Corticotropin releasing factor (CRF) systems: Promoting cocaine pursuit without distress via incentive motivation

Abstract

Corticotropin releasing factor (CRF) systems in limbic structures are posited to mediate stress-induced relapse in addiction, traditionally by generating distress states that spur drug consumption as attempts at hedonic self-medication. Yet evidence suggests that activating CRF-expressing neurons in the central amygdala (CeA) or nucleus accumbens (NAc) can magnify incentive motivation in absence of distress, at least for sucrose rewards. However, traditional CRF hypotheses in addiction neuroscience are primarily directed toward drug rewards. The question remains open whether CRF systems can similarly act via incentive motivation mechanisms to promote pursuit of drug rewards, such as cocaine. Here we tested whether optogenetic excitation of CRF-containing neurons in either NAc medial shell, lateral CeA, or dorsolateral BNST of transgenic Crh-Cre+ rats would spur preference and pursuit of a particular laser-paired cocaine reward over an alternative cocaine reward, and whether excitation served as a positively-valenced incentive itself, through laser self-stimulation tests. We report that excitation of CRF-containing neurons in either NAc or CeA recruited mesocorticolimbic circuitry to amplify incentive motivation to pursue the laser-paired cocaine: focusing preference on the laser-paired cocaine reward in a two-choice task, and spurred pursuit as doubled breakpoint in a progressive ratio task. Crucially indicating positive-valence, excitation of CRF neurons in NAc and CeA also was actively sought after by most rats in self-stimulation tasks. Conversely, CRF neuronal activation in BNST was never self-stimulated, but failed to enhance cocaine consumption. Collectively, we find that NAc and CeA CRF-containing neurons can amplify pursuit and consumption of cocaine by positively-valenced incentive mechanisms, without any aversive distress.

Fig 1. Photomicrograph of virus expression and local Fos plumes.

Photomicrographs display ChR2 virus expression (green) in neurons of Crh-Cre rats, and laser-induced local Fos expression in neurons (purple) located within plumes surrounding optogenetic fiber tips (mapped at right). Fiber/virus sites are in A) nucleus accumbens (NAc) medial shell, B) central nucleus of the amygdala (CeA), and C) bed nucleus of the stria terminalis (BNST). We previously reported co-expression of Crh mRNA and Cre mRNAs in the same neurons within NAc, CeA and BNST using fluorescent in situ hybridization to validate this transgenic Crh-Cre rat line (19). Laser Fos plume diagrams at right show the average plume diameter and % elevation intensity of local Fos expression immediately surrounding the fiber tips, induced by laser stimulation of CRF-containing neurons in CRF ChR2 rats, compared to control levels measured at the same box sites in eYFP rats after identical laser illuminations. Light blue reflects >150% Fos elevation and dark blue reflects >200% Fos elevation over eYFP baseline levels measured in the inactive virus control group. Scale bars show 0.1mm for reference. NAcSh, nucleus accumbens shell; NAcC, nucleus accumbens core; LS, lateral septum; opt, optic tract; CeL, lateral central amygdala; BLA, basolateral amygdala; mBNST, medial bed nucleus of stria terminalis; dlBNST, dorsolateral bed nucleus of stria terminalis.

Fig 2. Localization of function maps for Laser+Cocaine preference.

Localization of function maps show the magnitude of behavioral effects induced at corresponding histological sites of optic fibers in individual ChR2 CRF rats: A) nucleus accumbens (NAc) medial shell, B) central nucleus of the amygdala (CeA), and C) dorsolateral bed nucleus of the stria terminalis (BNST). Colors indicate intensity (%) of laser-induced pursuit with reds indicating stronger preference for Laser+Cocaine in the two-choice task. Conversely, blue colors indicate avoidance of Laser+Cocaine (or preference for Cocaine alone). Symbol sizes are scaled to match maximum 0.7mm diameter measured for illuminated ChR2 Fos plumes. Also see S1 Table in S1 File. LS, lateral septum; LV, lateral ventricle; CPu, caudate putamen; NAc, nucleus accumbens; VP, ventral pallidum; ac, anterior commissure; ic, internal capsule; MeA, medial amygdala; GP, globus pallidus; IntC, intercalated amygdala; BMA, basomedial amygdala; BLA, basolateral amygdala; fx, fornix; Shy, septohypothalamic nucleus, MPA, medial preoptic area; LPO, lateral preoptic area.

Fig 3. Distant Fos recruitment of mesocorticolimbic activity.

A) After NAc medial shell ChR2 excitation of CRF-containing neurons, distant 150–300% increases in Fos expression over eYFP control levels were recruited in several mesocorticolimbic structures related to incentive motivation for rewards: nucleus accumbens core (NAcC), anterior and posterior ventral pallidum (aVP, pVP), posterior bed nucleus of stria terminalis (pBNST), anterior and posterior lateral hypothalamus (aLH, pLH), medial amygdala (MeA), central amygdala (CeA), and ventral tegmentum (VTA). B) After CeA ChR2 excitation of CRF neurons, 150–300% increases in distant Fos were recruited in: NAcC, anterior and posterior NAc shell (aNAcSh, pNAcSh), aVP, pVP, anterior BNST (aBNST), pBNST, aLH, pLH, MeA and VTA. C) In contrast for BNST, ChR2 excitation of CRF-containing neurons recruited 150–300% increases in Fos in several distinct structures related to avoidance/pain/distress: paraventricular nucleus of the hypothalamus (PVN) and midbrain periaqueductal grey (PAG). All analyses were performed on raw cell counts (see S2-S4 Tables in S1 File). Bar graphs depict means and SEMs of percent change from eYFP control rats and colors denote the degree of change. Circles show individual ChR2 CRF rats. SN, substantia nigra; IF, infralimbic cortex. *p<0.05, **p<0.01, ***p<0.001.

