-5048-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wöhr M, Schwarting RKW, 2009. Ultrasonic communication in rats: Effects of morphine and naloxone on vocal and behavioral responses to playback of 50-kHz vocalizations. Pharmacol. Biochem. Behav 94, 285–295. 10.1016/j.pbb.2009.09.008 [DOI] [PubMed] [Google Scholar]

RESOURCES