Researchers led by a BBRF grantee report new research extending knowledge about how a naturally occurring hormone called FGF21 helps to regulate alcohol consumption. Using a synthetic analogue of the hormone, they showed in mouse experiments how it appears to impact behaviors relevant in alcohol consumption as well as how it impacts the activation of neurons involved in the initiation and termination of drinking.

The research is of particular interest because FGF21 and the pathways it impacts are targets of interest in the development of new treatments for alcohol use disorder (AUD). A number of medications for AUD are in use today (naltrexone, nalmefene, acamprosate, and topiramate), but their effectiveness varies widely and the search for new treatments continues.

FGF21 is one of a number of peptides (protein fragments) that operate in the body as hormones and play key roles in metabolic health (glucose regulation, insulin sensitivity) and energy balance. These include GLP-1, produced in the gut and the target of weight-loss and diabetes medicines such as Ozempic and Mounjaro. (Others are orexin, produced in the brain, and leptin, generated within fatty tissue.) FGF21 (fibroblast growth factor 21) is generated in the liver in response to various metabolic stressors, including alcohol.

In a paper appearing in Neuropsychopharmacology, a team led by 2020 BBRF Young Investigator E. Zayra Millan, Ph.D., of the University of New South Wales, Sydney, Australia, point to three lines of evidence implicating FGF21 in the regulation of alcohol consumption. One is experiments in mice in which induced overexpression of FGF21 as well as pharmaceutical administration of an FGF21 analogue both lead to reduced preference for alcohol in animals. Second, in large-scale genome studies in people, DNA variations affecting the gene that encodes FGF21 and its cellular receptor are statistically associated with alcohol consumption and risk of AUD. Third, studies in which FGF21 signaling is disrupted or blocked result in increased alcohol consumption by laboratory mice.

The latter experiments serve to remind that FGF21 is involved in signaling that normally acts to curtail alcohol consumption. A pathway carrying this signal has been localized in mice, and involves neurons in the basolateral amygdala (BLA) that project to the nucleus accumbens (NAc). Stimulation of the relevant NAc neurons inhibits consumption, while a pause in activation of these neurons is required when an individual initiates and maintains alcohol consumption. How FGF21 affects these NAc neurons was unknown prior to the current study.

Dysregulation of FGF21’s normal function could be one way of understanding how chronic and habitual alcohol consumption can lead to AUD. Mammals began consuming alcohol from fermented fruit long before humans developed methods to distill alcohol. It is therefore not surprising that multiple bodily systems in mammals, including humans, evolved over time to sense and regulate alcohol consumption. The prevalence of AUD in humans indirectly suggests that naturally evolved regulatory systems can become dysfunctional (due to genetic and/or environmental factors), removing the evolutionary "brake" on excessive or health-impairing alcohol intake.

In their newly reported research, Dr. Millan and colleagues used an FGF21 analogue called PF-05231023 to confirm FGF21’s ability to reduce voluntary alcohol consumption and preference for alcohol (vs. other fluids offered—sweetened water, in the mouse experiments). But the results that are most notable concerned the influence of the FGF21 analogue on behaviors involving alcohol consumption. Notable among these are “approach behaviors,” i.e., those an individual takes toward a stimulus perceived to be positive or rewarding. In alcohol consumption, a variety of cues, such as time of day, a specific activity, or suggestion by peers can trigger approach behaviors, i.e, actions required to obtain alcohol.

The team’s experiments showed that the FGF21 analogue directly reduced alcohol consumption in male mice, but not females. The reason for the sex specificity is not clear and will be pursued in future studies. The team suggests the difference may be due to sex-specific metabolic effects of synthetic FGF21 (i.e., the analogue, as opposed to the naturally occurring hormone), and/or may be related to a difference in expression of FGF21 receptors in males and females or in liver status relative to diet.

In both sexes, the FGF21 analogue weakened the intensity of responses following the presentation of alcohol-related cues, and also reduced the motivation of individual animals to seek alcohol. Importantly, the drug did not affect the animals’ pursuit of sugar when sucrose-related cues were given or when motivation to seek sucrose solution was tested. This is one of several pieces of evidence suggesting to the team that the FGF21 analogue’s effects on consumption was reward-specific—it did not perturb the reward response globally.

The experiments also showed that the FGF21 analogue’s impact on alcohol drinking in males appeared to be associated with “pre-ingestive evaluative processes and reward palatability.” In other words, the hedonic, or pleasure-driven urge to consume alcohol appeared to be altered. This was linked in the team’s research with the drug’s ability to alter the activity of neurons in the nucleus accumbens—it increased the recruitment of NAc neurons involved in termination of alcohol drinking behavior.

The team administered the FGF21 analogue in concert with a sub-therapeutic dose of a GLP-1 stimulating drug called Exendin-4, which targets signaling between the gut and brain to control metabolism. Their joint administration had the effect of augmenting the impact of the FGF21 analogue on alcohol-seeking behavior. Exendin-4, alone, had no such effect.

To the team, this was evidence of a “complementary and inter-dependent mechanism of action” in FGF21 and GLP-1. “Our findings suggest that combination agonist [i.e., hormone-stimulating] approaches may be of benefit in the treatment of AUD.” Such an approach, they note, is already being tested in context of weight loss and insulin sensitivity.