Orexin/hypocretin neuron activation is correlated with alcohol seeking and preference in a topographically specific manner

Abstract

Orexin (ORX) (also known as hypocretin) neurons are located exclusively in the posterior hypothalamus, and are involved in a wide range of behaviours, including motivation for drugs of abuse such as alcohol. Hypothalamic subregions contain functionally distinct populations of ORX neurons that may play different roles in regulating drug-motivated and alcohol-motivated behaviours. To investigate the role of ORX neurons in ethanol (EtOH) seeking, we measured Fos activation of ORX neurons in rats following three different measures of EtOH seeking and preference: (i) context-induced reinstatement, or ABA renewal; (ii) cue-induced reinstatement of extinguished responding for EtOH; and (iii) a home cage task in which preference for EtOH (vs. water) was measured in the absence of either reinforcer. We found significant activation of ORX neurons in multiple subregions across all three behavioural tests. Notably, ORX neuron activation in the lateral hypothalamus correlated with the degree of seeking in context reinstatement and the degree of preference in home cage preference testing. In addition, Fos activation in ORX neurons in the dorsomedial hypothalamic and perifornical areas was correlated with context and home cage seeking/preference, respectively. Surprisingly, we found no relationship between the degree of cue-induced reinstatement and ORX neuron activation in any region, despite robust activation overall during reinstatement. These results demonstrate a strong relationship between ORX neuron activation and EtOH seeking/preference, but one that is differentially expressed across ORX field subregions, depending on reinstatement modality.

Keywords: Fos; alcoholism; lateral hypothalamus; reinstatement; reward.

Figure 1

High-power photomicrograph of coronal section through the lateral hypothalamus immunolabelled for Fos-protein (black nuclear stain) and orexin-A (brown cytoplasmic stain). Arrows indicate co-labeled neurons. Other orexin neurons without Fos staining and Fos staining in non-orexin neurons are shown. Scale bar = 20 µm.

Figure 2

Context-induced reinstatement of ethanol seeking. Rats were trained to respond for EtOH in two separate groups in similar operant chambers located in two distinct lab rooms, to provide different contexts. There were no behavioral differences between groups in terms of self-administration, extinction or reinstatement behavior. (A) Animals exhibited a strong preference for responding on the active lever (versus the inactive lever) during self-administration (left). Lever responding was then extinguished in the opposite room (middle). On test day, animals were returned to the room in which they originally learned to respond for EtOH, resulting in a significant reinstatement of responding on the active lever only (right). (B) There was significant variability in reinstatement behavior (active lever responding) as a result of returning to the original self-administration context. (C–D) Responding on the active lever during context-induced reinstatement was significantly, and positively, correlated with the percentage of Fos-positive orexin neurons in DMH (C) and LH (E), but not the PF (D). (F–H) Activation of orexin neurons was not correlated with the number of non-rewarded well entries animals made throughout the reinstatement session, a measure of non-goal-directed locomotor activity. *** p<0.001.

Figure 3

Cue-induced reinstatement of ethanol seeking. (A) Rats were trained to respond for EtOH paired with light-tone cues. Lever pressing was extinguished in sessions where both EtOH and cues were withheld. On test day, lever responses resulted in delivery of light-tone cues but no EtOH, and resulted in significant reinstatement of active lever responding. (B) As with context-induced reinstatement, there was significant variability in active lever responding during cued reinstatement test sessions. (C–E) Responding on the active lever during cue-induced reinstatement was not correlated with the percentage of Fos-positive orexin neurons in any of the subregions examined. (F–H) Similarly, there was no relationship between the activation of orexin neurons and the number of non-rewarded well entries the animals made throughout the reinstatement session. ** p<0.01.

Figure 4

Orexin neurons are activated during expression of ethanol preference. (A,B) Preference for EtOH was measured in the home cage using a junction potential detection method, as per previous studies (Hayar et al., 2006). This method allowed us to quantify the number of licks each animal made on the EtOH versus water bottles. (C) Rats sampled both EtOH and water tubes. Sampling of EtOH vs. water was variable across rats. (D) Preference for EtOH on test day was positively, and significantly, correlated with animals’ preference for EtOH on the 2-bottle choice test during the week prior to testing. (E– G) Animals’ preference for EtOH was positively, and significantly, correlated with percentage of Fos-positive orexin neurons in PF (F) and LH (G), but not DMH (E), subdivisions. (H–J) Activation of orexin neurons in PF (I) and LH (J) was also positively, and significantly, correlated with the number of licks animals’ made on the EtOH (black), but not water (gray), bottles.