Role of BNST CRFR1 Receptors in Incubation of Fentanyl Seeking

Abstract

The time-dependent increase in cue-triggered opioid seeking, termed "incubation of opioid craving," is modeled in rodents by examining responding for opioid-associated cues after a period of forced abstinence. With opioid drugs, withdrawal symptoms may heighten cue reactivity by recruiting brain systems involved in both reward seeking and stress responses. Corticotropin releasing factor (CRF) in the bed nucleus of the stria terminalis (BNST) is a critical driver of stress-induced relapse to drug seeking. Here, we sought to determine whether BNST CRF receptor 1 (CRFR1) signaling drives incubation of opioid craving in opioid dependent and non-dependent rats. First, we tested whether BNST CRFR1 signaling drives incubation of opioid craving in rats with short-access fentanyl self-administration experience (2.5 μg/kg/infusion, 3 h/day for 10 days). On Day 1 of forced abstinence, we gave bilateral intra-BNST vehicle injections to all rats and measured lever responding for opioid cues in the absence of fentanyl infusions. On Day 30 of forced abstinence, we gave an identical test after bilateral intra-BNST injections of vehicle or CRFR1 receptor antagonist, R121919 (1 μg/0.3 μL/hemisphere). Vehicle treated rats showed greater responding for opioid associated cues on Day 30 relative to Day 1, and this incubation effect was prevented by intra-BNST R121919 on Day 30. Next, we incorporated an opioid-dependence procedure to investigate whether BNST CRFR1 signaling drives opioid cue-reactivity to a greater extent in opioid-dependent relative to non-dependent rats. We trained rats to self-administer fentanyl for 5 days before initiating the dependence phase and resuming daily fentanyl self-administration sessions for 10 days. We gave intra-BNST R121919 or vehicle injections before testing during acute (Day 5) or protracted (Day 30) withdrawal. During acute withdrawal, antagonizing BNST CRFR1 decreased the number of press bouts without affecting bout size or duration. These patterns of responding with R121919 treatment resulted in less fentanyl-associated conditioned reinforcement during test. Together, these findings suggest a role for BNST CRFR1 signaling in driving cue-reinforced opioid seeking after periods of forced abstinence.

Keywords: BNST; CRF; incubation; opioid dependence; withdrawal.

FIGURE 1

Experiment 1: BNST CRFR1 antagonist effect on incubation of fentanyl craving. (A) Experimental timeline: we trained rats to self-administer fentanyl for 3 h/day for 10 days. On forced abstinence Day 1 and Day 30, we measured lever responding under extinction conditions. (B) Fentanyl infusions + cues earned across the 10 daily self-administration sessions. (C) Rats discriminated Active from Inactive Lever during self-administration training. (D,E) Incubation test data showing cues earned on FR1, 20 s TO schedule (D) and Active/Inactive Lever pressing (E) on Day 1 and Day 30 of forced abstinence for vehicle (open bars) treated and R121919 treated (filled bars) conditions. (F) Time course of Active and Inactive Lever presses on Day 30 test for both treatment groups. *p < 0.05, **p < 0.01, different from Session 1 in (B), Lever main effect in (C), different from Day 1 in (D), and Treatment main effect in (F). R12 = R121919. n = 13 Veh, n = 15 R12. Data are mean ± SEM.

FIGURE 2

Experiment 2: Fentanyl self-administration training in opioid dependent versus non-dependent rats. (A) Experimental timeline: We trained rats to self-administer fentanyl for 5 days before inducing opioid dependence. Rats resumed self-administration for 10 more days while dependence was maintained. On forced abstinence Day 5 and Day 30, we measured lever responding under extinction conditions. (B) Change in body weight across days for opioid-dependent and non-dependent rats after start of dependence phase. (C) Number of wet-dog shakes for opioid-dependent and non-dependent rats. (D) Fentanyl infusions + cues earned across the 15 3 h self-administration sessions, sessions 1–5: before dependence and no infusion maximum criteria, 6–15: during dependence with infusion maximum criteria. (E) Latency to reach infusion maximum criteria in minutes. (F) Active/Inactive Lever pressing for dependent and non-dependent rats during training. (G) Active Lever press rate. *p < 0.05, main effect of dependence, ***p < 0.001, different from non-dependent. SA, Self-administration; ND, non-dependent; D, dependent. Data are mean ± SEM.

FIGURE 3

Experiment 2: BNST CRFR1 antagonist effect on fentanyl seeking during acute withdrawal (Day 5) test. (A) Day 5 incubation test data showing cues earned on FR1, 20 s to schedule. (B) Inter cue interval (ICI), the time between presses that result in cues. (C) Cumulative cues earned pattern of a representative pair of vehicle ($\overline{X}$ = 1.52) and R121919 ($\overline{X}$ = 2.75) injected rat. (D) Number of bouts. A bout is defined as two or more presses for which the interval between successive presses did not exceed 20 s. *p < 0.05, **p < 0.01, main effect of treatment. All the inset graphs indicate mean ± SEM when collapsed across dependence. ND, non-dependent; D, dependent. Data are mean ± SEM.

FIGURE 4

Placement for injector tips in rats with BNST cannulae implants. (A) Gray dots represent sites of R121919 infusions on coronal sections as a distance from Bregma (mm). Drawings are adapted from Paxinos and Watson (2007). (B) A representative brain section depicting cannulae placement in the BNST.