The role of neuropeptide systems in stress-induced relapse to alcohol seeking

Abstract

Alcohol use disorders are chronic relapsing conditions that represent one of the leading causes of preventable death. The enduring propensity to relapse is one of the fundamental features of alcohol use disorders and stress is a major precipitant of relapse behaviour. The pharmacological stressor, yohimbine, has been widely used to elicit relapse-like behaviour in rodents and craving in human addicts, however, the neurochemical mechanisms in which yohimbine exert its actions are still elusive. Neuropeptides and their receptors offer potential as novel treatments of psychiatric and other disorders as their signalling powerfully modulates important functions such as stress responses, anxiety and social behaviours. Therefore I examined the mechanism in which yohimbine precipitates relapse-like behaviour, and region-specific involvement of five neuropeptide systems; corticotropin releasing factor (CRF), orexin (OX), dynorphin (DYN), cocaine and amphetamine regulated transcript (CART) and relaxin-3 in yohimbine-induced reinstatement of alcohol seeking.

Firstly, to determine brain regions in which yohimbine-induced relapse to alcohol seeking specifically activates, the pattern of neural activation following home cage yohimbine (1 mg/kg i.p.) and yohimbineinduced relapse to alcohol seeking was examined. Fos immunohistochemistry (IHC) across 40 brain regions revealed a distinct pattern of neural activation. A number of regions were activated specifically following yohimbine-induced reinstatement of alcohol seeking including the dorsal hippocampus and hypothalamus, while the prefrontal cortex and extended amygdala were activated by both home cage yohimbine administration and yohimbine-induced reinstatement of alcohol seeking. These data suggest the prefrontal cortex and extended amygdala may be activated by yohimbine to drive reinstatement of alcohol seeking, while other brain regions are recruited during this process. In light of these novel results I examined the neurochemical phenotype of the activated neurons within the extended amygdala and hypothalamus. Yohimbine-induced reinstatement of alcohol seeking increased Fos activation in central nucleus of the amygdala (CeA) CRF, DYN, CART and GABA neurons; bed nucleus of the stria terminalis (BNST) CRF and CART neurons; and hypothalamic CART and OX neurons compared to naïve and vehicle controls.

The neuropeptide relaxin-3 and its cognate receptor, relaxin family peptide receptor 3 (RXFP3), modulate stress-induced reinstatement of alcohol seeking in rats and while the BNST has been implicated in this regard, the CeA also receives a relaxin-3 innervation and CeA neurons densely express RXFP3 mRNA. Therefore, the role of the relaxin-3/RXFP3 signalling within the CeA in yohimbine-induced reinstatement of alcohol seeking was examined. Bilateral intra-CeA injections of the selective RXFP3 antagonist, R3(B1-22)R (1 μg/side), attenuated yohimbine-induced reinstatement of alcohol seeking, suggesting relaxin-3/RXFP3 signalling in the CeA as another node that mediates yohimbine-induced reinstatement of alcohol seeking.

The major source of relaxin-3 is confined to a small stress-sensitive hindbrain region, the nucleus incertus (NI). Relaxin-3 NI neurons express receptors for stress neuropeptides including CRF1, CRF2, OX1 and OX2. Previous reports implicate OX2, but not OX1 receptors within the NI to mediate yohimbine-induced reinstatement of alcohol seeking. Although OX neurons directly innervate the NI, the pattern of innervation is unknown, therefore to determine the identify the OX inputs to the NI that are activated following yohimbine-induced reinstatement of alcohol seeking, the retrograde tracer cholera toxin subunit β (CTβ; 80 nl) was discretely injected into the NI and activation of OX inputs to the NI examined. Initial findings revealed back-labelled neurons evenly throughout OX neuron-containing regions of the hypothalamus, some of which were activated following stress-induced reinstatement of alcohol seeking. To examine the contribution of NI CRF1 and CRF2 receptors in yohimbine-induced relapse reinstatement of seeking, bilateral injections of the CRF1 antagonist (CP376395, 500 ng) or CRF2 antagonist (astressin-2B, 200 ng) were administered discretely within the NI. Intra-NI antagonism of CRF1 receptors attenuated yohimbine-induced reinstatement of alcohol seeking. In contrast, CRF2 receptor antagonism had no significant effect. In line with these data, CRF1, but not CRF2, receptor mRNA was upregulated in the NI following chronic ethanol intake. Our quantitative PCR analysis also identified CRF mRNA within the rat NI, and the existence of a newly identified population of CRF-containing neurons was subsequently confirmed by detection of CRF immunoreactivity in rat and mouse NI. These data suggest that NI neurons contribute to reinstatement of alcohol seeking, via an involvement of CRF1 signalling. Furthermore, chronic ethanol intake leads to neuroadaptive changes in CRF and relaxin-3 systems within rat NI.

The source of CRF to the NI remains elusive. This in part is hindered by the difficulties in the detection of CRF peptide immunoreactivity without the prior administration of colchicine. Therefore to examine the putative CRF inputs to the NI, a Crf-IRES-Cre::Ai14 reporter mouse was employed. Immunohistochemical evaluation of the Crf-IRES-Cre mouse showed good validity in the distribution and concordance of tdTomato reporter distribution with CRF immunoreactivity. Further, tdTomato neurons in the PVN were responsive to both acute forced swim stress and yohimbine administration; while chronic stress led to a baseline increase in tdTomato activation across several stress sensitive brain regions. Using Allen Brain Institute resources to determine CRF innervation of the NI revealed that very little CRF innervation to the NI originates in CRF-rich regions, suggesting that either more diffuse CRF+ regions send projections to the NI, or that other mechanisms, such as intrinsic CRF release or volume transmission are involved.

These findings represent a major contribution to the field and provide further evidence that the CRF, DYN, OX and relaxin-3 systems are involved in stress-induced alcohol seeking behaviours in rats. These studies also provide the first evidence that the CART system may be involved in stress-induced relapse behaviours, and may provide a novel target to treat relapse behaviours. These studies also warrant further examination of the relaxin-3/RXFP3 system particularly in stress and reward related disorders.