Orexin-1 receptor signaling in ventral tegmental area mediates cue-driven demand for cocaine

Abstract

Drug-associated sensory cues increase motivation for drug and the orexin system is importantly involved in this stimulus-enhanced motivation. Ventral tegmental area (VTA) is a major target by which orexin signaling modulates reward behaviors, but it is unknown whether this circuit is necessary for cue-driven motivation for cocaine. Here, we investigated the role of VTA orexin signaling in cue-driven motivation for cocaine using a behavioral economics (BE) paradigm. We found that infusion of the orexin-1 receptor (Ox1R) antagonist SB-334867 (SB) into VTA prior to BE testing reduced motivation when animals were trained to self-administer cocaine with discrete cues and tested on BE with those cues. SB had no effect when animals were trained to self-administer cocaine without cues or tested on BE without cues, indicating that learning to associate cues with drug delivery during self-administration training was necessary for cues to recruit orexin signaling in VTA. These effects were specific to VTA, as injections of SB immediately dorsal had no effect. Moreover, intra-VTA SB did not have an impact on locomotor activity, or low- or high-effort consumption of sucrose. Finally, we microinjected a novel retrograde adeno-associated virus (AAVretro) containing an orexin-specific short hairpin RNA (OxshRNA) into VTA to knock down orexin in the hypothalamus-VTA circuit. These injections significantly reduced orexin expression in lateral hypothalamus (LH) and decreased cue-driven motivation. These studies demonstrate a role for orexin signaling in VTA, specifically when cues predict drug reward.

Fig. 1. Removal of discrete drug-paired cues reduced motivation in animals trained to associate drug delivery with cues.

A A schematic of cue and no-cue BE testing. Animals were trained on FR1 either with (+) or without (−) cues, as described in the text and indicated in the diagram. Animals were then trained on the BE procedure with cues (Group 1) or without cues (Groups 2 and 3) until stability criteria were reached. Cues were then removed (Group 1) or added (Groups 2 and 3) and behavior was re-stabilized for a minimum of 3 days to test the effects of cues on demand after training on the new BE procedure. B, C Animals trained and tested on the BE paradigm with cues (Group 1; n = 6) displayed increased α (decreased motivation) when cues were removed; Q₀ was not affected (C). D, E Animals trained on FR1 self-administration with cues and the BE paradigm without cues (Group 2; n = 6) showed no change in α (D) or Q₀ values (E) when cues were added during BE. F, G Animals trained on FR1 and BE without cues (Group 3; n = 7) showed no change in α (F) or Q₀ values (G) when cues were introduced. *p < 0.05. Bar graphs indicate means ± SEMs.

Fig. 2. Intra-VTA SB lowered motivation for cocaine paired with discrete cues in animals trained to self-administer cocaine with cues.

A A schematic of testing with intra-VTA SB/aCSF during BE with or without cues. Animals were trained on FR1 either with (Groups 1 and 2) or without cues (Group 3), and then tested with intra-VTA aCSF/SB microinjections during BE either with (Group 1) or without cues (Groups 2 and 3). The same animals were then re-tested on subsequent sessions with aCSF/SB on BE either with (Groups 2 and 3) or without cues (Group 1). B–D Intra-VTA SB (injection sites in B) reduced motivation (increased α) in Group 1 animals (n = 6) only when cues were present (C). Intra-VTA SB had no effect on Q₀ in these animals when tested with or without cues (D). Panels E-G - Intra-VTA SB (injection sites in E) did not reduce motivation in animals trained with cues and tested on BE first without cues (Group 2, n = 6; F). Intra-VTA SB had no effect on Q₀ in Group 2 animals (G). H–J In Group 3 animals (trained on FR1 without cues; n = 5–8), SB (injection sites in H) did not have an impact on motivation (α; I) or low-effort consumption (Q₀) on BE with or without cues present (J). *p < 0.05. Bar graphs indicate means ± SEMs.

Fig. 3. SB increased cocaine demand elasticity (decreased motivation) when microinjected into VTA specifically.

A Plots of microinjection locations in VTA (a separate group than those in Figs. 1 and 2). Control microinjections (not depicted here) were made first in each of these animals, 1.8 mm dorsal to these injection sites. B Intra-VTA SB microinjections increased demand elasticity (α) compared to dorsal control SB or VTA aCSF microinjections (n = 9). C None of the intraparenchymal microinjections impacted low-effort consumption (Q₀). *p < 0.05, **p < 0.01. Bar graphs represent means ± SEM.

Fig. 4. SB did not have an impact on locomotor activity or motivation for natural (sucrose) reward.

A, B Locomotor activity following intra-VTA microinjection of SB. SB did not have an impact on locomotor activity in any of the 10-min testing bins (n = 9; A) or total locomotor activity (B). C Intra-VTA SB also did not have an impact on active lever presses for sucrose pellets on an FR1 schedule of reinforcement (n = 6). D–F Effects of intra-VTA SB (injection sites in D) on demand for sucrose in a BE test (n = 8). SB did not impact motivation (α; E) nor low-effort consumption (Q₀; F) for sucrose pellets. Bar graphs indicate means ± SEMs.

Fig. 5. Retrograde knockdown of orexin in neurons projecting to VTA increased cocaine demand elasticity.

A Schematic of experiments using microinjection of the AAVretro-OxshRNA into VTA to knock down orexin in VTA-projecting neurons in the lateral hypothalamic area (LHA). B Frontal sections through hypothalamus stained with an antibody for orexin A, indicating orexin knockdown in rats with AAVretro-OxshRNA microinjected into VTA, compared to control animals with AAVretro-scRNA microinjected into VTA. Midline is at the left, dorsal is upwards, scale bars indicate 100 µm. C–E Data from specificity validation experiments (1 µL of intra-VTA AAVretro-OxshRNA or AAVretro-scRNA virus infused unilaterally; n = 2–4/group). C Plots of microinjection locations in VTA in animals infused with the AAVretro-OxshRNA (green circles) or -scRNA viruses (black circles) for specificity validation experiments. There was a significant difference in orexin cell numbers at 2, 4, and 6 weeks post intra-VTA injection in AAVretro-OxshRNA animals, compared to rats with control (AAVretro-scRNA) injections into VTA (cells counted in both hemispheres; D). No changes were observed in the number of MCH-containing neurons in the same hypothalamic area for these same injections (E). F–L Data from behavioral experiments (0.3 µL of AAVretro-OxshRNA or AAVretro-scRNA virus infused intra-VTA bilaterally) in a different cohort of animals. F Plots of microinjection locations in VTA in animals infused with the AAVretro-OxshRNA (green circles) and -scRNA viruses (black circles) for behavioral experiments. G AAVretro-OxshRNA microinjected into VTA reduced the total bilateral number of orexin-immunoreactive neurons in animals tested on behavior, compared to AAVretro-scRNA animals (n = 7/group). H AAVretro-OxshRNA significantly reduced the number of orexin-expressing cells in LH compared to AAVretro-scRNA. Results were not significant for orexin cells in perifornical/dorsomedial hypothalamus (Pef/DMH). I Rats with AAVretro-OxshRNA microinjection into VTA exhibited increased α (reduced motivation) 5 weeks post injection. No differences were observed in AAVretro-scRNA animals. J–L AAVretro-OxshRNA did not have an impact on Q₀ (J) or locomotor activity (K, L; n = 7–8/group). *p < 0.05, **p < 0.01. Bar graphs indicate means ± SEMs.