Fig 4. NAc ChR2 stimulation of CRF-containing neurons amplifies motivation and focuses preference on laser-paired cocaine.

A) In the two-choice task, ChR2 excitation of CRF-containing neurons in NAc medial shell focused preference and cocaine pursuit on the Laser+Cocaine option by a 4:1 ratio over Cocaine alone in the two-choice task (n = 7). B) In comparison, control NAc eYFP rats chose equally between the two cocaine options (n = 6). C) NAc ChR2 rats also escalated total cocaine intake (Laser+Cocaine plus Cocaine alone options combined) to higher magnitudes than eYFP controls in two-choice task. D) In the progressive ratio (PR) task that measures effort breakpoint, NAc CRF-containing neuron excitation nearly doubled breakpoint in the Laser+Cocaine session over the Cocaine alone session (n = 7; left panel). Laser did not affect the breakpoint of NAc eYFP control rats (n = 6). Right panel depicts timeframe of behavioral responses for cocaine during the 1hr PR sessions. Means and SEM reported. *p<0.05, **p<0.01.

Fig 5. CeA ChR2 stimulation of CRF-containing neurons amplifies motivation and focuses preference on laser-paired cocaine.

A) In the two-choice task, ChR2 excitation of CRF-containing neurons in the lateral CeA amplified cocaine pursuit and focused preference on the Laser+Cocaine option by a 4:1 ratio over Cocaine alone test (n = 6). B) In comparison, control CeA eYFP rats chose equally between options (n = 6). C) CeA ChR2 rats also had higher total cocaine intake in the two-choice task across all days than eYFP controls (combined intakes from both Laser+Cocaine plus Cocaine alone). D) In a separate progressive ratio (PR) breakpoint test of incentive motivation intensity, CeA CRF neuronal excitation doubled effort breakpoint for cocaine in the Laser+Cocaine session over breakpoint of the same CeA ChR2 rats in their Cocaine alone session (n = 6; left panel). Laser did not affect breakpoint of CeA eYFP control rats (n = 6). Right panel depicts timeframe of behavioral responses for cocaine during the 1hr PR sessions. Means and SEM reported. *p<0.05, **p<0.01.

Fig 6. BNST ChR2 stimulation of CRF-containing neurons fails to change cocaine pursuit.

Pairing ChR2 excitation of CRF neurons in BNST with one cocaine option failed to alter cocaine pursuit in the two-choice task for A) BNST ChR2 rats (n = 6); or B) BNST eYFP inactive control rats, (n = 6). C) BNST ChR2 stimulation had little no effect on total cocaine consumption in the two-choice task compared to eYFP controls plus Cocaine alone infusions). D) In progressive ratio (PR) tests, CRF-containing neuronal excitation failed to alter effort breakpoint of BNST ChR2 rats (n = 5; left panel), which remained similar also to the unchanging breakpoint of BNST eYFP control rats (n = 6). Right panel depicts timeframe of behavioral responses for cocaine during the 1hr PR sessions. Means and SEM reported.

Fig 7. Laser self-stimulation of CRF-containing neurons in NAc and CeA, but not BNST.

A) ChR2 rats could self-stimulate laser to excite CRF-containing neurons by touching a designated Laser spout to earn 3sec illuminations, whereas touching an Inactive spout delivered nothing. NAc ChR2 rats as an entire group demonstrated significant self-stimulation across all 3 days of testing (n = 7). NAc ChR2 individuals that met criteria for laser self-stimulation on day 1 (n = 5, ChR2 SS) continued to self-stimulate on days 2–3 by touching their Laser-spout >400% more than the Inactive-spout. B) CeA ChR2 rats as an entire group showed only a nonsignificant trend toward laser self-stimulation (n = 8). However, CeA ChR2 individuals that met self-stimulation criteria on day 1 (n = 6; ChR2 SS) continued to self-stimulate on days 2–3. touching their Laser-spout 400% more than the Inactive-spout. By contrast, control NAc eYFP (n = 6) and CeA eYFP rats (n = 6) failed to self-stimulate, and merely touched randomly at low rates. C) BNST ChR2 rats failed as a group to self-stimulate for laser excitation of CRF-containing neurons (n = 8) across all 3 test days, similar to eYFP controls (n = 7). No BNST ChR2 individuals met criteria for self-stimulation on any day. Means and SEM reported. *p <0.05, **p<0.01.

Fig 8. Laser-paired cocaine preference predicts cocaine breakpoint, not self-stimulation in NAc and CeA ChR2 rats.

Laser self-stimulation by NAc ChR2 rats (red/purple squares & dashed line) or CeA ChR2 rats (blue squares & dashed line) did not correlate with their degree of laser-induced amplification of cocaine value. A) Lack of correlation between self-stimulation and paired cocaine preference in two-choice task. B) Lack of correlation between self-stimulation and intensity of paired cocaine motivation in breakpoint task. Even rats that met criteria for laser self-stimulation failed to show correlation with strength of laser-paired cocaine preference or pursuit. C) The magnitude of laser-induced increase in cocaine value was stable across both progressive ratio and two-choice tasks for NAc ChR2 and CeA ChR2 rats, as indicated by a simple linear regression analysis. Regression lines with β-coefficients and R² values depicted.