Fentanyl Addiction: Individual Differences, Neural Circuitry, and Treatment with a GLP-1 Receptor Agonist

Project Summary Opioid use disorder (OUD) is defined in the DSM-V as a “problematic pattern of opioid use leading to problems or distress”. More than 36,000 people overdosed on synthetic opioids including fentanyl in 2019 and, with the COVID-19 pandemic, there has been a 30% increase in overdose deaths.1,2 Current treatments for OUD include therapy and medication-assisted treatments (MATs) such as methadone, buprenorphine, and naltrexone. Naltrexone has low compliance rates and there is stigma associated with the use of methadone and buprenorphine, as they are opioids used to treat OUD. Access to these drugs, then, is limited, and relapse rates remain high.3-6,46 Relapse often is precipitated by withdrawal and withdrawal, we hypothesize, is a need state.7 Thus, as the need for fluid is sated by water, for example, the need for drug (i.e., withdrawal) is sated by drug. In accordance, we further hypothesized that glucagon-like peptide-1 (GLP-1), a satiety drug, could be utilized to reduce drug seeking and taking in rodent models of OUD. In support, GLP-1 targeted treatments are effective in reducing responding for many substances of abuse in rodent models.8,55-56,72 Additionally, we found that GLP-1 receptor agonists (GLP-1RAs) can reduce heroin self-administration, cue-induced heroin seeking, and drug-induced reinstatement of heroin seeking.9,24 GLP-1RAs also reduce oxycodone taking and seeking.15 Finally, our preliminary data suggest that a GLP-1RA can reduce cue- and drug-induced seeking of not only heroin, but fentanyl as well (Urbanik et al., in preparation). This finding is consistent with a recent report.13 With fentanyl contributing to the majority of opioid overdoses1, it is critical that we understand where in the brain fentanyl is acting and how treatment with a GLP-1RA might mitigate these effects. Here we will use rodent models, light sheet microscopy, qRT-PCR, and pharmacology to address these questions. For our rodent model, we will utilize an extended access drug self-administration paradigm.22 We predict that: (1) As with other drugs of abuse tested,16,22-23 half of the rats tested will be high drug takers and these rats will have higher cue-induced seeking and greater inhibition of the GLP-1 ‘satiety’ pathway and greater activation of reward substrates compared to low fentanyl taker/seekers. (2) Fentanyl intake, fentanyl seeking, brain activation patterns, and gene expression will be attenuated by treatment with the GLP-1RA, liraglutide. (3) The protective effects of GLP-1RA treatment on behavior and brain will be blocked by the administration of the GLP-1R antagonist, exendin-9 (Ex-9), directly into the lateral hypothalamus. These hypotheses will be tested across three specific aims. If our hypotheses are supported, we will have identified the most vulnerable of fentanyl taking and seeking rats, rescued these subjects from fentanyl seeking via treatment with a GLP-1RA, verified that fentanyl, particularly in the most vulnerable, inhibits the GLP-1 ‘satiety’ pathway and activates reward/seeking substrates, and implicated a key role for GLP-1 receptors in the lateral hypothalamus – a structure critically involved in homeostasis and motivated behavior.